Sleep and Cognitive Functioning: The Moderating Role of Vagal Indices 
 
by 
 
Lori D. Elmore-Staton 
 
 
 
 
A dissertation submitted to the Graduate Faculty of 
Auburn University 
in partial fulfillment of the 
requirements for the Degree of 
Doctor of Philosophy  
 
Auburn, Alabama 
December 12, 2011 
 
 
 
 
Keywords: sleep, actigraphy, child development, respiratory sinus arrhythmia,  
cognitive functioning  
 
Copyright 2011 by Lori Elmore-Staton 
 
Approved by 
 
Mona El-Sheikh, Chair, Leonard Peterson & Co. Inc.  
Professor of Human Development and Family Studies 
Brian E. Vaughn, Professor of Human Development and Family Studies 
Margaret K. Keiley, Associate Professor of Human Development and Family Studies 
Joseph A. Buckhalt, Wayne T. Smith Distinguished Professor of Special Education, 
Rehabilitation, and Counseling 
 
 
 ii 
 
 
 
 
 
 
Abstract 
 
This dissertation contains two studies. In the first study, vagal tone and vagal reactivity 
(indexed by respiratory sinus arrhythmia; RSA) were examined as possible moderators of the 
associations between both actigraphy and subjective sleep disruptions and multiple aspects of 
children?s cognitive performance. Children (M age = 8.70 years; N = 166; 3rd grade) wore 
actigraphs for seven consecutive nights and completed a subjective measure of sleep via 
interview. To evaluate cognitive functioning, children completed the Woodcock-Johnson III 
Tests of Cognitive Abilities (WJ III; Woodcock, McGrew, & Mather, 2001) and a computerized 
reaction time task (Psych/Lab; Abrams, 2004) in the laboratory. Vagal indices were assessed in 
the lab; RSA was obtained during a three minute baseline and RSA reactivity (RSA-R) was 
derived as the difference between RSA during a mild cognitive stressor (i.e., star tracing task) 
and initial RSA. Results suggest that children who exhibited RSA withdrawal to the laboratory 
stressor (lower RSA during stressor than baseline), and to a lesser extent those who had higher 
levels of baseline RSA, were protected from the negative effects of sleep disruptions on various 
executive functioning tests.  
Building on the first study and following the same procedures, the second study further 
delineated the role of the parasympathetic nervous system by examining children?s RSA activity 
(i.e., baseline levels, reactivity to a lab challenge and/or recovery following the challenge) as 
potential moderators in the link between children?s sleep (self-report and actigraphy-based) and 
multiple domains of children?s cognitive functioning both cross-sectionally (N  = 132; 80 boys; 5th 
grade) and longitudinally over two years (from 3rd to 5th grade). Cross-sectional results indicate 
that actigraphy-based sleep variables (i.e., sleep minutes, sleep efficiency, and long wake 
 iii 
minutes) interact with RSA reactivity (RSA-R) to a problem-solving task to predict children?s 
Decision Speed scores on the WJ III; cross-sectional findings did not yield significant 
moderation effects for baseline RSA or RSA recovery. Longitudinal findings show that RSA-R, 
and to a lesser extent baseline RSA, interact with both self-reported and actigraphy-based sleep 
parameters to predict multiple WJ III test scores two years later. Results build on the scant 
literature examining individual differences in the sleep-cognitive functioning link and suggest 
that interactions between physiological and biological systems are important for the prediction of 
children?s cognitive functioning.  
 
 ix 
 
 
 
 
 
 
Acknowledgments 
 
 
 I am grateful to my mentor and friend, Mona El-Sheikh, who believed in me more than I 
believed in myself.  Without her this dissertation would not be possible. Special thanks to my 
committee members for their direction, assistance, and guidance throughout my graduate 
career. Thanks to the Sleep, Health, and Development Lab, particularly Bridget Wingo and 
Ryan Kelly, for their encouragement and support over the years. I am indebted to my parents 
(Danny & Lisa Elmore), in-laws (Terry and Jane Staton), and Aunt Gail Satterfield for all their 
love and assistance. And finally, words alone cannot express how grateful I am for my husband 
(Brad) and children (Ashton and Ella) for their love, patience, and encouragement.  
 v 
 
 
 
 
 
Table of Contents 
 
 
Abstract .................................................................................................................................... ii 
Acknowledgments  ................................................................................................................... iv 
List of Tables .......................................................................................................................... vii 
List of Figures ........................................................................................................................ viii 
List of Abbreviations ................................................................................................................ ix 
Introduction   ............................................................................................................................. 1 
 Sleep   ........................................................................................................................... 1 
 Sleep and Cognition? ? ? .? ? ? . ? . ......................................................................... 2 
   Parasympathetic Nervous System?. ? ????????????????????..7  
 Parasympathetic Nervous System and Cognition? ??????????????? 10 
 Parasympathetic Activity as a Moderator?.. ??????????????????12  
 Sleep Measurement??? ?????????????????????????..13  
 Potential Influential Variables?? ??????????????????????..14  
 Research Objectives?? ???????????????????? ??????.21  
 References??. ??????????????????????????????25  
Abstract Manuscript 1   ........................................................................................................... 45 
Sleep and cognition: The moderating role of RSA and RSA reactivity   ................................... 46 
            Vagal Tone and RSA.?? ?????????????????????????...48  
 Sleep and RSA Functioning??. ??????? ????? ???????????50  
 RSA and Cognition?? ...?????????????????????????? .50 
            New Directions??.. ? .??????????????????????????....51  
 v 
            Method.?? ?? ?????????????????????????????.53  
            Results??? ?? ????????????????????????????..61  
            Discussion?? ????????? ?????????????????????69  
            References.? .??? ???????? .??????????????????....76  
Abstract Manuscript 2 ............................................................................................................. 94 
Sleep and cognition: The moderating role of respiratory sinus arrhythmia activity   ................. 95 
             PNS Activity??.. ?? ..??? ??????????????????????? 100 
             Sleep and RSA Functioning..??.. ?????????????????????.102  
             RSA and Cognition? ..? ??????????????????????? ??..103  
             New Directions..? ? ???????????????????????????.103  
             Method...??. ? ...?????????? ??????????????????.105  
           Results......? ??? ???????????????????????????..113  
              Discussion? .? ? ??????????????????? .?????????.123  
             References? .?? ??????????????????????? ?????.130  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  
vi 
 vii 
 
 
 
 
 
 
List of Tables 
 
 
Table 1 Manuscript 1 Descriptive Table .................................................................................. 85 
Table 2 Manuscript 1 Correlation Table .................................................................................. 86 
Table 1 Manuscript 2 Descriptive Table of T1 Study Variables ............................................. 142 
Table 2 Manuscript 2 Descriptive Table of T2 Study Variables ............................................. 143 
Table 3 Manuscript 2 Correlation Table ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ..144 
 viii 
 
 
 
 
 
 
List of Figures 
 
 
Figure 1 Manuscript 1 Conceptual Model ................................................................................ 88 
Figure 2 Manuscript 1 Example Path Model............................................................................ 89 
Figure 3 Manuscript 1 Sleep X RSA Predicting Concept Formation ........................................ 90 
Figure 4 Manuscript 1 Sleep X RSA Predicting Numbers Reversed ....................................... 91 
Figure 5 Manuscript 1 Sleep X RSA Predicting Auditory Working Memory ............................. 92 
Figure 6 Manuscript 1 Sleep X RSA Predicting Decision Speed ............................................. 93 
Figure 1 Manuscript 2 Conceptual Model .............................................................................. 148 
Figure 2 Manuscript 2 Example Path Model???????? ??????????? ? ...149 
Figure 3 Manuscript 2 Cross-Sectional Sleep X RSA Predicting Decision Speed..????. .150 
Figure 4 Manuscript 2 Longitudinal Sleep X RSA Predicting Concept Formation.? ???? 151 
Figure 5 Manuscript 2 Longitudinal Sleep X RSA Predicting Numbers Reversed?. ???. ..152 
Figure 6 Manuscript 2 Longitudinal Sleep X RSA Predicting Auditory Working Memory?? ? ? 153 
Figure 7 Manuscript 2 Longitudinal Sleep X RSA Predicting Decision Speed? ?? .?????154  
 ix 
 
 
 
 
 
 
List of Abbreviations 
 
 
RSA Respiratory Sinus Arrthymia    
ANS Autonomic Nervous System 
 
 
1 
 
Introduction 
 
Sleep disruptions, such as short sleep duration, trouble falling asleep, or prolonged night 
wakings are common occurrences in Americans? lives, and this issue has been recognized by 
the medical community and the media. Although this is the case, much of the sleep research 
has focused on clinical sleep problems (e.g., obstructive sleep apnea) or has been conducted 
with samples of adolescents or adults. Recent estimates by the National Sleep Foundation 
(2004) report about 20 - 40% of children have sleep disruptions, but there is a scant amount of 
information available regarding the role of sleep in typically developing school-aged children?s 
adaptation and maladaptation. The limited research conducted with school-aged children 
indicates that children get less than the recommended amount of sleep for optimal functioning 
(National Sleep Foundation, 2004; 2006; 2008). Specifically, the National Sleep Foundation 
recommends that children in middle childhood get 10-11 hours of sleep. However, most studies 
report that children are getting one to two hours less than the recommended amount of sleep 
per night (Buckhalt, El-Sheikh, & Keller, 2007; Sleep in America Poll, 2004). Even though this 
discrepancy may seem insignificant, recent research conducted with children indicates that 
even small restrictions in sleep amount (e.g., 1 hour per night), and/or mildly disrupted sleep, 
can have deleterious effects on children?s overall functioning (Sadeh, Gruber, & Raviv, 2003), 
including reductions in cognitive performance (Buckhalt, El-Sheikh, & Keller, 2007). This, in 
combination with the high rate of sleep disruptions, suggests that American children may be at 
an increased risk for adjustment, academic, and cognitive problems. It also brings to the 
forefront the importance of studying how sleep problems impact school-aged children?s 
outcomes, especially cognitive functioning domains.    
 
 
2 
 
Sleep 
The definition of sleep has changed drastically over time, with earlier perceptions of 
sleep encompassing the notion that it is a state of inactiveness for the body and mind. Today, 
we know that this is not the case; instead, sleep can be defined as an active, reversible state of 
unresponsiveness and disengagement with the environment (Carskadon & Dement, 2005). In 
this paper, sleep problems are conceptualized on a continuum and are evidenced by shorter 
sleep durations, restlessness or high activity during sleep, multiple and extended night waking, 
and/or highly variable sleep schedules. These kinds of sleep problems may be the most likely 
forms of sleep disruptions experienced by typically developing children, and addressing sleep 
using a continuum can contribute substantially to the understanding of developmental 
outcomes. The  terms sleep disruptions and sleep problems will be used interchangeably, 
however, it is important to note that these terms do not refer to clinical levels of sleep disorders 
(e.g., sleep apnea) unless noted.  
Sleep and Cognition 
 The importance of sleep in learning and memory is well established in the literature 
across the lifespan (Sadeh et al., 2003) and in recent years the connection between sleep and 
memory has gained increased interest from the scientific community. Evidence suggests that 
memory is improved following sleep (Kopasz, Loessl, Hornyak, Riemann, Nissen, Piosczyk, & 
Voderholzer, 2010), and that attention, cognition, and memory are significantly altered following 
episodes of sleep deprivation (Adam, Retey, Khatami, & Landolt, 2006; Drummond, Brown, 
Gillin, Stricker, Wong, & Buxton, 2000; Durmer & Dinges, 2005; Pilcher & Huffcut, 1996). 
Recent work incorporating imaging techniques (i.e., fMRI) suggests that one function of sleep is 
the consolidation of novel material into memory (Gais et al., 2007; Kopasz et al., 2010; Walker & 
Stickgold, 2006). Thus, studying sleep and the effects it has on cognitive functioning, particularly 
during the school-years when the brain is still developing and basic building blocks of 
knowledge are being acquired is of utmost importance. 
 
3 
 
 The role of sleep seems to be particularly important for declarative or explicit memory 
which is defined as factual memories that are consciously accessible (Kopasz et al., 2010). For 
example, fMRI sleep studies link the connection between the hippocampus and the prefrontal 
cortex, especially during slow-wave sleep, to improved learning or memory of novel information 
(Born, Rasch, & Gais, 2006). However, if sleep deprivation occurs following a learning task, it is 
hypothesized that poorer memory of the novel material may be a result of a malfunction in the 
communication between these two brain regions (Born, Rasch, & Gais, 2006). Another possible 
explanation is that increased disruptions of sleep or shorter sleep amounts may reduce the 
brain activity that occurs overnight and that is essential for neurocognitive functioning, such as 
the integration of novel material (Dewald, Meijer, Oort, Kerkhof, & Bogels, 2009). These higher 
order cognitive functions are heavily influenced by the prefrontal cortex, which is especially 
sensitive to poor sleep quality and quantity (Curcio, Ferrara, & De Gennaro, 2006; Harrison & 
Horne, 1998). 
It is important to note that the aforementioned imaging research refers to adult samples, 
and it is unclear if these relations would hold true for children. Specifically, it is known that 
children and adult?s sleep differ in both the duration of sleep as well as the percentage of time 
spent in each stage. For example, children sleep longer and cycle into slow-wave sleep (SWS) 
and rapid eye movement (REM) sleep more often than adults (Anders, 1994; Hill, Hogan, & 
Karmiloff-Smith, 2007; Ohayon, Carskadon, Guilleminault, & Vitiello, 2004). The specific 
reasons for these differences throughout stages of development are unclear. But, taking into 
account the large amount of knowledge children are processing, especially at young ages when 
the basic building blocks of knowledge are forming, it is likely that sleep may play a different or 
more significant role in the development of short and long term cognitive functioning for children 
in comparison to adults. 
Multiple studies report links between sleep disruptions (e.g., low sleep amount, 
increased night wakings) and poorer cognitive performance, including difficulties in executive 
 
4 
 
functioning, working memory and attention (Blunden, Lushington, & Kennedy, 2001; Curcio, 
Ferrara, & De Gennaro, 2006; Gozel, 1998; Huntley & Lewin, 2006; Sadeh, 2007; Sadeh, 
Gruber, & Raviv, 2002; Sadeh et al., 2003; ~ 2 ? 7% of the variance in cognitive functioning is 
accounted for by sleep in non-clinical populations). Two primary components of executive 
functions are sustained attention (i.e., ability to maintain focus on a task for extended periods; 
Johnsen, Laberg, Eid, & Hugdahl, 2002) and working memory (i.e., continuous updating and 
prolonged storage of information; Logie, Zucco, & Baddeley, 1990). Executive functions stem 
from the neural systems of the prefrontal cortex and are indicated as important contributors to 
behavior regulation (Cole, Usher, & Cargo, 1993; Pennington & Ozonoff, 1996). The prefrontal 
cortex coordinates executive functions and it is particularly susceptible to sleep disruptions 
(Curcio, Ferrara, & De Gennaro, 2006; Dahl, 1996; Drummond & Brown, 2001; Harrison, & 
Horne, 1998).  
  A meta analysis of research with adults concluded that: (1) cognitive performance was 
affected more than motor skills when sleep deprivation occurred; (2) as the task lengthened, 
sleep deprived participants showed marked decreases in performance levels and (3) the range 
of effects from sleep debt suggests that there are a number of individual differences yet to be 
explored (Pilcher & Huffcutt, 1996). An experimental study indicated that children (M age = 10.6 
years) with extended sleep (i.e., an additional one hour of sleep per night for three nights) had 
improved performance on a memory test and a continuous performance task (baseline to post), 
whereas sleep-restricted children (i.e., one less hour of sleep per night for three nights) had a 
significant decline in their performance on a simple reaction time task (Sadeh et al., 2003). A 
study conducted by Huntley and Lewin (2006) found acute sleep restriction (one night of four 
hours of sleep) had a significant impact on children?s (N = 93; M age = 7.9, SD =1.4 years) 
neurocognitive functioning. Specifically, more errors of omission, reduced signal detection, and 
longer response times were recorded during the neurocognitive task following the sleep 
restriction night in comparison to the baseline measure taken two weeks prior. These results 
 
5 
 
suggest that sleep impacts children?s attention and processing abilities. Another study with 11- 
to 13- year- olds found prolonged sleep restriction (6.5 hours per night for 6 nights) was related 
to attention impairments (Fallone, Seifer, Acebo, & Carskadon, 2000). Acute sleep restriction 
during one night (5 hours sleep) for 10- to 14- year-olds was found to impair higher cognitive 
functions related to attention (Randazzo, Muehlbach, Schweitzer, & Walsh, 1998). 
 Overall, there are more than 30 adult and child studies addressing sleep problems 
(defined and measured in a variety of ways) and cognitive or academic performance (Buckhalt, 
Wolfson, & El-Sheikh, 2009; Curcio, Ferrara, & De Gennaro, 2006), with the majority, but not all, 
reporting links between sleep disruptions and poorer cognitive performance. For example, 
Steenari and colleagues (2003) studied 60 children (6- to 13- years-old) and found that 
actigraphically derived sleep variables were related to auditory and visual tasks assessed on 
separate days. Specifically, reduced sleep efficiency and increased sleep latency was related to 
more incorrect responses on both the visual and auditory tasks, and shorter sleep times were 
only related to reduction in working memory performance at the highest memory load level. 
From these findings, the authors speculate that sleep quality more than sleep duration is more 
closely related to working memory performance.  
On the other hand, Meijer and colleagues (2000) sampled 449 Dutch children (9-to 14- 
years-old) and found no relation between subjective reports of sleep and scores on an attention 
task. Considering that attention is crucial for successful school performance, and that 
insufficient, poor, or disorganized sleep has been linked to academic achievement (Epstein, 
Chillag, & Lavie, 1998; Gozal, 1998; Randazzo et al., 1998; Wolfson & Carskadon, 1998), it is 
important to delineate specific aspects of both objective and subjective sleep problems related 
to problems in cognition (Sadeh, 2007), including executive functions, working memory, and 
processing speed difficulties.  
Recent research with children indicates links between insufficient or poor quality sleep 
and impaired neurobehavioral functions, especially executive functions, such as sustained 
 
6 
 
attention and working memory (Blunden, Lushington, Lorenzen, Martin, & Kennedy, 2005; 
Sadeh et al., 2003). These cognitive functions are essential to learning and may be especially 
important at ages when skills are being acquired (Van Leijenhorst, Crone, & Van der Molen, 
2007). Understanding how and why sleep is related to children?s cognitive functioning is 
essential for children?s future academic achievement. Specifically, sleep debt or deprivation, 
created when the optimal and actual amounts of sleep are not congruent (Van Dongen, Rogers 
& Dinges, 2003), has lasting effects on functioning and adjustment (Carskadon, 2002). Reviews 
of the literature (Buckhalt, Wolfson, & El-Sheikh, 2009; Curcio, Ferrara, & De Gennaro, 2006; 
Mitru, Millrood, & Mateika, 2002; Sadeh, 2007; Wolfson & Carskadon, 2003) show reduced 
amounts of sleep, varying sleep schedules, delayed bedtimes and wake times, and active or 
fragmented sleep influence many aspects of cognitive performance and functioning, including 
those incorporated in comprehensive intelligence tests and narrow tests of neurocognitive 
functioning.   
Further evidence of associations between sleep problems and poor cognitive functioning 
stems from research conducted with clinical samples. Over 50 studies have been reported with 
sleep disordered children since 2000, and studies have linked sleep-disordered breathing with a 
variety of cognitive and academic problems in children (Beebe, 2006; Blunden & Beebe, 2006; 
Blunden, Lushington, Lorenzen, Martin, & Kennedy, 2005; Halbower & Mahone, 2006). 
Recently, studies have also linked treatments of sleep disordered breathing (i.e., 
adenotonsillectomy) with improvements in attention and cognition one year following the surgery 
(Chervin et al., 2006). In addition, children diagnosed with learning problems, such as mental 
retardation and autism spectrum disorders, have higher incidences of sleep problems than their 
typically developing counterparts (Richdale, Francis, Gavidia-Payne, & Cotton, 2000; Stores & 
Wiggs, 2001; Williams, Sears, & Allard, 2004).   
Although sleep restrictions and disrupted sleep patterns are linked to deficits in cognitive 
performance, not all children with restricted or disrupted sleep display these negative outcomes. 
 
7 
 
Further, in the clinical research realm, investigators have indicated that clinical sleep 
interventions work well for some individuals but not at all for others. This brings to the forefront 
the importance of delineating possible individual difference variables that may contribute to 
these variations.  
Parasympathetic Nervous System 
Researchers have begun to recognize the importance of individual differences in 
psychophysiology and how those differences potentially influence child adjustment (Cicchetti & 
Dawson, 2002), including cognitive performance. One such individual difference that may be an 
important factor in the association between sleep and cognitive performance is parasympathetic 
nervous system (PNS) activity. The PNS is the down-regulating branch of the autonomic 
nervous system (ANS) and it works to maintain homeostasis under normal conditions (Porges, 
1991). The term vagal tone is used as an indicator of the influence of the PNS on the heart. 
Specifically, the vagus nerve is the 10th cranial nerve that provides an inhibitory influence on the 
heart (i.e., decrease in heart rate). Respiratory sinus arrthymia (RSA), the changes in heart rate 
associated with the normal cycles of respiration (heart rate decreases during exhalation and 
increases during inhalation), is used as a marker of vagal tone. The vagus nerve?s influence on 
RSA has been demonstrated through its significant attenuation or elimination by 
pharmacological blockade (Berntson et al., 1997; Hayano et al., 1991; Sherwood, Allen, Obrist, 
& Langer, 1986). Within a wide range of normal physiological functioning, RSA is an acceptable 
measure of vagal tone (Grossman &Taylor, 2007). Baseline RSA, also termed resting RSA, 
reflects the status of the PNS at rest, the ability to maintain homeostasis under normal 
circumstances, and perhaps the ability to sustain attentional focus (Porges, 1991). Vagal 
pathways may reduce (vagal suppression) or increase (vagal augmentation) cardiac output to 
promote regulatory and goal-directed behavior (Porges, 1995, 2007; Thayer & Lane, 2000), 
referred to as RSA regulation. Specifically, RSA typically decreases during stressors and 
engagements that require increases in attention (Porges, 2007). RSA is an appropriate index of 
 
8 
 
vagal activity for research with normally developing children (Grossman &Taylor, 2007) and can 
be obtained non-invasively.  
The Polyvagal Theory?s propositions (Porges, 2007) link the ANS with emotion 
regulation, cognition, and behavior, and will be used as a guiding theory for the potential role of 
vagal functioning in the relation between sleep parameters and facets of cognitive performance. 
Empirical research supporting the theory indicates that higher levels of baseline vagal tone 
positively correlate with optimal adjustment and health outcomes (Rottenberg, 2007). That is, 
higher vagal tone maintains a low heart rate (Porges, 1991) under normal conditions, which is 
optimal for health and adaptation. Stability of vagal tone has been examined at different points 
in childhood. Specifically, Calkins and Keane (2004) conducted a two-point longitudinal study 
with preschoolers (age 2 at T1; age 4.5 at T2) and found high stability (r = .57) in baseline vagal 
tone. Kennedy, Rubin, and Hastings (2004) also reported stability in preschool children?s 
cardiac vagal tone over two years (r = .47). In addition, El-Sheikh (2005) examined stability of 
vagal tone in middle childhood and reported that baseline vagal tone was a fairly stable (r =. 49) 
individual difference variable over a two year span.  
The reactivity of the PNS in response to a challenge is also an important aspect of PNS 
activity and is typically referred to as vagal regulation or vagal reactivity (Porges, 2007). When 
faced with a challenge, the PNS responds in one of two ways: (1) it can be withdrawn or 
suppressed, frequently referred to as vagal suppression or vagal withdrawal, which leads to an 
increase in heart rate, or (2) it can be activated or augmented, frequently termed vagal 
augmentation, which leads to a decrease in heart rate. Vagal suppression is thought to be an 
optimal response when confronted with a challenge because it allows for the potential to 
mobilize attentional resources when needed to respond to environmental demands (Huffman et 
al., 1998). Research indicates that vagal suppression likely facilitates a shift in metabolic 
resources that support information processing (Bornstein & Suess, 2000). However, it has been 
noted in the literature that while moderate vagal suppression appears adaptive, vagal 
 
9 
 
withdrawal that is extreme or prolonged may index over-reactivity (Beauchaine, 2001), which 
could have implications on the ability to focus attention. Therefore, more clarification is needed 
to truly understand the process associated with attention and other aspects of cognition, 
particularly the role of vagal regulation.  
Vagal regulation has shown stability across tasks occurring within one session 
(preschoolers; Calkins & Keane, 2004) and across time points (middle childhood; El-Sheikh, 
2005). Vagal regulation in response to cognitive tasks (i.e., star-tracing) has shown some 
stability (r = -.23) over two years (El-Sheikh, 2005). Although vagal tone and vagal regulation 
are often moderately correlated, each vagal marker can have differential relations to child 
outcomes (Calkins, 1997). For example, higher levels of vagal augmentation to a reaction time 
task have been linked with increases in both objective and subjective sleep problems, whereas 
baseline RSA with the same sample was correlated only with subjective sleep measures (El-
Sheikh & Buckhalt, 2005). In a paper published with the current sample, vagal suppression in 
response to listening to an argument between two adults interacted with sleep parameters and 
functioned as a moderator of risk in the link between sleep problems and adjustment problems. 
Specifically, higher levels of vagal suppression functioned as a protective factor in the sleep-
child adjustment link (i.e., externalizing and depression). Conversely, lower levels of vagal 
suppression or augmentation to the stressor acted as a vulnerability factor in the link between 
sleep problems and children?s adjustment.  
Lastly, the Polyvagal Theory postulates that once the challenge has subsided, PNS 
activity needs to return to baseline levels, and the percentage or recovery time is termed vagal 
recovery. The theory posits that a faster return to baseline is more optimal (i.e., less wear and 
tear on the heart), but few studies have addressed this assumption empirically. Adult studies 
have linked slower vagal recovery times to an increase in cardiovascular disease (Mezzacappa, 
Kelsey, Katkin, & Sloan, 2001). However, only one study to our knowledge has examined vagal 
 
10 
 
recovery in children (ages 4 to 7), with results indicating that less vagal recovery predicted 
reduced attention (Santucci, Silk, Shaw, Gentzler, Fox, & Kovacs, 2008).                                                                           
Parasympathetic Nervous System and Cognition 
 RSA indices are recognized as both a correlate and predictor of cognitive functioning 
(Morgan, Aikins, Steffian, Coric, & Southwick, 2007; Porges, 2007). Specifically, Beauchaine 
(2001) found those with a greater capacity to suppress vagal tone had more optimal cognitive 
functioning. Additionally, better performance on working memory and attention tasks have been 
noted for persons with higher variability in baseline RSA levels (Hansen, Johnsen, Thayer, 
2003). A dynamic range of vagal tone (i.e., high resting levels of vagal tone and the ability to 
suppress to low levels when met with a stressor) is a correlate of better cognitive performance, 
and it is believed that these persons have a greater ability to appropriately engage and 
disengage with the environment as needed (Thayer & Lane, 2000), as well as exhibit a more 
rapid recovery pattern (Porges, 2001).  
 Of importance to note is that the majority of the literature in this area has been 
conducted with adults or young children. Low intensity stimulation to the vagus system is shown 
in adults to be related to increased verbal memory storage (Clark, Naritoku, Smith, Browning, & 
Jensen, 1999) and improvements in alertness (Rizzo, Beelke, De Carli, Canovaro, Nobili, 
Robert, et al., 2003). Further, when examinations are conducted regarding focused and divided 
attention, results show a reduction in PNS activity as attention load increases (Backs & Ryan, 
1992). Attention processes are found to be mediated by vagal suppression (Suess, Newlin, & 
Porges, 1997; Weber, Van der Molen, & Molenaar, 1994).  
 The limited literature based on infants and children overall support the association 
between RSA and cognitive functioning. For example, Richards (1985) linked basal vagal tone 
and vagal suppression with infants? ability to sustain attention during a continuous performance 
task. This study also found higher levels of baseline RSA to be correlated with more ?mature? 
visual fixation lengths. Similarly, infants that were more distracted during tasks had lower basal 
 
11 
 
RSA than their more focused counterparts (Richards, 1987). More pronounced vagal 
suppression is associated with taking less time to orient to a visual stimulus or reach a learning 
criterion than those infants with less pronounced vagal suppression.  
 Scant information is available regarding RSA and cognitive performance in typically 
developing children beyond infancy, and of those that have been conducted the range of 
cognitive outcomes and/or the sample size has been limited. Preschool aged children (N= 42) 
with more RSA suppression had better on-task behaviors in the classroom as reported by their 
teachers, which was significantly correlated with better executive functioning (Blair, 2003). A 
small study (N = 32) with fourth and fifth grade children found links between higher baseline 
RSA and a greater ability to sustain attention on a continuous performance task than children 
with lower levels of basal RSA (Suess, Porges, & Plude, 1994). Additionally, Mezzacappa and 
colleagues (1998) conducted a study with 10-year-olds (41 of 42 participants were boys) and 
results indicated that those children with greater vagal variability showed more competence in 
executive control. A pilot study conducted by this author (Staton, El-Sheikh, & Buckhalt, 2009) 
with a small sample (N = 41) of children ages 6- to 13-years old (M age = 10.06; SD = 1.74) 
addressed links between RSA and RSA regulation and cognitive outcomes derived through the 
Woodcock-Johnson III Tests of Cognitive Ability (WJ III; Woodcock, McGrew, & Mather, 2001) 
and a version of the Sternberg memory scanning paradigm (Sternberg, 1966). Results indicated 
that even after controlling for child age, sex, and pubertal status, RSA was a robust predictor of 
multiple aspects of cognitive functioning (i.e., accounted for 5 ? 11% of the variance in cognitive 
measures). Specifically, those with higher levels of baseline RSA had higher scores of 
intellectual ability, processing speed, working memory, and cognitive efficiency. In addition, 
these children had better reaction time scores than their counterparts with lower levels of basal 
RSA. On the other hand, RSA reactivity was not found to be a significant predictor of WJ III 
scores, and only approached statistical significance for Brief Intellectual Ability and reaction 
time. The reason for the lack of association between RSA reactivity and cognitive functioning 
 
12 
 
variables was unclear, however it is possible that the challenge task (i.e., reaction time task) did 
not function as a stressor since there was no consequence based on their performance. 
Specifically, the mean level of RSA during the baseline and challenge task did not vary 
significantly, and a large variation of individual differences in RSA reactivity was observed.  
Parasympathetic Activity as a Moderator 
Sleep problems can disrupt neurobiological regulatory systems, especially those largely 
governed by the prefrontal cortex (PFC; Archbold, Giordani, Ruzicka, & Chervin, 2004; Dahl, 
1996; Horne, 1993; Sadeh et al., 2003), such as executive functions and PNS regulatory 
functions (El-Sheikh & Buckhalt, 2005, El-Sheikh, Erath, & Keller, 2007a; Irwin, 2008). Although 
sleep may influence vagal functioning, the neural networks impacted by sleep are not the only 
influence on vagal functioning. Thus, it is likely that some variability in vagal functioning is 
independent of sleep. Therefore, vagal functioning may not only act as a mechanism through 
which sleep influences cognitive functioning, but it may also intensify or weaken the impact of 
sleep problems on children?s cognitive functioning. Higher vagal tone and suppression have 
been found to function as protective factors against a number of adjustment outcomes, including 
cognitive impairments, in the context of family stress (El-Sheikh, 2005). Similarly, optimal vagal 
functioning may also serve as a protective factor in the context of biological stress, such as 
sleep disruptions. The moderation hypothesis is based on the mutual influence of sleep and 
vagal functioning on the regulation of attention. Dahl (1996) posits that sleep is affected by not 
only the levels of arousal and vigilance, but also the ability to regulate arousal and emotions 
during waking hours. Furthermore, the Polyvagal Theory (Porges, 2007), states that vagal 
functioning provides one with the ability to efficiently regulate attention to meet with the 
demands of the environment. Thus, both sleep and vagal functioning assist or obstruct the 
regulation of attention. Due to the similar influences on attention, more optimal vagal functioning 
could compensate for the lack of attention due to sleep disruptions, whereas less adaptive vagal 
functioning could exacerbate the attention problems related to sleep disruptions. In turn, 
 
13 
 
regulatory capacities have implications for children?s cognitive functioning, but this has not been 
addressed in the context of sleep disruptions. Recent empirical evidence based on the present 
sample indicated that children with less optimal vagal functioning (i.e., lower baseline vagal 
tone, less vagal suppression, or vagal augmentation) were at an increased risk for externalizing 
and internalizing symptoms when their sleep was disrupted (increased wake minutes and 
reduced sleep efficiency) in comparison to children with more optimal vagal regulation (El-
Sheikh et al., 2007a). Of note that in the aforementioned study there were no direct relations 
apparent between the various sleep parameters and children?s vagal tone or reactivity to the lab 
stressor. The present investigation will build on this literature by examining not only vagal tone 
and regulation, but also vagal recovery in the context of sleep disruptions and by focusing on 
the effects of sleep on multiple facets of cognitive functioning.  
Sleep Measurement 
 Information regarding the quality and quantity of sleep can be obtained in two ways: (1) 
objectively (e.g., polysomnography-PSG; actigraphy) or (2) subjectively (e.g., questionnaires, 
sleep diaries). Historically, sleep research has been conducted in medical settings in which 
physicians asked patients questions regarding their sleep quality and quantity. If sleep problems 
were significant, physicians usually referred patients to a sleep lab, which uses PSG (invasive 
and time-consuming method) to assess a wide range of clinical sleep problems, such as 
obstructive sleep apnea. Recently, new technological developments have allowed for the 
assessment of sleep activity (i.e., movement) via a non-invasive wristwatch (i.e., actigraphy). 
This small device has been validated against PSG (agreements ranging from 85-90% between 
the two methods; Sadeh, Sharkey, & Carskadon,1994; Sadeh, Acebo, Seifer, Aytur, & 
Carskadon,1995), and found to be an acceptable proxy for sleep (Sadeh, 2008). The actigraphy 
watch is a great research tool for a more in depth study of sleep in the natural setting (Sadeh, et 
al., 1995).  
 
14 
 
Subjective reports of sleep of children are typically completed by either the parent or the 
child, depending on the developmental stage of the child. Although parents tend to be fairly 
reliable in their reporting of children?s sleep schedules (Tikotzky & Sadeh, 2001), they do not 
always know the number of night wakings, how long it takes the child to go back to sleep after 
he or she has awoken, or how much sleep activity occurred, leading to discrepancies in sleep 
quality measures between objective and subjective assessments (Sadeh, Sharkey, & 
Carskadon, 1994). Children, on the other hand, may be more accurate regarding how many 
times they get up during the night, but they may not be good at reporting how long it took them 
to fall asleep or their sleep schedule (e.g., time perception may be more difficult for them). Child 
and adolescent reports of subjective sleep quality have been found to have low or non-
significant relations with objective measures of sleep (Liu & Zhou, 2002; Meijer, Habekothe, & 
Van Den Wittenboer, 2000), however both types of sleep assessments have been found to be 
related to child adjustment and health (Dewald et al., 2009; Sadeh, 2008; Sadeh, Gruber, & 
Raviv, 2002; Wolfson & Carskadon, 2005). This provides evidence for the idea that subjective 
and objective measures represent two separate domains (Dewald et al., 2009). Although 
actigraphy provides more precise information (e.g., sleep onset) and is related to a number of 
adjustment and cognitive outcomes, subjective reports have been found to show stronger 
associations with behavior and cognitive function (Mitru, Millrood, & Mateika, 2002; Moore & 
Meltzer, 2008). Thus, subjective reports, especially those related to the amount of sleepiness, 
could be tapping into an aspect of sleep that objective assessments currently cannot access, 
such as differences in individual needs of sleep amounts. The inclusion of both objective and 
subjective sleep measures is therefore essential given their differing links with child outcomes 
(Sadeh, 2008; Sadeh, Gruber, & Raviv, 2002; Wolfson & Carskadon, 2005). Therefore, in the 
following studies we will use both subjective child reports and objective actigraphy measures to 
provide better estimates of sleep parameters than reliance on either measure alone.  
 
