CAN A HORSE LEARN WHILE UNDER THE INFLUENCE OF A TRANQUILIZER (ACEPROMAZINE MALEATE)? Samantha C. Griffith Except where reference is made to the work of others, the work described in this thesis is my own or was done in collaboration with my advisory committee. This thesis does not include proprietary or classified information. _______________________________________ Samantha Christine Griffith Certificate of Approval: ___________________________ ___________________________ Joseph J. Molnar Cynthia A. McCall, Chair Professor Professor Agriculture Economics and Rural Animal Science Sociology ___________________________ ___________________________ Robert Lishak Joe F. Pittman Associate Professor Interim Dean Biology Graduate School CAN A HORSE LEARN WHILE UNDER THE INFLUENCE OF A TRANQUILIZER (ACEPROMAZINE MALEATE)? Samantha C. Griffith A Thesis Submitted to the Graduate Faculty of Auburn University in Partial Fulfillment of the Requirements for the Degree of Master of Science Auburn, AL December 15, 2006 iii CAN A HORSE LEARN WHILE UNDER THE INFLUENCE OF A TRANQUILIZER (ACEPROMAZINE MALEATE)? Samantha C. Griffith Permission is granted to Auburn University to make copies of this thesis at its discretion, upon request of individuals or institutions and at their expense. The author reserves all publication rights ________________________________ Signature of Author ________________________________ Date of Graduation iv VITA Samantha Christine (Lambertus) Griffith, daughter of John and Barbara Lambertus, was born July 11, 1981 in Terre Haute, Indiana. She and her siblings Amanda, Randy and John Lambertus grew up in Terre Haute, Indiana. She graduated from Terre Haute South High School in 1999. She attended Purdue University in West Lafayette, Indiana for four years and graduated with a Bachelor of Science degree in Animal Science and a minor in Psychology in May, 2002. After graduation Samantha moved to Notasulga, Alabama where she began a graduate program in Animal Sciences at Auburn University under the direction of Dr. Cynthia A. McCall to study horse behavior. She married Lucas Griffith, son of Greg and Kelli (Hayes) Griffith, on October 16, 2004. v THESIS ABSTRACT CAN A HORSE LEARN WHILE UNDER THE INFLUENCE OF A TRANQUILIZER (ACEPROMAZINE MALEATE)? Samantha C Griffith Master of Science, December 15, 2006 (M.Ed., Auburn University, 2006) (B.S., Purdue University, 2002) 54 Typed Pages Directed by Cynthia A. McCall Although tranquilizers commonly are used as ?training aids,? little is known about how this practice affects efficiency of learning in the horse. Forty mature horses were used to compare learning performance of tranquilized and non-tranquilized horses on spatial and discrimination learning tests. The spatial test was a one-choice ?T? maze in which only one side was rewarded with feed, and the discrimination test was a choice between a white bucket and a black bucket in which only one color was rewarded with feed. On the test day horses were deprived of their usual morning concentrate ration until testing was completed as motivation to perform the tests. Horses were randomly assigned to either a control group, receiving 3ml of 0.9% saline, IM, or a tranquilized vi group, receiving 0.044 mg/kg acepromazine maleate IM. Horses were fitted with heart rate monitors during testing to establish that the horse had calmed down from the injection, and their heart rate had returned to resting value. Each horse then performed both learning tests on the same day. The horses? selections during the test were recorded and analyzed using a t-test. Mean heart rates for control and tranquilized groups (49.8+1.4 and 53.8+1.5 respectively), in the discrimination test, were not significantly different (P=0.06). Likewise, during the spatial test no difference was detected (P=0.36) between heart rates for control and tranquilized horses (55.1+4.0 and 51.3+1.5, respectively). Significant heart rate differences between the control and tranquilized horses, were not expected and the heart rate was used to verify that all horses began testing in a calm state. Mean percent correct responses for tranquilized and control horses in the spatial test (80.0+3.4 and 72.5+3.6, respectively) did not differ (P>0.13). Similarly, the discrimination test detected no difference (P>0.43) in mean percent correct responses for tranquilized and control horses (69.8+2.0 and 67.6+2.0, respectively). Results indicate tranquilized horses had similar learning performance on simple spatial and discrimination tests as those that were not tranquilized. Tranquilization makes the horse more tractable without significantly affecting learning performance. Thus, it is a tool that may be utilized effectively by less skilled horse handlers on fractious horses to perform aversive procedures, e.g. trailer loading or clipping, while allowing the horse to learn to tolerate these procedures. vii ACKNOWLEDGMENTS The author would especially like to thank her major professor Dr. Cynthia A. McCall for all of her patience, support, and guidance throughout her graduate program. The author also would like to thank Dr. Wendell McElhenney for all of his help with statistical analysis The author would like to thank all of the graduate students for their support and friendship (Larry Garcia, Taylor Parsons, Jennifer Cofield, Carrisa Wickens and Jarod Eddy). She would also like to thank the undergraduate students who helped with data collection for her research, especially Heather Dull and Leah Shaddix. A debt of gratitude is owed to both Jarod Eddy and Carrisa Wickens for their assistance in data collection. These two people were vital to the data collection portion of this project as well as to the preliminary stages of this project. A special debt of gratitude goes out to Greg Williams and the entire horse unit staff for all of their help with the author?s project and throughout her graduate program. The author most importantly thanks her family for all of their support throughout the years. Each of my family members played a vital part in helping me finish this program. She would especially like to thank her grandparents, John and Jane Lambertus, for having faith in her when others did not. Without the support of my husband, Lucas Griffith, I may have never completed the graduate program. viii Style manual or journal used_______Applied Animal Behavior Science__________ Computer Software used _________Microsoft Word 2000 and Microsoft Excel 2000 ix TABLE OF CONTENTS LIST OF FIGURES .......................................................................................................... vii I. INTRODUCTION.......................................................................................................1 II. LITERATURE