 
15 
 
Potential Influential Variables 
 Study Variables and Age. Sleep parameters, including amount (duration) and quality 
(number of night wakings) change with age (Sadeh, Raviv, & Gruber, 2000). According to the 
National Sleep Foundation, infants need, on average, a total of 15 hours of sleep a day, but 
children in middle childhood typically require about 10 hours of sleep, whereas adults require 
much less (~ 8 hrs). By the time children attend school, they would have spent more time 
asleep than in any other waking activity (Dahl, 1996). Sleep is an important regulatory process 
for individuals of all ages, but it seems especially fundamental to brain development at young 
ages. Research on sleep patterns of typically developing school-aged children is growing but 
small, and suggests that sleep problems at early ages have a significant impact on 
developmental outcomes at later ages. For example, sleep difficulties at age 4 predicted 
behavioral, emotional, and attention problems at age 15 (Gregory & O?Connor, 2002). 
Additionally, Buckhalt and colleagues (2009) found sleep at age 8 to significantly predict 
cognitive functioning a year later, controlling for children?s initial cognitive functioning. 
 In addition to age influencing aspects of children?s sleep, age also plays a role in levels 
of understanding and cognitive functioning abilities. Throughout life, but particularly during the 
younger years of development when the basic building blocks of knowledge (e.g., vocabulary, 
addition) are being developed, neural plasticity and refinement are constantly occurring in 
response to challenges presented by the environment (Kopasz et al., 2010).  
 Furthermore, relations between age and vagal suppression have been reported in the 
literature. Older school-aged children have been found to show less RSA suppression to 
laboratory tasks in comparison to their younger counterparts (El-Sheikh, 2005). 
 Both studies included in this dissertation focused on children of a narrow age range (8 to 
9 yrs at T1; 10 to 11 yrs at T2) to minimize potential confounds associated with developmental 
and sleep requirement differences among participants. Further, I consider this investigation an 
initial step towards understanding associations among sleep, vagal regulation, and children?s 
 
16 
 
cognitive functioning. Age is used as a control variable in all path models in which age is a 
correlate of any of the main study variables.   
    Study Variables and Ethnicity. Ethnic practices impact sleep (Redline et al., 1999), but 
variations in sleep patterns among ethnic groups are relatively unexplored. For example, the 
majority of African American (AA) sleep literature has appeared within the last 5 years. In a 
review of 30 AA adult sleep studies, Durrence and Lichstein (2006) generally concluded that: (1) 
AAs take longer to fall asleep, (2) nap longer, (3) have higher rates of sleep-disordered 
breathing (SDB), and (4) have basic differences in their sleep architecture (e.g., depth of sleep). 
Furthermore, studies have reported that AA adults were more likely than European American 
(EA) adults to have short sleep duration or sleep less than seven hours per night (Stamatakis, 
Kaplan, & Roberts, 2007; Krueger & Friedman, 2009).  
Sleep and ethnicity research conducted with children is sparse, but suggests that 
asthma and SDB are more prevalent in AA children than EA children (Redline et al., 1999; 
Rosen et al., 2003). A recent study (Hale, Berger, LeBourgeois, & Brooks-Gunn, 2009) found 
that even after controlling for potential SES and family composition confounds, African-
American children went to bed later and had less regular bedtimes or bedtime routines than 
European-American children. A longitudinal study (N = 1,043) conducted with 2- to 8-year-olds 
examined ethnic differences in parent reported sleep (Crosby, LeBourgeois, & Harsh, 2005). 
Results indicated that the AA children (26% of the sample) had shorter nighttime sleep, more 
napping, and more weekend ?oversleep? than their EA counterparts, the latter of which is 
considered a marker of sleep debt (Crosby et al., 2005). Crosby and colleagues (2006) 
conducted another study with 715 caregiver reports regarding the sleep behaviors of children 
ages 2- to 12-years old. Results indicated differences in EA and AA children?s sleep patterns. 
For example, EA children were found to quit napping by age 6, whereas 64% of the AA children 
continued to nap between the ages of 6 and 12. In addition, EA children began to sleep about 
25 minutes longer on the weekends compared to the weekdays around the age that school 
 
17 
 
attendance began. For AA children, this weekend ?oversleep? began at age 3. Further, El-
Sheikh and colleagues have reported AA children to have significantly shorter sleep durations, 
less sleep activity, and more variability in their sleep onset time in comparison to their EA 
counterparts. For example, AA children slept on average 23 minutes less a day and got up 
about 17 minutes earlier on the weekdays than EA children (Buckhalt, El-Sheikh, & Keller, 
2007). Although I would usually examine ethnicity as a moderating variable because of the 
aforementioned differences, it was determined that it would be best to control for race in these 
studies due to issues associated with power.  
Study Variables and SES. According to the National Sleep Foundation 
(www.sleepfoundation.org), sleep hygiene (recommendations for obtaining good sleep), 
includes practices such as maintaining a cool, comfortable sleep environment and having a 
consistent sleep-wake schedule. These recommendations may be harder to obtain for those of 
lower SES than higher SES homes. For example, chaotic living conditions of economically 
disadvantaged preschool children, such as less structured routines, overcrowding, and family 
instability, is significantly related to mothers? reports of their children?s subjective sleep problems 
(Brown & Low, 2008). Furthermore, smoking is more common in lower SES homes (Barbeau, 
Krieger, & Soobader, 2004) and children from lower SES homes are at an increased risk of 
second-hand smoke exposure especially when the size of the home and the ventilation system 
is accounted. Research indicates that children from low-income families have higher rates of 
sleep disordered breathing, shorter sleep durations, reduced sleep quality, and more frequent 
weekend napping in comparison to their more affluent counterparts (Crosby, LeBourgeois, & 
Harsh, 2005; Redline et al., 1999; Rosen et al., 2003; Stepanski, Zayyad, Nigro, Lopata, & 
Basner, 1999). Other SES factors that could contribute to a reduction in optimal sleep are: 
resources (money) or the capability (central heat and air) to adjust home temperature settings, 
increased home allergens, room sharing, school/work schedules (parents may work swing 
 
18 
 
shifts; children may have to wake early for the school), low quality mattresses and bedding, as 
well as the added stressors that tend to aggregate disproportionally in poor populations.  
The potential role of SES in children?s sleep is important to address, yet few studies 
have examined these effects (Buckhalt et al., 2007; Buckhalt & Staton, in press). Of those that 
have, results suggest that children from higher SES homes report fewer sleep-wake problems 
and slept longer in comparison to those of lower SES homes (El-Sheikh, Buckhalt, Keller, 
Cummings, & Acebo, 2007b). Cross-sectional examinations conducted with the current sample 
indicated that when children were in third grade, SES moderated the link between sleep and 
cognitive performance. Specifically, when sleep patterns and quality were optimal, no 
differences were found in cognitive functioning of lower and higher SES children. However, 
when sleep problems were present, children from lower SES homes performed more poorly 
than their higher SES counterparts (Buckhalt, et al., 2007). When examinations were conducted 
longitudinally, SES, particularly paternal education, was again found to be a moderator of the 
sleep-cognitive link (Buckhalt, et al., 2009). However, the extent to which specific ecological 
risks predict child sleep disruptions is unknown.   
Cognitive performance is greatly impacted by poverty and lower SES (Pungello, 
Kupersmidt, & Burchinal, 1996).  A ?gap? in cognitive functioning between those of higher and 
lower SES has consistently been recognized in the literature (Buckhalt et al., 2007; Buckhalt & 
Staton, in press). Children from lower SES homes are more likely to have poor school 
performance (Duncan, Brooks-Gunn, & Klebanov, 1994; Sirin, 2005), are referred for special 
education services more frequently (Blair & Scott, 2002), and are at a greater risk to repeat 
grades or drop out of school than their more economically advantaged peers (Alexander, 
Entwisle, & Dauber, 1996; Sirin, 2005), although the role of sleep in the SES-cognitive link has 
relatively been unexplored (Buckhalt & Staton, in press). One of the few studies examining the 
role of sleep in the SES-cognitive functioning link was conducted with children (Bates, Viken, 
Alexander, Beyers, & Stockton, 2002), and findings suggest that variability in sleep schedule is 
 
19 
 
related to academic performance only for preschoolers (90% EA) from low income homes. Due 
to the relations of the study variables with SES, this demographic variable was controlled in all 
analyses.  
 Study Variables and Sex. Sex differences have been reported in the sleep literature 
(El-Sheikh, Erath, & Keller, 2007; Sadeh et al., 2000). For example, the 2005 Sleep in America 
poll conducted by the National Sleep Foundation found that women were more likely than men 
to experience problems falling and staying asleep and to report more daytime sleepiness. In a 
study with adults conducted by Lindberg and colleagues (1997), women reported longer sleep 
times than men, but difficulties with maintaining sleep and daytime sleepiness were more 
common among women. Sleep disordered breathing is more likely to occur in men than women 
(Redline, Kump, Tishler, Browner, & Ferrette, 1994).  
Although few studies have examined gender differences with typically developing 
children, some have found girls to have longer sleep duration and less active sleep than boys 
(Gaina, Sekine, Hamanishi, Chen, & Kagamimori, 2005; Sadeh, Raviv, & Gruber, 2000; Ito, 
Kondo, Kanbayashi, Ohnuma, Kanayama, Aizawa, Kaneko, Shimizu, & Manber, 2006). In the 
current sample, initial analyses indicated that girls slept longer (M = 7:34 hrs:min) than boys 
(7:05), and had better sleep efficiency (88%) and less activity during their sleep (41) than boys 
(M sleep efficiency = 85%; sleep activity = 50) at T1. During the second wave of data collection, 
girls were still found to sleep longer (7:32) than boys (7:16) and have less sleep activity (M = 37) 
than boys (M = 43).  
Sex differences have also been noted in the vagal literature although results have not 
been consistent across studies. Specifically, some studies have found boys to have higher 
baseline RSA than girls (El-Sheikh, 2005; Salomon, 2005), whereas others have reported 
higher baseline levels in girls (Fabes, Eisenberg, Karbon, Troyer, & Switzer, 1994). In addition, 
others have found no gender differences (Suess et al., 1994). Due to the relations between 
gender and the study variables, gender was considered a covariate in all path models.  
 
20 
 
Study Variables and Puberty. A change in circadian rhythm, known as sleep phase 
delay, has been noted to occur during puberty (Carskadon, Vieira, & Acebo, 1993). This change 
delays the time that adolescents feel sleepy, which in turn alters their sleep patterns (Wolfson & 
Carskadon, 1998) and can reduce their sleep duration especially during the school year when 
wake times are more consistent. Additionally, research has shown that even when teens obtain 
optimal amounts of sleep, they show an increase in daytime sleepiness (Carskadon, Vieri, & 
Acebo, 1993). 
In addition, pubertal development has been found to be related to RSA suppression, in 
that more physically mature children have been found to have less RSA suppression to 
laboratory tasks (El-Sheikh, 2005). To determine if puberty should be controlled in our analyses, 
we examined correlations between children?s pubertal status and all study variables. Results 
indicated that there were no significant relations between pubertal status and the study 
variables at T1 or T2; therefore, puberty was not used as a covariate.      
Study Variables and Body Mass Index. Sleep problems and body mass index (BMI) 
have been associated positively both cross-sectionally (Chervin et al., 2003; Gupta, Muller, 
Chan, & Meininger, 2002; Von Kries, Toschke, Wurmser, Sauerwald, & Koletzko, 2002; Sekine 
et al., 2002) and longitudinally (Agras, Hammer, McNicholas, & Kraemer, 2004; Reilly et al., 
2005; Sugimori et al., 2004). Overweight individuals have more symptoms of sleep disordered 
breathing, later sleep times, shorter sleep durations, and more fragmented sleep than healthy 
controls (Beebe et al., 2007). Further, in comparison to lean children, children with weight 
problems are at a significantly higher risk of having obstructive sleep apnea (Chay et al., 2000; 
Marcus, 1996; Wing et al., 2003).  
Obesity is linked to reductions in heart rate variability, a marker of PNS influence on the 
heart (Connolly, Kraus, Short & Carno, 2006). Specifically, researchers examined 54 
participants (24 normal BMI; 10 at risk for obesity; 20 overweight) with a mean age of 9.2 years 
(SD = 4.1 yrs). Results demonstrated that the degree of obesity was more closely linked to 
 
21 
 
decreases in heart rate variability associated with sympathetic over-drive than the presence of 
obstructive sleep apnea syndrome. 
Additionally, relations between children?s BMI and baseline RSA have been noted in the 
literature. El-Sheikh and colleagues (2007a) found a significant negative association between 
BMI and RSA, in that school-aged children with a higher body mass index displayed lower 
baseline RSA. The links between BMI and the variables of interest in this study indicate a need 
to control for children?s BMI in all analyses. 
 Study Variables and Season of Sleep Assessment. The change of seasons brings 
about alterations in the length of daylight during a given 24-hour period. Considering that light 
patterns impact the release of melatonin, it is important to consider the season of the year when 
examining children?s sleep parameters (Ozaki, Ono, Ito, & Rosenthal, 1995). Season was 
dichotomized into two; winter and all other seasons. The season of actigraph watch wear was 
found to be related to the objective sleep variables at T1, however not at T2; therefore, in all 
regression equations at T1 the season of watch wear was controlled. 
 The aforementioned child and demographic variables were controlled for or examined as 
possible controls to increase the rigor in testing the proposed research questions.  
Research Objectives 
This dissertation contains two studies. In Study 1 (manuscript 1), I will test whether the 
objective or subjective sleep markers (Sleep Minutes, Long Wake Episodes, Sleep Efficiency, 
Sleepiness, or Sleep/Wake Problems) interact with either vagal tone or vagal reactivity cross-
sectionally to predict one of the following cognitive outcomes: Concept Formation, Auditory 
Working Memory, Numbers Reversed, and Decision Speed assessed via the Woodcock-
Johnson III Tests of Cognitive Abilities (WJIII; Woodcock et al., 2001). The aim of the study is 
to determine if vagal tone or vagal reactivity is a potential individual difference variable that 
should be considered in future sleep research and in the development of treatment and 
 
22 
 
intervention programs aimed at lessening the effects of sleep disruptions on child outcomes, 
specifically cognitive performance.  
Building on the first study, the second study (manuscript 2) addresses the 
aforementioned research question not only cross-sectionally at T2, but also longitudinally (3rd to 
5th grade) while controlling for the autoregressive effects. In addition, I add vagal recovery as a 
third marker of vagal functioning with a potential to moderate the sleep disruptions-poor 
cognitive functioning connection cross-sectionally. Of note, this is the first study to examine links 
between vagal recovery and sleep in children. My aims for the second manuscript are to follow 
up on significant moderation findings presented in manuscript 1 two years later. This follow up 
could provide insight into possible critical stages in which the interaction of sleep and vagal 
functioning may be most influential in the prediction of cognitive functioning. Further, the 
longitudinal component of the study will allow for clarification of a possible cumulative effect in 
cognitive functioning associated with the interaction of two biological systems, namely sleep and 
parasympathetic nervous system functioning.  
 Based on available evidence, I hypothesize that higher vagal tone, more pronounced 
vagal suppression to the challenge, and/or faster vagal recovery will act as a protective factor 
against reduced cognitive functioning in the context of sleep disruptions (Sleep Minutes, Long 
Wake Episodes, Sleep Efficiency, Sleepiness, or Sleep/Wake Problems). That is, I expect an 
interaction between sleep and vagal functioning in that children with more optimal vagal 
functioning indices will be protected from reduced cognitive functioning (i.e., these children will 
perform better on the WJIII tests) in the context of sleep disruptions. Conversely, children with 
lower levels of baseline vagal tone, less pronounced vagal suppression, and/or slower recovery 
from stress are expected to show less optimal cognitive performance scores when sleep 
disruptions are present. That is, I expect children with less optimal vagal functioning (i.e., low 
baseline vagal tone, less vagal suppression or vagal augmentation, and/or slower vagal 
recovery) to have reduced cognitive functioning when sleep disruptions are present than their 
 
23 
 
more optimally regulating counterparts. These studies are the first of their kind and therefore no 
hypotheses regarding which sleep indices (quality, quantity, sleepiness, or sleep/wake 
problems) will interact with which vagal parameter (vagal tone, vagal reactivity, or vagal 
recovery) most strongly to influence cognitive functioning. However, I do anticipate that the 
cognitive outcomes that are most tied to vagal functioning (executive functions and working 
memory) will be the most likely candidates for the moderation hypothesis.  
For a more thorough examination of the research question, the objective sleep variables 
were examined in separate models for each of the cognitive outcomes. The subjective reports of 
sleep were examined together for each cognitive outcome. These decisions were made due to 
multi-collinearity of the objective sleep variables and because this is a first step in understanding 
what types of cognitive functions are impacted by the interaction of sleep and vagal indices. 
Further, a call in the sleep literature has been made to examine quality, quantity and sleepiness 
separately because they are considered to represent separate domains of sleep (Dewald et al., 
2009). In addition, it was suggested that research compare the effects of subjective and 
objective assessments within the same investigation to examine possible parameter 
assessment differences (Sadeh, 2007).  
In the two studies presented in this document, reference will be made to vagal 
functioning when no specificity is required regarding the various facets of vagal activity. 
However, when more specificity is needed, we use the terms vagal tone, vagal regulation 
(suppression or augmentation for added precision regarding the direction of vagal regulation), 
and vagal recovery as pertinent. Vagal tone and regulation, that is higher vagal tone and vagal 
suppression to tasks, has been established in the literature regarding social (e.g., Eisenberg, 
Fabes, & Murphy, 1996), emotional (e.g., Calkins, 1997), and behavioral (e.g., Pine et al., 1996) 
domains. However, vagal recovery literature is scant, especially with children. This will be the 
first study to our knowledge to examine the role of vagal tone, vagal regulation, and vagal 
recovery (the latter used only at T2) in the relation between sleep parameters (i.e., quality, 
 
24 
 
quantity, subjective reports) and several cognitive performance measures (e.g., Concept 
Formation, Auditory Working Memory, Numbers Reversed, and Decision Speed).  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
25 
 
References 
Acebo, C., & Carskadon, M. (2001). Scoring actigraph data using ACTION-W 2. Providence, RI: 
Bradley Sleep Center, Brown University. 
Acebo, C., Sadeh, A., Seifer, R., Tzischinsky, O., Wolfson, A., Hafer, A., & Carskadon, M. A. 
(1999). Estimating sleep patterns with actigraphy monitoring in children and adolescents: 
How many nights are necessary for reliable measures? Sleep, 22, 95-103. 
Adam, M., Retey, J. V., Khatami, R., & Landolt, H. P. (2006). Age-related  
 changes in the time course of vigilant attention during 40 hours without  
 sleep in men. Sleep, 29, 55-57. 
Adam, E. K., Snell, E. K., & Pendry, P. (2007). Sleep timing and quantity in ecological and 
family context: A nationally representative time-diary study. Journal of Family 
Psychology, 21(1), 4-19. 
Agras, W., Hammer, L., Mcnicholas, F., & Kraemer, H. C. (2004). Risk factors for childhood 
overweight: A prospective study from birth to 9.5 years. Journal of Pediatrics, 145, 20-
25. 
Alexander, K. L., Entwisle, D. R., & Dauber, S. L. (1996). Children in motion: School transfers 
and elementary school performance. The Journal of Educational Research, 90, 3-12. 
Anders, T. F. (1994). Infant sleep, nighttime relationships, and attachment. Psychiatry, 57, 11-
21. 
Archbold, K. H., Giordani, B., Ruzicka, D. L., & Chervin, R. D. (2004). Cognitive executive 
dysfunction in children with mild sleep-disordered breathing. Biological Research For 
Nursing, 5(3), 168-176. 
Archbold, K. H., Pituch, K. J., Panahi, P., & Chervin, R. D. (2002). Symptoms of sleep 
disturbances among children at two general pediatric clinics. Journal of Pediatrics, 140, 
97-102. 
 
26 
 
Backs, R. W. & Ryan, A. M. (1992). Multimodal measures of mental workload during dual-task 
performance: Energetic demands of cognitive processes. Presented at the Human 
Factors Society 36th Annual Meeting, Santa Monica, CA, pp. 1413-1417. 
Barbeau, E. M., Krieger, N., & Soobader, M. (2004). Working class  
 matters: Socioeconomic disadvantage, race/ethnicity, gender, and 
 smoking in NHIS 2000. American Journal of Public Health, 94, 269- 
 278. 
Bar-Haim, Y., Marshall, P. J., & Fox, N. A. (2000). Developmental changes in heart period and 
high-frequency heart period variability from 4 months to 4 years of age. Developmental 
Psychobiology, 37, 44-56. 
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social 
psychological research: Conceptual, strategic, and statistical consideration. Journal of 
Personality and Social Psychology, 51, 1173-1182. 
Bates, J. E., Viken, R.J., Alexander, D.B., Beyers, J., & Stockton, L. (2002). Sleep and 
adjustment in preschool children: Sleep diary reports by mothers relate to behavior 
reports by teachers. Child Development, 71, 62-74. 
Beauchaine, T. (2001). Vagal tone, development, and Gray's Motivational Theory: Toward an 
integrated model of autonomic nervous system functioning in psychopathology. 
Development & Psychopathology, 13, 183-214. 
Beebe, D. W. (2006). Neurobehavioral morbidity associated with disordered  
 breathing during sleep in children: A comprehensive review. Sleep, 29,  
 1115?1134. 
Beebe, D. W., Lewin, D., Zeller, M., McCabe, M., MacLeod, K., Daniels, S. R., &  
 Amin, R. (2007). Sleep in overweight adolescents: Shorter sleep, poorer  
 sleep quality, sleepiness, and sleep-disordered breathing. Journal of  
 
27 
 
 Pediatric Psychology, 32, 69-79. 
Berntson, G. G., Bigger, J. T., Eckberg, D. L., Grossman, P., Kaufmann, P. G., Malik, M., et al. 
(1997). Heart rate variability: Origins, methods, and interpretive caveats. 
Psychophysiology, 34, 623-648. 
Blair, C. (2003). Behavioral inhibition and behavioral activation in young children: Relations with 
self-regulation and adaptation to preschool in children attending Head Start. 
Developmental Psychobiology, 42, 301-311. 
Blair, C. & Scott, K. G. (2002). Proportion of LD placements associated with low 
 Socioeconomic status: Evidence for a gradient? Journal of Special Education, 36, 14-22. 
Blunden, S. L., & Beebe, D. W. (2006). The contribution of intermittent hypoxia, sleep debt and 
sleep disruption to daytime performance deficits in children: Consideration of respiratory 
and non-respiratory sleep disorders. Sleep Medicine Reviews, 10, 109-118. 
Blunden, S., Lushington, K., & Kennedy, D. (2001). Cognitive and behavioural performance in 
children with sleep related obstructive breathing disorders. Sleep medicine Reviews, 5, 
447-461. 
Blunden, S., Lushington, K., Lorenzen, B., Martin, J., & Kennedy, D. (2005). Neuropsychological 
and psychosocial function in children with a history of snoring or behavioral sleep 
problems. The Journal of Pediatrics, 146(6), 780-786. 
Born, J., Rasch, B., & Gais, S. (2006). Sleep to remember. The  
 Neuroscientist, 12, 410-424. 
Bornstein, M. H., & Suess, P.E. (2000). Child and mother cardiac vagal tone: Continuity, 
stability, and concordance across the first 5 years. Developmental Psychology, 36, 54-
65. 
Brown, E. D., & Low, C. M. (2008). Chaotic living conditions and sleep problems 
 Associated with children?s responses to academic challenge. Journal of Family 
Psychology, 22, 920-923. 
 
28 
 
Buckhalt, J. A., El-Sheikh, M., & Keller, P. (2007). Children's sleep and cognitive functioning: 
Race and socioeconomic status as moderators of effects. Child Development, 78(1), 
213-231. 
Buckhalt, J. A., El-Sheikh, M., Keller, P. S., & Kelly, R. J. (2009). Concurrent and longitudinal 
relationships between children's sleep and cognitive functioning: The moderating role of 
parent education. Child Development, 80, 875-892. 
Buckhalt, J. A., & Staton, L. E. (in press). Children?s sleep, school performance, and family 
socioeconomic status. In M. El-Sheikh (Ed.), Sleep and development: Familial and 
sociocultural considerations. New York: Oxford University Press. 
Buckhalt, J., Wolfson, A., & El-Sheikh, M. (2009). Children's sleep and school psychology 
practice. School Psychology Quarterly, 24, 60-69. 
Brugha, T. S., Bebbington, P. E., Tennant, C., & Hurry, J. (1985). The list of threatening 
experiences: A subset of 12 event categories with considerable long-term contextual 
threat. Psychological Medicine, 15, 189-194. 
Calkins, S. D. (1997). Cardiac vagal tone indices of temperamental reactivity and behavioral 
regulation in young children. Developmental Psychobiology, 31, 125-135. 
Calkins, S. D., & Keane, S. P. (2004). Cardiac vagal regulation across the preschool period: 
Stability, continuity, and implications for childhood adjustment. Developmental 
Psychobiology, 45(3), 101-112. 
Carskadon, M. A. (1992). Yawning elicited by reading: Effects of sleepiness. Sleep Research, 
21, 101. 
Carskadon, M. A. (2002). Factors influencing sleep patterns of adolescents. In Adolescent sleep 
patterns: Biological, social, and psychological influences (pp. pp.4-26). New York: 
Cambridge University Press. 
Carskadon, M. A., & Acebo, C. (1992). Relationship of a morningness/eveningness scale to 
sleep patterns in children. Sleep Research, 21, 367. 
 
29 
 
Carskadon, M. A., &  Dement, W. C. (2005). Normal human sleep: An overview. In M. H. 
Kryger, T. Roth, & W. C. Dement (Ed.), Principles and Practice of Sleep Medicine (4 ed., 
pp. 1359-1377). Philadelphia: Elsevier Saunders. 
Carskadon, M. A., Viera, C., & Acebo C. (1993). Association between puberty and delayed 
phase preference. Sleep, 16, 258-262. 
Chay, O. M., Goh, A., Abisheganaden, J., Tang, J., Lim, W. H., Chan, Y. H., et al.  
 (2000). Obstructive sleep apnea syndrome in obese Singapore children.  
 Pediatric Pulmonology, 29, 284?290. 
Chervin, R. D., Clarke, D. F., Huffman, J. L., Szymanski, E., Ruzicka, D. L., Miller, V., et al. 
(2003). School performance, race, and other correlates of sleep-disordered breathing in 
children. Sleep Medicine, 4, 21-27. 
Chervin, R., Ruzicka, D., Giordani, B., Weatherly, R., Dillon, J., Hodges, E., Marcus, C., & 
Guire, K. (2006). Sleep-disordered breathing, behavior, and cognition in children before 
and after adenotonsillectomy. Pediatrics, 117, e769-e778. 
Cicchetti, D., & Dawson, G. (2002). Editorial: Multiple levels of analysis. Development and 
Psychopathology, 14, 417-420. 
Clark, K. B., Naritoku, D. K., Smith, D. C., Browning, R. A., & Jensen, R. A. (1999). Enhanced 
recognition memory following vagus nerve stimulation in human subjects. Nature 
Neuroscience, 2, 94-98. 
Cole, D. A., & Maxwell, S. E. (2003). Testing mediational models with longitudinal data: 
Questions and tip in the use of structural equation modeling. Journal of Abnormal 
Psychology, 112, 558-577. 
Cole, p., Usher, B., & Cargo, A. (1993). Cognitive risk and its association with risk for disruptive 
behavior disorder in preschoolers. Journal of Clinical Child Psychology, 22, 154-165. 
Crosby, B., LeBourgeois, M. K., & Harsh, J. (2005). Racial differences in reported napping and 
nocturnal sleep in 2-to-8-year-old children. Pediatrics, 115, 225-232. 
 
30 
 
Curcio, G., Ferrara, M., & De Gennaro, L. (2006). Sleep loss, learning capacity and academic 
performance. Sleep Medicine Reviews, 10, 323-337. 
Dahl, R. E. (1996). The regulation of sleep and arousal: Development and psychopathology. 
Development and Psychopathology, 8, 3-27. 
Davies, P. T., Forman, E.M., Rasi, J.A., & Stevens, K.I. (2002). Assessing children's emotional 
security in the interparental relationship: The security in the interparental subsystem 
scales. Child Development, 73, 544-562. 
Dewald, J. F., Meijer, A. M., Oort, F. J., Kerkhof, G. A, & Bogels, S. M. (2009). The influence of 
sleep quality, sleep duration and sleepiness on school performance in children and 
adolescents: A meta-analytic review. Sleep Medicine Reviews, Advanced online 
publication. PMID: 20093054.  
Duncan, G., Brooks-Gunn, J., & Klebanov, P. (1994). Economic deprivation and early childhood 
development. Child Development, 65, 296-318. 
Durrence, H. H., & Lichstein, K. L. (2006). The sleep of African-Americans: A comparative 
review. Behavioral Sleep Medicine, 4, 29-44. 
Drummond, S., & Brown, G. (2001). The effects of total sleep deprivation on cerebral response 
to cognitive performance. Neuropsychopharmacology, 25(S5), S68-S73. 
Drummond, S., Brown, G. G., Gillin, J. C., Stricker, J. L., Wong, E. C., Buxton, R. B. (2000). 
Altered brain response to verbal learning following sleep deprivation. Nature, 403, 655-
657. 
Durmer, J. S., & Dinges, D. F. (2005). Neurocognitive consequences of sleep  
 deprivation. Seminars in Neurology, 25, 117-129. 
Efron, B., & Tibshirani, R. J. (1993). An introduction to the Bootstrap. In Monographs on 
Statistics and Applied Probability (Vol. 57). New York: Chapman & Hall. 
 
31 
 
Eisenberg, N., Fabes, R. A., & Murphy, B. C. (1996). Parents' reactions to children's negative 
emotions: Relations to children's social competence and comforting behavior. Child 
Development, 67(5), 2227-2247. 
El-Sheikh, M. (2005). The role of emotional responses and physiological reactivity in the marital 
conflict-child functioning link. Journal of Child Psychology and Psychiatry, 46, 1191-
1199. 
El-Sheikh, M., & Buckhalt, J. A. (2005). Vagal regulation and emotional intensity predict 
children's sleep problems. Developmental Psychophysiology, 46, 307-317. 
El-Sheikh, M., Buckhalt, J., Mize, J., & Acebo, C. (2006). Marital conflict and disruption of 
children's sleep. Child Development, 77, 31-43. 
El-Sheikh, M., Buckhalt, J. A., Keller, P. S., Cummings, E. M., & Acebo, C. (2007b). Child 
emotional insecurity and academic achievement: The role of sleep disruptions. Journal 
of Family Psychology, 21, 29-38. 
El-Sheikh, M., Erath, S. A., & Keller, P. S. (2007a). Children's sleep and adjustment: The 
moderating role of vagal regulation. Journal of Sleep Research, 16, 396-405. 
El-Sheikh, M., & Harger, J. (2001). Appraisals of marital conflict and children's adjustment, 
health, and physiological reactivity. Developmental Psychology, 37, 875-885. 
El-Sheikh, M., & Whitson, S. A. (2006). Longitudinal relations between marital conflict and child 
adjustment: Vagal regulation as a protective factor. Journal of Family Psychology, 20, 
30-39. 
Epstein, R., Chillag, N., & Lavie, P. (1998). Starting times of school: Effects of daytime 
functioning of fifth grade children in Israel. Sleep, 21, 250-256. 
Erdfelder, E., Faul, F., & Buchner, A. (1996). GPOWER: A general power analysis program. 
Behavior Research Methods, Instruments & Computers, 28, 1-11. 
Evans, G. W. (2003). A multimethodological analysis of cumulative risk and allostatic load 
among rural children. Developmental Psychology, 39, 924-933. 
 
32 
 
Fabes, R. A., Eisenberg, N., Karbon, M., Troyer, D., & Switzer, G. (1994). 
 The relations of children?s emotion regulation to their vicarious emotional responses and 
comforting behaviors. Child Development, 65, 1678-1693.  
Fallone, G., Seifer, R., Acebo, C., & Carskadon, M. (2000). Prolonged sleep restriction in 11- 
and 12-year-old children: Effects on behavior, sleepiness and mood. Sleep, 23S2, A:28. 
Fredriksen, K., Rhodes, J., Reddy, R., & Way, N. (2004). Sleepless in Chicago: Tracking the 
effects of adolescent sleep loss during the middle school years. Child Development, 75, 
84-95. 
Gaina, A., Sekine, M., Hamanishi, S., Chen, X., & Kagamimori, S. (2005). Gender and temporal 
differences in sleep-wake patterns in Japanese school children. Sleep, 28, 337-342. 
Gais, S., Albouy, G., Boly, M., Dang-Vu, T. T., Darsaud, A., Desseilles, M., et al. (2007). Sleep 
transforms the cerebral trace of declarative memories. Proceedings of the National 
Academy of Sciences, 107, 18778-18783. 
Galles, S. J., Miller, A., Cohn, J. F., & Fox, N. A. (2002). Estimating parasympathetic control of 
heart rate variability: Two approaches to quantifying vagal tone. Psychophysiology, 
39(Suppl. 1), S37. 
Gianaros, P. J., Van Der Veen, F. M., & Jennings, R. (2004). Regional cerebral blood flow 
correlates with heart period and high-frequency heart period variability during working-
memory tasks: Implications for the cortical and subcortical regulation of cardiac 
autonomic activity. Psychophysiology, 41, 521-530. 
Gozal, D. (1998). Sleep-disordered breathing and school performance in children. Pediatrics, 
102, 616-620. 
Gregory, A. M., & O'Connor, T. G. (2002). Sleep problems in childhood: A longitudinal study of 
developmental change and association with behavioral problems. Journal of the 
American Academy of Child & Adolescent Psychiatry, 41, 964-971. 
 
33 
 
Greenberg, M., & Lengua, L. (1995). Scale Construction for the Neighborhood Questionnaire. 
Fast Track Project Technical Report. Seattle, WA: University of Washington. 
Grossman, P., & Taylor, E. W. (2007). Toward understanding respiratory sinus arrhythmia: 
Relations to cardiac vagal tone, evolution and biobehavioral functions. Biological 
Psychology, 74, 263-285. 
Grossman, P., & Wientjes, K. (1986). Respiratory sinus arrhythmia and parasympathetic cardiac 
control: Some basic issues concerning quantification, application, and implications. In K. 
J. P. Grossman, & D. Vaitl (Eds.) (Ed.), Cardiorespiratory and cardio somatic 
psychophysiology  (pp. 117-138). New York: Plenum. 
Gupta, N. K., Mueller, W. H., Chan, W., & Meininger, J. C. (2002). Is obesity associated with 
poor sleep quality in adolescents? American Journal of Human Biology, 14(6), 762-768. 
Halbower, A. C., & Mahone, E. M. (2006). Neuropsychological morbidity linked to  
 childhood sleep disordered breathing. Sleep Medicine Reviews, 10, 97? 
 107. 
Hale, L., Berger, L. M., LeBourgeois, M., & Brooks-Gunn, J. (2009). Social and  
 Demographic predictors of preschoolers? bedtime routines. Journal of  
 Developmental and Behavioral Pediatrics, 30, 394-402. 
Hansen, A. L., Johnsen, B. H., & Thayer, J. F. (2003). Vagal influence on working memory and 
attention. International Journal of Psychophysiology, 48, 263-274. 
Harper, G. W., & Ottinger, D. R. (1992). The performance of hyperactive and control 
preschoolers on a new computerized measure of visual vigilance: The preschool 
vigilance task. Journal of Child Psychology & Psychiatry, 33, 1365-1372. 
Harrison, Y., & Horne, J. A. (1998). Sleep loss impairs short and novel language 
 tasks having a prefrontal focus. Journal of Sleep Research, 7, 95?100. 
 
34 
 
Hastings, P., Sullivan, C., McShane, K., Coplan, R., Utendale, W., & Vyncke, J. (2008). Parental 
socialization, vagal regulation, and preschoolers' anxious difficulties: Direct mothers and 
moderated fathers. Child Development, 79, 45-64. 
Hayano, J., Sakakibara, Y., Yamada, A, Yamada, M., Mukai, S., Fujinami, T., et al. (1991). 
Accuracy of assessment of cardiac vagal tone by heart rate variability in normal 
subjects. American Journal of Cardiology, 67, 199-204. 
Hill, C. M., Hogan, A. M., & Karmiloff-Smith, A. (2007). To sleep, perchance to enrich learning? 
Archives of Disease in Childhood, 92 (7), 637-643. 
Hollingshead, A. B. (1975). Four factor index of social status. Unpublished manuscript. 
Horne, J. A. (1993). Human sleep, sleep loss and behaviour. Implications for the prefrontal 
cortex and psychiatric disorder. British Journal of Psychiatry, 162, 413-419. 
Huffman, L. C., Bryan, Y. E., del Carmen, R., Pederson, F. A., Doussard-Roosevelt, J. A., & 
Porges, S. W. (1998). Infant temperament and cardiac vagal tone: Assessments at 
twelve weeks of age. Child Development, 69, 624-635. 
Huntley, E., & Lewin, D. S. (June, 2006). The effects of sleep restriction on performance testing 
on children with obstructive sleep apnea and a healthy comparison group. Poster 
presented at the 20th Annual Meeting of the Associated Professional Sleep Societies, 
Salt Lake City, UT. 
Irwin, M. (2008). Human psychoneuroimmunology: Twenty years of discovery. Brain, Behavior, 
and Immunity, 22, 129-139. 
Irwin, M. R., Valladares, E. M., Motivala, S., Thayer, J. F., & Ehlers, C. L. (2006). Association 
between nocturnal vagal tone and sleep depth, sleep quality, and fatigue in alcohol 
dependence. Psychosomatic  Medicine, 68, 159-166. 
Irwin, M., & Ziegler, M. (2005). Sleep deprivation potentiates activation of cardiovascular and 
catecholamine responses in abstinent alcoholics. Hypertension, 45, 252-257. 
 
35 
 
Jean-Louis, G., Kripke, D. F., Ancoli-Israel, S., Klauber, M., & Sepulveda, R. (2000). Sleep 
duration, illumination, and activity patterns in a population sample: Effects of gender and 
ethnicity. Biological Psychiatry, 47, 921-927. 
Johnsen, B. H., Laberg, J. C., Eid, J., & Hugdahl, K. (2002). Dichotic listening and sleep 
deprivation: Vigilance effects. Scandinavian Journal of Psychology, 43, 413-417. 
Katz, L. (2007). Domestic violence and vagal reactivity to peer provocation. Biological 
Psychology, 74, 154-164. 
Kennedy, A. E., Rubin, K. H., Hastings, P. D., & Maisel, B. (2004). Longitudinal relations 
between child vagal tone and parenting behavior: 2 to 4 years. Developmental 
Psychobiology, 45(1), 10-21. 
Kopasz, M., Loessl, B., Hornyak, M., Riemann, D., Nissen, C., Piosczyk, H., & Voderholzer, U. 
(2010). Sleep and memory in healthy children and adolescents-A critical review. Sleep 
Medicine Reviews. 
Kraemer, H., Kiernan, M., Essex, M., & Kupfer, D. (2008). How and why criteria defining 
moderators and mediators differ between the Baron & Kenny and MacArthur 
approaches. Health Psychology, 27, S101. 
Krueger, P. M., & Friedman, E. M. (2009). Sleep duration in the United States: A cross-sectional 
population-based study. American Journal of Epidemiology, doi: 10.1093/aje/kwp1023. 
Lane, R. D., Reiman, E. M., Ahern, G. L., & Thayer, J. F. (2001). Activity in medial prefrontal 
cortex correlates with vagal component of heart rate variability during emotion. Brain and 
Cognition, 47, 97-100. 
Lindberg, E., Janson, C., Gilason, T., Bjomsson, E., Hetta, J., & Boman, G. Sleep disturbances 
in a young adult population: Can gender differences be explained by differences in 
psychological status? Sleep, 20, 213-229. 
Linnemeyer, S., & Porges, S. (1986). Recognition memory and cardiac vagal tone in 6-month-
old infants. Infant Behavior and Development, 9, 43-56. 
 
36 
 
Logie, R., Zucco, G., & Baddeley, A. (1990). Interference with visual short-term memory. Acta 
Psychiatrica 75, 55-74. 
Mackinnon, D., Lockwood, C., & Williams, J. (2004). Confidence limits for indirect effect: 
Distribution of the product and resampling methods. Multivariate Behavioral Research, 
39, 99-128. 
Marcus, C. L. (1996). Evaluation of pulmonary function and polysomnography in  
 obese children and adolescents. Pediatric Pulmonology, 21, 176?183. 
Marcus, R. (1990). The Parent/Child Reunion Inventory: A measure of attachment for children 
beyond the infancy years. Psychological Reports, 67, 1329-1330. 
Meijer, A. M., Habekothe, H.T., & van den Wittenboer, G. L. H. (2000). Time in bed, quality of 
sleep and school functioning of children. Journal of Sleep Research, 9, 145-153. 
Mezzacappa, E. S., Kelsey, R., Katkin, E., & Sloan, R. (2001). Vagal rebound and recovery from 
psychological stress. Psychosomatic Medicine, 63, 650-657. 
Mezzacappa, E., Kindlon, D., Saul, J. P., & Earls, F. (1998). Executive and motivational control 
for performance task behavior, and autonomic heart-rate regulation in children: 
Physiologic validation of two-factor solution inhibitory control. Journal of Child 
Psychology and Psychiatry, 39, 525-531. 
Milan, S., Snow, S., & Belay, S. (2007). The context of preschool children's sleep: Racial/ethnic 
differences in sleep locations, routines, and concerns. Journal of Family Psychology, 21, 
20-28. 
Mitru, G., Millrood, D. L., & Mateika, J. H. (2002). The impact of sleep on learning and behavior 
in adolescents. Teachers College Record, 104, 704-726. 
Moore, M., & Meltzer, L. J. (2008). The sleep adolescent: Causes and consequences of sleep 
problems. Pediatric Respiratory Reviews, 9, 114-121. 
 
37 
 
Morgan, C. A., Aikins, D. E., Steffian, G., Coric, V., & Southwick, S. (2007). Relation between 
cardiac vagal tone and performance in male military personnel exposed to high stress: 
Three prospective studies. Psychophysiology, 44, 120-127. 
National Sleep Foundation. (2004). 2004 Sleep in America Poll. Retrieved November 12, 2004, 
from http://www.sleepfoundation.org/2004poll.cfm 
Ohayon, M., Carskadon, M., Guilleminault, C., & Vitiello, M. (2004). Meta-analysis of 
quantitative sleep parameters from childhood to old age in healthy individuals; 
Developing normative sleep values across the human lifespan. Sleep, 27 (7), 1255-
1273. 
Owens, J. A., Spirito, A., & McGuinn, M. (2000). The Children's Sleep Habits Questionnaire 
(CSHQ): Psychometric properties of a survey instrument for school-aged children. 
Sleep, 23, 1-9. 
Ozaki, N., Ono, Y., Ito, A., & Rosenthal, N. E. (1995). The prevalence of  
  
 seasonal difficulties among Japanese civil servants. American Journal of  
  
 Psychiatry, 152, 1225-1227. 
Pa Van Dongen, H., Rogers, N. L., & Dinges, D. F. (2003). Sleep debt:  
 Theoretical and empirical issues. Sleep and Biological Rhythms, 1, 5- 
 13. 
Pennington, B. F., & Ozonoff, S. (1996). Executive functions and developmental 
psychopathology. Journal of Child Psychology and Psychiatry and Allied Disciplines, 37, 
51-87. 
Pilcher, J., & Huffcutt, A. (1996). Effects of sleep deprivation on performance: A meta-analysis. 
Sleep, 19, 318-326. 
 
38 
 
Pine, D. S., Wasserman, G., Coplan, J., Fried, J., Sloan, R., Myers, M., et al. (1996). 
Serotonergic and cardiac correlates of aggression in children. Annuals of the New York 
Academy of Sciences, 796, 391-393. 
Porges, S. W. (Ed.). (1991). Vagal tone: An autonomic mediator of affect. New York: Cambridge 
University Press. 
Porges, S. W. (1995). Cardiac vagal tone: A physiological index of stress. Neuroscience and 
Biobehavioral Reviews, 19, 225-233. 
Porges, S. W. (1998). Love: An emergent property of the mammalian autonomic nervous 
system. Psychoneuroendocrinology, 23(8), 837-861. 
Porges, S. W. (2001). The Polyvagal theory: Phylogenetic substrates of a social nervous 
system. International Journal of Psychophysiology, 42, 123-146. 
Porges, S. W. (2007). The polyvagal perspective. Biological Psychology, 74(2), 116-143. 
Porges, S. W., Heilman, K. J., Bazhenova, O. V., Bal, E., Doussard-Roosevelt, J. A., & Koledin, 
M. (2007). Does motor activity during psychophysiological paradigms confound the 
qualification and interpretation of heart rate and heart rate variability measures in young 
children? Developmental Psychobiology, 49, 485-494. 
Pungello, E. P., Kupersmidt, J. B., & Burchinal, M. R. (1996). Environmental risk factors and 
children's achievement from middle childhood to early adolescence. Developmental 
Psychology, 32, 755-767. 
Randazzo, A., Muehlbach, M., Schweitzer, P., & Walsh, J. (1998). Cognitive function following 
acute sleep restriction in children ages 10-14. Sleep, 21, 861-868. 
Redline, S., Kump, K., Tishler, P. V., Browner, I., & Ferrette, V. (1994). Gender differences in 
sleep disordered breathing in a community-based sample. American Journal of 
Respiratory Critical Care Medicine, 149(3), 722-726. 
 
39 
 
Redline, S., Tishler, P. V., Schluchter, M., Aylor, J., Clark, K., & Graham, G. (1999). Risk factors 
for sleep-disordered breathing in children: Associations with obesity, race, and 
respiratory problems. American 
Journal of Respiratory Critical Care Medicine, 159, 1527-1532. 
Reilly, J. J., Armstrong, J., Dorosty, A. R., Emmett, P. M., Ness, A., Rogers, I., et  
 al. (2005). Early life risk factors for obesity in childhood: Cohort study.  
 British Medical Journal, 330, 1357. 
Richards, J. E. (1985). The development of sustained visual attention in infants from 14 to 26 
weeks of age. Psychophysiology, 22, 409-416. 
Richards, J. E. (1987). Infant visual sustained attention and respiratory sinus arrhythmia. Child 
Development, 58, 488-496. 
Richdale, A., Francis, A., Gavidia-Payne, S., & Cotton, S. (2000). Stress,  
 behaviour, and sleep problems in children with an intellectual disability.  
 Journal of Intellectual and Developmental Disability, 25, 147?161. 
Rizzo, P., Beelke, M., De Carli, F., Canovaro, P., Nobili, L, & Robert, A. et al.  
 (2003). Chronic vagus nerve stimulation improves alertness and reduces  
 rapid eye movement sleep in patients affected by refractory epilepsy.  
 Sleep, 26, 607-611. 
Rosen, C. L., Larkin, E. K., Kirchner, H. L., Emancipator, J. L., Bivins, S. F., Surovec, S. A., et 
al. (2003). Prevalence and risk factors for sleep-disordered breathing in 8- to 11-year-old 
children: association with race and prematurity. The Journal of Pediatrics, 142, 383-389. 
Rossy, L., & Thayer, J. (1998). Fitness and gender-related differences in heart period variability. 
Psychosomatic Medicine, 60, 773-781. 
Rottenberg, J. (2007). Cardiac vagal control in depression: A critical analysis. Biological 
Psychology, 74, 200-211. 
 
40 
 
Sadeh, A. (2007). Consequences of sleep loss or sleep disruption in children. Sleep Medicine 
Clinics, 2(3), 513-520. 
Sadeh, A. (2008). Commentary: Comparing actigraphy and parental report as measures of 
children's sleep. Journal of Pediatric Psychology, 33, 406-407. 
Sadeh, A., & Acebo, C. (2002). The role of actigraphy in sleep medicine. Sleep Medicine 
Reviews, 6, 113-124. 
Sadeh, A., Alster, J., Urbach, D., & Lavie, P. (1989). Actigraphically based automatic bedtime 
sleep-wake scoring: Validity and clinical applications. Journal of Ambulatory Monitoring, 
2, 209-216. 
Sadeh, A., Gruber, R., & Raviv, A. (2002). Sleep, neurobehavioral functioning, and behavior 
problems in school-age children. Child Development, 73, 405-417. 
Sadeh, A., Gruber, R., & Raviv, A. (2003). The effects of sleep restriction and extension on 
school-age children: what a difference an hour makes. Child Development, 74, 444-455. 
Sadeh, A., Acebo, C., Seifer, R., Aytur, S., & Carskadon, M. (1995). Activity-based assessment 
of sleep-wake patterns during the 1st year of life. Infant Behavior and Development, 18, 
329-337. 
Sadeh, A., Raviv, A., & Gruber, R. (2000). Sleep patterns and sleep disruptions in school-age 
children. Developmental Psychology, 36, 291-301. 
Sadeh, A., Sharkey, K. M., & Carskadon, M. A. (1994). Activity-based sleep-wake 
indentification: An empirical test of methodological issues. Sleep, 17, 201-207. 
Santucci, A. K., Silk, J. S., Shaw, D. S., Gentzler, A., Fox, N. A., & Kovacs, M. (2008). Vagal 
tone and temperament as predictors of emotion regulation strategies in young children. 
Developmental Psychobiology, 50, 205-216. 
Sekine, M., Yamagami, T., Handa, K., Saito, T., Nanri, S., et al. (2002). A dose-response 
relationship between short sleeping hours and childhood obesity: Results of the Toyama 
birth cohort study. Child: Care, Health and Development, 28, 163-170. 
 
41 
 
Sherwood, A., Allen, M., Obrist, P., & Langer, A. (1986). Evaluation of beta-adrenergic 
influences on cardiovascular and metabolic adjustments to physical and psychological 
stress. Psychophysiology, 23, 89-104. 
Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: New 
procedures and recommendations. Psychological Methods, 7, 422-445. 
Sirin, S. R. (2005). Socioeconomic Status and academic achievement: A meta- 
 analytic review of research. Review of Educational Research, 75, 417- 
 453. 
Spruyt, K., O'Brien, L. M., Cluydts, R., Verleye, G. B., & Ferri, R. (2005). Odds, prevalence and 
predictors of sleep problems in school-age normal children. Journal of Sleep Research, 
14, 163-176. 
Stamatakis, K. A., Kaplan, G. A., & Roberts, R. E. (2007). Short sleep duration across income, 
education, and race/ethnic groups: Population prevalence and growing disparities during 
34 years of follow-up. Annals of Epidemiology, 17, 948-955. 
Staton, L., El-Sheikh, M., & Buckhalt, J. (2009). Respiratory sinus arrhythmia and cognitive 
functioning in children. Developmental Psychobiology, 51, 249-258. 
Steenari, M., Vuontela, V., Paavonen, E. J., Carlson, S., Fjallberg, M., & Aronen, E. T. (2003). 
Working memory and sleep in 6- to 13-year-old school children. Journal of the American 
Academy of Child and Adolescent Psychiatry, 42(1), 85-92. 
Stepanski, E., Zayyad, A., Nigro, C., Lopata, M., & Basner, R. (1999). Sleep-disordered 
breathing in a predominatly African-American pediatric population. Journal of sleep 
Research, 8, 65-70. 
Sternberg, S. (1966). High-speed scanning in human memory. Science, 153,  
 652?654. 
Stores, G., & Wiggs, L. (Eds.). (2001). Sleep disturbance in children and  
 adolescents with disorders of development: Its significance and  
 
42 
 
 management. Oxford, United Kingdom: MacKeith. 
Suess, P. E., Newlin, D. B., & Porges, S. W. (1997). Motivation, sustained attention, and 
autonomic regulation in school-age boys exposed in utero to opiates and alcohol. 
Experimental and Clinical Psychopharmacology, 5(4), 375-387. 
Suess, P. E., Porges, S. W., & Plude, D. J. (1994). Cardiac vagal tone and sustained attention 
in school-age children. Psychophysiology, 31, 17-22. 
Sugimori, H., Yoshida, K., Izuno, T., Miyakawa, M., Suka, M., Sekine, M., et al.  
 (2004). Analysis of factors that influence body mass index from ages 3?6  
 years: A study based on the Toyama cohort study. Pediatrics  
 International, 46, 302?310. 
Thayer, J. F., & Brosschot, J. F. (2005). Psychosomatics and psychopathology: Looking up and 
down from the brain. Psychoneuroendocrinology, 30, 1050-1058. 
Thayer, J. F., & Lane, R. D. (2000). A model of neurovisceral integration in emotion regulation 
and dysregulation. Journal of Affective Disorders, 61, 201-216. 
Thayer, J. F., & Lane, R. D. (2007). The role of vagal function in the risk for cardiovascular 
disease and mortality. Biological Psychology, 74, 224-242. 
Thiedke, C. C. (2001). Sleep disorders and sleep problems in childhood. American Family 
Physician, 63, 277-284. 
Thompson, B. (2001). Significance, effect sizes, stepwise methods, and other issues: Strong 
arguments move the field. Journal of Experimental Education, 70, 80-93. 
Tikotzky, L., & Sadeh, A. (2001). Sleep patterns and sleep disruptions in kindergarten children. 
Journal of Clinical Child Psychology, 30, 579-589. 
Van Dongen, H. (2006). Shift work and inter-individual differences in sleep and sleepiness. 
Chronobiology International, 23(6), 1139-1147. 
 
43 
 
Van Leijenhorst, L., Crone, E., & Van der Molen, M. (2007). Developmental trends for object 
and spatial working memory: A psychophysiological analysis. Child Development, 78, 
987-1000. 
Von Kries, R., Toschke, A. M., Wurmser, H., Sauerwald, T., Koletzko, B. (2002). Reduced risk 
for overweight and obesity in 5- and 6-y old children by duration of sleep-a cross-
sectional study. International Journal of Obesity and Related Metabolic Disorders, 26, 
710-716. 
Walker, M. P., & Stickgold, R. (2006). Sleep, memory, and plasticity. Annual Review of 
Psychology, 57, 139-166. 
Weber, E. J., Van der Molen, M. W., & Molenaar, P. C. (1994). Heart rate and sustained 
attention during childhood: Age changes in anticipatory heart rate, primary bradycardia, 
and respiratory sinus arrhythmia. Psychophysiology, 31(2), 164-174. 
Wechsler, D. (2002). Wechsler Primary and Preschool Scale of Intelligence (3rd ed.). San 
Antonio: Harcourt Assessment. 
Williams, D. R. (1999). Race, socioeconomic status, and health: The added effects of racism 
and discrimination. Annals of the New York Academy of Sciences, 896, 173-188. 
Williams, P. G., Sears, L. L., & Allard, A. (2004). Sleep problems in children with  
 autism. Journal of Sleep Research, 13, 265-268. 
Wing, Y. K., Hui, S. H., Pak, W. M., Ho, C. K., Cheung, A., Li, A. M., et al. (2003).  
 A controlled study of sleep related disordered breathing in obese children.  
 Archives of Diseases in the Child, 88, 1043?1047. 
Wolfson, A. R., & Carskadon, M. A. (1998). Sleep schedules and daytime functioning in 
adolescents. Child Development, 69, 875-887. 
Wolfson, A., & Carskadon, M. (2003). Understanding adolescent's sleep patterns and school 
performance: A critical appraisal. Sleep Medicine Reviews, 7(6), 491-506. 
 
44 
 
Wolfson, A. R., & Carskadon, M. A. (2005). A survey of factors influencing high school start 
times. NASSP Bulletin, 89 (642), 47-66. 
Woodcock, R. W., McGrew, K. S., & Mather, M. (2001). Woodcock-Johnson III tests of 
cognitive ability. NY: Riverside Publishing. 
Zhong, X., Hilton, H. J., Gates, G. J., Jelic, S., Stern, Y., Bartels, M. N., et al. (2005). Acute 
sleep deprivation is associated with increased sympathetic and decreased 
parasympathetic cardiovascular modulation in normal humans. Journal of Applied 
Physiology, 98, 2024-2032. 
 
 
 
 
 
 
  
 
   
45 
 
                                            Abstract Manuscript 1 
The role of children?s vagal tone and vagal reactivity (indexed by respiratory sinus 
arrhythmia; RSA) as a possible moderator of the associations between both actigraphy and 
subjective sleep disruptions and multiple aspects of children?s cognitive performance was 
examined. Children (N = 166; M age = 8.70 years old) wore actigraphs for seven consecutive 
nights and completed a subjective measure of sleep via interview. To evaluate cognitive 
functioning, children completed several tests in the the Woodcock-Johnson III Tests of Cognitive 
Abilities (WJ III; Woodcock, McGrew, & Mather, 2001) and a computerized reaction time task 
(Psych/Lab; Abrams, 2004) in the laboratory. Vagal indices were assessed in the lab; RSA was 
obtained during a three minute baseline and RSA reactivity (RSA-R) was derived as the 
difference between RSA during a mild cognitive stressor (i.e., star tracing task) and initial RSA. 
Results suggest that children who exhibited RSA withdrawal to the laboratory stressor (lower 
RSA during stressor than baseline), and to a lesser extent those who had higher levels of 
baseline RSA, were protected from the negative effects of sleep disruptions on various 
executive functioning tests. Findings demonstrate the importance of examining individual 
difference variables, such as physiological regulation, when considering the effects of sleep 
disruptions on children?s cognitive functioning.  
 
 
 
 
 
 
   
46 
 
Sleep and cognition: The moderating role of RSA and RSA reactivity 
Non-clinical sleep disruptions, such as reductions in sleep quality and duration, have 
become causes of concern due to their high prevalence (25 to 50%) in typically developing 
children (Mindell, Owens, & Carskadon, 1999; Stein, Mendelsohn, Obermeyer, Amromin, & 
Benca, 2001). Recommendations of pediatric specialists (Mindell & Owens, 2009) suggest that 
children in middle childhood get approximately 10 hours of sleep per night, and recent  studies 
indicate they are obtaining much less than the recommended amount (e.g., Buckhalt, El-Sheikh, 
& Keller, 2007; National Sleep Foundation, 2004; Spilsbury et al., 2005). Reductions in sleep as 
small as one hour per night and mild disruptions in sleep quality are linked to poorer cognitive 
performance in children, including slower reaction times (Buckhalt et al., 2007) and reduced 
attention (Sadeh, Gruber, & Raviv, 2003). These findings, in conjunction with the high rate of 
chronic sleep reduction in American children, suggest that many American children may be at 
risk for cognitive and academic problems associated with sleep problems. Further, although 
sleep disruptions do not affect all persons in the same way (Buckhalt et al., 2007), little is known 
about individual difference variables that may act as protective or vulnerability factors in these 
relations. The goal of this study is to explicate the possible moderating role of individual 
differences at the physiological level, namely vagal indices (i.e., resting respiratory sinus 
arrhythmia [RSA] and changes in RSA from baseline to challenge conditions [RSA reactivity]).  
This study focuses on relations between non-clinical sleep disruptions, characterized by 
reduced sleep duration or increased sleep activity and awakenings (indicative of poor quality) 
and non-clinical problems in children?s cognitive performance (e.g., slower reaction time 
responses, inattention, short term memory lapses), unless otherwise noted. Sleep and cognitive 
functioning problems are defined on a continuum and in relation to other children within this 
community sample.   
   
47 
 
 In recent years, researchers have begun to ask whether children?s sleep disruptions 
show similar effects on cognitive functioning as those that have been extensively documented 
with adults. Now more than 30 studies link sleep parameters of typically developing children and 
aspects of cognitive functioning and or achievement (Buckhalt, Wolfson, & El-Sheikh, 2009b). 
Reviews of the science indicate that multiple sleep parameters, including reduced sleep 
duration, poor sleep quality and daytime sleepiness are related to an array of cognitive 
functioning outcomes, including difficulties in executive functioning and slower reaction times 
(Buckhalt et al., 2009b; Curcio, Ferrara, & De Gennaro, 2006; Sadeh, 2007). For example, 
insufficient or highly disrupted sleep in children is known to impair aspects of executive 
functioning, such as working memory and the ability to sustain attention (Blunden, Lushington, 
Lorenzen, Martin, & Kennedy, 2005; Sadeh et al., 2003). Low sleep efficiency (a marker of 
sleep quality) and longer sleep latency (time it takes to fall asleep) in typically developing 
elementary school-aged children correlate with higher error rates on both visual and auditory 
tasks (Steenari, et al., 2003). Furthermore, experimental studies indicate that acute and 
prolonged sleep restriction in children are related to significant declines in reaction time (Sadeh 
et al., 2003) and deficits in attention (Fallone, Seifer, Acebo, & Carskadon, 2000). 
Although sleep disruptions are associated with poorer cognitive performance, not all 
children or adults who experience sleep problems are at an equal risk for reduced cognitive 
outcomes (Buckhalt et al., 2007; Buckhalt, El-Sheikh, Keller, & Kelly, 2009a; Dorrian, Rogers, & 
Dinges, 2005; Van Dongen, 2006). This heterogeneity in outcomes highlights the importance of 
moderators of effects in the association between sleep and cognitive performance. For 
example, Van Dongen (2006) concluded that basal level of alertness, the amount of sleep 
needed to optimally function, and the timing of the circadian process were all individual 
differences that may influence the extent of reduced cognitive performance otherwise 
associated with sleep deprivation in adults. For children, recent studies examined social class 
(e.g., father education level) and minority status (Buckhalt et al., 2007; Buckhalt et al., 2009a) or 
   
48 
 
daytime sleepiness (Geiger, Achermann, & Jenni, 2010) as possible individual differences 
variables that may explain the conditions to which poor sleep is most highly associated with 
poor cognitive functioning. For example, Buckhalt et al. (2007, 2009a) found ethnic minority 
children and those from more disadvantaged socioeconomic situations to be more vulnerable 
than their counterparts to the effects of sleep disruptions on cognitive functioning. Differences in 
psychophysiology are recognized as important individual differences that can potentially 
influence child adjustment (Cicchetti & Dawson, 2002). For example, El-Sheikh and colleagues 
(2007) found that individual differences at the physiological level moderated the effects of 
disrupted or shortened sleep on children?s internalizing and externalizing problems. Yet, 
physiological indices as moderators of the association between sleep disruptions and cognitive 
functioning have not been investigated.  
Vagal Tone and RSA  
The parasympathetic nervous system (PNS) is the down-regulating component of the 
autonomic nervous system (ANS), which works to maintain homeostasis under normal 
conditions (Porges, 1991). An indicator of the PNS?s influence on the heart is termed vagal tone, 
which refers to the role of the vagus nerve, the 10th cranial nerve that provides an inhibitory 
influence on the heart (i.e., decreases in heart rate). Vagal tone in this study is indexed by 
respiratory sinus arrhythmia (RSA), which represents the rhythmic changes in heart period 
related to the respiratory cycle (Grossman, Karemaker, & Wieling, 1991). Specifically, heart rate 
decreases during exhalation and increases during inhalation. RSA is an appropriate index of 
vagal activity for research with normally developing children (Grossman & Taylor, 2007) and can 
be obtained non-invasively. The vagus nerve has been demonstrated to influence RSA through 
its significant attenuation or elimination by pharmacological blockade (Bernston, et al., 1997; 
Hayano et al., 1991; Sherwood, Allen, Obrist, & Langer, 1986).  
The Polyvagal Theory?s (Porges, 2007) propositions connect the ANS with emotion 
regulation, cognition, and behavior. Specifically, the Polyvagal Theory proposes that when a 
   
49 
 
mammal is confronted with a challenge, it automatically responds by disengaging the vagal 
brake (i.e., vagal suppression/withdrawal) which inhibits parasympathetic influence (i.e., 
increased heart rate). This response allows the individual to focus attention on the environment, 
obtain information, and/or engage social strategies to reduce the challenge/stress. Once the 
challenge has subsided, the PNS influence can reengage to reduce arousal. The ability to 
quickly engage and disengage as the environmental conditions become more challenging and 
to be able respond to metabolic demands, including increased attention and information 
processing (Suess, Porges, & Plude, 1994), are adaptive processes vital for behavior. We utilize 
the Polyvagal Theory as a guiding framework for our assessment of vagal functioning as a 
moderator in the link between disrupted sleep in children and multiple facets of their cognitive 
functioning.   
  Baseline vagal tone (i.e., RSA) indexes the activity of the PNS at rest, and is posited to 
reflect the ability to sustain attention and maintain homeostasis under normal conditions 
(Porges, 1991). Empirical research supporting the Polyvagal Theory has indicated that higher 
levels of baseline vagal tone are generally positively associated with more optimal adjustment 
outcomes (Rottenberg, 2007). That is, higher vagal tone maintains a low heart rate under 
normal conditions (Porges, 1991), which reduces the wear and tear on the heart and leads to 
more optimal health and adaptation.  
The reduction or increase in heart rate by vagal pathways to promote regulatory and 
goal-directed behavior is referred to as vagal reactivity (Porges, 1995, 2007; Thayer & Lane, 
2000). More specifically, RSA typically decreases during challenges that require focused 
attention (Porges, 2007). When a challenge is encountered, the PNS can respond in one of two 
ways: it can be (1) withdrawn or suppressed, frequently referred to as vagal withdrawal, which 
leads to an increase in heart rate; or (2) activated or augmented, which leads to a decrease in 
heart rate. Vagal withdrawal is thought to be the most optimal response when confronted with a 
challenge because it mobilizes attentional resources when the environment demands a 
   
50 
 
response (Huffman et al., 1998). Vagal withdrawal likely facilitates a shift in metabolic resources 
that support information processing (Bornstein & Suess, 2000). However, there is a small 
amount of literature, specifically with infants, indicating that  vagal augmentation in response to 
a surprising stimulus (i.e., jack-in-the-box) leads to more attentiveness than vagal suppression 
(DiPietro, Porges, Uhly, 1992); very high levels of vagal withdrawal have also been associated 
with adjustment problems (Beauchaine, 2001). Research on vagal reactivity (i.e., RSA-R) and 
cognitive functioning in children is scant, and further clarification of this association is warranted. 
Although vagal tone and vagal reactivity are often moderately correlated, each vagal marker can 
have differential relations to child outcomes (Calkins, 1997), which supports the assessment of 
both forms of vagal activity. 
Sleep and RSA Functioning 
 Few investigations have examined links between sleep and RSA functioning in children. 
Of those that have, vagal augmentation to a laboratory challenge has been shown to be a risk 
factor, while vagal withdrawal to a stressor functioned as a protective factor. For example, 
increases in sleep problems are linked to higher levels of vagal augmentation (less optimal 
response) during a reaction time task (El-Sheikh & Buckhalt, 2005). In a previous paper with the 
current sample, increased vagal withdrawal to an inter-adult argument served as a protective 
factor against externalizing and depression symptoms in the context of increased sleep 
disruptions (El-Sheikh et al., 2007). Conversely, children who responded to the interadult 
argument with reduced vagal withdrawal or increased vagal augmentation were at greater risk 
for externalizing and depression symptoms in the context of sleep problems (El-Sheikh et al., 
2007). 
RSA and Cognition 
 More optimal vagal functioning (e.g., higher RSA and increased RSA withdrawal) is 
related to a greater ability to engage and disengage with the environment as needed (Porges, 
2001; Thayer & Lane, 2000), which is necessary for optimal cognitive functioning. The majority 
   
51 
 
of the studies that have examined the associations between vagal functioning and cognitive 
performance have been conducted with infants or adults; however, a few small studies have 
addressed this question with school- aged children. Specifically, higher baseline RSA was 
significantly correlated with greater sustained attention during a continuous performance task in 
a sample (N = 32) of fourth and fifth graders (Suess et al., 1994). Further, 10-year-olds (N = 42) 
with greater vagal variability (i.e., wider range of responses; higher baseline and greater vagal 
withdrawal) exhibited more competence in executive control than their counterparts with lower 
vagal variability (Mezzecappa, Kindlon, Saul, & Earls, 1998).  Staton, El-Sheikh, and Buckhalt 
(2009) found that, higher baseline RSA was associated with better Woodcock-Johnson III 
(WJIII; Woodcock et al., 2001) scores of overall intellectual ability, processing speed, working 
memory, and cognitive efficiency with a sample of 6- to 13-year olds (N = 41; M age = 10.06; 
SD = 1.74). However in the latter study, RSA reactivity (derived during a memory task) was not 
predictive of cognitive functioning as measured with the WJ III (Woodcock et al., 2001) or the 
memory task (a variation of the Sternberg Memory Task; Sternberg, 1966). 
New Directions 
 In the present study, we advance prior work conducted in the sleep disruptions-cognitive 
performance literature by examining the moderating role of two physiological individual 
difference variables: vagal tone (RSA) and vagal reactivity (RSA-R). Sleep problems can disrupt 
neurobiological regulatory systems, especially those governed largely by the prefrontal cortex 
(PFC; Archbold, Giordani, Ruzicka, & Chervin, 2004; Dahl, 1996; Horne, 1993; Sadeh et al., 
2003), such as executive functions and PNS regulatory functions (El-Sheikh & Buckhalt, 2005; 
El-Sheikh et al., 2007; Irwin, 2008). Although sleep may influence vagal functioning, the neural 
networks impacted by sleep are not the only influence on vagal functioning. Therefore, some 
variability in vagal functioning exists independent of sleep. In addition, baseline RSA and to a 
lesser extent RSA reactivity (RSA-R) are considered stable individual difference variables for 
   
52 
 
school-aged children (El-Sheikh, 2005; Salomon, 2005), which supports the examination of RSA 
indices as moderators of effects. 
It is our hypothesis that vagal functioning may strengthen or weaken the influence of 
sleep problems on children?s cognitive functioning (see El-Sheikh et al., 2007). Dahl (1996) 
posits that sleep is affected not only by the levels of arousal and vigilance at night, but also by 
the ability to regulate arousal and emotions during waking hours. Furthermore, the Polyvagal 
Theory (Porges, 2007) states that vagal functioning facilitates efficient regulation of attention in 
response to environmental demands. Thus, both sleep and vagal functioning can assist or 
obstruct the regulation of attention.  
 We hypothesize that both RSA and RSA-R will function as moderators of effects in the 
association between children?s sleep disruptions and their cognitive performance. Based on the 
extant literature, we expect higher levels of RSA and RSA withdrawal to serve as protective 
factors in the context of sleep disruptions, as it was shown to do among other areas of 
adjustment (e.g., externalizing and internalizing problems; El-Sheikh et al., 2007), thereby 
attenuating the connection between sleep problems and reductions in cognitive functioning. 
Conversely, and consistent with a conceptualization of protection and vulnerability along a 
continuum (Brown & Barlow, 2005), low RSA and RSA augmentation are expected to function 
as vulnerability factors and exacerbate the associations between sleep disruptions and poor 
cognitive performance. In the context of this emerging literature, no specific hypotheses are 
advanced regarding the moderating role of vagal indices in relation to specific forms of sleep 
disruptions (e.g., duration, quality). However, based on literature supportive of vagal indices as 
markers of attention (Porges, 2007), and the role of the prefrontal cortex in executive functioning 
(Dahl, 1996), we expect that those tests tapping into higher levels of executive functioning (e.g., 
working memory) will be the best candidates for the moderation hypothesis. Of note, all of our 
cognitive outcomes tested in this study require focused attention to perform well, and all of the 
cognitive tests tap varying levels of executive functioning.  
   
53 
 
 In the present study, we examine both actigraphy and children?s self-reported sleep 
(Sadeh, 2008) and assess two important sleep parameters (duration, quality). Both actigraphy 
(Buckhalt et al., 2007; Buckhalt et al., 2009a; Sadeh, Raviv, & Gruber, 2000) and subjective 
(Buckhalt et al., 2007; Buckhalt et al., 2009a; Wolfson & Carskadon, 2003) sleep problems have 
been linked to reduced cognitive performance and thus their assessment in the context of vagal 
functioning was deemed important. Similarly, for a more comprehensive explication of children?s 
cognitive functioning in relation to vagal functioning and sleep, we administered an extensive 
and well-validated battery of cognitive abilities and a well-established reaction time task. The 
examination of multiple sleep parameters (e.g., actigraphy-based duration and quality, and 
subjective reports of sleep quality and daytime sleepiness) as they interact with two vagal 
functioning indices (RSA, RSA-R) is imperative for a more nuanced explication of physiological 
and biological domains linked with various cognitive performance parameters. The proposed 
conceptual model is depicted in Figure 1.  
Method 
Participants        
        The sample consisted of 166 children (92 girls) and their mothers and fathers from the 
Southeastern United States. To recruit participants, an agreement with three local school 
systems was obtained. The school systems provided researchers with names, phone numbers, 
and addresses of children in the third grade. Researchers then contacted parents via letter and 
phone. Of the 254 families contacted that were eligible to participate, 66% participated in the 
study, 28% declined participation, and 6% were interested but said they were too busy at the 
time. Children were unable to participate in the study if they had a diagnosed chronic illness, 
mental retardation, attention-deficit/hyperactivity disorder (ADHD), or a diagnosed sleep 
problem. If children were suffering from an acute illness, such as the flu, their participation was 
postponed until they were well. If children became sick during the week of actigraphy and took 
   
54 
 
medications that influence sleep (e.g., Benadryl), the nights they took medications were 
excluded from actigraphy analyses.  
Participants were eight-years-old (M = 8.70; SD =.30), in the third grade, and from two-
parent homes when they were contacted to participate in the study. The majority of children 
(77%) lived with both biological parents; 21% lived with their biological mother and a step-father; 
2% lived with their biological father and a step-mother. Parents of these children had been living 
together for an average of 10.2 years (SD = 5.3). According to parental reports, the ethnic 
composition of the sample was similar to the community from which they were drawn. 
Specifically, 69% were European-American (EA) and 31% were African-American (AA). We 
used oversampling procedures to obtain AA and EA participants across all socio-economic 
backgrounds. Based on the Hollingshead (1975) criteria, 30% of the families were classified in 
level 4 or 5 (e.g., professional), 44% in level 3 (e.g., skilled labor), and 26% in level 1 or 2 (e.g., 
unskilled or semi-skilled laborers). Mothers reported the annual family income at the following 
percentages: 4% less than $10,000; 7% between $10,000 and $20,000; 25% between $20,000 
and $35,000; 28% between $35,000 and $50,000; 25% between $50,000 and $75,000 and 11% 
more than $75,000. Parental education attainment was as follows for fathers and mothers, 
respectively: 12% and 8% partial high-school or less; 47% and 36% high school diploma; 30% 
and 43% partial college or specialized training; 9% and 12% bachelor?s degree; and 3% and 2% 
graduate degree.                    
Procedures 
 This study is part of a larger longitudinal study examining the role of the family 
environment (e.g., marital conflict, parent-child attachment) in relation to children?s adjustment, 
sleep, and health. Only the procedures pertinent to the current study are detailed. The study 
was approved by the institution?s Internal Review Board (IRB), and parents provided consent 
and children provided assent for participation. Families received monetary compensation for 
participating in the study.    
   
55 
 
 To reduce confounds associated with variability in sleep schedules, all objective sleep 
data (actigraphy) were obtained during the regular school year (i.e., not during breaks or 
summer months). Actigraphs, watch-like devices that measure movement, were used to obtain 
objective measures of sleep duration and quality. The watches were delivered by researchers to 
the participants? homes. Parents were instructed to place the actigraph on the child?s non-
dominant wrist just prior to bedtime and to remove it upon waking. Actigraphy took place over 
seven consecutive nights, which is considered to be more than acceptable for obtaining reliable 
measurements of typical sleep patterns (Acebo et al., 1999). During the week of actigraphy 
assessment researchers called parents daily to obtain information regarding children?s sleep 
and wake times, which was used to validate actigraphic recordings (Acebo & Carskadon, 2001).  
 Typically, on the day after the actigraph data collection children and their parents came 
to our research laboratory located on a university campus. Children were administered the brief 
version of the WJ III. Immediately following the WJIII, children completed a 15-minute reaction 
time task on the computer. After a short break, children were accompanied into another lab 
room for a physiological assessment session.  
 To reduce any anxiety associated with attaching electrodes to obtain RSA data, 
researchers conversed with the child and a parent (typically the mother) while the electrodes 
and a pneumatic bellows belt were attached. This process took approximately ten minutes, 
which allowed a sufficient amount of time for the child to relax prior to the collection of 
physiological data. The parent and researcher then left the room and the child was given two 
additional minutes to acclimate to the setting without others present before data collection 
began (the child was asked to relax while sitting down quietly). Immediately following, a three 
minute baseline (resting level) was obtained. To accurately assess RSA activity, the duration of 
the conditions must be sufficient to characterize the signal and three minutes are considered 
more than adequate for assessment of RSA (Bar-Haim, Marshall, & Fox, 2000). To assess 
RSA-R, children completed a mildly stressing problem-solving task, namely the star-tracing task 
   
56 
 
(Matthews, Woodall, & Stoney, 1990). This task requires the child to trace the design of a star 
on a piece of paper by only seeing the star in a mirror (Lafayette Instrument Company, 
Lafayette, Indiana, United States). In addition to the difficulty associated with the reverse 
properties that the mirror adds, children were only given three minutes to complete the task. 
This task has been used in numerous investigations to evoke ANS reactivity in children (El-
Sheikh, et al., 2009; Matthews, Rakaczky, Stoney, & Manuck, 1987; Matthews, Woodall, & 
Stoney, 1990). Following the physiological session, children were given a break and a snack 
and were then interviewed by a research assistant regarding their sleep habits.  
Measures 
 Children?s Objective Sleep. The actigraph watch used was the Motionlogger Octagonal 
Basic (Ambulatory Monitoring, Inc., Ardsley, NY). This device is small (36 mm X 38 mm X 13 
mm) and lightweight (45 g) and was worn like a wristwatch on the non-dominant arm. Movement 
during sleep was continuously recorded and raw data obtained from the actigraph were 
processed with the Octagonal Motionlogger Interface with ACTme Software, and the Analysis 
Software Package (ActionW2 or AW2). Using the Sadeh algorithm, which has established 
validity with children (Sadeh, Sharkey, & Carskadon, 1994), one-minute epochs using zero 
crossing mode were scored as either sleep or wake. Guidelines for determining sleep onset 
were developed at the E.P. Bradley Sleep Laboratory at Brown University (Acebo & Carskadon, 
2001). Good reliability has been established with this actigraph and the Sadeh algorithm (Acebo 
et al., 1999). 
 Sleep variables derived via actigraphy included: (a) Sleep Minutes - total minutes scored 
as sleep during the Sleep Period (time in bed), providing a measure of sleep duration; (b) Sleep 
Efficiency - percentage of minutes in bed that are scored as sleep, providing a measure of sleep 
quality; and (c) Long-Wake Episodes- number of wake episodes during the night that last at 
least 5 minutes, providing a second measurement of sleep quality.    
   
57 
 
 Stability of sleep variables was examined through intraclass correlations. Analyses 
indicated high levels of stability for sleep duration and sleep quality (.78 for Sleep Minutes, .86 
for Sleep Efficiency, .85 Long Wake Episodes) over the week of actigraphy. Thus, the reliability 
for these measures exceeded recommended minimum levels (.70; Acebo et al., 1999).  
 The majority of children (75%) had all seven nights of actigraphic data. However, due to 
a malfunctioning actigraph, forgetting to wear the actigraph, or exclusion of nights during which 
children took medication, 13% of children had six nights of data, 7% had five nights of data, 2% 
had four nights of data, 1% had either two or three nights of data, and 2% of children had no 
actigraphy data. This proportion of valid data for children is comparable to similar studies and is 
considered very good (Acebo et al., 1999). Sleep parameters were based on averaged data 
across all available nights of actigraphic assessments.                               
Children?s Subjective Sleep. Researchers interviewed children during the lab session 
to obtain children?s perceptions of their sleep quality and daytime sleepiness. Specifically, 
children completed the Sleep Habits Survey (SHS; Wolfson & Carskadon, 1998), which inquires 
about sleep during the past two weeks. This questionnaire is commonly used with children 
(Acebo & Carskadon, 2002), and has good reliability and validity (Carskadon, Seifer, & Acebo, 
1991; El-Sheikh & Buckhalt, 2005; Wolfson et al., 2003). Two subscales were used in analyses: 
(1) the Sleep/Wake Problems Scale, which is composed of 10 items that assess the frequency 
of sleep problems (e.g., oversleeping and staying up late at night); and (2) the Sleepiness scale 
(comprised of 10 items), which measures the extent to which children struggle to stay awake 
during the day. One item (i.e., sleepiness while driving) on the Sleepiness scale was deleted 
due to the child?s age. Items were rated on a 5-point Likert scale, and sums for each subscale 
were computed. Higher scores are indicative of more sleepiness and sleep-related problems. In 
the current sample, internal reliability coefficients were acceptable for both scales: .67 for 
Sleep/Wake Problems and .70 for Sleepiness.                                                                                                        
   
58 
 
RSA and RSA-R. To obtain heart rate, electrodes were placed axially on the left and 
right rib cage, approximately 10 to 15 cm below the armpits. A pneumatic bellows was placed 
around the chest and held in place using a metal beaded chain. Physiological acquisition 
equipment and software were from James Long Company. The bioamplifier was set for band-
pass filtering with half power cut-off frequencies of 0.1 and 1000 hertz, and the signal was 
amplified with a gain of 500 for electrocardiography (ECG). The pneumatic bellows, used to 
assess respiration, was attached to a pressure transducer with a band-pass of direct current to 
4000 hertz, to minimize phase or time shifts in the measurement. Data were digitized at a 
sampling rate of 1000 Hz with a 12-bit analog-to-digital board in a laboratory computer.                                                   
An automated algorithm was used to identify the R-waves in the ECG data. In the rare 
case it was needed, manual correction of missed or misidentified R-waves was conducted using 
an interactive graphical program. R-wave times were converted to inter-beat intervals (IBIs) and 
resampled into equal time intervals of 125 ms. Prorated IBI values were used for analysis of the 
mean and variance of heart period as well as the processing of heart period variability due to 
RSA. The mean RSA was calculated over the entire 3-min of baseline and the 3-min of the star 
tracing challenge task. To derive RSA-R in response to the star tracing task, a change score 
between RSA during the star tracing task and RSA at baseline (i.e., task minus baseline) was 
computed. Lower scores during the star tracing as compared to the baseline indicate RSA 
suppression to the challenge task (i.e., negative score for RSA-R), whereas RSA augmentation 
was indicated by a higher task score as compared to baseline (i.e., positive score for RSA-R). 
RSA was controlled in all assessments of RSA-R. 
  The peak-valley method was used to compute RSA; RSA measurement units reported 
in this paper are in seconds. The peak-valley method is one of several acceptable procedures 
for quantifying RSA (Berntson et al., 1997) and this procedure highly correlates with spectrally 
derived measures of RSA (Galles, Miller, Cohn, & Fox, 2002). To identify inspiration and 
expiration onset times and relative tidal volume, the respiration signal was used. RSA was 
   
59 
 
calculated as the difference in IBI between inspiration and expiration onset for consecutive 
respiratory cycles. Specifically, RSA was computed twice per breath, once from inspiration to 
expiration and again from expiration to inspiration, and then averaged across respiratory cycles 
to ensure that any underlying trends in heart period had little impact on the computed mean 
RSA.  
 Cognitive Functioning. The Woodcock-Johnson III (WJIII; Woodcock et al., 2001) is a 
well-normed measurement of general intellectual ability (g), crystallized intelligence (Gc), fluid 
reasoning (Gf), and processing speed (Gs). General intellectual ability refers to non-specific 
abilities that affect the performance of most tasks, whereas Gc is thought to reflect skills 
acquired through knowledge and experience (e.g., verbal ability). Fluid reasoning is the capacity 
to think logically and solve novel problems, and processing speed involves the ability to 
automatically and fluently perform tasks. In the current study, the following tests were used: 
Concept Formation (CF); Numbers Reversed (NR); Auditory Working Memory (AWM); and 
Decision Speed (DS). CF is a novel controlled-learning task that requires categorical reasoning 
based on principles of inductive logic. It is a test of fluid reasoning and also measures an aspect 
of executive functioning, flexibility in shifting mental set. NR is a test of short-term memory span 
and also engages working memory, as the numbers must be held in working memory while 
performing a mental operation (reversing the sequence). AWM also measures short-term 
auditory memory and working memory while also requiring the use of divided attention. A series 
of numbers mixed with words is presented, and one must recall the series reporting the words in 
correct order first, then the numbers. DS measures visual processing speed by requiring that 
two conceptually similar items are selected on multiple trials presenting rows of alternatives. 
These four tests were chosen to obtain scores reflecting simple processing speed, attention and 
varying levels of executive functioning, which allowed for a comprehensive assessment of 
cognitive functioning domains that may be impacted by disruptions in sleep.   
   
60 
 
Reaction Time. Speed and variability of response speed across a number of trials was 
measured via a computerized reaction time task based on the classic Donders task adapted by 
Sternberg (1969) (Psych/Lab; Abrams, 2004). Specifically, the child was seated at a computer 
and a researcher instructed the child to focus on a ?1? in the center of the computer screen. The 
child placed the right index finger on the ?/? key, and the left index finger was placed on the ?Z? 
key. The child was told to follow the instructions presented on the screen, which were also read 
to the child by the researcher. If the child responded too early, too late, not at all, or if a key was 
pressed when one was not required, a short tone was sounded.  
To assess simple reaction time (Donders Type A), the child was told to press the ?/? key 
as quickly as possible when the ?x? appears in the box on the right. For the assessment of 
choice reaction time (Donders Type C), two boxes were presented and either could have the ?x? 
appear, but children were told to only respond (by pressing ?/?) when the ?x? appears in the right 
box. The difficulty is increased in this task due to the monitoring of two boxes and having to 
inhibit a response with the left finger. In a third condition, choice reaction time 2 (Donders Type 
B), two boxes were presented and the ?x? could appear in either box. Children were instructed to 
press the ?/? key with the right finger when the ?x? appears in the right box or press the ?z? key 
with the left finger if the ?x? appears in the left box. Conditions were presented in a fixed order 
with the following parameters: 10 practice trials per condition; 20 test trials per condition; a 500 
ms inter-trial interval; 100 ms minimum allowed RT; and 1,500 ms maximum allowed RT.  Trials 
were repeated if responses were faster or slower than the criteria. Test trial averages and 
standard deviations were computed for each condition and were derived for analyses. The 
mean scores represent the average response time, whereas the standard deviation scores 
represent vigilance to the task. Examination of children?s responses during the three conditions 
(types A, B, and C) indicated that both the means (rs = .66 ? .76, ps < .001), and SDs (rs = .37 
? .42, ps < .001) were significantly correlated. Thus, to reduce the number of analyses, one 
averaged mean and one SD were derived and used in all subsequent analyses.   
   
61 
 
Body Mass Index. Children?s body mass index was calculated using height and weight 
measurements obtained by a researcher during the laboratory session. A Sunbeam scale was 
used to obtain the child?s weight. A standard formula was used to compute BMI [703 X (weight ? 
height2); http://www.cdc.gov]. Based on the Centers for Disease Control criteria, 1% of the 
sample was classified as underweight, 61% were of normal weight, 23% were overweight, and 
15% were considered to be obese.  
Demographic Information. Children?s gender, ethnicity, age and parental education 
and occupation were obtained by phone from a parent prior to participation. Parental education 
and occupation were used to derive a socioeconomic status index via the Hollingshead Index 
(Hollingshead, 1975).  
Season of Assessment. The change of seasons brings about alterations in the length 
of daylight during a given 24-hour period. Considering that light patterns impact the release of 
melatonin, it is important to consider the season of the year when examining children?s sleep 
parameters (Ozaki, Ono, Ito, & Rosenthal, 1995). Season was dichotomized into two; winter and 
all other seasons. 
Results 
Preliminary Analyses  
 Table 1 displays means and standard deviations of study variables. Descriptive analyses 
indicated that on average, children had a fairly low sleep efficiency of 87% (? 90% is considered 
optimal; Sadeh, Raviv, & Gruber, 2000). Similarly, children's average sleep duration was 7.42 
hours per night, well below the recommended average for children of this age (Mindell & 
Owens, 2009). The majority of the children (69%) exhibited RSA withdrawal in response to the 
star tracing task; 13% showed no change and 18% exhibited RSA augmentation to the task. Of 
the WJ III tests examined, children in this sample scored the lowest on concept formation and 
the highest on auditory working memory, but scores on all tests were in the normative range.      
   
62 
 
Additional analyses explored relations between the main study variables (sleep, vagal indices, 
cognitive functioning and reaction time) and several possible confounding variables: BMI, child 
age, gender, ethnicity, socioeconomic status (SES), season of actigraph assessment, and 
pubertal status. Higher baseline RSA was related to lower BMI. In comparison to younger 
children, older children showed better cognitive performance on all WJ III tests (rs = .16 - .21, ps 
< .05), with the exception of decision speed. Further, in comparison to boys, girls had higher 
sleep efficiency, fewer long wake episodes, more sleep minutes, and less self-reported 
sleepiness (rs = .15 - .24, ps < .05). Boys had significantly lower decision speed scores (r =       
-.17, p < .05) and a faster mean reaction time (r = -.31, p < .01) than girls. African-Americans 
were more likely to be from lower SES backgrounds (r = -.25, p < .01), to have less long wake 
episodes (r = - .19, p < .05), and to report greater sleepiness (r = .16, p < .05) and sleep/wake 
problems (r = .25, p < .01) than European-Americans. African-Americans also scored lower on 
the WJ III concept formation (r = -.19, p < .05) and numbers reversed tests (r = -.15, p < .05) 
than their counterparts. The season of actigraph assessment was related to sleep duration (r = 
.17, p < .05). Pubertal status was not found to be related to any of the study variables and was 
therefore not considered in any further analyses.  
Table 2 denotes relations among study variables. As expected, actigraphy-based sleep 
variables were strongly related to each other; so were self-reported measures of sleep. 
Similarly, cognitive functioning variables (WJ III and reaction time measures) were highly 
correlated. Neither sleep nor cognitive variables were associated with either RSA or RSA-R.  
Analysis Plan 
 A series of path models were fit using AMOS 17.0 (Arbuckle, 2007). The use of path 
analysis in AMOS improves the power (i.e., uses the full sample size) by employing maximum 
likelihood estimation to deal with missing data, unlike multiple regression. Listwise deletion is 
typically used in multiple regression software such as SPSS and it relies on the assumption that 
   
63 
 
the data are missing completely at random (MCAR; Little & Rubin, 2002). The assumption 
regarding missing at random (MAR) is more relaxed in path analysis than in multiple regression.  
 In each path model, error terms of endogenous variables and significantly related 
exogenous variables were allowed to correlate. Covariates for each model included child 
gender, race/ethnicity, age, body mass index, familial socioeconomic status, and season of 
actigraph assessment (these were the last paths included in each model; Keiley, personal 
communication). Sleep parameters, RSA, RSA-R or cognitive performance data that exceeded 
4 SDs (n = 4) were considered outliers and were removed from analyses. Of note, sensitivity 
analyses were conducted and results with and without outlier removal were similar.  An example 
of the path models is presented in Figure 2. 
Separate models were fit for each objective sleep parameter (sleep efficiency, long wake 
episodes, sleep minutes) and each cognitive outcome (concept formation, numbers reversed, 
auditory working memory, decision speed, Donders M, Donders SD) resulting in 18 models. 
Additionally, subjective sleep parameters were examined simultaneously (Sleep/Wake 
Problems, Sleepiness) for each cognitive outcome; a total of six models. All models contained 
the two vagal indices (RSA, RSA-R), the main effects and the associated interactions with the 
sleep parameter(s).  
Separate analyses of the actigraphic sleep variables were conducted based on 
recommendations in the pediatric sleep literature (Berger, Wielgus, Young-McCaughan, 
Fischer, Farr, & Lee, 2008; DeWald, Meijer, Oort, Kerkhof, & Bogels, 2010) and towards a more 
nuanced understanding of associations between sleep parameters and child functioning. 
Further, a call in the sleep literature has been made to examine duration, quality and sleepiness 
separately because they are considered to represent separate sleep domains (Dewald et al., 
2010). In addition, it is recommended that researchers examine multiple sleep parameters 
including actigraphy-based and self-report measures of sleep (Sadeh, 2011). Thus, due to 
multicollinearity and power considerations (Babyak, 2004; Whisman & McClelland, 2005), each 
   
64 
 
cognitive outcome was fit separately. Due to the large number of models fit (n = 24) and to 
decrease the likelihood of Type 1 error, we only report and interpret moderation findings when 
significant interactions represented at least 10% of the number of interactions examined for that 
cognitive outcome (e.g., 10% of  significant findings across all models associated with Concept 
Formation). Models were considered an acceptable fit if at least two of the three following 
criteria were met: ?2/df < 2, CFI > .95, and RMSEA < .08 (Arbuckle & Wothke, 1999; Browne & 
Cudeck, 1993). All fitted models reported in this study satisfied these criteria.     
As recommended by Aiken and West (1991), we plotted significant interactions at +1 SD 
and -1 SD of the predictor and the moderator to understand the influence of the moderator. Of 
note, the majority of the children exhibited RSA augmentation to the task. Thus, the depiction of 
the less optimal RSA reactivity group (i.e., Low RSA Withdrawal) includes children who 
exhibited no change, augmentation, or less vagal withdrawal in comparison to their 
counterparts. The interaction utility developed by Preacher, Curran, and Bauer (2006) was used 
to plot the interactions; estimates were obtained from the fitted models. Significant interactions 
signify a statistically significant difference between the plotted lines (e.g., the association 
between sleep and cognitive functioning at high levels of the moderator is significantly different 
from the link at low levels of the moderator). We also examined whether each slope 
representing high and low levels of the moderator was significantly different from zero; 
significant slopes are indicated in the figures.  
Examination of RSA Indices as Moderators of Associations between Children?s Sleep 
and Concept Formation Scores 
 To examine whether RSA or RSA-R to the cognitive laboratory task moderated the paths 
among children?s sleep parameters (actigraphy-based and self-reported) and their scores on the 
WJ III concept formation test, interaction terms were added to the path models after the 
inclusion of the main effects. Three of the four models tested (75% of possible relations) 
   
65 
 
indicated a significant moderating role of RSA-R in the relation between sleep problems and 
performance on the concept formation test; no moderation effects were found for RSA.  
 The path between sleep efficiency and concept formation was moderated by RSA-R (? = 
.20, p < .05) and explained 3% of unique variance in concept formation (?2 = 45.45, p =.29, ?2/df 
= 1.10, CFI = .96, RMSEA = .03). The full model accounted for 16% of the variance in concept 
formation. As depicted in Figure 3A, in the context of low sleep efficiency, children who 
exhibited RSA withdrawal to the star tracing task showed better performance on the concept 
formation test than those who exhibited RSA augmentation to the lab challenge. Furthermore, 
for children who exhibited RSA withdrawal, sleep efficiency was negatively associated with 
concept formation and the slope representing this association was significantly different from 
zero. Children who exhibited RSA augmentation had lower concept formation scores regardless 
of their sleep efficiency.  
  Similar to the pattern of effects observed for sleep efficiency, RSA-R moderated the 
association between sleep minutes and concept formation (? = .14, p = .09) and explained 2% 
of the unique variance in concept formation (?2 = 40.28 p = .41, ?2/df = 1.03, CFI = .99, RMSEA 
= .01). The full model accounted for 15% of the variance in concept formation. Specifically, in 
the context of reduced sleep minutes, children who exhibited RSA withdrawal performed better 
on the concept formation test than those who exhibited RSA augmentation (Fig. 3B). 
Furthermore, a significant slope indicated a negative association between sleep minutes and 
concept formation scores for children who exhibited RSA withdrawal. Conversely, children who 
showed RSA augmentation had similar scores on concept formation regardless of the amount of 
sleep minutes they obtained. 
 RSA-R also played a significant moderating role in the link between long wake episodes 
and concept formation scores (? = .24, p < .01) and explained 2% of the unique variance in 
concept formation (?2 = 68.75, p < .01, ?2/df = 1.56, CFI = .74, RMSEA = .06). The full model 
accounted for 15% of the variance in concept formation. As shown in Figure 3C, in the context 
   
66 
 
of increased long wake episodes, children who exhibited RSA withdrawal to the laboratory task 
performed better than their counterparts who exhibited RSA augmentation. Further, a significant 
positive association between long wake episodes and concept formation was observed for 
children who showed RSA withdrawal task. However, for those who exhibited RSA 
augmentation, cognitive performance was similar regardless of sleep problems.   
 Neither RSA nor RSA-R moderated the link between subjective measures of children?s 
sleep (sleepiness; sleep/wake problems) and concept formation.  
Examination of RSA Indices as Moderators of Associations between Children?s Sleep 
and Numbers Reversed Scores 
 Analyses of children's performance on the WJ III numbers reversed test indicated that 
RSA was a significant moderator in one of the four models examined (25% of possible 
relations), and RSA-R was a significant moderator in one of the four models (25% of the 
possible relations).  
 Neither RSA nor RSA-R moderated the paths between actigraphy-based sleep 
parameters (i.e., sleep efficiency, sleep minutes, or long wake episodes) and numbers reversed.   
 RSA moderated the pathway between sleepiness and numbers reversed (? = -.17, p 
 < .05) and explained 2% of the unique variance in numbers reversed (?2 = 112.63, p < .01, ?2/df 
= 1.56, CFI = .81, RMSEA = .06). A total of 18% of the variance in numbers reversed was 
accounted for by the full model. Specifically, children who showed low RSA had average and 
stable performance regardless of their sleepiness (Figure 4A). However, children who exhibited 
high RSA and experienced low sleepiness, performed much better than all other children. 
 RSA-R also moderated the association between sleepiness and numbers reversed (? =  
-.23, p < .05) and explained 3% of the unique variance (?2 = 112.63, p < .01, ?2/df = 1.56, CFI = 
.81, RMSEA = .06). The full model accounted for 18% of the variance in numbers reversed. As 
shown in Figure 4B, children who exhibited RSA withdrawal performed similarly regardless of 
level of sleepiness. The pattern of effects was only partially consistent with expectations: at high 
   
67 
 
levels of sleepiness, those exhibiting RSA withdrawal performed better than children who 
exhibited augmentation; yet, the pattern was reversed at lower levels of sleepiness.  
 RSA did not moderate the link between sleep/wake problems and numbers reversed, but 
RSA-R did (? = .24, p < .01) and explained 4% of the unique variance (?2 = 112.63, p < .01, ?2/df 
= 1.56, CFI = .81, RMSEA = .06) in numbers reversed. A total of 18% variance in numbers 
reversed was accounted for by the full model. Again, this relation was not consistent with 
expectations. Specifically, children who showed RSA withdrawal performed significantly worse 
in the context of high sleep/wake problems than in the context of low sleep/wake problems. 
Further, there was no significant difference in children?s performance on the numbers reversed 
test for those children who exhibited RSA augmentation at either high or low levels of 
sleep/wake problems.  
Examination of RSA Indices as Moderators of Associations between Children?s Sleep 
and Auditory Working Memory Scores 
 Analyses of children's performance on the WJ III auditory working memory test indicated 
that RSA was a significant moderator in one of the four models examined (25% of possible 
relations), and RSA-R was a significant moderator in two of the four models (50% of the 
possible relations).  
RSA-R moderated the link between sleep efficiency and auditory working memory (? = 
.15, p < .09), which explained 1% of unique variance (?2= 45.22, p < .30, ?2/df = 1.10, CFI = .96, 
RMSEA = .03) and 11% of the variance in auditory working memory was accounted for by the 
full model. As depicted in Figure 5A, when sleep efficiency was low, children who exhibited RSA 
withdrawal performed better than their counterparts. On the other hand, those children who 
exhibited RSA augmentation performed better in the context of high sleep efficiency than 
children who showed RSA withdrawal.  
  RSA moderated the link between sleep minutes and auditory working memory (? =        
- .17, p = .05) and explained 2% of the unique variance in auditory working memory (?2 = 40.26, 
   
68 
 
p < .42, ?2/df = 1.03, CFI = .99, RMSEA = .01). The full model accounted for 12% of the 
variance in auditory working memory. In the context of short sleep, children with higher levels of 
RSA performed better on the WJ III auditory working memory test than those who exhibited 
lower baseline RSA (Figure 5B). Those who exhibited high RSA did not significantly differ on the 
auditory working memory test regardless of their sleep duration. However, for those who with 
lower RSA, sleep minutes was negatively associated with auditory working memory, with the 
worst performance in the context of shortened sleep. No moderation effects were found for 
RSA-R in the association between sleep minutes and auditory working memory.  
 RSA did not moderate the link between long wake episodes and auditory working 
memory, however RSA-R was a moderator of this pathway (? = -.15, p = .06; 1% of the unique 
variance explained; ?2 = 68.77, p = .01, ?2/df = 1.56, CFI = .73, RMSEA = .06; the full model 
accounted for 12% of the variance in auditory working memory). As shown in Figure 5C, when 
sleep was poor, children who exhibited RSA withdrawal performed better than their counterparts 
who showed RSA augmentation; the opposite pattern was observed when sleep was more 
optimal.  
 Neither RSA nor RSA-R moderated the link between subjective measures of children?s 
sleep and auditory working memory.  
Examination of RSA Indices as Moderators of Associations between Children?s Sleep 
and Decision Speed Scores 
 Examination of children's scores on the WJ III decision speed test yielded two significant 
moderation effects for RSA-R (50% of possible relations); no moderation effects were found for 
RSA. 
 Neither RSA nor RSA-R moderated associations between sleep efficiency or sleep 
minutes and decision speed.  
 RSA-R moderated the link between long wake episodes and decision speed (? = -.15, p 
< .05), which explained 4% of unique variance (?2 = 68.71, p = .01, ?2/df = 1.56, CFI = .74, 
   
69 
 
RMSEA = .06) and 17% of the variance in decision speed was accounted for by the full model. 
When sleep was optimal (fewer long wake episodes) all children had similar predicted means on 
decision speed (Figure 6A). However, when sleep was less optimal, children who exhibited RSA 
withdrawal performed much better than children who expressed RSA augmentation.  
 RSA-R also acted as a moderator in the pathway between sleepiness and decision 
speed (? = - .28, p < .01) and explained 4% of unique variance in decision speed (?2 = 111.83, p 
< .01, ?2/df = 1.55, CFI = .82, RMSEA = .06). The full model accounted for 19% of the variance 
in decision speed. Figure 6B depicts that when sleepiness is high, children who exhibited RSA 
withdrawal performed much better on the WJ III decision speed test than those who showed 
RSA augmentation. Further, for children exhibiting RSA augmentation, a negative association 
was observed between sleepiness and cognitive performance; however, this pattern was 
reversed for those exhibiting RSA withdrawal.  
 No moderation effects were found for the link between sleep/wake problems and 
decision speed.  
Examination of RSA Indices as Moderators of Associations between Children?s Sleep 
and Reaction Time Means and SDs 
Analyses revealed that neither RSA nor RSA-R moderated the link between children?s 
sleep and their performance on the reaction time task (M or SD scores).  
Discussion 
We examined whether vagal tone and vagal reactivity, as indexed by RSA, moderated 
associations between disruptions in children?s sleep and varying domains of cognitive 
performance in a large sample of healthy elementary school-age children. Results are among 
the first to indicate that vagal functioning interacts with sleep disruptions (objective and 
subjective) to statistically predict children?s cognitive performance. Specifically, in the context of 
poor sleep, children with more optimal vagal functioning characterized by either higher RSA or 
more pronounced RSA withdrawal to lab challenges tended to perform better on several 
   
70 
 
cognitive functioning measures (i.e., WJIII) than children who exhibited lower RSA or less 
optimal RSA reactivity. Although the amount of variance accounted for by the interaction term 
was small, the consistency of the findings strengthens confidence in the results. Further, the 
percent of variation explained is similar to studies that have investigated interactions between 
sleep and vagal activity in the prediction of child adjustment (e.g., El-Sheikh, Erath, & Keller, 
2007). Findings highlight the importance of individual differences in physiological functioning 
and the aggregation of risk across physiological and biological domains.  
The impact sleep problems have on the prefrontal cortex?s ability to exercise executive 
control over behaviors and thought was first suggested by Horne (1988, 1993) and based on 
observations in the lab (i.e., novel language and creativity were significantly impaired by sleep 
loss). Dahl (1996) extended this model by proposing that sleep loss directly affects cognitive 
tasks that require efficient working memory and executive functioning, and/or the ability to 
synchronize attention and arousal to perform abstract or complex goals. Here we found that the 
direct relations between either the objective or subjective sleep parameters and cognitive 
performance were not robust. However, some significant associations in the expected directions 
were observed: sleep minutes and long wake episodes were related to decision speed, 
sleepiness was associated with auditory working memory, and sleep-wake problems were 
related to numbers reversed, auditory working memory and a higher variability in reaction time. 
Further, using the same sample, Buckhalt and colleagues (2007) reported similar associations 
between sleep problems and cognitive performance albeit with different cognitive measures. 
The direct relations observed in the current study, as well as those reported by Buckhalt and 
colleagues, provide some support for Dahl?s (1996) hypothesis that facets of executive 
functioning such as working memory are directly impacted by sleep disruptions.    
A major contribution of the present investigation is the delineation of RSA and RSA-R as 
moderators in the link between sleep problems and cognitive performance. While the pattern of 
effect varied based on the sleep parameter, the vagal activity index, and the cognitive 
   
71 
 
performance domain, one general conclusion can be made: in the context of poor sleep, 
children with more optimal vagal functioning (higher RSA or more pronounced RSA withdrawal) 
tended to perform better than their counterparts with less optimal vagal functioning. For 
example, in the context of reduced sleep duration or increased reports of daytime sleepiness 
children with lower RSA were at a greater risk for poor cognitive performance especially on 
tasks requiring working memory (i.e., auditory working memory, numbers reversed). This 
pattern of effects illustrates aggregation of risk in that children who have poor sleep in 
conjunction with lower levels of RSA are the ones most at risk for poorer cognitive performance. 
Overall, results are consistent with those of El-Sheikh and colleagues (2007) conducted with the 
present sample in that RSA interacted with multiple sleep parameters to predict children's 
adjustment and body mass index; children at most risk were those who had poor sleep and 
lower levels of RSA.     
Similarly, children with less optimal RSA-R (i.e., augmentation) were particularly at risk 
for poorer cognitive performance in the context of sleep disruptions. Findings pertaining to 
reactivity were much more robust than those found with RSA. Specifically, children who had 
higher levels of sleep problems (actigraphy-based duration and quality as well as subjective 
reports of sleep/wake problems and sleepiness), and exhibited RSA augmentation were at 
increased risk for cognitive problems in comparison to their more physiologically regulated 
counterparts (i.e., RSA withdrawal). However, neither RSA nor RSA-R interacted with any of the 
sleep variables to predict children?s reaction time or reaction time variability. This finding was 
unexpected given that vagal activity is considered to be a marker of attention (Porges, 2007), 
which is essential to perform optimally on the reaction time task; replication is necessary and 
caution should be exercised not to accept the null hypotheses. Overall, results are novel and 
suggest an interaction between biological (sleep) and physiological (PNS activity) systems in 
the prediction of children?s ability to store novel information while performing a task (i.e., 
   
72 
 
numbers reversed, auditory working memory) and to apply previously learned rules about 
grouping a class of objects or ideas (i.e., concept formation, decision speed).  
Children with poor vagal functioning (i.e., low RSA and/or RSA augmentation) may have 
trouble obtaining good sufficient sleep (Dahl, 1996; El-Sheikh  et al., 2007). Of note, in the 
current sample there were no direct relations between any of the sleep variables and RSA or 
RSA-R. Lack of significant associations contradicts findings reported by El-Sheikh and Buckhalt 
(2005) in which multiple objectively and subjectively derived sleep parameters were associated 
with children's RSA-R to a continuous performance task. It is not evident why discrepant results 
were observed and sample characteristics (i.e., children in El-Sheikh and Buckhalt 's study were 
older) or study design (i.e., RSA-R was derived during task performance in El-Sheikh and 
Buckhalt, whereas RSA and cognitive performance were examined during different time periods 
in the present study) may have contributed to the inconsistencies. Further, it is not unusual that 
vagal indices have no direct relations with study outcome variables (e.g., externalizing and 
internalizing behaviors), but rather act as moderators of relations (e.g., El-Sheikh et al., 2007; 
El-Sheikh & Whitson, 2006; Leary & Katz, 2004). In addition, the lack of association among 
sleep and vagal indices may be due to the normative sample used in this study. It is possible 
that samples of children with clinically diagnosed problems in sleep or vagal functioning may 
show direct links between the two biophysiological systems. Examinations of associations 
between vagal indices during non-nocturnal conditions and sleep parameters are scarce. Thus, 
further testing of these relations among diverse populations of children is likely to delineate the 
conditions under which RSA and sleep problems are related.  
 The protective function of higher RSA and increased RSA withdrawal to stressors are 
consistent with the Polyvagal Theory (Porges, 1998), which proposes that vagal functioning 
influences the regulation of arousal, affect and attention (Porges, 2007). An adaptive 
physiological response when confronted with stress (e.g., sleep disruptions), is considered to be 
a disengagement of the vagal brake (i.e., vagal withdrawal), which inhibits PNS influence (i.e., 
   
73 
 
increased heart rate). This response allows the individual to focus attention on the environment, 
obtain information, and/or reduce the challenge by using social strategies to regulate emotions. 
Once the challenge has subsided, the PNS influence can reengage to reduce arousal. The 
ability to quickly engage and disengage as the environmental conditions become more 
challenging and to be able respond to metabolic demands, including increased attention and 
information processing (Suess et al., 1994), are adaptive processes vital for behavior.  
Several of the moderation effects were difficult to comprehend especially those 
pertaining to either subjective sleep measures or the pattern of effects observed in the context 
of more optimal sleep. For example based on the large literature regarding RSA withdrawal, one 
would expect it children who exhibited this pattern of reactivity to a stressor would obtain more 
optimal outcomes than those who showed augmentation regardless of the context. However, in 
five of the ten significant moderation findings children who exhibited higher levels of vagal 
withdrawal performed significantly better on several cognitive assessments when sleep was 
poor than when it was more optimal. While the meaning of these inconsistent effects is not 
apparent, we conducted post-hoc analyses to examine some possible three-way interaction 
effects. Child gender and ethnicity were examined through three-way interactions involving 
sleep, vagal functioning, and cognitive performance; no significant interactions emerged.  
The meaning of these unexpected findings remains unclear and warrants further investigation.   
Although the results suggest that more optimal PNS activity may play an important role 
in the link between sleep disruptions and children?s cognitive performance, the PNS is one 
branch of the ANS  and the second branch, namely the sympathetic nervous system (SNS)  is 
an important physiological system associated with sleep (El-Sheikh & Arsiwalla, 2011). 
Furthermore, PNS and SNS activity can have additive and interactive effects in the prediction of 
child functioning (e.g., El-Sheikh et al., 2009). Thus, future research would benefit from 
examining the potential influence of both branches of the ANS in relation to sleep and cognitive 
functioning. Similarly, vagal recovery (the change in RSA once a challenge has subsided) is a 
   
74 
 
third component of vagal activity that was not measured in this study. According to the 
Polyvagal Theory (Porges, 2007), once the stressful event has ended, it is pertinent for heart 
rate to return to normal levels to reduce wear and tear on the heart, which would correspond 
with a return of RSA to baseline or pre-stress levels. The inclusion of this third index of PNS 
functioning (i.e., recovery) may also assist in further explication of the role of the PNS in the 
sleep-cognitive functioning link. 
 The reliance on cross-sectional data did not allow us to determine the causality or 
examine the possible meditational role of vagal functioning in the link between sleep problems 
and cognitive performance; note however, that sleep was not directly associated with vagal 
activity limiting mediation or intervening variable pathways in this cross-sectional study. Results 
should be interpreted with caution and considered as a first step in explicating how sleep is 
differentially associated with cognitive performance based on an individual?s physiological 
reactivity. Additionally, these results must be interpreted in relation to the developmental age of 
the children. It may be that the role of vagal functioning in the link between sleep disruptions 
and cognitive functioning may be more or less pronounced for individuals at different stages of 
development (e.g., infancy, adolescence). This sample contained only children from the third 
grade, which allowed for a more focused examination; however, the limited age range reduces 
the generalizability of the findings to a more diverse population. Further, the normative level of 
sleep and physiological functioning in this community sample, and exclusion/inclusion criteria 
(e.g., two-parent home, lack of children with mental retardation, developmental delays, ADHD, 
chronic illnesses) does not allow for understanding examined questions in the context of more 
severe or clinically significant problems.  
 Although this study used both actigraphy and subjective measures of children?s sleep, it 
is important to note that daytime naps were not taken into consideration. Sleeping during the 
day may compensate for some of the problems associated with reduced sleep duration and we 
recommend that future studies incorporate napping into the total daily sleep duration to further 
   
75 
 
delineate the relations between sleep and cognition. Although the use of actigraphy and 
subjective sleep measures constitute an advance in this field of inquiry, and are more likely to 
explicate relations between sleep problems and cognitive outcomes (Sadeh, 2008), these 
assessments do not have the ability to assess sleep staging. An important next step for 
researchers is to investigate sleep staging through polysomnography (PSG) and its association 
with both cognitive outcomes and physiological reactivity. The amount of time spent in different 
stages of sleep may play a role in the type of cognitive functioning that is impaired by 
disruptions in the duration and quality of children?s sleep. For example, research has shown that 
slow-wave sleep (stages 3 and 4) is implicated in declarative memory consolidation, and rapid 
eye movement sleep (REM) is implicated in procedural memory consolidation (Hornung, 
Danker-Hopfe, Regen, Schredl, & Heuser, 2007). Thus, the use of PSG in future studies is 
warranted.  
 In summary, support for interactions between biological (e.g., sleep) and physiological 
(e.g., RSA) regulatory systems in predicting children?s cognitive performance abilities was found 
with a large sample of typically developing third grade children. Results extend work in the area 
of sleep and cognitive outcomes of typically developing school-aged children by (1) assessing 
possible individual differences variables; (2) incorporating both actigraphy and self-reports as a 
measure of sleep; and (3) examining multiple aspects of executive functioning using 
standardized tests. The multifaceted nature of this study lends strength to the moderation 
effects. Results suggest that poor physiological regulation (i.e., low baseline RSA or RSA 
augmentation) may undermine children?s cognitive performance, namely executive functions, in 
the context of normative levels of sleep problems.  
 
 
 
 
   
76 
 
References 
Abrams, R. A. (2004). Psych/Lab for windows (Computer software and manual). 
Acebo, C., & Carskadon, M. (2001). Scoring actigraph data using ACTION-W 2. Providence, RI: 
Bradley Sleep Center, Brown University. 
Acebo, C., & Carskadon, M. A. (2002). Influence of irregular sleep patterns on waking behavior. 
In M. A. Carskadon (Ed.), Adolescent sleep patterns:  Biological, social, and 
psychological influences (pp. 220-235). New York: Cambridge University Press. 
Acebo, C., Sadeh, A., Seifer, R., Tzischinsky, O., Wolfson, A., Hafer, A., & Carskadon, M. A. 
(1999). Estimating sleep patterns with actigraphy monitoring in children and adolescents: 
How many nights are necessary for reliable measures? Sleep, 22, 95-103. 
Aiken, L. S., & West, S. G. (1991). Multiple regression:  Testing and interpreting interactions. 
Newbury Park, CA: Sage. 
Arbuckle, J. L. (2007). Amos 16.0 User's Guide. Spring House, PA: Amos Development 
Corporation. 
Arbuckle, J., & Wothke, W. (1999). AMOS 4 User's reference guide. Chicago: Smallwaters 
Corp. 
Archbold, K. H., Giordani, B., Ruzicka, D. L., & Chervin, R. D. (2004). Cognitive executive 
dysfunction in children with mild sleep-disordered breathing. Biological Research for 
Nursing, 5, 168-176. 
Babyak, M. A. (2004). What you see may not be what you get: A brief, nontechnical introduction 
to overfitting in regression-type models. Psychosomatic Medicine, 66, 411-421. 
Bar-Haim, Y., Marshall, P. J., & Fox, N. A. (2000). Developmental changes in heart period and 
high-frequency heart period variability from 4 months to 4 years of age. Developmental 
Psychobiology, 37, 44-56. 
   
77 
 
Beauchaine, T. (2001). Vagal tone, development, and Gray's Motivational Theory: Toward an 
integrated model of autonomic nervous system functioning in psychopathology. 
Development & Psychopathology, 13, 183-214. 
Berger, A. M., Wielgus, K. K., Young-McCaughan, S., Fischer, P., Farr, L., & Lee, K. A. (2008). 
Methodological Challenges When Using Actigraphy in Research. [doi: 
10.1016/j.jpainsymman.2007.10.008]. Journal of Pain and Symptom Management, 
36(2), 191-199. 
Berntson, G. G., Bigger, J. T., Eckberg, D. L., Grossman, P., Kaufmann, P. G., Malik, M., ?van 
der Molen, M. W. (1997). Heart rate variability: Origins, methods, and interpretive 
caveats. Psychophysiology, 34, 623-648. 
Blunden, S., Lushington, K., Lorenzen, B., Martin, J., & Kennedy, D. (2005). Neuropsychological 
and psychosocial function in children with a history of snoring or behavioral sleep 
problems. The Journal of Pediatrics, 146, 780-786. 
Bornstein, M. H., & Suess, P. E. (2000). Child and mother cardiac vagal tone: Continuity, 
stability, and concordance across the first 5 years. Developmental Psychology, 36, 54-
65. 
Brown, T. A., & Barlow, D. H. (2005). Dimensional versus categorical classification of mental 
disorders in fifth edition of the diagnostic and statistical manual of mental disorders  
and beyond: Comment on the special section [Special Issue]. Journal of Abnormal 
Psychology, 114, 551-556 
Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen & 
J. S. Long (Eds.), Testing structural equation models (pp. 136-162). Newbury Park, CA: 
Sage. 
Buckhalt, J. A., El-Sheikh, M., & Keller, P. (2007). Children's sleep and cognitive functioning: 
Race and socioeconomic status as moderators of effects. Child Development, 78, 213-
231. 
   
78 
 
Buckhalt, J. A., El-Sheikh, M., Keller, P. S., & Kelly, R. J. (2009). Concurrent and longitudinal 
relationships between children's sleep and cognitive functioning: The moderating role of 
parent education. Child Development, 80, 875-892. 
Buckhalt, J., Wolfson, A., & El-Sheikh, M. (2009). Children's sleep and school psychology 
practice. School Psychology Quarterly, 24, 60-69. 
Calkins, S. D. (1997). Cardiac vagal tone indices of temperamental reactivity and behavioral 
regulation in young children. Developmental Psychobiology, 31, 125-135. 
Carskadon, M. A., Seifer, R., & Acebo, C. (1991). Reliability of six scales in a sleep 
questionnaire for adolescents. Sleep Research, 20, 421. 
Cicchetti, D., & Dawson, G. (2002). Editorial: Multiple levels of analysis. Development and 
Psychopathology, 14, 417-420. 
Curcio, G., Ferrara, M., & De Gennaro, L. (2006). Sleep loss, learning capacity and academic 
performance. Sleep Medicine Reviews, 10, 323-337. 
Dahl, R. E. (1996). The regulation of sleep and arousal: Development and psychopathology. 
Development and Psychopathology, 8, 3-27. 
Dewald, J. F., Meijer, A. M., Oort, F. J., Kerkhof, G. A, & Bogels, S. M. (2010). The influence of 
sleep quality, sleep duration and sleepiness on school performance in children and 
adolescents: A meta-analytic review. Sleep Medicine Reviews, Advanced online 
publication. PMID: 20093054.  
DiPietro, J. A., Porges, S. W., & Uhly, B. (1992). Reactivity and developmental competence in 
preterm and fullterm infants. Developmental Psychology, 28, 831-841. 
Dorrian, J., Rogers, N. L., & Dinges, D. F. (2005). Psychomotor vigilance performance: 
Neurocognitive assay sensitive to sleep loss. In C. Kushida (Ed.), Sleep Deprivation (pp. 
39-63).  New York: Marcel Dekker.  
   
79 
 
El-Sheikh, M. (2005). The role of emotional responses and physiological reactivity in the marital 
conflict-child functioning link. Journal of Child Psychology and Psychiatry, 46, 1191-
1199. 
El-Sheikh, M., & Arsiwalla, D. D. (2011). Children?s sleep, skin conductance level and mental 
health. Journal of Sleep Research, 20(2), 326-337. 
El-Sheikh, M., & Buckhalt, J. A. (2005). Vagal regulation and emotional intensity predict 
children's sleep problems. Developmental Psychophysiology, 46, 307-317. 
El-Sheikh, M., Erath, S. A., & Keller, P. S. (2007). Children's sleep and adjustment: The 
moderating role of vagal regulation. Journal of Sleep Research, 16, 396-405. 
El-Sheikh, M., Kouros, C. D., Erath, S., Cummings, E. M., Keller, P., & Staton, L.  (2009). 
Marital conflict and children?s externalizing behavior: Interactions between 
parasympathetic and sympathetic nervous system activity. Monographs of the Society 
for Research in Child Development, 74(1, Serial No. 231). 
El-Sheikh, M., & Whitson, S. A. (2006). Longitudinal relations between marital conflict and child 
adjustment: Vagal regulation as a protective factor. Journal of Family Psychology, 20, 
30-39. 
Fallone, G., Seifer, R., Acebo, C., & Carskadon, M. (2000). Prolonged sleep restriction in 11- 
and 12-year-old childre: Effects on behavior, sleepiness and mood. Sleep, 23S2, A:28. 
Galles, S. J., Miller, A., Cohn, J. F., & Fox, N. A. (2002). Estimating parasympathetic control of 
heart rate variability: Two approaches to quantifying vagal tone. Psychophysiology, 
39(Suppl. 1), S37. 
Geiger, A., Achermann, P., & Jenni, O. G. (2010). Association between sleep duration and  
 intelligence scores in healthy children. Developmental Psychology, 46, 949-954. 
Grossman, P., Karemaker, J., & Wieling, W. (1991). Prediction of tonic sarasympathetic cardiac 
control using respiratory sinus arrhythmia: The need for respiratory control. 
Psychophysiology, 28(2), 201-216. 
   
80 
 
Grossman, P., & Taylor, E. W. (2007). Toward understanding respiratory sinus arrhythmia: 
Relations to cardiac vagal tone, evolution and biobehavioral functions. Biological 
Psychology, 74, 263-285. 
Hayano, J., Sakakibara, Y., Yamada, A, Yamada, M., Mukai, S., Fujinami, T., ?Takata, K. 
(1991). Accuracy of assessment of cardiac vagal tone by heart rate variability in normal 
subjects. American Journal of Cardiology, 67, 199-204. 
Hollingshead, A. B. (1975). Four factor index of social status. unpublished manuscript. 
Horne, J. A. (1988). Why we sleep: The functions of sleep in humans and other animals.  
 Oxford: Oxford University Press. 
Horne, J. A. (1993). Human sleep, sleep loss and behaviour. Implications for the prefrontal 
cortex and psychiatric disorder. British Journal of Psychiatry, 162, 413-419. 
Hornung, O. P., Regen, F., Danker-Hopfe, H., Schredl, M., & Heuser, I. (2007). The 
Relationship Between REM Sleep and Memory Consolidation in Old Age and Effects of 
Cholinergic Medication. [doi: 10.1016/j.biopsych.2006.08.034]. Biological Psychiatry, 
61(6), 750-757. 
Huffman, L. C., Bryan, Y. E., del Carmen, R., Pederson, F. A., Doussard-Roosevelt, J. A., & 
Porges, S. W. (1998). Infant temperament and cardiac vagal tone: Assessments at 
twelve weeks of age. Child Development, 69, 624-635. 
Irwin, M. (2008). Human psychoneuroimmunology: Twenty years of discovery. Brain, Behavior, 
and Immunity, 22, 129-139. 
Kahn, A., Van de Merckt, C., Rebuffat, E., Mozin, M.J., Sottiaux, M., Blum, D., & Hennart, P. 
(1989). Sleep problems in healthy preadolescents. Pediatrics, 84, 542-546. 
Leary, A., & Katz, L. F. (2004). Coparenting, family-level processes, and peer outcomes: The 
moderating role of vagal tone. Development and Psychopathology, 16, 593-608. 
Little, R. J., & Rubin, D.B. (2002). Statistical analysis with missing data. (2nd ed.). New York: 
John Wiley, & Sons. 
   
81 
 
Matthews, K. A., Rakaczky, C. J., Stoney, C. M., & Manuck, S. B. (1987). Are cardiovascular 
responses to behavioral stressors a stable individual difference variable in childhood? 
Psychophysiology, 24, 464-473. 
Matthews, K. A., Woodall, K. L., & Stoney, C. M. (1990). Changes in and stability of 
cardiovascular responses to behavioral stress: results from a four-year longitudinal study 
of children. Child Development, 61, 1134-1144. 
Mezzacappa, E., Kindlon, D., Saul, J. P., & Earls, F. (1998). Executive and motivational control 
for performance task behavior, and autonomic heart-rate regulation in children: 
Physiologic validation of two-factor solution inhibitory control. Journal of Child 
Psychology and Psychiatry, 39, 525-531. 
Mindell, J. A., & Owens, J. A. (2009). A clinical guide to pediatric sleep: Diagnosis and 
management of sleep problems. New York: Lippincott, Williams, & Wilkins. 
Mindell, J. A., Owens, J. A., & Carskadon, M. A. (1999). Developmental features of sleep. Child 
and Adolescent Psychiatric Clinics of North America, 8, 695-725. 
National Sleep Foundation. (2004). 2004 Sleep in America Poll. Retrieved November 12, 2004, 
from http://www.sleepfoundation.org/2004poll.cfm 
Ozaki, N., Ono, Y., Ito, A., & Rosenthal, N. E. (1995). The prevalence of  
 seasonal difficulties among Japanese civil servants. American Journal of  
 Psychiatry, 152, 1225-1227. 
Porges, S. W. (Ed.). (1991). Vagal tone: An autonomic mediator of affect. New York: Cambridge 
University Press. 
Porges, S. W. (1995). Cardiac vagal tone: A physiological index of stress. Neuroscience and 
Biobehavioral Reviews, 19, 225-233. 
Porges, S. W. (1998). Love: An emergent property of the mammalian autonomic nervous 
system. Psychoneuroendocrinology, 23(8), 837-861. 
   
82 
 
Porges, S. W. (2001). The polyvagal theory: Phylogenetic substrates of a social nervous 
system. International Journal of Psychophysiology, 42, 123-146. 
Porges, S. W. (2007). The polyvagal perspective. Biological Psychology, 74, 116-143. 
Preacher, K. J., Curran, P. J., & Bauer, D. J. (2006). Computational tools for probing 
interactions in multiple linear regression, multilevel modeling, and latent curve analysis. 
Journal of Educational and Behavioral Statistics, 31, 437-448. 
Rottenberg, J. (2007). Cardiac vagal control in depression: A critical analysis. Biological 
Psychology, 74, 200-211. 
Sadeh, A. (2007). Consequences of sleep loss or sleep disruption in children. Sleep Medicine 
Clinics, 2, 513-520. 
Sadeh, A. (2008). Commentary: Comparing actigraphy and parental report as measures of 
children's sleep. Journal of Pediatric Psychology, 33, 406-407.  
Sadeh, A. (2011). Sleep assessment methods. In M. El-Sheikh (Ed.), Sleep and development: 
Familial and socio-cultural considerations (pp. 355-374). New York, NY: Oxford 
University Press. 
Sadeh, A., Gruber, R., & Raviv, A. (2003). The effects of sleep restriction and extension on 
school-age children: What a difference an hour makes. Child Development, 74, 444-455. 
Sadeh, A., Raviv, A., & Gruber, R. (2000). Sleep patterns and sleep disruptions in school-age 
children. Developmental Psychology, 36, 291-301. 
Sadeh, A., Sharkey, K. M., & Carskadon, M. A. (1994). Activity-based 
Sleep/Wakeindentification: An empirical test of methodological issues. Sleep, 17, 201-
207. 
Salomon, K.  (2005). Respiratory sinus arrhythmia during stress predicts resting respiratory 
sinus arrhythmia 3 years later in a pediatric sample. Health Psychology, 24, 68-76. 
   
83 
 
Sherwood, A., Allen, M., Obrist, P., & Langer, A. (1986). Evaluation of beta-adrenergic 
influences on cardiovascular and metabolic adjustments to physical and psychological 
stress. Psychophysiology, 23, 89-104. 
Spilsbury, J. C., Storfer-Isser, A., Drotar, D., Rosen, C. L., Kirchner, L., & Redline, S. (2005). 
Effects of the home environment on school-aged children?s sleep. Sleep, 28, 1419-1427. 
Staton, L., El-Sheikh, M., & Buckhalt, J. (2009). Respiratory sinus arrhythmia and cognitive 
functioning in children. Developmental Psychobiology, 51, 249-258. 
Sternberg, S. (1969). Memory-scanning: Mental processes revealed by reaction-time 
experiments. American Scientist, 57, 421-457. 
Steenari, M., Vuontela, V., Paavonen, E. J., Carlson, S., Fjallberg, M., & Aronen, E. T. (2003). 
Working memory and sleep in 6- to 13-year-old school children. Journal of the American 
Academy of Child and Adolescent Psychiatry, 42, 85-92. 
Stein, M. A., Mendelsohn, J., Obermeyer, W. H., Amromin, J., & Benca, R. (2001). Sleep and 
behavior problems in school-aged children. Pediatrics, 107, e60. 
Sternberg, S. (1966). High-speed scanning in human memory. Science, 153, 652-654. 
Suess, P. E., Porges, S. W., & Plude, D. J. (1994). Cardiac vagal tone and sustained attention 
in school-age children. Psychophysiology, 31, 17-22. 
Thayer, J. F., & Lane, R. D. (2000). A model of neurovisceral integration in emotion regulation 
and dysregulation. Journal of Affective Disorders, 61, 201-216. 
Van Dongen, H. (2006). Shift work and inter-individual differences in sleep and sleepiness. 
Chronobiology International, 23(6), 1139-1147. 
Wolfson, A. R., & Carskadon, M. A. (1998). Sleep schedules and daytime functioning in 
adolescents. Child Development, 69, 875-887. 
Wolfson, A., & Carskadon, M. (2003). Understanding adolescent's sleep patterns and school 
performance: A critical appraisal. Sleep Medicine Reviews, 7, 491-506. 
   
84 
 
Wolfson, A. R., Carskadon, M. A., Acebo, C., Seifer, R., Fallone, G., Labyak, S. E., & Martin, J. 
L. (2003). Evidence for the validity of a sleep habits survey of adolescents. Sleep, 26, 
213-216. 
Woodcock, R. W., McGrew, K. S., & Mather, M. (2001). Woodcock-Johnson III Tests of 
Cognitive Ability. NY: Riverside Publishing. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
   
85 
 
 
Table 1. Means, Standard Deviations and Range of Study Variables. 
 
 
 
 
 
 
 
 
 
 
 
               
Note: N ranges from 159 to 165 due to missing data. WJIII = Woodcock Johnson III; M = mean;    
SD = standard deviation; ms = milliseconds. 
 Mean SD Range 
Child Age-years 8.72 .30 8.08 - 9.83 
Family Socioeconomic Status  35.95 9.35 17 - 63 
Body Mass Index 20.34 4.63 12.87 - 41.70 
Sleep Efficiency-actigraphy 86.78 7.94 48.70 - 97.86 
Long Wake Episodes-actigraphy 3.92 2.20 .43 - 10.57 
Sleep Minutes-actigraphy 445.10 54.78 254.00 - 539.71 
Sleepiness-child report 14.11 4.50 0 - 27 
Sleep/WakeProblems-child report 17.04 4.95 10 - 34 
RSA .13 .06 .02 - .36 
RSA-R -.02 .05 -.16 - .31 
Concept Formation-WJ III 99.32 12.67 54 - 142 
Numbers Reversed- WJ III 104.03 15.92 52 - 150 
Auditory Working Memory- WJ III 112.01 13.08 75 - 151 
Decision Speed- WJ III 107.65 13.12 76 - 145 
Reaction Time M-Donders (ms) 464.74 75.44 327.04 - 754.30 
Reaction Time SD-Donders (ms) 140.11 37.16 66.72 - 227.39 
   
86 
 
Table 2. Correlations among study variables.  
 
 
 
 
 
 
 
 
 
 
 
 
 
Note: Season of watch wear, 1= winter, 0= fall, spring; RSA = respiratory sinus arrhythmia; RSA-R = respiratory sinus arrhythmia 
reactivity; WJ III = Woodcock Johnson III.  
*p < .05; **p < .01; ***p < .001. 
 
 
 
 
 
 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 
1.  Child Age -           
2.  Child Gender; 1 = Boy; 0 = Girls -.02 -          
3.  Ethnicity; 1 = AA, 0 = EA -.10 -.08 -         
4.  Family Socioeconomic Status .18*  .05 - .24** -        
5.  Body Mass Index -.11 -.09 .07 -.13 -       
6.  Season of actigraphy  .36** -.06 .05    .20** .04 -      
7.  Sleep Efficiency-actigraphy  .17* -.20 .09 .10   -.18*  .15* -     
8.  Long Wake Episodes-actigraphy -.19*   .18*  -.19* -.04    .11  .07 -.85** -    
9.  Sleep Minutes-actigraphy  .18*   -.24**  -.09    .20**  -.21** -.17* .83** -.61** -   
10. Sleepiness-child report -.10 .15*  .16* -.10   -.01 -.07  -.07 .01 -.08 -  
11. Sleep/Wake Problems-child report -.09 -.06   .25**  -.16*    .06 -.02  -.07 .02 -.12   .43** - 
12. RSA -.08 .07 .05 -.03  -.24** -.02   .03  -.05 .09   .04 .10 
13. RSA-R   .02 .05 .04 -.02 .04 -.05  -.04  -.02 -.08  -.03 .02 
14. Concept Formation-WJ III  .16 -.03 -.19*    .25**   -.03  .19*  -.04 .04 .00  -.05   -.08 
15. Numbers Reversed-WJ III .20** .01   -.15*   .15*   -.07  .11   .02  -.07 .03  -.13 -.16* 
16. Auditory Working Memory-WJ III -.26** .06   -.11    .23**   -.10  .05   .10  -.12 .11  -.16* -.18* 
17.  Decision Speed- WJ III  .12 -.17*   -.01    .20**   -.05  .03   .13 -.15*  .16*  -.03 .02 
18.  Reaction Time M-Donders -.24** -.31** .07 -.11 .13 -.04   .01 -.04 .04  -.01 .12 
19.  Reaction Time SD-Donders -.24** -.10 .08 -.13 .08 -.05  -.04 -.02 -.04   .05  .17* 
   
87 
 
 
Table 2 continued. 
 
 
 
 
 
 
Note: Season of watch wear, 1= winter, 0= fall, spring; RSA = respiratory sinus arrhythmia; RSA-R = respiratory sinus arrhythmia 
reactivity; WJ III = Woodcock Johnson III.  *p < .05; **p < .01; ***p < .001. 
 
 
 
 
 
 12. 13. 14. 15. 16. 17. 18. 19. 
13. RSA-R  -.13 -       
14. Concept Formation-WJ III -.02 -.01 -      
15. Numbers Reversed-WJ III  .07 -.03 .37** -     
16. Auditory Working Memory-WJ III -.08 .06 .33** .38**      -    
17.  Decision Speed- WJ III -.11 -.06 .24** .20**    .21**   -   
18.  Reaction Time M-Donders .01 .01 -.21** -.24** -.24** -.23**   -  
19.  Reaction Time SD-Donders .05 .09 -.28** -.28** -.20* -.20* .72** - 
 
88 
 
Figure 1.  
 
 
 
Figure 1. Conceptual model depicting children?s parasympathetic nervous system activity as a 
moderator of the association between children?s sleep and their cognitive functioning. 
. 
 
 
 
 
 
 
 
 
 
 
Parasympathetic 
Nervous System 
Activity 
 
 
 
 
 
Children?s 
Sleep 
 
 
Cognitive 
Functioning 
 
89 
 
Figure 2. 
 
 
Figure 2. An example of the path models that examined the possible moderating role of RSA 
and RSA-R in the sleep-cognitive functioning link. Of note, control variables are not included in 
the model above however all controls (sex, race/ethnicity, age, body mass index, and season of 
watch wear) were entered as the last paths in the model. 
 
 
 
 
 
 
 
 
 
 
 
 
 
90 
 
Figure 3. 
 
 
 
 
 
 
 
 
A. 
95
97
99
101
103
105
107
109
111
1 1 3
1 1 5
- 1  S D + 1  S D
C
o
n
c
e
p
t 
F
o
r
m
a
ti
o
n
T1
Sl e e p  Eff i c i e n c y   T 1
H i g h  R SA  
W i th d r a w a l
L o w  R SA   
W i th d r a w a l
 
B. 
95
97
99
101
103
105
107
109
111
1 1 3
1 1 5
- 1  S D + 1  S D
C
o
n
c
e
p
t 
F
o
r
m
a
ti
o
n
T1
Sl e e p  Mi n u te s   T 1
H i g h  R SA       
W i th d r a w a l
L o w  R SA        
W i th d r a w a l
 
C. 
95
97
99
101
103
105
107
109
111
1 1 3
1 1 5
- 1  S D + 1  S D
C
o
n
c
e
p
t 
F
o
r
m
a
ti
o
n
T1
L o n g  W a k e  Ep i s o d e s   T 1
H i g h  R SA     
W i th d r a w a l
L o w  R SA      
W i th d r a w a l
 
 
 
 
Figure 3. (A) Interaction between Sleep Minutes and RSA-R in the prediction of 
children?s Concept Formation. (B) Interaction between Sleep Efficiency and RSA-R in 
the prediction of children?s Concept Formation. (C) Interaction between Long Wake 
Episodes and RSA-R in the prediction of children?s Concept Formation. 
 
91 
 
Figure 4. 
 
 
 
 
 
 
 
 
 
A. 
95
97
99
101
103
105
107
109
111
1 1 3
1 1 5
- 1  S D + 1  S D
N
u
m
b
e
r
s
 
R
e
v
e
r
s
e
d
 
T1
Sl e e p i n e s s   T 1
H i g h  R SA
L o w  R SA
 
B. 
95
97
99
101
103
105
107
109
111
1 1 3
1 1 5
- 1  S D + 1  S D
N
u
m
b
e
r
s
 
R
e
v
e
r
s
e
d
 
T1
Sl e e p i n e s s   1
H i g h  R SA     
W i th d r a w a l
L o w  R SA    
W i th d r a w a l
 
C. 
95
97
99
101
103
105
107
109
111
1 1 3
1 1 5
- 1  S D + 1  S D
N
u
m
b
e
r
s
 
R
e
v
e
r
s
e
d
 
T1
Sl e e p / W a k e  Pr o b l e m s   T 1
H i g h  R SA     
W i th d r a w a l
L o w  R SA        
W i th d r a w a l
 
 
 
 
Figure 4. (A) Interaction between Sleepiness and RSA in the prediction of children?s 
Numbers Reversed scores. (B) Interaction between Sleepiness and RSA-R in the prediction 
of children?s Numbers Reversed scores. (C) Interaction between Sleep/Wake Problems and 
RSA-R in the prediction of children?s Numbers Reversed scores. 
 
92 
 
Figure 5.  
 
A. 
100
102
104
106
108
1 1 0
1 1 2
1 1 4
1 1 6
1 1 8
120
- 1  SD +1  SD
A
u
d
i
to
r
y
 
W
o
r
k
i
n
g
 
Me
m
o
r
y
 
T1
Sl e e p  Eff i c i e n c y  T 1
H i g h  R SA      
W i th d r a w a l
L o w  R SA       
W i th d r a w a l
 
B. 
100
102
104
106
108
1 1 0
1 1 2
1 1 4
1 1 6
1 1 8
120
- 1  SD +1  SD
A
u
d
i
to
r
y
 
 
W
o
r
k
i
n
g
 
Me
m
o
r
y
 
T1
Sl e e p  Mi n u te s  T 1
H i g h  V a g a l  
T o n e
L o w  V a g a l  
T o n e
 
C. 
 
100
102
104
106
108
1 1 0
1 1 2
1 1 4
1 1 6
1 1 8
120
- 1  SD +1  SD
A
u
d
i
to
r
y
 
 
W
o
r
k
i
n
g
 
Me
m
o
r
y
 
T1
L o n g  W a k e  Ep i s o d e s  T 1
H i g h  R SA      
W i th d r a w a l
L o w  R SA      
W i th d r a w a l
 
 
 
 
 
 
Figure 6. 
Figure 5. (A) Interaction between Sleep Efficiency and RSA-R in the prediction of children?s 
Auditory Working Memory scores. (B) Interaction between Sleep Minutes and RSA in the 
prediction of children?s Auditory Working Memory scores. (C) Interaction between Long Wake 
Episodes and RSA-R in the prediction of children?s Auditory Working Memory scores. 
 
 
93 
 
Figure 6.  
 
 
A. 
95
97
99
101
103
105
107
109
111
1 1 3
1 1 5
- 1  SD +1  SD
D
e
c
i
s
i
o
n
 
Sp
e
e
d
 
T
1
L o n g  W a k e  Ep i s o d e s  T 1
H i g h  R SA      
W i th d r a w a l
L o w  R SA     
W i th d r a w a l
 
 
B. 
95
97
99
101
103
105
107
109
111
1 1 3
1 1 5
- 1  SD +1  SD
D
e
c
i
s
i
o
n
 
Sp
e
e
d
 
T
1
Sl e e p i n e s s  T 1
H i g h  R SA       
W i th d r a w a l
L o w  R SA       
W i th d r a w a l
 
 
 
 
 
 
 
 
 
Figure 6. (A) Interaction between Long Wake Episodes and RSA-R in the prediction of 
children?s Decision Speed scores. (B) Interaction between Sleepiness and RSA-R in the 
prediction of children?s Decision Speed scores.  
94 
 
                                                Abstract Manuscript 2 
Children?s respiratory sinus arrhythmia (RSA) indices (i.e., baseline levels, reactivity to a 
lab challenge and/or recovery following the challenge), markers of the parasympathetic nervous 
system activity, were examined as potential moderators of the association between sleep 
parameters (self-report and actigraphy-based) and multiple domains of children?s cognitive 
functioning (i.e., concept formation, working memory, processing speed, attention) both cross-
sectionally and longitudinally. African- and European-American third grade children (N = 166; 
55% boys; M age = 8.72) participated at T1 and again two years later (T2; N = 132; 80 boys; M 
age = 10.79). At both time points: (1) children?s sleep was assessed via seven nights of 
actigraphy and by interview; (2) cognitive abilities were derived from the Woodcock-Johnson II 
Tests of Cognitive Abilities (WJ III) and a reaction time task administered individually; and (3) 
RSA indices (tone, reactivity, and/or recovery) were assessed in the lab. Cross-sectional results 
indicate that actigraphy-based sleep variables (i.e., sleep minutes, sleep efficiency, and long 
wake minutes) interact with RSA reactivity (RSA-R) to a problem-solving task to predict 
children?s decision speed scores on the WJ III; cross-sectional findings did not yield significant 
moderation effects for baseline RSA, RSA recovery. Longitudinal findings show that RSA-R, 
and to a lesser extent baseline RSA, interact with both self-reported and actigraphy-based sleep 
parameters to predict multiple WJ III test scores two years later. Results build on the scant 
literature examining individual differences in the sleep-cognitive functioning link and suggest 
that interactions between physiological and biological systems are important for the prediction of 
children?s cognitive functioning.  
 
 
 
95 
 
Sleep and cognition: The moderating role of respiratory sinus arrhythmia activity 
The effects of sleep on children?s and adolescents? functioning are receiving increased 
attention in both the research and public media realms. Studies have provided strong evidence 
that sleep problems, both clinical and non-clinical, can lead to a number of problems, including 
emotion dysregulation, behavior problems, and trouble performing cognitive tasks (Sadeh, 
2007). In addition, adult and child research has linked sleep problems with psychiatric disorders 
and physical health problems (Cohen-Zion & Ancoli-Israel, 2004; Dahl & Harvey, 2007; 
Williams, Sears, & Allard, 2004). However, in the context of sleep disruptions, not everyone 
shows the same cognitive functioning decrements (Buckhalt, El-Sheikh, & Keller, 2007; 
Buckhalt, El-Sheikh, Keller, & Kelly, 2009), yet we know little about individual differences that 
may modulate these associations. Respiratory sinus arrhythmia (RSA), a marker of 
parasympathetic nervous system activity, is an important individual difference variable that is 
linked to children?s cognitive functioning (Staton, Buckhalt, & El-Sheikh, 2009). RSA interacts 
with sleep problems to predict child adjustment outcomes (El-Sheikh, Erath, & Keller, 2007), 
including cognitive functioning (manuscript 1). Thus, in this study we aim to examine how 
indices of biological regulation (i.e., sleep) and physiological activity (i.e., RSA) interact to 
predict various parameters of children?s executive functioning (i.e., concept formation, numbers 
reversed, auditory working memory, decision speed) and reaction time (i.e., reaction time, 
variability in reaction time), both cross-sectionally and longitudinally two years later. Figure 1 
depicts the conceptualization of the moderating role of vagal activity in the association between 
sleep disruptions and cognitive performance.  
The majority of sleep research with children has been conducted with clinical samples; 
however, there has been a recent focus on typically developing children. Non-clinical sleep 
disruptions such as reduced sleep quality and duration in typically developing school-aged 
children are estimated at rates ranging from 20% to 50% (Kahn et al., 1989; Mindell, Owens, & 
Carskadon, 1999; Smaldone, Honig, & Byrne, 2007; Stein, Mendelsohn, Obermeyer, Amromin, 
96 
 
& Benca, 2001). Epidemiologic and laboratory studies suggest that an adequate amount of 
sleep for school-aged children is around 9 to 10 hours per night (Anders, Sadeh, & Appareddy, 
1995; Heussler, 2005; Mindell & Owens, 2009). However, studies with American school-aged 
children show that they are getting less than the recommended amount (Buckhalt, El-Sheikh, & 
Keller, 2007; Buckhalt, El-Sheikh, Keller, & Kelly, 2009; Sleep in America Poll, 2004; Spilsbury, 
Storfer-Isser, Drotar, Rosen, Kirchner, & Redline, 2005). Given consistent links between sleep 
problems (i.e., short sleep duration, reduced sleep quality) and numerous domains of 
psychological adjustment (Gregory, Caspi, Eley, Moffitt, O?Connor, & Pouton, 2005), physical 
health (Redwine, Hauger, Gillin, & Irwin, 2000), social problems (Bates, Viken, Alexander, 
Beyers, & Stockton, 2002) and cognitive functioning (Buckhalt, El-Sheikh, & Keller, 2007; 
Sadeh, 2007), sleep of typically developing children including predictors and sequelae warrants 
further study. Thus, in this study we focus on associations between non-clinical sleep 
disruptions (i.e., reduced sleep amount and worse sleep quality) and non-clinical problems in 
children?s cognitive performance (e.g., inattention, reduced reaction time responses, short term 
memory lapses). That is, sleep and cognitive functioning problems in this paper are defined 
along a continuum and in comparison to other children within this community sample.                                     
 Multiple studies conducted with adults and children report links between sleep 
disruptions (e.g., short sleep, increased night wakings) and poorer cognitive performance, 
including difficulties with executive functioning (Blunden, Lushington, & Kennedy, 2001; Gozal, 
1998; Huntley & Lewin, 2006; Sadeh et al., 2002; Sadeh et al., 2003). Two primary components 
of executive functioning are sustained attention (i.e., ability to maintain focus on a task for 
extended periods; Johnsen et al., 2002) and working memory (i.e., continuous updating and 
prolonged storage of information; Logie, Zucco, & Baddeley, 1990). Executive functions stem 
from the neural systems of the prefrontal cortex, a part of the brain that is particularly 
susceptible to sleep disruptions (Curcio, Ferrara, & De Gennaro, 2006; Dahl, 1996; Drummond 
97 
 
& Brown, 2001; Horne, 1988; Harrison, & Horne, 1998), and are important contributors to 
behavior regulation (Cole, Usher, & Cargo, 1993; Pennington & Ozonoff, 1996). 
Research conducted with adults consistently links impairments in attention and cognition 
with restricted or fragmented sleep (Pilcher & Huffcut, 1996). Adult fMRI studies suggest that 
sleep may play an important role in the consolidation of newly formed memories (Walker & 
Stickgold, 2006). In regards to links between children?s sleep and cognitive functioning, clinical 
research indicates improvements in attention and working memory when sleep disorders are 
treated (Chervin, Ruzicka, Giordan, Weatherly, Dillon, Hodges et al., 2006). Further, research 
with non-clinical child populations suggests that small reductions in sleep duration or mildly 
disrupted sleep can impact children?s attention, processing speed, working memory, and 
response times (Buckhalt, El-Sheikh, & Keller, 2007; Sadeh, Gruber, & Raviv, 2003). Further, 
poor sleep may impact cognitive functioning two-years later (Buckhalt, El-Sheikh, Keller, & 
Kelly, 2009), and worsening of sleep problems over time predict poorer cognitive functioning 
later in development (Bub, Buckhalt, & El-Sheikh, in press). Sleep debt or deprivation, occurring  
when the optimal and actual amounts of sleep are not congruent (Van Dongen, Rogers & 
Dinges, 2003), has lasting effects on functioning and adjustment (Carskadon, 2002). Reviews of 
the literature suggest that reduced amounts of sleep and active or fragmented sleep influence 
many aspects of cognitive functioning, including tests of neurocognitive functioning and reaction 
time (Buckhalt, Wolfson, & El-Sheikh, 2009; Curcio, Ferrara, & De Gennaro, 2006; Mitru, 
Millrood, & Mateika, 2002; Sadeh, 2007; Wolfson & Carskadon, 2003). Problems with 
neurobehavioral functions such as sustained attention are particularly important at ages when 
skills are being acquired (Van Leijenhorst, Crone, & Van der Molen, 2007). Thus, understanding 
how and why sleep is related to children?s cognitive functioning is essential for children?s future 
achievement.  
    Experimental sleep-cognitive studies conducted with non-clinically diagnosed children 
are limited, yet results are similar to the findings with adults. Specifically from pre- to post-test, 
98 
 
children (M age = 10.6 years) in the extended sleep group (i.e., an additional one hour of sleep 
per night for three nights) showed memory and attention improvement, whereas the children in 
the sleep restriction group (i.e., one less hour of sleep per night for three nights) exhibited 
significantly slower response times on a reaction time task (Sadeh, et al., 2003). In addition to 
reduced reaction times, acute sleep restriction (i.e., one night of four hours of sleep) in children  
(N = 93; M age = 7.9, SD =1.4 years) is linked to increased omission errors and reduced signal 
detection on a neurocognitive task as compared to a baseline taken two weeks prior to the 
sleep restriction (Huntley & Lewin, 2006). Both acute (i.e., 5 hours of sleep one night) and 
prolonged (6.5 hours per night for 6 nights) sleep restriction in children are associated with 
attention impairments (Fallone, Seifer, Acebo, & Carskadon, 2000; Randazzo, Muehlbach, 
Schweitzer, & Walsh, 1998). 
Although a growing literature with typically developing school-aged children consistently 
links disrupted or shortened sleep with reductions in cognitive functioning (Buckhalt, Wolfson, & 
El-Sheikh, 2009; Curcio, Ferrara, & De Gennaro, 2006), the relation between specific facets of 
sleep (e.g., quality, duration) and cognitive domains (e.g., reaction time, working memory load) 
is less understood. Further, findings stemming from actigraphy do not yield the same 
associations found with self-reported sleep measures. For example, actigraphic measures of 
sleep including reduced sleep efficiency (a measure of sleep quality) and increased sleep 
latency (how long it takes to fall asleep once actively trying) in children (N = 60; 6- to 13- years-
old) were related to incorrect responses on visual and auditory cognitive tasks; short sleep 
duration was associated with a reduction in working memory at the highest memory load 
(Steenari et al., 2003). From these findings, the authors speculate that sleep quality may be 
more closely related to working memory performance than sleep duration. However, Meijer and 
colleagues (2000) sampled 449 Dutch children (9-to 14- years-old) and found no relation 
between subjective reports of sleep quality or duration and scores on a short attention task. 
Using polysomnography (PSG), Randazzo and colleagues (1998) found a single night of sleep 
99 
 
restriction for children ages 10 to 14 years impaired their verbal creativity and abstract thinking; 
no effect was found for less complex cognitive tests assessing attention and reaction time. 
These variations in findings suggest that studies should incorporate objective (i.e., actigraphy; 
PSG) and self-reports of sleep, while also examining a wide array of cognitive outcomes with 
varying levels of difficulty.  
Adult and child research indicates individual difference variables contribute to the 
strength of the connection between sleep and cognitive performance. For example, VanDongen 
(2006) found the association between adult sleep deprivation and poor cognitive performance to 
be moderated by the individual amount of sleep need, the timing of the circadian process, and 
the basal level of alertness. For children, research examining possible moderators of this 
relation are limited to race/ethnicity and socio-economic status (SES) (e.g., Buckhalt, Keller, & 
El-Sheikh, 2007; Buckhalt, Keller, Kelly, & El-Sheikh, 2009), gender (Bub et al., in press) and 
daytime sleepiness (Geiger, Achermann, & Jenni, 2010). For example, Buckhalt and colleagues 
(2007; 2009) examined demographic characteristics (i.e., multiple markers of socioeconomic 
status, race) as possible moderators in the sleep-cognitive link and results show that children 
from economically disadvantaged backgrounds or from African-American ethnicity were at a 
greater risk for cognitive problems (crystallized and fluid intelligence) when sleep disruptions 
were present than did their more advantaged and European-American counterparts. In Bub et 
al. (in press), girls were found to be more susceptible than boys to the negative impact of 
increases in sleep problems over time on cognitive performance. 
The evaluation of possible individual difference variables at the physiological level are 
important to consider because they can modulate the type and degree of maladjustment 
associated with stress exposure (Steinberg & Avenevoli, 2000); chronic sleep disruptions are 
considered a significant biological stressor (McEwen, 2006). It is important to understand how 
physiological regulation influences adjustment and cognitive outcomes (e.g., Cicchetti & 
Dawson, 2002) to develop appropriate intervention and prevention strategies. Parasympathetic 
100 
 
nervous system  activity (PNS), the down-regulating component of the autonomic nervous 
system (ANS),  has been established as a correlate and predictor of developmental outcomes 
(Beauchaine, 2001; Calkins & Dedmon, 2000; El-Sheikh, 2005; El-Sheikh & Whitson, 2006), 
the), including those related to sleep disruptions (see El-Sheikh, Erath, & Keller, 2007).  
PNS Activity         
The ANS is comprised of two branches, the PNS and the sympathetic nervous system 
(SNS; Porges, 1991). The PNS provides an inhibitory influence on the cardiac muscle (i.e., 
slows heart rate) via the vagus or 10th cranial nerve. Respiratory sinus arrhythmia (RSA), a 
measure of vagal tone, is used in this study to index the PNS?s influence on the heart. RSA 
represents the changes in heart rate associated with normal cycles of respiration (i.e., heart rate 
decreases during exhalation and increases during inhalation). RSA is considered an appropriate 
index of vagal nerve activity under normal conditions (e.g., sitting, breathing normally; Porges, 
2007) and is a suitable measure for research with typically developing children because it can 
be assessed non-invasively (Grossman & Taylor, 2007). RSA is a well-validated physiological 
marker of emotion regulation (Beauchaine, 2001; El-Sheikh & Erath, 2011; Porges, 2007; 
Thayer & Lane, 2009), is a correlate of a number of child adjustment outcomes, including 
cognitive functioning variables (Staton, El-Sheikh, & Buckhalt, 2009), and is often found to 
function as a physiological risk or protective factor during childhood (Beauchaine, 2001; Calkins 
& Dedmon, 2000; El-Sheikh, 2005; El-Sheikh & Whitson, 2006).  
The Polyvagal Theory (Porges, 2007), the guiding theory for this study, is the prominent 
theory in the area of vagal research and it posits connections between the ANS and cognition; 
also similar to the neuroanatomical integration of neural networks developed by Thayer and 
Lane (2007). Specifically, the Polyvagal Theory postulates that mammal?s first response when 
confronted with a challenge is to disengage the vagal brake (i.e., vagal withdrawal), which 
allows the individual to focus attention on the environment, obtain information, and/or engage 
social strategies to reduce the challenge/stress. Once the challenge is over, the PNS influence 
101 
 
can reengage the vagal brake thus reducing arousal. The ability to physiologically engage and 
disengage as the environment demands (e.g., the need for focused attention; Suess, Proges, & 
Plude, 1994) is important for optimal cognitive performance. We utilize the Polyvagal Theory 
and our extension of the theory to consider vagal tone as a moderator of effects (El-Sheikh & 
Erath, 2011) as guiding frameworks for our examination of vagal functioning as it interacts with 
child sleep disruptions to predict cognitive functioning.  
Baseline vagal tone reflects PNS activity at rest and the ability to sustain attention and 
maintain homeostasis under normal conditions (Porges, 1991). Supporting empirical research 
indicates that higher levels of baseline vagal tone are positively associated with more optimal 
adjustment outcomes (Rottenberg, 2007). That is, higher vagal tone maintains a low heart rate 
under normal conditions (Porges, 1991), which reduces the wear and tear on the heart and 
leads to more optimal health and adaptation (Whitson & El-Sheikh, 2003).  
Increases and reductions in heart rate by vagal pathways to promote regulatory and 
goal-directed behavior are referred to as RSA reactivity (Porges, 1995, 2007; Thayer & Lane, 
2000). When a challenge is encountered, the PNS can respond in one of two ways: (1) it can be 
withdrawn or suppressed, frequently referred to as vagal withdrawal, which leads to an increase 
in heart rate; or (2) it can be activated or augmented, which leads to a decrease in heart rate. 
Vagal withdrawal facilitates focused attention when needed (Huffman et al., 1998; Porges, 
2007) and the shift in metabolic resources to support information processing (Bornstein & 
Suess, 2000). Thus, vagal withdrawal is generally considered the most optimal response when 
confronted with a challenge. However, infants' attentiveness has been associated with vagal 
augmentation (i.e., jack-in-the-box play; DiPietro et al., 1992), and very high levels of vagal 
withdrawal have been linked with adjustment problems (Beauchaine, 2001). Examinations of 
associations between vagal reactivity and cognitive outcomes in children are few and warrant 
further investigation. Although vagal tone and vagal reactivity are often moderately correlated, 
each has shown differential relations to child outcomes (Calkins, 1997), therefore supporting the 
102 
 
inclusion of both forms of vagal activity towards a more thorough examination of child 
functioning.  
Vagal recovery, the change in RSA once a challenge has subsided, is a third component 
of vagal activity. The Polyvagal Theory proposes that a faster return to baseline levels following 
a challenge is an optimal response and empirical research on health and adaptation is 
supportive. Vagal recovery has not received much attention in the literature; however, adult 
research indicates low recovery (i.e., slower return to baseline levels) is associated with an 
increase in cardiovascular disease (Mezzacappa, Kelsey, Katkin, & Sloan, 2001). Only one 
study to our knowledge has examined vagal recovery in children (ages 4 to 7), and results 
indicate low vagal recovery predicts the inability to focus attention (Santucci, Silk, Shaw, 
Gentzler, Fox, & Kovacs, 2008). Further examination of vagal recovery and cognitive functioning 
is needed and our assessment of three indices of vagal functioning in this study is more likely to 
explicate relations examined than any single vagal index alone. 
Sleep and RSA Functioning 
  Few child studies have investigated associations between sleep and RSA functioning. 
Of those few investigations, vagal withdrawal to a lab stressor functioned as a protective factor, 
while vagal augmentation was shown to be a risk factor. For example, vagal augmentation 
during a reaction time task was linked to a higher occurrence of sleep problems (El-Sheikh & 
Buckhalt, 2005). In a previous paper with the first wave of this sample, vagal withdrawal to an 
inter-adult argument was a protective factor against child externalizing and depression 
symptoms in the context of sleep problems (El-Sheikh, Erath, & Keller, 2007). Further, children 
who exhibited less vagal withdrawal or increased vagal augmentation were more at risk for 
adjustment problems (e.g., externalizing symptoms) in the context of increased sleep 
disruptions (El-Sheikh et al., 2007).  
 
 
103 
 
RSA and Cognition 
  A greater ability to engage and disengage with the environment as needed is 
associated with more optimal vagal functioning (Porges, 2001; Thayer & Lane, 2000) and is 
necessary for optimal cognitive performance. Researchers who have examined associations 
between vagal indices and cognitive performance have addressed this question with infant and 
adult samples; few have assessed these relations with typically developing school-aged 
children. Of those few studies, higher baseline RSA was linked to greater sustained attention 
(Suess, Porges, & Plude, 1994) and better cognitive performance on standardized tests of 
cognitive abilities (Staton, El-Sheikh, & Buckhalt, 2009). Additionally, greater vagal variability 
(i.e., wider range of responses; higher baseline and greater vagal withdrawal) was predictive of 
better executive control (Mezzecappa, Kindlon, Saul, & Earls, 1998). However, Staton and 
colleagues (2009) did not find vagal reactivity during a continuous performance task to predict 
cognitive performance on the task or on the standardized cognitive assessments (i.e., WJ III). 
Only one study to our knowledge has demonstrated relations between longer vagal recovery 
and worse cognitive outcomes (i.e., inattention; Santucci et al., 2008). Thus, more examinations 
are needed to delineate relations among vagal indices and cognitive functioning parameters in 
school-aged children.  
New Directions 
This study advances prior sleep-cognitive literature by examining the moderating role of 
three important physiological regulation indicators: vagal tone (RSA), vagal reactivity (RSA-R), 
and vagal recovery (RSA recovery). Executive functions and PNS regulatory functions are 
known to be disrupted by sleep problems (El-Sheikh & Buckhalt, 2005; El-Sheikh, Erath et al., 
2007; Irwin, 2008), however sleep is not the only contributor to the neural networks that 
influence vagal functioning. Supportive of our examination of RSA indices as moderators of 
effects, variability in vagal functioning exists independent of sleep and RSA parameters are 
considered stable individual difference variables for school-aged children (El-Sheikh, 2005; 
104 
 
Salomon, 2005). Polyvagal Theory states optimal vagal functioning facilitates efficient regulation 
of attention and Dahl (1996) considers daytime regulation of arousal and emotion to affect 
sleep. In accordance with the Polyvagal Theory (Porges, 1998; Porges, 2007) and Dahl (1996), 
and our extension of these views, we expect vagal functioning may exacerbate or attenuate the 
effect of sleep problems on children?s cognitive functioning (see El-Sheikh et al., 2007; 
manuscript 1).  
It is our hypotheses that lower RSA, RSA augmentation (or reduced vagal withdrawal) in 
response to the challenge task, or a reduced RSA recovery following the lab task will interact 
with sleep problems to exacerbate the risk of poor cognitive functioning. Conversely, and 
consistent with a conceptualization of protection and vulnerability along a continuum (Brown & 
Barlow, 2005), higher RSA, RSA withdrawal, and a greater RSA recovery are expected to 
attenuate the association between sleep problems and poor cognitive performance. Given the 
scarcity of this literature, no specific hypotheses are advanced regarding the moderating role of 
the various vagal indices in relation to specific types of sleep disruptions (e.g., duration, quality). 
Of note, all of the cognitive outcomes that we examine involve executive functioning and require 
attention for better performance. We expect that cognitive tests tapping higher levels of 
executive functioning (e.g., highest working memory load) will be most likely influenced by vagal 
tone and its moderating effects (Staton et al., 2009). 
To conduct a rigorous test of our hypotheses, we examine actigraphy and children?s self-
reported sleep problems and assess multiple sleep parameters (i.e., duration, quality) as 
recommended in the literature (Sadeh, 2008). Actigraphy (Buckhalt et al., 2007; 2009; Sadeh, 
Gruber, & Raviv, 2000) and subjective reports of sleep (Buckhalt et al., 2007; 2009; Wolfson & 
Carskadon, 2003) have been linked to cognitive performance and thus their examination in the 
context of vagal functioning was considered important. Similarly, we administered an extensive 
and well-validated cognitive abilities test and a well-established reaction time task to allow for a 
more thorough examination of cognitive functioning in relation to sleep and vagal activity. 
105 
 
Examinations of multiple sleep parameters as they interact with three vagal functioning indices 
is imperative for a more nuanced explication of physiological and biological domains linked with 
various cognitive performance parameters.  
Method 
Participants 
 Three school systems in the Southeastern United States provided contact information for 
children enrolled in the third grade. Researchers then contacted parents by letter or phone. Via 
initial phone contact, researchers assessed if the children were eligible for participation in the 
study based on the following criteria: (1) they lived in a two-parent home for at least two years; 
(2) had no chronic physical illness; (3) were not diagnosed with ADHD, learning disability or 
mental retardation; or (4) did not have a history of diagnosed sleep problems. Of those eligible, 
66% agreed to participate. Children with acute illnesses (e.g., flu) were asked to postpone their 
participation until they were well.  
The sample at T1 consisted of 166 third grade children (45% boys). Children had a 
mean age of 8.72 (SD = .30), with 94% of the sample considered prepubertal and 6% early 
pubertal (Petersen et al., 1988). The majority of the children (77%) lived with both biological 
parents; 21% lived with their biological mother and a step-father; 2% lived with their biological 
father and a step-mother and marital partners had been living together for an average of 10.20 
years (SD = 5.30 years). Consistent with the recruiting school districts? demographic 
characteristics, 69% of parents reported being European-American (EA) and 31% reported 
being African-American (AA). Oversampling methods were used to ensure that families from 
both ethnic groups represented a wide range of socioeconomic levels. The sample was diverse 
in relation to socioeconomic status: 26% of the families were unskilled or semiskilled workers, 
44% were skilled workers, and 30% were minor professionals or professionals (based on 
Hollingshead 1975 criteria). Mothers reported the annual familial income at the following ranges: 
4% less than 10,000; 7% between $10,000 and $20,000; 25% between $20,000 and $35,000; 
106 
 
28% between $35,000 and $50,000; 25% between $50,000 and $75,000; and 11% more than 
$75,000. In terms of years of parental education, mothers had an average of 13.80 years (SD = 
2.82) and fathers had an average of 13.20 years (SD = 2.82), indicative of one year of post high 
school education or specialized training. 
 Of the original 166 participants, 132 children and their families (80 boys) participated two 
years later (T2; 80% of the original sample). Attrition was due to the inability to be located, 
hectic schedule, lack of interest, and geographic relocation.  At T2, children?s mean age was 
10.80 years (SD = .36). The ethnic composition of the sample was 70% EA and 30% AA. 
Pubertal status was 2.07 (SD = .56) for girls, which is indicative of an early pubertal status 
(Petersen et al., 1988). Boys were considered to be in prepubertal stages, with a mean pubertal 
status score of 1.49 (SD = .42). 
 In comparison to children who participated in the longitudinal follow up, those who 
dropped out of the study had lower scores on the Woodcock Johnson Test of Cognitive Abilities 
(WJ III, Woodcock, McGrew, & Mather, 2001) for Cognitive Efficiency, t(164) = - 2.73, p < .01, 
and came from families characterized by lower overall SES, t(164) = - 3.37, p < .001. Missing 
and retained participants did not differ in terms of ethnicity, gender, sleep parameters, or vagal 
indices. 
Procedures                                                                                                         
 This study is part of a larger longitudinal investigation of the role of the family 
environment (e.g., marital conflict) in relation to children?s adjustment, sleep, and health. Only 
the pertinent procedures are detailed. The study was approved by the institution?s Internal 
Review Board (IRB), and parents provided consent and children provided assent for 
participation. Families received monetary compensation for their participation. All procedures 
were identical at T1 and T2 unless otherwise indicated.  
 During both waves of data collection, research assistants delivered the actigraph (watch-
like devices that measure movement) to the child?s home and instructed the parents to place the 
107 
 
watch on the child?s non-dominant wrist just prior to bedtime and remove it upon waking for 
seven consecutive nights. A week of actigraphic assessments exceeds the recommended 
number of nights for the reliable measurement of typical sleep parameters (Acebo et al., 1999). 
During the week of actigraphy, researchers called parents daily to obtain children?s bed and 
wake times, which were used to corroborate actigraphy measures. Actigraphic data were 
collected only during the regular school year to avoid confounding variables (i.e., irregular sleep 
patterns due to less structured wake times). Due to the influence medications have on sleep 
(Qureshi & Lee-Chong, 2004), only data from medication-free nights were used in analyses.  
 Typically on the day following the last night of actigraphic sleep assessment, children 
and their families came to the university laboratory where children were individually 
administered the brief version of the Woodcock-Johnson III Cognitive Abilities Battery (WJ III; 
Woodcock et al., 2001), completed a 15-minute computerized reaction time task (i.e., Donders), 
participated in a physiological assessment session and completed an interview.  
 To reduce anxiety associated with attaching electrodes to obtain RSA data, researchers 
conversed with the child and a parent (typically the mother) while the electrodes and a 
pneumatic bellows belt were attached. This process took approximately ten minutes, which 
allowed a sufficient amount of time for the child to relax prior to the collection of physiological 
data. The parent and researcher then left the room and the child was given two additional 
minutes to acclimate to the setting without others present before data collection began (the child 
was asked to relax while sitting down quietly). Immediately following, a three minute baseline 
(resting level) was obtained. Three minutes are considered more than adequate for assessment 
of RSA (Bar-Helm et al., 2000). To assess RSA reactivity, children completed a mildly stressing 
problem-solving task, namely the star-tracing task (Matthews, Woodall, & Stoney, 1990). This 
task requires the child to trace a star on a sheet of paper while only seeing the star in a mirror. 
In addition to the difficulty associated with the reverse properties that the mirror adds, children 
were only given three minutes to trace the star, which increases task demands. This task has 
108 
 
been used in numerous investigations to evoke ANS reactivity in children (El-Sheikh, et al., 
2009; Matthews, Rakaczky, Stoney, & Manuck, 1987; Matthews, Woodall, & Stoney, 1990). At 
T2, an additional three minute assessment of RSA following the star tracing task allowed for the 
examination of RSA recovery. Following the physiological session, children were given a break 
and a snack and were then interviewed by a research assistant regarding their sleep habits.  
Measures                                                                                           
 Children?s Objective Sleep. The actigraph wristwatch, Motionlogger Octagonal Basic 
(Ambulatory Monitoring, Inc., Ardsley, NY), is small (36 mm X 38 mm X 13 mm) and lightweight 
(45 g). Movement during sleep was continuously recorded and collected in 1-min epochs using 
zero crossing mode. The raw data was processed with the Octagonal Motionlogger Interface 
with ACTme Software, and the Analysis Software Package (ActionW2 or AW2). Using the 
Sadeh algorithm, which has established validity with children (Sadeh, Sharkey, & Carskadon, 
1994) and good reliability with this actigraph (Acebo, Sadeh, Seifer, Tzischinsky, Wolfson, 
Hafer, et al., 1999), one-minute epochs were scored as either sleep or wake. Guidelines used 
for determining sleep onset were developed at the E.P. Bradley Sleep Laboratory at Brown 
University (Acebo & Carskadon, 2001).  
 Sleep variables derived via actigraphy included: (a) Sleep Minutes - total minutes scored 
as sleep during the Sleep Period, providing a measure of sleep duration; (b) Sleep Efficiency - 
percentage of minutes scored as sleep, providing a measure of sleep quality; and (c) Long 
Wake Minutes ? number of wake episodes during the night at least 5 minutes in duration, 
providing a second measurement of sleep quality.   
 Cronbach?s alphas were used to assess the stability of children's sleep over the week of 
actigraphy. The reliability coefficients for T1 and T2 were: Sleep Minutes = .78 and .86; Sleep 
Efficiency = .86 and .93; Long Wake Episodes = .85 and .88.  All of the actigraphy variables 
exceed the recommendation of Acebo and colleagues (1999) for achieving a stability of .70 or 
greater.                                                                   
109 
 
 At T1, 74% of the children had actigraphy data for all seven nights. Due to actigraph 
malfunctions, four children (2%) did not have any actigraphy data. The remaining 24% had data 
for fewer than seven nights (M = 5.33 days; SD = 0.97) due to use of allergy medication, 
forgetting to put the watch on prior to bed, or sleeping while riding in a car.  At T2, the majority 
of the children (56%) had all seven nights of actigraphy data, 25% had data for six nights, 14% 
had data for five nights, and 5% had data for four or fewer nights. These percentages of valid 
actigraphy data are considered to be very good (Acebo et al., 1999) and are comparable to 
similar studies. Sleep parameters used in analyses were based on averaged data across all 
available nights of actigraphic assessments. 
       Children?s Subjective Sleep. Children completed the Sleep Habits Survey (SHS; 
Wolfson & Carskadon, 1998) via interview, which assesses children?s perceptions of their sleep 
within the last two weeks. This questionnaire is commonly used with children and adolescents 
(Acebo & Carskadon, 2002; Buckhalt, El-Sheikh, & Keller, 2007; Buckhalt, El-Sheikh, Keller, & 
Kelly, 2009), and has good reliability and validity (Carskadon, Seifer, & Acebo, 1991; El-Sheikh 
& Buckhalt, 2005; Wolfson, Carskadon, Acebo, Seifer, Fallone, Labyak, et al., 2003). Two 
subscales were used in analyses: (1) the Sleep/Wake Problems Scale, which is composed of 10 
items that assess the frequency of sleep problems (e.g., oversleeping and staying up late at 
night); and (2) the Sleepiness scale, comprised of 10 items, which measures the extent to which 
children struggle to stay awake during their daily activities. One item (i.e., sleepiness while 
driving) on the Sleepiness scale was deleted due to the child?s age. Items were rated on a 5-
point Likert scale, and sums for each subscale were computed. Higher scores are indicative of 
more sleep-related problems and daytime sleepiness. For this sample, the Sleep/Wake 
Problems Scale coefficient alphas were .67 and .94, respectively. The Sleepiness Scale had a 
reliability coefficient of .70 at T1 and .79 at T2.  
 RSA, RSA-R, and RSA Recovery. To obtain heart rate, electrodes were placed axially 
on the left and right rib cage, approximately 10 to 15 cm below the armpits. A pneumatic bellows 
110 
 
was placed around the chest and held in place using a metal beaded chain. Physiological 
acquisition equipment and software were from James Long Company. The bioamplifier was set 
for band-pass filtering with half power cut-off frequencies of 0.1 and 1000 hertz, and the signal 
was amplified with a gain of 500 for electrocardiography (ECG). The pneumatic bellows, used to 
assess respiration, was attached to a pressure transducer with a band-pass of direct current to 
4000 hertz, to minimize phase or time shifts in the measurement. Data were digitized at a 
sampling rate of 1000 Hz with a 12-bit analog-to-digital board in a laboratory computer.                
An automated algorithm was used to identify the R-waves in the ECG data. Manual 
correction of missed or misidentified R-waves occurred in the rare event that it was needed 
using an interactive graphical program. R-wave times were converted to inter-beat intervals 
(IBIs) and resampled into equal time intervals of 125 ms. Prorated IBI values were used for 
analysis of the mean and variance of heart period as well as the processing of heart period 
variability due to respiratory sinus arrhythmia (RSA).  
The peak-valley method was used to compute RSA, and RSA is reported in seconds. 
The peak-valley method is one of several acceptable procedures for quantifying RSA (Berntson 
et al., 1997) and this procedure highly correlates with spectrally derived measures of RSA 
(Galles, Miller, Cohn, & Fox, 2002). To identify inspiration and expiration onset times and 
relative tidal volume, the respiration signal was used. RSA was calculated as the difference in 
IBI between inspiration and expiration onset for consecutive respiratory cycles. Specifically, 
RSA was computed twice per breath, once from inspiration to expiration and again from 
expiration to inspiration, and then averaged across respiratory cycles to ensure that any 
underlying trends in heart period had little impact on the computed mean RSA. To derive RSA-R 
in response to the star tracing task, a change score between RSA during the task and RSA at 
baseline (i.e., task minus baseline) was computed. Negative RSA-R scores indicate RSA 
withdrawal to the task, whereas positive RSA-R scores represent augmentation to the star 
tracer. In all analyses examining RSA-R, we controlled for RSA. RSA recovery was calculated 
111 
 
as the difference between RSA during the three minutes after the star tracing task and RSA 
during the initial baseline (Santucci, Silk, Shaw, Gentzler, Fox, & Kovacs, 2008). A higher RSA 
recovery score represents a faster return to individual baseline levels of vagal tone after the 
challenge had subsided, whereas a lower recovery score represents slower return to basal RSA 
levels.   
 Cognitive Functioning. The Woodcock-Johnson III Tests of Cognitive Abilities (WJ III; 
Woodcock et al., 2001) is a well-normed measure of general intellectual ability (g), crystallized 
intelligence (Gc), fluid reasoning (Gf), and processing speed (Gs). General intellectual ability 
refers to non-specific abilities that affect the performance of most tasks, whereas Gc is thought 
to reflect skills acquired through knowledge and experience (e.g., verbal ability). Fluid reasoning 
is the capacity to think logically and solve novel problems and processing speed involves the 
ability to automatically and fluently perform tasks. The following tests were administered: 
Concept Formation (CF); Numbers Reversed (NR); Auditory Working Memory (AWM); and 
Decision Speed (DS). CF is a novel controlled-learning test that requires the use of categorical 
reasoning based on principles of inductive logic. This test assesses fluid reasoning and one?s 
flexibility in shifting mental sets; the latter is an aspect of executive functioning. NR requires that 
numbers be held in working memory while executing a mental operation (i.e., reversing the 
sequence), and examines short-term memory. AWM is another test of short-term and working 
memory, but it also requires the use of divided attention. Specifically, a series of numbers and 
words are presented in a random order, and the respondent is asked to recall the words in the 
correct order they were stated, and then the numbers. For the last test, DS, one is asked to 
select two conceptually similar items from a row of alternatives. Multiple rows of items are 
presented and one is only given 3 minutes to complete as many rows as possible, thus 
assessing visual processing speed. All tests started at a basic skill level and increased in 
complexity as the test progressed. These tests were chosen to comprehensively examine 
112 
 
cognitive functioning processes that may be impacted by sleep disruptions, specifically 
attention, working memory and simple processing speed.     
Reaction Time. Speed and variability of response speed across a number of trials was 
assessed using a computerized reaction time task known as Donders (Psych/Lab; Abrams, 
2004). Multiple conditions were used to obtain reaction times. Specifically, the child was seated 
at a computer and a researcher instructed the child to focus on a ?1? in the center of the 
computer screen. The child placed the right index finger on the ?/? key, and the left index finger 
was placed on the ?Z? key. The child was told to follow the instructions presented on the screen, 
which were also read to the child by the researcher. A short tone sounded if the child responded 
too early, too late, not at all, or if a key was pressed when one was not required, a short tone 
was sounded.  
To assess simple reaction time (Donders Type A), the child was told to press the ?/? key 
as quickly as possible when the ?x? appears in the box on the right. For the assessment of 
choice reaction time (Donders Type C), two boxes were presented and either could have the ?x? 
appear, but children were told to only respond (by pressing ?/?) when the ?x? appears in the right 
box. The difficulty is increased in this task due to the monitoring of two boxes and having to 
inhibit a response with the left finger. In a third condition, choice reaction time 2 (Donders Type 
B), two boxes were presented and the ?x? could appear in either box. Children were instructed to 
press the ?/? key with the right finger when the ?x? appears in the right box or press the ?z? key 
with the left finger if the ?x? appears in the left box. Conditions were presented in a fixed order 
with the following parameters: 10 practice trials per condition; 20 test trials per condition; a 500 
ms inter-trial interval; 100 ms minimum allowed RT; and 1,500 ms maximum allowed RT.  Trials 
were repeated if responses were faster or slower than the criteria. Test trial averages and 
standard deviations were computed for each condition and were derived for analyses. The 
mean scores represent the average response time, whereas the standard deviation scores 
represent vigilance to the task. Examination of children?s responses during the three conditions 
113 
 
(types A, B, and C) indicated that both the means (rs = .66 ? .76, ps < .001), and SDs (rs = .37 
? .42, ps < .001) were significantly correlated. Thus, to reduce the number of analyses, one 
averaged mean and one SD were derived and used in all subsequent analyses.   
 Body Mass Index. Children?s height and weight (via a Sunbeam scale) were measured 
during the laboratory visit and used to calculate BMI [703 x (weight ? height2); 
http://www.cdc.gov]. Based on the Center for Disease Control?s classifications, 1% of the T1 
sample was categorized as underweight, 61% was of normal weight, 23% were overweight, and 
15% were obese. At T2, 2% were classified as underweight, 72% were of normal weight, 13% 
were overweight, and 13% were obese.  
Demographic Variables. Child age, ethnicity, sex, and familial SES were obtained from 
a parent over the phone prior to participation. Familial SES was derived using parental 
education and occupation as indexed via the Hollingshead Index (Hollingshead, 1975).  
Results 
Preliminary Analyses 
 Means and standard deviations for primary study variables are presented in Table 1. 
Descriptive analyses indicated on average, children had fairly low sleep efficiency (87% at T1; 
89% at T2); ? 90% is optimal (Sadeh et al., 2000). Similarly, at both time points children 
obtained well below the recommended average amount of sleep for children of this age (~10 
hours; www.cdc.gov/features/sleep/), with the average sleep duration similar across time points 
(~7.25 hours per night). Most children exhibited RSA withdrawal to the star tracing task (69% at 
T1; 49% at T2); 13% at T1 and 12% at T2 showed no change and 18% at T1 and 39% at T2 
exhibited RSA augmentation to the task. At T2 (recall that recovery data were not derived at 
T1), 45% of children had RSA levels during the post-task baseline that were lower than those 
during the initial baseline; 16% fully recovered (the difference between the initial and post-task 
baselines was zero); and 38% had RSA levels higher during the post-task than the initial 
baseline. Children in the sample scored the lowest on Concept Formation at both waves and the 
114 
 
highest on Auditory Working Memory at T1 and Decision Speed at T2; all scores were in the 
normative range. 
 Relations between the main study variables (sleep, vagal indices, cognitive functioning, 
reaction time) and several confounding variables (BMI, child age, gender, ethnicity, 
socioeconomic status, season of actigraph collection, and pubertal status) were examined. 
Lower BMI was associated with higher RSA at both time points. Child age was associated with 
most of the cognitive functioning variables at T1 (i.e., older children typically performed better 
than younger children), however this association was not found at T2. Gender differences in 
actigraphic and self-reported sleep patterns were found at T1, with boys having more sleep 
problems than girls. AT T2, these gender differences were not apparent. At both assessments, 
AA children reported more sleepiness and sleep/wake problems than their EA counterparts. 
Actigraphic results indicated that AA children had fewer long wake episodes than EA children at 
T1 and less sleep duration at T2. At T2, girls had higher decision speed scores and a slower 
mean reaction time than boys at T1, with reaction time associations holding across waves.  
 Table 2 shows associations among study variables. As expected at cross-sectionally 
and longitudinally, actigraphy-based sleep variables were highly intercorrelated and so were the 
self-report measures of sleep. Although sleep efficiency and long wake minutes were related, 
these are considered to be separate constructs in that long wake minutes can be used as an 
index of insomnia, whereas sleep efficiency is percentage of active versus non-active sleep over 
the course of the sleep onset to sleep offset. The WJ III variables and the reaction time 
measures were also highly correlated within and across time points. RSA indices were 
correlated within each timepoint, however there were no significant associations between RSA-
R at T1 and RSA-R at T2; RSA was correlated over time with itself. RSA and RSA-R were not 
related to either sleep or cognitive variables at T1, however at T2 RSA-R was cross-sectionally 
related to children's sleep efficiency, sleep minutes, and self-reported sleepiness.  
 
115 
 
Analysis Plan                                                                                                                                                                                                            
 AMOS 17.0 (Arbuckle, 2007) was used to fit a series of path models. Path analysis was 
chosen because unlike multiple regression it employs maximum likelihood estimation to handle 
missing data, which improves the power. Additionally, the assumption that the data are missing 
completely at random (MCAR; Little & Rubin, 2002) is more relaxed in path analysis than in 
multiple regression.  
In each path model, residuals of endogenous variables and significantly related 
exogenous variables were allowed to correlate. The covariates for each model were entered as 
the last paths in the model. Covariates for each model included: season of actigraphic 
assessment, child gender, ethnicity, age, BMI, and familial SES; longitudinal models also 
controlled for the autoregressive effects. Data points that were considered to be outside the 
normal distribution (i.e., SDs greater than 4) were removed from the dataset (n = 2 at T1; n =1 
at T2). After removing the univariate outliers, variables were also checked for multivariate 
outliers. Results did not indicate any multivariate outliers. 
Cross-sectional models were fit for each objective sleep parameter (sleep efficiency, 
long wake episodes, sleep minutes) and each cognitive outcome (concept formation, numbers 
reversed, auditory working memory, decision speed, reaction time M, reaction time SD) 
resulting in 18 models. Moreover, self-reported sleep parameters were examined 
simultaneously (sleepiness, sleep/wake problems) for each of the six aforementioned cognitive 
outcomes; a total of six models. All models contained three vagal indices (RSA, RSA-R, RSA 
recovery), the main effects and the associated interactions with the sleep parameter(s). Control 
variables were added to the models: season of sleep assessment, child age, sex, ethnicity, BMI, 
and SES status. All predictor and control variables were centered prior to inclusion in the path 
model.  
Longitudinal models were also fit separately for each objective sleep parameter at T1 
predicting each cognitive outcome variable at T2; a total of 18 models. An additional six models 
116 
 
were fit in which sleepiness and sleep/wake problems at T1 predicted T2 concept formation, 
numbers reversed, auditory working memory, decision speed, reaction time mean, and reaction 
time variability. All longitudinal models included two T1 vagal parameters (RSA, RSA-R), the 
main effects, the associated interactions with the sleep parameter(s), and the autoregressive 
effect (T1 cognitive outcome). Controlling for prior levels of the predicted outcome helps to 
reduce bias in parameter estimates, allows for conclusions about change in the predicted 
variable, and provides knowledge about the direction of effects (Cole & Maxwell, 2003). An 
example of the models is presented in Figure 2.  
Actigraphy-based sleep variables were examined separately based on calls in the 
pediatric sleep literature to include multiple specific sleep variables in research to facilitate 
comparisons across studies (Berger, Wielgus, Young-McCaughan, Fischer, Farr, & Lee, 2008; 
DeWald, Meijer, Oort, Kerkhof, & Bogel, 2010) and to further delineate associations between 
sleep parameters and child functioning. Additionally, separate examinations of duration, quality 
and sleepiness have been recommended because research indicates that these are separate 
sleep domains (DeWald et al., 2010). Further, it is suggested that researchers examine multiple 
sleep parameters including actigraphy-based and self-reported measures of sleep (Sadeh, 
2011). Due to multicollinearity and power considerations (Babyak, 2004; Whisman & 
McClelland, 2005), each cognitive outcome was fit separately.  
Considering the large number of models fit (n = 24 cross-sectional; n = 24 longitudinal) 
and to reduce the likelihood of Type 1 error, we only report and interpret findings when 
significant interactions represented at least 10% of the number of interactions examined for that 
cognitive outcome (e.g., ? 10% of significant findings across all models associated with concept 
formation). Models were considered an acceptable fit if at least two of the three following criteria 
were met: ?2/df < 2, CFI > .95, and RMSEA < .08, p <.05 (Arbuckle & Wothke, 1999; Browne & 
Cudeck, 1993). All fitted models reported satisfied these criteria.    
117 
 
Significant interactions were plotted at high (+1 SD) and low (-1 SD) levels of the 
predictor and moderator as recommended by Aiken and West (1991). Of note, a very small 
percentage of the sample exhibited RSA augmentation to the task. Thus, the depiction of the 
less optimal RSA reactivity group (i.e., Low RSA Withdrawal) includes children who exhibited no 
change, augmentation, or less vagal withdrawal in comparison to their counterparts. The 
interaction utility developed by Preacher, Curran, and Bauer (2006) was used to graph the 
interactions and estimates plotted were obtained from the fitted models. Significant interactions 
indicate that there is a statistically significant difference between the plotted lines (e.g., the 
association between sleep and cognitive functioning at high levels of the moderator is 
significantly different from the link at low levels of the moderator).  We also tested each slope to 
determine whether it was significantly different from zero; significant slopes are indicated in the 
figures.                                
Cross-Sectional Examinations of T2 RSA Indices as Moderators of Associations 
between Children?s Sleep and their Cognitive Performance  
 Concept Formation. To examine whether RSA, RSA-R, or RSA recovery following the 
task moderated the cross-sectional pathway between children?s sleep parameters and their CF 
scores, interaction terms were added to the model. None of the four models fit indicated a 
significant moderating role of any of the three RSA indices.  
 Numbers Reversed. Analyses yielded no significant moderation effects for RSA, RSA-
R, or RSA recovery in the link between children?s sleep and their cognitive performance indexed 
by NR scores.    
 Auditory Working Memory Scores. No moderation effects were observed for any of 
the vagal indices.  
 Decision Speed Scores. Analyses of children?s performance on the WJ III decision 
speed test indicated that RSA-R was a significant moderator in three of the four models 
118 
 
examined (75% of possible relations); no moderation effects were found for either RSA or RSA 
recovery. 
 RSA-R moderated the link between sleep efficiency and DS (? = - .33, p < .05), 
explained 6% of unique variance in DS (?2 = 59.45, p < .06, ?2/df = 1.35, CFI = .92, RMSEA = 
.00). The full model explained 14% of the variance in DS. As depicted in Figure 3A, regardless 
of sleep efficiency, children who exhibited RSA withdrawal to the star tracing task showed better 
performance on the DS test than those who exhibited RSA augmentation. Further, for children 
who exhibited RSA augmentation, sleep efficiency was negatively associated with DS.  
 Similar to the pattern of effects observed for sleep efficiency, the association between 
long wake episodes and DS was moderated by children?s RSA-R (? = .31, p < .05), which 
explained 4% of unique variance in DS (?2 = 62.29 p = .26, ?2/df = 1.11, CFI = .96, RMSEA = 
.00); the full model accounted for 17% of the variance in DS. Specifically, regardless of the 
quality of sleep, children who exhibited RSA withdrawal to the lab challenge performed better 
than their counterparts on the DS test (Fig. 3b). Children who showed RSA augmentation during 
the lab challenge had a positive association between long wake episodes and DS and the slope 
representing this association was significantly different from zero.   
 RSA-R also moderated the relation between sleep minutes and DS (? = - .29, p = .05) 
and explained 4% of unique variance in DS (?2 = 61.78 p = .04, ?2/df = 1.40, CFI = .89, RMSEA 
= .04). The full model accounted for 14% of the variance in DS. As shown in Figure 3C, in the 
context of longer sleep duration children who exhibited RSA withdrawal to the star tracing task 
performed better on the DS test than those who exhibited RSA augmentation.  
RSA activity did not moderate relations between Sleepiness or Sleep/Wake Problems 
and DS.  
 Reaction Time Means and SDs. RSA parameters did not moderate the link between 
children?s sleep (objective and subjective) and their performance on the reaction time task (M or 
SD scores).  
119 
 
Longitudinal Examinations of T1 RSA Indices as Moderators of Associations between 
Children?s Sleep at T1 and their Cognitive Performance at T2 
 Concept Formation. To examine whether RSA or RSA-R moderated the longitudinal 
pathway between children?s sleep parameters (both objective and subjective) and their CF 
scores, interaction terms were added to the path models after the inclusion of the main effects. 
One of the four models (25%) was significant for RSA and 50% were significant for RSA-R.  
 The relation between sleep efficiency at T1 and children?s scores on the CF test at T2 
varied based on children?s RSA (? = - .14, p = .05) and explained 2% of unique variance in CF 
(?2 = 74.85, p = .02, ?2/df = 1.43, CFI = .88, RMSEA = .05). A total of 45% of the variance in CF 
was accounted for by the full model. Specifically, in the context of low sleep efficiency, children 
who had high levels of RSA performed better on concept formation two years later than children 
who exhibited low RSA (Fig. 4A). 
 RSA did not function as a moderator in the longitudinal pathway between long wake 
episodes at T1 and children?s CF at T2, but RSA-R did (? = - .20, p < .01) and explained < 1% 
of unique variance  in CF (?2 = 94.13 p = .001, ?2/df = 1.71, CFI = .76, RMSEA = .07). The full 
model accounted for 44% of the variance in CF. Specifically, children who expressed RSA 
withdrawal to the lab task had a significant positive association between T1 long wake episodes 
and T2 CF (Fig. 4B). The pattern of effects was only partially consistent with expectations: at 
high levels of sleep problems, children who showed RSA withdrawal performed better than 
children who exhibited augmentation; yet the no clear pattern was evident at lower levels of 
sleep problems.  
The association between sleep minutes at T1 and CF at T2 was not moderated by either 
RSA or RSA-R. 
RSA-R (but not RSA) moderated the link between self-reported sleepiness at T1 and CF 
scores at T2 (? = - .31, p < .001) and explained 5% of unique variance in CF (?2 = 134.92 p = 
.001, ?2/df = 1.56, CFI = .83, RMSEA = .06); a total variance of 51% was accounted for by the 
120 
 
full model. As depicted in Figure 4C, children who exhibited RSA withdrawal had similar CF 
scores at T2 regardless of their level of sleepiness reported at T1. However, consistent with 
expectations, a negative association was found between sleepiness and CF for children who 
showed RSA augmentation and the slope representing this relation was significantly different 
from zero.   
No moderation effects were found between children?s self-reported sleep/wake problems 
at T1 and their concept formation scores at T2.  
 Numbers Reversed. Analyses of children?s performance on the WJ III NR test indicated 
that RSA-R, not RSA, moderated the longitudinal pathway between sleep efficiency and NR (? 
= .20, p < .05) and explained < 1% of unique variance in NR (?2 = 67.98 p = .07, ?2/df = 1.31, 
CFI = .89, RMSEA = .04); a total of 34% of the variance in NR was accounted for by the full 
model. Results indicated a negative longitudinal association between children?s sleep problems 
and NR test scores for children who exhibited RSA withdrawal and the slope representing this 
association was significantly different from zero (Fig. 5A). The slope representing the 
association between sleep efficiency and NR was not significantly different from zero for 
children who showed RSA augmentation.   
 The longitudinal relation between long wake episodes and NR was not moderated by 
RSA, but RSA-R did act as a moderator of this association, (? = - .21, p < .001) and explained 
2% of unique variance in NR (?2 = 90.05 p = .00, ?2/df = 1.63, CFI = .73, RMSEA = .06). The full 
model accounted for 6% of the variance in NR. Specifically, in the context of high sleep 
disruptions at T1, children who showed RSA withdrawal to the laboratory challenge performed 
better on the NR scores than their counterparts with RSA augmentation (Fig. 5B). Further, the 
longitudinal association between long wake episodes and children?s cognitive performance was 
only significant for those exhibiting RSA withdrawal. 
 The link between sleep minutes at T1 and children?s scores on the NR test T2 was not 
moderated by either RSA or RSA-R.  
121 
 
 RSA did not moderate the longitudinal association between child-reported sleepiness 
and NR. However, RSA-R did moderate the relation, (? = - .27, p < .001) and explained 2% of 
unique variance in NR (?2 = 144.86 p = .00, ?2/df = 1.68, CFI = .77, RMSEA = .06). A total of 
37% of the variance in NR was accounted for by the full model. As depicted in Figure 5C, in the 
context of high rates of sleepiness at T1, children who exhibited RSA withdrawal to the task 
performed better showed better cognitive performance two years later than their counterparts 
who showed RSA augmentation. For those children who showed RSA augmentation, there was 
a negative association between child-reported sleepiness and NR scores and this slope was 
significantly different from zero.    
 No moderation findings were evident for RSA in the longitudinal pathway between 
sleep/wake problems and NR. Conversely, RSA-R did moderate the sleep-cognitive link, (? = 
.16, p = .05) and explained 1% of unique variance in NR (?2 = 40.28 p = .41, ?2/df = 1.68, CFI = 
.77, RMSEA = .06); 37% of the total variance in NR was accounted for by the model. The 
pattern of effects was not consistent with our hypotheses (Fig. 5D).  
 Auditory Working Memory. Examination of children?s scores on the WJ III AWM test 
yielded one significant moderation effect for RSA-R (25% of the possible relations); no 
moderation effects were found for RSA. 
 Neither RSA nor RSA-R moderated the association between any of the actigraphy-
based sleep measures (sleep efficiency, long wake episodes, sleep minutes) at T1 and 
children?s AWM scores two years later.  
 RSA-R moderated the link between subjective reports of sleepiness and AWM, (? =  
- .32, p < .001) and explained 6% of unique variance in AWM (?2 = 136.745 p = .00, ?2/df = 1.59, 
CFI = .78, RMSEA = .06); the full model accounted for 28% of the variance in AWM. Regardless 
of the amount of self-reported sleepiness, children who exhibited RSA withdrawal to the star 
tracing task performed similarly on the AWM test two years later (Fig. 6A). However, consistent 
with hypotheses, a negative association between sleepiness and AWM was found for children 
122 
 
who showed RSA augmentation and the slope representing this relation was significantly 
different from zero.  
 RSA-R also acted as a moderator in the longitudinal pathway between sleep/wake 
problems and AWM scores, (? = .31, p < .001) and explained 7% of unique variance in AWM  
(?2 = 136.745 p = .00, ?2/df = 1.59, CFI = .78, RMSEA = .06); the full model accounted for 28% 
of the variance in AWM. Figure 6B depicts that when sleep/wake problems are high, children 
who exhibited RSA withdrawal performed worse on the working memory task than those who 
showed RSA augmentation. Further, for children exhibiting RSA augmentation, a positive 
association was observed between sleep problems and cognitive performance; however, this 
pattern was reversed for those exhibiting RSA withdrawal. Both slopes representing these 
associations were significantly different from zero.  
 Decision Speed Scores. Analyses revealed that RSA was a significant moderator in 
one of the four models examined (25% of possible relations); RSA-R did not moderate the link 
between children?s sleep and DS scores over time.  
 The pathway between sleep efficiency at T1 and children?s scores on the WJ III DS test 
at T2 was not moderated by either RSA or RSA-R.  
 The longitudinal association between long wake episodes and DS was moderated by 
children?s RSA (? = - .18, p < .05), which explained 3% of unique variance in DS (?2 = 94.80 
 p = .02, ?2/df = 1.72, CFI = .70, RMSEA = .07). The full model accounted for 34% of the 
variance in DS. As depicted in Figure 7A, a positive association between children?s disrupted 
sleep and their DS was found for children who exhibited low RSA and the slope representing 
that relation was significantly different from zero. The performance of children with higher RSA 
did not seem to be impacted by sleep problems.  
RSA activity did not moderate relations between sleep minutes, sleepiness or 
sleep/wake problems at T1 and DS scores two years later.  
123 
 
 Reaction Time Means and SDs. RSA activity did not moderate in the link between 
children?s sleep (objective and subjective) at T1 and their performance on the reaction time task 
(M or SD scores) at T2.  
Discussion 
We investigated whether multiple indices of vagal activity as indexed by RSA (baseline, 
reactivity, and/or recovery), moderated the associations between children?s sleep disruptions 
and multiple domains of cognitive performance cross-sectionally and longitudinally in a large 
sample of healthy elementary school-age children. Cross-sectionally, RSA-R was the only index 
of vagal activity that functioned as a moderator and all significant moderation effects were for 
associations between actigraphy-based sleep parameters and children?s decision speed. 
Consistent with hypotheses, RSA withdrawal to the star tracing task functioned as a protective 
factor, in that, regardless of the sleep context children who expressed withdrawal performed 
better on the decision speed test than their counterparts who exhibited RSA augmentation. This 
protective function was observed in relation to multiple sleep parameters (actigraphyically-
derived sleep efficiency, long wake episodes, and sleep minutes). Although the amount of 
variance accounted for by the interaction term was small, the consistency of the findings 
strengthens confidence in the results. In addition, the amount of variance accounted for in this 
study is similar to other studies that have examined interactions between sleep and vagal 
activity in the prediction of child adjustment (e.g., El-Sheikh, Erath, & Keller, 2007). Taken 
together, these findings are novel and build on the literature by identifying vagal withdrawal as a 
protective factor for cognitive functioning in the context of sleep disruptions.  
Longitudinal analyses revealed that vagal tone and reactivity moderated relations 
between children?s sleep problems and their cognitive functioning as indexed by multiple WJ III 
test scores. In 10 out of the 18 models examined, vagal activity was a significant moderator; 
RSA-R was the most robust moderating vagal index (8 of the 10 significant models examined 
were moderated by RSA-R). Although the overall pattern of effects indicates that children who 
124 
 
exhibit RSA withdrawal, and to a lesser extent higher resting RSA, perform better than those 
who show RSA augmentation or exhibit lower resting RSA in the context of poor sleep, the 
pattern of effects is not clear when sleep is more optimal. The fact that RSA-R emerged as a 
more robust moderator of relations between children's sleep and their cognitive functioning is 
similar to previous studies examining these moderation effects in relation to children's cognitive 
performance (manuscript 1) and psychological adjustment (El-Sheikh, Erath, & Keller, 2007). 
Overall, findings highlight the importance of individual differences in physiological functioning 
and the aggregation of risk across physiological and biological domains.  
Horne (1988, 1993) was the first to suggest sleep impacted cognitive functions, as his 
lab experiments indicated novel language and creativity were significantly impaired by sleep 
loss. This notion was extended by Dahl?s research (1996), which suggested a direct effect of 
sleep on executive functions and/or the synchronization of attention and arousal to perform 
abstract or complex tasks. In this study we found few direct relations between objective or 
subjective sleep parameters and cognitive performance. However, some significant associations 
in the expected directions were observed. Specifically, greater sleep duration and reduced night 
waking were associated with better decision speed performance at T1. At T2, children?s self-
reported sleep problems, particularly sleepiness, was related to poor performance on the WJ III 
tests.  
The novel contribution of this study is the examination of vagal activity indices as 
moderators of the link between reductions in sleep duration and quality and cognitive 
performance cross-sectionally and longitudinally. In the cross-sectional investigation three 
aspects of vagal functioning were examined as moderators of effects: vagal tone, reactivity, and 
recovery. Results indicate that both objective sleep duration and quality measures interacted 
with RSA-R to predict children?s decision speed. That is, regardless of the sleep context children 
who exhibited vagal withdrawal performed had better decision speed scores than their 
counterparts. For children who exhibited vagal augmentation to the challenge, the pattern was 
125 
 
unexpected in that under optimal sleep conditions decision speed scores were lower than when 
sleep quality was less optimal. In regards to sleep duration, children who showed RSA 
augmentation performed similarly on the decision speed test regardless of the amount of sleep 
they received. No other cross-sectional moderation patterns emerged with any other WJ III tests 
or vagal activity (RSA, RSA recovery).  
While the pattern of effect for the longitudinal findings varied based on the sleep 
parameter, the vagal activity index, and the cognitive performance domain, in 7 of the 10 
significant moderation models it was more optimal vagal functioning (RSA withdrawal and to a 
lesser extent high RSA) that functioned as a protective factor in the context of poor sleep. For 
example, in all three significant models involving sleepiness, children who showed RSA 
withdrawal to the task performed similarly on the three tests (i.e., concept formation, auditory 
working memory, numbers reversed) regardless of the level of reported sleepiness. On the 
other hand, children who exhibited RSA augmentation had a significant negative association 
between sleepiness and cognitive performance, indicating an aggregation of risk. That is, the 
pattern suggests that children who have poor sleep in conjunction with less optimal vagal 
reactivity (augmentation or lower levels of vagal withdrawal) are the ones most at risk for poorer 
cognitive performance two years later, particularly tasks involving working memory. Of note, 
there were three models in which this pattern of effects were not observed and it is not evident 
why this is the case. Thus, although there were some consistent results supportive of vagal 
withdrawal as a protective factor in the context of poor sleep, findings should be viewed as 
tentative pending replication. Overall, results are somewhat consistent with the few studies that 
have examined interactions between sleep and vagal activity in the prediction of healthy 
children?s adjustment; children at most risk were those who had poor sleep and less optimal 
vagal functioning (manuscript 1; El-Sheikh et al., 2007). 
Children with less optimal vagal functioning (i.e., low RSA and/or RSA augmentation) 
may have difficulty obtaining sufficient, good quality sleep (Dahl, 1996; El-Sheikh, Erath, & 
126 
 
Keller, 2007). There were no direct relations found with this sample between the objective or 
self-reported sleep and RSA parameters at T1, however RSA-R at T2 was cross-sectionally 
related to all actigraphy-based sleep parameters and self-reported sleepiness. Specifically, 
RSA-R was negatively associated with sleep minutes and sleep efficiency and positively 
associated with long-wake episodes and child-reported sleepiness. The cross-sectional 
relations are consistent with results from El-Sheikh and Buckhalt?s (2005) indicating multiple 
actigraphy-based and subjective measures of sleep to be related with children?s vagal 
withdrawal during a continuous performance task, with better sleep associated with increased 
vagal withdrawal. It is unclear as to why these direct effects were observed at T2 but not T1. 
Sample characteristics (i.e., children in El-Sheikh & Buckhalt?s study were older than those at 
T1) or the study design (i.e., parent-reported vs. child-reported sleep problems) may have 
contributed to the discrepant findings. However, it is not unusual that vagal activity measures 
are not directly related to the study outcome variables (e.g., internalizing and externalizing 
behaviors), but rather function as moderators of the associations (e.g., El-Sheikh et al., 2007; 
El-Sheikh & Whitson, 2006; Leary & Katz, 2004). Further, the normative sample used in this 
study may have contributed to the lack of association among sleep and vagal indices. Children 
with clinically diagnosed sleep or vagal functioning problems may show stronger direct 
correlations between the two biophysiological systems (Hsieh, Chen, McAfee, & Kifle, 2008; 
Marzec, Edwards, Sagher, Fromes, & Malow, 2003).  
Consistent with Polyvagal Theory (Porges, 1998) propositions that vagal functioning 
impacts arousal, affect and attention (Porges, 2007), increased RSA withdrawal and to a lesser 
extent higher baseline levels of RSA served a protective function in the link between sleep 
problems and cognitive functioning. According to the theory, an adaptive physiological reaction 
to stress (e.g., sleep disruptions), is considered to be a disengagement of the vagal break (i.e., 
vagal withdrawal), which suppresses the PNS?s influence on the heart and increases heart rate. 
This response allows one to become focused on the task at hand, obtain information, and/or 
127 
 
reduce the challenge by engaging social strategies that regulate emotions. After the challenge 
has subsided, the PNS influence can reengage to reduce arousal. The ability to engage and 
disengage as environmental demands change and to respond to metabolic demands, including 
increased attention and information processing (Suess, Porges, & Plude, 1994) are processes 
vital for adaptive behavior.  
 A novel aspect of this study was the inclusion of vagal recovery as a possible moderator 
in the pathway between children?s sleep disruptions and cognitive performance during the fifth 
grade. Vagal recovery was not a moderator in this study; however we were only able to examine 
this index of vagal functioning cross-sectionally at T2. In the current study, 45% of children had 
RSA levels during the post-task baseline that were lower than those during the initial baseline; 
16% fully recovered (the difference between the initial and post-task baselines was zero); and 
38% had RSA levels higher during the post-task than the initial baseline. Although there was 
variation in the amount of recovery, it is important to note that vagal recovery can be computed 
in a number of ways (e.g., amount of time before baseline levels are reached, percentage of 
recovery that has occurred within a specified timeframe) and it is possible that findings could 
differ based on how the variable was computed. Future research is encouraged to incorporate 
multiple calculations of vagal recovery into studies investigating physiological activity.    
Similar to the first manuscript of this dissertation, vagal markers did not interact with 
sleep parameters to predict children?s reaction time mean or the variability in response time. 
Considering the importance of attention and vigilance needed to perform well on the reaction 
time task, and that vagal activity is a marker of attention (Porges, 2007) and vigilance is hard to 
obtain when sleepy (Dahl, 1996), this finding was unexpected. The study design may have 
contributed to the results (e.g., no incentives for performance were given), thus caution should 
be exercised to not accept the null hypotheses. 
Several of the moderation effects were hard to interpret and inconsistent with 
expectations. Specifically, and consistent with the cross-sectional findings from manuscript 1, 
128 
 
three of the ten longitudinal plots revealed that children who exhibited higher levels of vagal 
withdrawal performed significantly better on several cognitive assessments when sleep was 
poor than when it was more optimal. While findings supportive of our hypotheses suggest that 
physiological regulation may play an important role in understanding how sleep impacts 
children?s cognitive functioning, we only examined one component ANS activity. The second 
branch, the sympathetic nervous system (SNS), has also been linked to sleep (El-Sheikh & 
Arsiwalla, 2011) and interactions between the PNS and SNS have additive and interactive 
effects in the prediction of child adaptation (e.g., El-Sheikh, Kouros, Erath, Cummings, Keller, & 
Staton, 2009). Therefore, for a more thorough explication of research questions, it may be 
fruitful for future studies to examine both branches of the ANS in relation to sleep and cognitive 
functioning.  
The results must be interpreted in relation to the age of the children. Cross-sectional 
findings when children were in the third grade (manuscript 1) were more robust than when the 
children were in the fifth grade. It is plausible that the interaction between sleep and vagal 
functioning may be more important at an early age when knowledge is being acquired than later 
on in development when the foundation of knowledge has been formed; it is also plausible that 
this association would be more apparent for adolescents who experience a higher rate of sleep 
problems (El-Sheikh, 2011). Thus, examination of our research questions with children across 
various developmental ages is imperative. Further, the community sample in conjunction with 
the exclusion/inclusion criteria (e.g., two-parent home, lack of children with mental retardation, 
developmental delays, ADHD, chronic illnesses, and diagnosed sleep problems) did not allow 
for understanding the research questions in the context of more severe or clinically significant 
problems.  
Although we had a thorough assessment of sleep (actigraphy and self-reports; Sadeh, 
2008) we did not take into account daytime naps nor did we have the ability to assess sleep 
staging. Sleeping during the day may compensate for sleep problems at night and it is 
129 
 
recommended that future research includes napping in the calculation of total sleep duration to 
further delineate the associations between sleep and cognition. An important next step in the 
field is the investigation of sleep staging through polysomnography (PSG) and its association 
with physiological activity and cognitive functioning. Adult research has implicated slow-wave 
sleep in declarative memory consolidation and REM sleep in the consolidation of procedural 
memory (Hornung, Danker-Hopfe, Schredl, & Heuser, 2007). Thus, the integration of PSG in 
child development studies is warranted.  
In summary, we found cross-sectional and longitudinal support for interactions between 
biological (i.e., sleep) and physiological (i.e., RSA) regulatory systems in predicting children?s 
cognitive performance with a large normative sample of school-aged children. Results extend 
the sleep-cognitive literature by (1) suggesting PNS indicators, primarily RSA-R, are important 
individual difference variables to be considered when assessing the role of sleep on child 
cognitive functioning; (2) including both actigraphy and self-reported assessments of sleep 
given the differential outcomes with the various sleep parameters; (3) assessing the moderating 
role of vagal indices both cross-sectionally and longitudinally in the link between sleep and 
cognitive functioning; and (4) examining multiple aspects of executive functioning using 
standardized tests and controlling for autoregressive effects. The moderation results consistent 
with hypotheses are strengthened by the multifaceted nature of the study. Results suggest that 
children who exhibit poor physiological regulation (i.e., low baseline RSA or RSA augmentation) 
may be at an increased risk for cognitive functioning problems otherwise associated with 
normative levels of poor sleep.  
 
 
 
 
130 
 
References 
Abrams, R. A. (2004). Psych/Lab for windows (Computer software and manual). 
Acebo, C., & Carskadon, M. (2001). Scoring actigraph data using ACTION-W 2. Providence, RI: 
Bradley Sleep Center, Brown University. 
Acebo, C., & Carskadon, M. A. (2002). Influence of irregular sleep patterns on waking behavior. 
In M. A. Carskadon (Ed.), Adolescent sleep patterns:  Biological, social, and 
psychological influences (pp. 220-235). New York: Cambridge University Press. 
Acebo, C., Sadeh, A., Seifer, R., Tzischinsky, O., Wolfson, A., Hafer, A., & Carskadon, M. A. 
(1999). Estimating sleep patterns with actigraphy monitoring in children and adolescents: 
How many nights are necessary for reliable measures? Sleep, 22, 95-103.  
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. 
Newbury Park, CA: Sage. 
Anders, T. F., Sadeh, A., & Appareddy, V. (1995). Normal sleep in neonates and children. In R. 
Ferber and M. Kryger (Eds.), Principles and practice of sleep medicine in the child, pp. 
7-18. Philadelphia: WB Saunders Company. 
Arbuckle, J. L. (2007) AMOS 17.0 User?s Guide. Chicago, IL: Amos Development Corporation.  
Arbuckle, J. & Wothke, W. (1999). AMOS 4.0 User?s Guide. Chicago, IL: Marketing 
Departement, SPSS Inc., Smallwaters Corporation.  
Babyak, M. A. (2004). What you see may not be what you get: A brief, nontechnical introduction 
to overfitting in regression-type models. Psychosomatic Medicine, 66, 411-421.  
Bates, J. E., Viken, R. J., Alexander, D. B., Beyers, J., & Stockton, L. (2002). Sleep and 
adjustment in preschool children: Sleep diary reports by mothers relate to behavior 
reports by teachers. Child Development, 73, 62-75.  
Bar-Haim, Y., Marshall, P. J., & Fox, N. A. (2000). Developmental changes in heart period and 
high-frequency heart period variability from 4 months to 4 years of age. Developmental 
Psychobiology, 37, 44-56. 
131 
 
Beauchaine, T. (2001). Vagal tone, development, and Gray?s motivational theory: Toward an 
integrated model of autonomic nervous system functioning in psychopathology. 
Development and Psychopathology, 13, 183-214. 
Berger, A. M., Wielgus, K. K., Young-McCaughan, S., Fischer, P., Farr, L., & Lee, K. A. (2008).  
Methodological challenges when using actigraphy in research. Journal of Pain and 
Symptom Management, 36, 191-199.  
Berntson, G. G., Bigger, J. T., Eckberg, D. L., Grossman, P., Kaufmann, P. G., Malik, M., ?Van 
Der Molen, M. W. (1997). Heart rate variability: Origins, methods, and interpretive 
caveats. Psychophysiology, 34, 623-648. 
Blunden, S., Lushington, K., & Kennedy, D. (2001). Cognitive and behavioural performance in 
children with sleep related obstructive breathing disorders. Sleep medicine Reviews, 5, 
447-461. 
Bornstein, M. H., & Suess, P. E. (2000). Child and mother cardiac vagal tone: Continuity, 
stability, and concordance across the first 5 years. Developmental Psychology, 36, 54-
65. 
Bub, K. L., Buckhalt, J., & El-Sheikh, M. (in press). Children?s sleep and cognitive  
 performance: A cross-domain analysis of change over time. Developmental Psychology. 
 
Brown, T. A & Barlow, D. H. (2005). Dimensional versus categorical classification of mental 
disorders in the fifth edition of the Diagnostic and Statistical Manual of the Mental 
Disorders. Journal of Abnormal Psychology, 114, 554-556.  
Browne, M. W. & Cudeck, R. (1993) Alternative ways of assessing model fit. In K. A. Bollen, J. 
S. Long (Eds.), Testing structural equation models (pp. 136-162). Thousand Oaks, CA: 
Sage.  
132 
 
Buckhalt, J. A., El-Sheikh, M., & Keller, P. (2007). Children's sleep and cognitive functioning: 
Race and socioeconomic status as moderators of effects. Child Development, 78, 213- 
231. 
Buckhalt, J. A., El-Sheikh, M., Keller, P., & Kelly, R. J. (2009). Concurrent and longitudinal 
relationships between children?s sleep and cognitive functioning: The moderating role of 
parent education. Child Development, 80, 875-892. 
Buckhalt, J., Wolfson, A., & El-Sheikh, M. (2009). Children's sleep and school psychology 
practice. School Psychology Quarterly, 24, 60-69. 
Calkins, S. D. (1997). Cardiac vagal tone indices of temperamental reactivity and behavioral 
regulation in young children. Developmental Psychobiology, 31, 125-135. 
Calkins, S. D. & Dedmon, S. E. (2000). Physiological and behavioral regulation in two-year-old 
children with aggressive/destructive behavior problems. Journal of Abnormal Child 
Psychology, 28, 103-118. 
Carskadon, M. A. (2002). Factors influencing sleep patterns of adolescents. In Adolescent sleep 
patterns: Biological, social, and psychological influences (pp. 4-26). New York: 
Cambridge University Press. 
Carskadon, M. A., Seifer, R., & Acebo, C. (1991). Reliability of six scales in a sleep 
questionnaire for adolescents. Sleep Research, 20, 421. 
Chervin, R. D., Ruzicka, D. L., Giordan, B. J., Weatherly, R. A., Dillon, J. E., Hodges, E. K.,. . . 
Guire, K. E. (2006). Sleep-disordered breathing, behavior, and cognition in children 
before and after adenotonsillectomy. Pediatrics, 117, e769-e778. 
Cicchetti, D., & Dawson, G. (2002). Editorial: Multiple levels of analysis. Development and 
Psychopathology, 14, 417-420. 
Cole, D. A. & Maxwell, S. E. (2003). Testing meditational models with longitudinal data: 
Questions and tips in the use of structural equation modeling. Journal of Abnormal 
Psychology, 112, 558-577. 
133 
 
Cole, P., Usher, B., & Cargo, A. (1993). Cognitive risk and its association with risk for disruptive 
behavior disorder in preschoolers. Journal of Clinical Child Psychology, 22, 154-165. 
Cohen-Zion, M., & Ancoli-Israel, S. (2004). Sleep in children with attention-deficit hyperactivity 
disorder (ADHD): A review of naturalistic and stimulant intervention studies. Sleep 
Medicine Reviews, 8, 379?402. 
Curcio, G., Ferrara, M., & De Gennaro, L. (2006). Sleep loss, learning capacity and academic 
performance. Sleep Medicine Reviews, 10, 323-337. 
Dahl, R. E. (1996). The regulation of sleep and arousal : Development and psychopathology. 
Development and Psychopathology, 8, 3-27.  
Dahl, R. E., & Harvey, A. G. (2007). Sleep in children and adolescents with behavioral and 
emotional disorders. Sleep Medicine Reviews, 2, 501-511. 
Dewald, J. F., Meijer, A. M., Oort, F. J., Kerkhof, G. A, & Bogels, S. M. (2010). The influence of 
sleep quality, sleep duration and sleepiness on school performance in children and 
adolescents: A meta-analytic review. Sleep Medicine Reviews, Advanced online 
publication. PMID: 20093054.  
DiPietro, J. A., Porges, S. W., & Uhly, B. (1992). Reactivity and developmental competence in 
preterm and fullterm infants. Developmental Psychology, 28, 831-841. 
Drummond, S., & Brown, G. (2001). The effects of total sleep deprivation on cerebral response 
to cognitive performance. Neuropsychopharmacology, 25, S68-S73. 
El-Sheikh, M. (2005). Stability of respiratory sinus arrhythmia in children and young 
adolescents: A longitudinal examination. Developmental Psychobiology, 46, 66-74. 
El-Sheikh, M. (2011). Sleep and Development: Familial and Socio-Cultural Considerations. New 
York: 
 Oxford University Press.  
El-Sheikh, M. & Arsiwalla, D. D. (2011). Children?s sleep, skin conductance level, and mental 
health. Journal of Sleep Research, 20, 326-337. 
134 
 
El-Sheikh, M., & Buckhalt, J. A. (2005). Vagal regulation and emotional intensity predict 
children's sleep problems. Developmental Psychophysiology, 46, 307-317. 
El-Sheikh, M., & Erath, S. A. (2011). Family conflict, autonomic nervous system functioning, and 
child adaptation: State of the science and future directions. Development and 
Psychopathology, 23, 723-721. 
El-Sheikh, M., Erath, S. A., & Keller, P. S. (2007). Children's sleep and adjustment: The 
moderating role of vagal regulation. Journal of Sleep Research, 16, 396-405. 
El-Sheikh, M., Kouros, C. D., Erath, S., Cummings, E. M., Keller, P., & Staton, L. (2009). Marital 
conflict and children?s externalizing behavior: Interaction between parasympathetic and 
sympathetic nervous system activity [Monograph]. Monographs of the Society for 
Research in Child Development, 74, 1-101. 
El-Sheikh, M. & Whitson, S. A. (2006). Longitudinal relations between marital conflict and child 
adjustment: Vagal regulation as a protective factor. Journal of Family Psychology, 20, 
30-39.  
Fallone, G., Seifer, R., Acebo, C., & Carskadon, M. (2000). Prolonged sleep restriction in 11- 
and 12-year-old children: Effects on behavior, sleepiness and mood. Sleep, 23S2, A:28. 
Galles, S. J., Miller, A., Cohn, J. F., & Fox, N. A. (2002). Estimating parasympathetic control of 
heart rate variability: Two approaches to quantifying vagal tone. Psychophysiology, 39, 
S37. 
Geiger, A., Achermann, P., & Jenni, O. G. (2010). Association between sleep duration and 
intelligence scores in healthy children. Developmental Psychology, 46, 949-954. 
Gozal, D. (1998). Sleep-disordered breathing and school performance in children. Pediatrics, 
102, 616-620. 
Gregory, A. M., Caspi, A., Eley, T. C., Moffitt, T. E., O?Connor, T. G., & Pouton, R. (2005). 
Prospective longitudinal associations between persistent sleep problems in childhood 
135 
 
and anxiety and depression disorders in adulthood. Journal o f Abnormal Child 
Psychology, 33, 157-163.  
Grossman, P., & Taylor, E. W. (2007). Toward understanding respiratory sinus arrhythmia: 
Relations to cardiac vagal tone, evolution and biobehavioral functions. Biological 
Psychology, 74, 263-285. 
Harrison, Y. & Horne, J. (1998). Sleep loss impairs short and novel language tasks having a 
prefrontal focus. Journal of Sleep Research, 7, 95-100.  
Heussler, H. S. (2005). Common causes of sleep disruption and daytime sleepiness: Childhood 
sleep disorders. The Medical Journal of Australia, 182, 484-489. 
Hollingshead, A. B. (1975). Four factor index of social status. Unpublished manuscript. 
Horne, J. A. (1988). Why we sleep: The functions of sleep in humans and other mammals. New 
York, NY, US: Oxford University Press.  
Horne, J. A. (1993). Human sleep, sleep loss and behaviour. Implications for the prefrontal 
cortex and psychiatric disorder. Br J Psychiatry, 162, 413-419. 
Hornung, O. P., Regen, F., Danker-Hopfe, H., Schredl, M., & Heuser, I. (2007). The relationship 
between REM sleep and memory consolidation in old age and effects of cholinergic 
medication. Biological Psychiatry, 61, 750-757. 
Hsieh, T., Chen, M., McAfee, A., & Kifle, Y. (2008). Sleep-related breathing disorder in children 
with vagal nerve stimulators. Pediatric Neurology, 38, 99-103.  
Huffman, L. C., Bryan, Y. E., del Carmen, R., Pederson, F. A., Doussard-Roosevelt, J. A., & 
Porges, S. W. (1998). Infant temperament and cardiac vagal tone: Assessments at 
twelve weeks of age. Child Development, 69, 624-635. 
Huntley, E. & Lewin, D. S. (2006). The effects of sleep restriction on performance testing on 
children with obstructive sleep apnea (OSA) and a healthy comparison guide. Submitted 
for the Associated Processional Sleep Societies 2006 annual meeting. 
136 
 
Irwin, M. (2008). Human psychoneuroimmunology: Twenty years of discovery. Brain, Behavior, 
and Immunity, 22, 129-139. 
Johnsen, B. H., Laberg, J. C., Eid, J., & Hugdahl, K. (2002). Dichotic listening and sleep 
deprivation: Vigilance effects. Scandinavian Journal of Psychology, 43, 413-417.  
Kahn, A., Van de Merckt, C., Rebuffat, E., Mozin, M.J., Sottiaux, M., Blum, D., Hennart, P. 
(1989). Sleep problems in healthy preadolescents. Pediatrics, 84, 542-546. 
Leary, A. & Katz, L. F. (2004). Coparenting, family-level processes, and peer outcomes: The 
moderating role of vagal tone. Development Psychopathology, 16, 593-608.  
Little, R. J. A., & Rubin, D. B. (2002). Statistical analysis with missing data, 2nd Edition. 
Hoboken, NJ: John Wiley & Sons, Inc.  
Logie, R., Zucco, G., & Baddeley, A. (1990). Interference with visual short-term memory. Acta 
Psychiatrica 75, 55-74. 
Marzec, M., Edwards, J., Sagher, O., Fromes, G., & Malow, B. A. (2003). Effects of vagus nerve 
stimulation on sleep-related breathing in epilepsy patients. Epilepsia, 44, 930-935.  
Matthews, K. A., Rakaczky, C. J., Stoney, C. M., & Manuck, S. B. (1987). Are cardiovascular 
responses to behavioral stressors a stable individual difference variable in childhood? 
Psychophysiology, 24, 464-473.  
Matthews, K. A., Woodall, K. L., & Stoney, C. M. (1990). Changes in and stability of 
cardiovascular responses to behavioral stress: results from a four-year longitudinal study 
of children. Child Development, 61, 1134-1144. 
McEwen, B. S. (2006). Sleep deprivation as a neurobiological and physiologic stressor: 
Allostasis and allostatic load. Metabolism, 55, S20-S23.  
Meijer, A. M., Habekothe, H. T., & van den Wittenboer, G. L. H. (2000). Time in bed, quality of 
sleep and school functioning of children. Journal of Sleep Research, 9, 145-153. 
Mezzacappa, E. S., Kelsey, R. M., Katkin, E. S., & Sloan, R. P. (2001). Vagal rebound and 
recovery from psychological stress. Psychosomatic Medicine, 63, 650-657.  
137 
 
Mindell, J. A., & Owens, J. A. (2009). A clinical guide to pediatric sleep: Diagnosis and 
management of sleep problems. New York: Lippincott, Williams, & Wilkins. 
Mindell, J. A., Owens, J. A., & Carskadon, M. A. (1999). Developmental features of sleep. Child 
and Adolescent Psychiatric Clinics of North America, 8, 695-725. 
Mitru, G., Millrood, D., & Mateika, J. H. (2002). The impact of sleep on learning and behavior in 
adolescents. Teachers College Record, 104, 704-726.  
National Sleep Foundation (NSF). (2004). 2004 Sleep in America Poll. Washington, DC: 
National Sleep Foundation. Retrieved August 14, 2005, from www.sleepfoundation.org 
Pennington, B. F., & Ozonoff, S. (1996). Executive functions and developmental 
psychopathology. Journal of Child Psychology and Psychiatry and Allied Disciplines, 37, 
51-87. 
Petersen, A. C., Crockett, L., Richards, M., Boxer, A. (1988). A self-report measure of pubertal 
status: Reliability, validity, and initial norms. Journal of Youth and Adolescence, 17, 117-
133.  
Pilcher, J. J., & Huffcut, A. I. (1996). Effects of sleep deprivation on performance: A meta-
analysis. Sleep, 19, 318-326. 
Porges, S. W. (Ed.). (1991). Vagal tone: An autonomic mediator of affect. New York: Cambridge 
University Press. 
Porges, S. W. (1995). Cardiac vagal tone: A physiological index of stress. Neuroscience & 
Biobehavioral Reviews, 19, 225-233. 
Porges, S. W. (1998). Love: An emergent property of the mammalian autonomic nervous 
system. Psychoneuroendocrinology, 23, 837-861.  
Porges, S. W. (2001). The polyvagal theory: Phylogenetic substrates of a social nervous 
system. International Journal of Psychophysiology, 42, 123-146. 
Porges, S. W. (2007). The polyvagal perspective. Biological Psychology, 74, 116-143. 
138 
 
Preacher, K. J., Curran, P. J., & Bauer, D. J. (2006). Computational tools for probing interaction 
in multiple linear regression, multilevel modeling, and latent curve analysis. Journal of 
Educational and Behavioral Statistics, 31, 437-448. 
Qureshi, A. & Lee-Chong, T. (2004). Medications and their effects on sleep. The Medical Clinics 
of North America, 88, 751-766. 
Randazzo, A. C., Muehlbach, M. J., Schweitzer, P. K., & Walsh, J. K. (1998). Cognitive function 
following acute sleep restriction in children ages 10-14. Sleep, 21, 861?868. 
Redwine, L., Hauger, R. L., Gillin, J. C., & Irwin, M. (2000). Effects of sleep and sleep 
deprivation on interleukin-6, growth hormone, cortisol, and melatonin levels in humans. 
The Journal of Clinical Endocrinology & Metabolism, 85, 3597. 
Rottenberg, J. (2007). Cardiac vagal control in depression: A critical analysis. Biological 
Psychology, 74, 200-211. 
Sadeh, A. (2007). Consequences of sleep loss or sleep disruption in children. Sleep Medicine 
Clinics, 2, 513-520. 
Sadeh, A. (2008). Commentary: Comparing actigraphy and parental report as measures of 
children's sleep. Journal of Pediatric Psychology, 33, 406-407. 
Sadeh, A. (2011). Sleep assessment methods. In M. El-Sheikh (Ed.), Sleep and development: 
Familial and socio-cultural considerations (pp. 355-374) New York, NY: Oxford 
University Press. 
Sadeh, A., Gruber, R., & Raviv, A. (2002). Sleep, neurobehavioral functioning, and behavior 
problems in school-age children. Child Development, 73, 405-417. 
Sadeh, A., Gruber, R., & Raviv, A. (2003). The effects of sleep restriction and extension on 
school-age children: What a difference an hour makes. Child Development, 74, 444-455. 
Sadeh, A., Raviv, A., & Gruber, R. (2000). Sleep patterns and sleep disruptions in school-age 
children. Developmental Psychology, 36, 291-301. 
139 
 
Sadeh, A., Sharkey, K. M., & Carskadon, M. A. (1994). Activity-based sleep-wake identification: 
An empirical test of methodological issues. Sleep, 17, 201-207. 
Salomon, K.  (2005). Respiratory sinus arrhythmia during stress predicts resting respiratory 
sinus arrhythmia 3 years later in a pediatric sample. Health Psychology, 24, 68-76. 
Santucci, A. K., Silk, J. S., Shaw, D. S., Gentzler, A., Fox, N. A., & Kovacs, M. (2008). Vagal 
tone and temperament as predictors of emotion regulation strategies in young children. 
Developmental Psychobiology, 50, 205-216. 
Smaldone, A., Honig, J. C., & Byrne, M. W. (2007). Sleepless in America: Inadequate sleep and 
relations to health and well-being of our nation?s children. Pediatrics, 119, S29-S37. 
Spilsbury, J. C., Storfer-Isser, A., Drotar, D., Rosen, C. L., Kirchner, H. L., & Redline, S. (2005). 
Effects of the home environment on school-aged children?s sleep. Sleep, 28, 1419-1427. 
Staton, L., El-Sheikh, M., & Buckhalt, J. A. (2009). Respiratory sinus arrhythmia and cognitive 
functioning in children. Developmental Psychobiology, 51 249-258. 
Steenari, M., Vuontela, V., Paavonen, E. J., Carlson, S., Fjallberg, M., & Aronen, E. T. (2003). 
Working memory and sleep in 6- to 13-year-old school children. Journal of the American 
Academy of Child and Adolescent Psychiatry, 42, 85-92. 
Steinberg, L., & Avenevoli, S. (2000). The role of context in the development of 
psychopathology: A conceptual framework and some speculative propositions. Child 
Development, 71, 66-74. 
Stein, M. A., Mendelsohn, J., Obermeyer, W. H., Amromin, J., & Benca, R. (2001). Sleep and 
behavior problems in school-aged children. Pediatrics, 107, e60. 
Suess, P. E., Porges, S. W., & Plude, D. J. (1994). Cardiac vagal tone and sustained attention 
in school-age children. Psychophysiology, 31, 17-22. 
Thayer, J. F., & Lane, R. D. (2000). A model of neurovisceral integration in emotion regulation 
and dysregulation. Journal of Affective Disorders, 61, 201-216. 
140 
 
Thayer, J. F. & Lane, R. D. (2007). The role of vagal function in the risk for cardiovascular 
disease and mortality. Biological Psychology, 74, 224-242. 
Thayer, J. F. & Lane R. D. (2009). Claude Bernard and the heart-brain connection: Further 
elaboration of a model of neurovisceral integration. Neuroscience & Biobehavioral 
Reviews, 33, 81-88.  
Van Dongen, H. P. A., Rogers, N. L., & Dinges, D. F. (2003). Sleep debt: Theoretical and 
empirical issues. Sleep and Biological Rhythms, 1, 5-13. 
Van Leijenhorst, L., Crone, E., & Van der Molen, M. (2007). Developmental trends for object 
and spatial working memory: A psychophysiological analysis. Child Development, 78, 
987-1000. 
Walker, M. P., & Stickgold, R. (2006). Sleep, memory, and plasticity. Annual Review of 
Psychology, 10, 139-166.  
Whisman, M. A. & McClelland, G. H. (2005). Designing, testing, and interpreting interactions 
and moderator effects in family research. Journal of Family Psychology, 19, 111-120.  
Whitson, S. & El-Sheikh, M. (2003). Marital conflict and health: Processes and protective 
factors. Aggression and Violent Behavior, 8, 283-312.  
Williams, P. G., Sears, L. L., & Allard, A. (2004). Sleep problems in children with autism. Journal 
of Sleep Research, 13, 265-268. 
Wolfson, A. R., & Carskadon, M. A. (1998). Sleep schedules and daytime functioning in 
adolescents Child Development, 69, 875-887. 
Wolfson, A., & Carskadon, M. (2003). Understanding adolescent's sleep patterns and school 
performance: A critical appraisal. Sleep Medicine Reviews, 7, 491-506. 
Wolfson, A. R., Carskadon, M. A., Acebo, C., Seifer, R., Fallone, G., Labyak, S. E., & Martin, J. 
L. (2003). Evidence for the validity of a sleep habits survey of adolescents. Sleep, 26, 
213-216. 
141 
 
Woodcock, R. W., McGrew, K. S., & Mather, M. (2001). Woodcock-Johnson III tests of cognitive 
ability. NY: Riverside Publishing. 
  
142 
 
 
Table 1. Means and standard deviations of study variables at T1. 
 
 
 
 
                                                          
 
 
 
 
 
 
 
 
Note: RSA = baseline vagal tone; RSA-R = vagal reactivity; Woodcock-Johnson III; M = mean; 
SD = standard deviation; ms = milliseconds.  
 
 
 
 
 
 
 
 
 
 
Variable Name M SD 
1. Sleep Minutes  7hrs:21min 0:59min 
2. Sleep Efficiency  86.78 7.94 
3. Long Wake Minutes  3.92 2.20 
4. Sleep/Wake Problems   17.04 4.95 
5. Sleepiness Scale  14.11 4.50 
6. RSA  .13 .06 
7. RSA-R  -.02 .05 
8. Concept Formation WJ III 99.32 12.67 
9. Numbers Reversed WJ III 104.03 15.92 
10. Auditory Working Memory  WJ III  112.01 13.08 
11. Decision Speed WJ III  107.65 13.12 
12. Reaction Time M  Donders (ms)  464.74 75.44 
13. Reaction Time SD Donders (ms) 140.11 37.16 
143 
 
Table 2. Means and standard deviations of study variables at T2. 
 
 
 
 
 
 
 
 
 
 
Note: RSA = baseline vagal tone; RSA-R = vagal reactivity; Woodcock-Johnson III; M = mean; 
SD = standard deviation; ms = milliseconds.  
 
 
 
 
  
Variable Name M SD 
1. Sleep Minutes  7hrs:25min 0:47min 
2. Sleep Efficiency  89.12 9.34 
3. Long Wake Minutes  3.09 2.22 
4. Sleep/Wake Problems   15.93 4.67 
5. Sleepiness Scale  12.63 3.53 
6. RSA  .14 .07 
7. RSA-R  -.02 .07 
8. Concept Formation WJ III 100.46 12.94 
9. Numbers Reversed WJ III 105.81 15.20 
10. Auditory Working Memory  WJ III  106.24 12.57 
11. Decision Speed WJ III  108.73 14.17 
12. Reaction Time M  Donders (ms)  124.14 39.93 
13. Reaction Time SD Donders (ms) 414.80 70.36 
 
144 
 
Table 3. Correlations among study variables.  
 
 
 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 
1.  Age T1 -              
2.  Gender; 1 = Boys, 0 = Girls - .02 -             
3.  Ethnicity; 1 = AA, 0 = EA - .10 -.08 -            
4.  Family Socioeconomic Status T1   .18*   .05 - .24** -           
5.  Body Mass Index T1 - .11 - .09   .07 - .13 -          
6.  Sleep Efficiency T1 a   .17* -  .20   .09   .10   .04 -         
7.  Long Wake Episodes T1 a - .19*   .18* - .19* - .04   .11 - .85** -        
8.  Sleep Minutes T1 a   .18* - .24** - .09   .20** - .21**   .83** - .61** -       
9.  Sleepiness T1 b - .10   .15*   .16* - .10 - .01 - .07   .01   .08 -      
10. Sleep/Wake Problems T1 b   .09 - .06   .25** - .16*   .06 - .07   .02 - .12   .43** -     
11. RSA T1 - .08   .07   .05 - .03 - .24**   .03 - .05   .09   .04   .10 -    
12. RSA-R T1   .16   .05   .04 - .02   .04 - .04 - .02 - .08 - .03   .02 - .13 -   
13. Concept Formation T1 c    .20** - .03 - .19*   .25** - .03 - .04   .04   .00 - .05 - .08 - .02 - .01 -  
14. Numbers Reversed T1 c   .20**   .01 - .15*   .15* - .07   .02 - .07   .03 - .13 - .16*   .07 - .03   .37** - 
15. Auditory Working Memory T1 c - .26**   .06 - .11   .23** - .10   .10 - .12   .11 - .16* - .18* - .08   .06   .33**   .38** 
16. Decision Speed T1 c   .12 - .17* - .01   .20** - .05   .13 - .15*   .16* - .03   .02 - .11 - .06   .24**   .20** 
17. Reaction Time Mean T1 d - .24** - .31**   .07 - .11   .13   .01 - .04   .04 - .01   .12   .01   .01 - .21** - .24** 
18. Reaction Time SD T1 d - .24** - .10   .08 - .13   .08 - .04 - .02 - .04   .05   .17*   .05   .09 - .28** - .28** 
19. Age T2   .69** - .08 - .16   .17 - .06   .00 - .00 - .03 - .12   .02 - .08   .07   .08   .10 
20. Family Socioeconomic Status T2   .11   .06 - .25**   .99*** - .18*   .11 - .04   .24**   .02 - .10   .05 - .01   .31**   .10 
21. Body Mass Index T2 - .05   .03   .02 - .12   .83** - .27**   .18* - .25** - .05   .05 - .26**   .02 - .01 - .07 
22. Sleep Efficiency T2 a - .01 - .14   .04   .05 - .12   .40** - .42**   .34**   .05 - .03   .05   .04   .09   .22** 
23. Long Wake Episodes T2 a - .05   .02 - .05 - .13   .09 - .54**   .55** - .42**   .01   .09 - .13   .03   .01 - .09 
24. Sleep Minutes T1 a   .02 - .17 - .21*   .23** - .16   .46** - .39   .57**   .05 - .01   .11 - .05   .05   .06 
25. Sleepiness T2 b   .01   .14   .20* - .01   .04   .12 - .13   .07   .20*   .21* - .05   .02 - .31** - .16 
26. Sleep/Wake Problems T2 b - .06   .07   .01 - .15   .08 - .09   .03 - .12   .20*   .43**   .11 - .08 - .12 - .13 
27. RSA T2 - .09   .12 - .03   .00 - .18 - .04   .05   .02   .03   .12   .36** - .18 - .08   .11 
28. RSA-R T2   .06 - .02   .15 - .07   .12 - .09 - .04 - .18   .15   .16 - .11   .09 - .12 - .16 
29. Concept Formation T2 c   .10   .04 - .31**   .22* - .17   .05   .06   .15 - .21* - .29** - .02   .10   .64**   .39** 
30. Numbers Reversed T2 c - .01 - .01 - .06   .09 - .11   .07 - .07   .13 - .10   .07   .18*   .07   .31**   .54** 
31. Auditory Working Memory T2 c   .01   .06 - .10   .05 - .04 - .04   .05   .01 - .22* - .13   .14 - .01   .37**   .40** 
32. Decision Speed T2 c - .03 - .16 - .03   .08 - .17   .08 - .09   .10 - .07 - .16 - .04   .01   .17   .15 
33. Reaction Time Mean T2 d - .02 - .26**   .01 - .08   .01 - .04   .02 - .03   .02   .22* - .13 - .01 - .23 - .15 
34. Reaction Time SD T2 d - .01 - .43**   .03 - .17   .03   .02 - .01   .03   .01   .24* - .12   .03 - .10 - .15 
 
145 
 
Table 3 continued.  
 
 
 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 
1.  Age T1               
2.  Gender; 1 = Boys, 0 = Girls               
3.  Ethnicity; 1 = AA, 0 = EA               
4.  Family Socioeconomic Status T1               
5.  Body Mass Index T1               
6.  Sleep Efficiency T1 a               
7.  Long Wake Episodes T1 a               
8.  Sleep Minutes T1 a               
9.  Sleepiness T1 b               
10. Sleep/Wake Problems T1 b               
11. RSA T1               
12. RSA-R T1               
13. Concept Formation T1 c                
14. Numbers Reversed T1 c               
15. Auditory Working Memory T1 c -              
16. Decision Speed T1 c   .21** -             
17. Reaction Time Mean T1 d - .24** - .23** -            
18. Reaction Time SD T1 d - .20* - .20*   .72** -           
19. Age T2   .08   .19* - .09 - .10 -          
20. Family Socioeconomic Status T2   .24**   .30** - .15 - .21*   .18* -         
21. Body Mass Index T2 - .08 - .18*   .14   .07 - .05 - .14 -        
22. Sleep Efficiency T2 a   .10   .09   .17   .10 - .02   .06 - .13 -       
23. Long Wake Episodes T2 a - .11   .03 - .14   .00   .01 - .14   .10 - .63** -      
24. Sleep Minutes T2 a   .14   .14   .18   .01 - .11   .23* - .19*   .45** - .60** -     
25. Sleepiness T2 b - .19*   .01 - .04   .01 - .04 - .03   .01  .06 - .03   .02 -    
26. Sleep/Wake Problems T2 b - .18* - .07 - .04   .09 - .02 - .14   .05 - .03   .08 - .13   .48** -   
27. RSA T2 - .16 - .09   .05   .15 - .05 - .00 - .22*   .08 - .12   .08   .04   .06 -  
28. RSA-R T2 - .29**   .01   .02 - .01   .08 - .06   .13 - .23*   .20* - .26**   .26**   .13 - .30** - 
29. Concept Formation T2 c   .35**   .21* - .25** - .31**   .13   .23* - .12   .10   .00 - .01 - .31** - .21* - .13 - .15 
30. Numbers Reversed T2 c   .38**   .14 - .33** - .31** - .01   .09 - .01   .07 - .00 - .02 - .20* - .15 - .01   .08 
31. Auditory Working Memory T2 c   .40**   .03 - .21** - .23** - .01   .06   .04   .01 - .00   .01 - .21* - .17 - .10   .06 
32. Decision Speed T2 c   .10   .50** - .10 - .13 - .14   .09 - .12 - .02   .07   .09 - .01* - .14 - .14   .10 
33. Reaction Time Mean T2 d - .09 - .15   .44**   .48** - .06 - .09 - .10  .08   .05   .05 - .03   .02 - .03   .04 
34. Reaction Time SD T2 d - .21* - .16   .62**   .47** - .08 - .17 - .04  .16 - .06   .10 - .01   .04   .50   .02 
 
146 
 
Table 3 continued.  
 
 
 
 
 
 
 
 
 
 
 
 
 29. 30. 31. 32. 33. 34. 
1.  Age T1       
2.  Gender; 1 = Boys, 0 = Girls       
3.  Ethnicity; 1 = AA, 0 = EA       
4.  Family Socioeconomic Status T1       
5.  Body Mass Index T1       
6.  Sleep Efficiency T1 a       
7.  Long Wake Episodes T1 a       
8.  Sleep Minutes T1 a       
9.  Sleepiness T1 b       
10. Sleep/Wake Problems T1 b       
11. RSA T1       
12. RSA-R T1       
13. Concept Formation T1 c        
14. Numbers Reversed T1 c       
15. Auditory Working Memory T1 c       
16. Decision Speed T1 c       
17. Reaction Time Mean T1 d       
18. Reaction Time SD T1 d       
19. Age T2       
20. Family Socioeconomic Status T2       
21. Body Mass Index T2       
22. Sleep Efficiency T2 a       
23. Long Wake Episodes T2 a       
24. Sleep Minutes T1 a       
25. Sleepiness T2 b       
26. Sleep/Wake Problems T2 b       
27. RSA T2       
28. RSA-R T2       
29. Concept Formation T2 c -      
30. Numbers Reversed T2 c   .27** -     
31. Auditory Working Memory T2 c   .40**   .56** -    
32. Decision Speed T2 c   .13   .17   .13 -   
33. Reaction Time Mean T2 d - .37** - .16 - .08 - .05 -  
34. Reaction Time SD T2 d - .22* - .22* - .10 - .10   .75** - 
 
147 
 
Note: T1 refers to cross-sectional analyses during the first wave of data collection; T2 refers to 
cross-sectional during the second wave of data collection. a = actigraphy; b = self-reported; c = 
Woodcock Johnson III; d = Donders; RSA = Respiratory Sinus Arrhythmia; RSA-R = RSA 
reactivity. RSA recovery was not found to be related to any of the study variables and thus was 
not included in the table.  
 *p < .05; **p < .01; ***p < .001. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
148 
 
 
Figure 1.  
 
 
 
Figure 1. Conceptual model depicting children?s parasympathetic nervous system activity as a 
moderator of the association between children?s sleep and their cognitive functioning. 
 
 
 
 
 
 
 
 
 
Parasympathetic 
Nervous System 
Activity 
 
 
 
  Children?s  
Sleep 
 
Cognitive 
Functioning 
 
149 
 
 
 
Figure 2. 
 
 
Figure 2. An example of the path models that examined the possible moderating role of RSA 
and RSA-R in the cross-sectional link sleep-cognitive functioning link. Of note, control variables 
are not included in the model above however all controls (sex, race/ethnicity, age, body mass 
index, and season of watch wear) were entered as the last paths in the model. Longitudinal 
models also included the pertinent T1 cognitive outcome (autoregressive effects).  
 
 
 
 
 
 
 
 
Sleep 
Efficiency 
X RSA Sleep 
Efficiency 
X RSA 
Reactivity 
RSA 
Reactivity 
RSA 
Sleep 
Efficiency 
Auditory 
Working 
Memory 
 
150 
 
 
 
Figure 3.  
 
 
 
 
 
 
A. 
 
100
102
104
106
108
110
112
114
116
118
120
- 1  S D + 1  S D
Sl e e p  Effi c i e n c y  T2
 
D
e
c
i
s
i
o
n
 
Sp
e
e
d
 
T
2
H i g h  R SA  
W i th d r a w a l
L o w  R SA   
W i th d r a w a l
 
B. 
100
102
104
106
108
110
112
114
116
118
120
- 1  S D + 1  S D
L o n g  W a k e  Ep i s o d e s  T 2
 
D
e
c
i
s
i
o
n
 
Sp
e
e
d
 
T
2
H i g h  R SA
W i th d r a w a l
L o w  R SA   
W i th d r a w a l
 
C. 
100
102
104
106
108
110
112
114
116
118
120
- 1  S D + 1  S D
Sl e e p  M i n u te s  T 2
 
D
e
c
i
s
i
o
n
 
Sp
e
e
d
 
T
2
H i g h  R SA
W i th d r a w a l
L o w  R SA   
W i th d r a w a l
 
 
 
 
Figure 3. (A) Cross-sectional interaction between Sleep Efficiency and RSA-R in the 
prediction of children?s Decision Speed. (B) Cross-sectional interaction between Long 
Wake Episodes and RSA-R in the prediction of children?s Decision Speed. (C) Cross-
sectional interaction between Sleep Minutes and RSA-R in the prediction of children?s 
Decision Speed. 
 
151 
 
 
Figure 4.  
 
 
 
 
 
 
 
A. 
 90
92
94
96
98
100
102
104
106
108
1 10
- 1 S D +1  S D
Conc
e
pt 
Fo
rma
tio
n 
T2
Sle e p Effic ie nc y  T1
High  RSA
Low  R SA
 
B. 
90
92
94
96
98
100
102
104
106
108
1 10
- 1 S D +1  S D
C
on
ce
pt 
Fo
rm
ati
on
 T
2
Lo ng  W ak e E pisod es  T 1
H i gh  R S A  
W i t hd raw al
Lo w  R S A   
W i t hd raw al
 
C. 
90
92
94
96
98
100
102
104
106
108
1 10
- 1 S D +1  S D
Conc
e
pt 
Fo
rma
tio
n 
T2
Sle e pi ne s s  T1
High  RSA  
W ith dra w a l
Lo w  RSA    
W ith dra w a l
 
 
 
 
Figure 4. (A) Longitudinal interaction between Sleep Efficiency and RSA-R in the 
prediction of children?s Concept Formation. (B) Longitudinal interaction between Long 
Wake Episodes and RSA-R in the prediction of children?s Concept Formation. (C) 
Longitudinal interaction between Sleepiness and RSA-R in the prediction of children?s 
Concept Formation. 
 
152 
 
 
Figure 5. 
 
 
 
 
 
 
 
A. 
 
95
97
99
101
103
105
107
109
111
1 13
1 15
- 1 S D +1 SD
Num
be
rs 
Rev
e
rse
d
T2
Sle e p Effic ie nc y  T1
High  RSA  
W ith dra w a l
Lo w  RSA    
W ith dra w a l
 
B. 
95
97
99
101
103
105
107
109
111
1 13
1 15
- 1 S D +1  S D
N
umbe
rs 
R
ev
erse
d
T2
Lo ng  W ak e E pisod es  T 1
H i gh  R S A  
W i t h d raw a l
Lo w  R S A    
W i t hd raw al
 
C. 
95
97
99
101
103
105
107
109
111
1 1 3
1 1 5
- 1 S D +1  S D
Num
b
e
rs
 Re
v
e
rs
e
d
T2
Sle e pi ne s s  T1
High  RSA  
W ithdraw al
Lo w  RSA    
W ith dra w a l
 
D.  
95
97
99
101
103
105
107
109
111
1 13
1 15
- 1 S D +1  S D
N
umbe
rs 
R
ev
erse
d
T2
S l ee p/W ak e P roblem s T 1
H i gh  R S A  
W i t hd raw al
Lo w  R S A    
W i t hd raw al
 
Figure 5. (A) Longitudinal interaction between Sleep Efficiency and RSA-R in the 
prediction of children?s Numbers Reversed. (B) Longitudinal interaction between Long 
Wake Episodes and RSA-R in the prediction of children?s Numbers Reversed. (C) 
Longitudinal interaction between Sleepiness and RSA-R in the prediction of children?s 
Numbers Reversed. (D) Longitudinal interaction between Sleep/Wake Problems and 
RSA-R in the prediction of children?s Numbers Reversed.  
 
153 
 
 
Figure 6.  
 
 
 
 
 
 
 
A. 
 95
97
99
101
103
105
107
109
111
1 13
1 15
- 1 S D +1  S D
A
ud
ito
ry
 W
ork
in
g
M
e
m
ory
 T2
Sle e pi ne s s  T1
High  RSA  
W ith dra w a l
Lo w  RSA    
W ith dra w a l 
 
B. 
95
97
99
101
103
105
107
109
111
1 13
1 15
- 1 S D +1  S D
A
ud
it
or
y
 W
orki
ng
M
emory
 T
2
S l ee p/W ak e P roblem s T 1
H i gh  R S A  
W i t hd raw al
Lo w  R S A    
W i t hd raw al  
 
 
 
 
 
 
Figure 6. (A) Longitudinal interaction between Sleepiness and RSA-R in the prediction 
of children?s Auditory Working Memory. (B) Longitudinal interaction between 
Sleep/Wake Problems and RSA-R in the prediction of children?s Auditory Working 
Memory.  
 
154 
 
 
Figure 7. 
 
A. 
 
95
97
99
101
103
105
107
109
111
1 13
1 15
- 1 S D +1 SD
Dec
is
io
n 
Spe
e
d
T2
L o n g  W a k e  Ep is o d e s  T 1
High  RSA
Lo w  RSA
 
 
 
 
 
 
 
 
Figure 3. (A) Longitudinal interaction between Long Wake Episodes and RSA in the 
prediction of children?s Decision Speed.