REVIEW............................................................................................3 Tranquilizers ................................................................................................................3 Discrimination Learning in Horses..............................................................................7 Spatial Learning in Horses...........................................................................................9 III. MATERIALS AND METHODS...............................................................................12 Subjects......................................................................................................................12 Testing Procedure ......................................................................................................12 Discrimination Learning Test ....................................................................................13 Spatial Learning Test .................................................................................................15 Statistical Analysis.....................................................................................................18 IV. RESULTS AND DISCUSSION................................................................................19 V. IMPLICATIONS .......................................................................................................27 LITERATURE CITED ......................................................................................................28 APPENDICES ...................................................................................................................32 x LIST OF FIGURES Acepromazine maleate 2 chemical structure .......................................................................3 Discrimination test stall. Solid walls and partitions are designated by unbroken lines Entryway is designated by a dashed line. Areas of feed buckets are designated by dot-dash lines. The divider is designated by a dotted line........................................................................................................................14 Spatial test apparatus. Solid walls are designated by unbroken lines. Entryway is designated by a dotted line. Areas of feed buckets are designated by dot-dash lines ..............................................................................................16 Mean heart rate (beats per minute) of control and tranquilized horses during the discrimination learning test ..............................................................................20 Mean heart rate (beats per minute) of control and tranquilized horses during the spatial learning test ...........................................................................................21 Mean correct response for control and tranquilized horses on the discrimination learning test................................................................................................23 Mean correct response for control and tranquilized horses on the spatial learning test........................................................................................................................24 1 INTRODUCTION A horse?s worth is determined by many factors including breed, sex, pedigree, conformation and training. However, a horse?s ability to demonstrate learned responses under saddle or in harness often has the greatest influence on its value. It is of particular interest to humans to develop methods for estimating learning ability in animals that they control and handle. Because horses serve such an important role in recreational activities, such as dressage, racing, show jumping, and carriage work they are learning abilities of particular interest. Equestrian activities require specialized training and good learning ability on the part of the horse. Recently, the equine research community has shown a great deal of interest in methods of measuring equine learning ability. Methods investigated include discrimination learning (Mader and Price, 1980; McCall, 1989; Dougherty and Lewis, 1991), reversal learning (Warren and Warren, 1962; Fiske and Potter, 1979), concept learning (Sappington and Goldman, 1994), maze learning (McCall et al., 1981; Houpt et al., 1982), avoidance learning (Haag et al., 1980; Rubin et al., 1980), place learning (Heird et al., 1981; Heird et al., 1986), and observational learning (Baer et al., 1983; Baker and Crawford, 1986). These various experimental measures have never been used to assess the learning ability of tranquilized horses. A maze is an useful method for studying learning ability in horses because it does not require manual dexterity. Mazes are also good tools for studying learning ability in 2 horses because they involve abilities which would have been necessary to the survival of horses in the wild, such as finding their way across countryside on paths. A discrimination test is also a useful method for studying learning ability in horses because aside from classical conditioning, simple discrimination is one of the most elementary forms of learning in animals and humans. A variety of discrimination tasks have been used as indicators of animals? abilities to learn. Horses are large, highly mobile and unpredictable creatures. Therefore, horses often are tranquilized during mildly aversive training (e.g., loading in a trailer) or management (e.g., dental care) procedures to improve the safety of both animal and handler. It currently is unknown whether horses can learn effectively while tranquilized, making subsequent attempts at a particular training or management procedure easier or if tranquilization simply makes the horse more tractable during the initial attempt without the benefit of learning. The purpose of the present study was to compare learning performance in tranquilized and non-tranquilized horses on two different types of learning tests. Results of this study will provide a better understanding of equine learning which may lead to more effective horse handling and training protocols. 3 LITERATURE REVIEW Tranquilizers Tranquilizers have been used widely as a way of calming animals and humans. The tranquilizer used in the present study was acepromazine maleate. Acepromazine maleate (ACE) formerly was known as actelypromazine and frequently is used in equine anesthesia. It is the 2-acetyl derivative of promazine (Figure 1) and has the chemical name 2-acetyl-10-(3-dimethylaminopropyl) phenothiazine (Hall et al., 2001). Figure 1: Acepromazine maleate 1 It is the most potent phenothiazine derivative (Riebold et al., 1995). A phenothiazine derivative is a strong neuroleptic (antipsychotic) agent (Hall et al., 2001) with a relatively low toxicity. Phenothiazine derivatives are dopamine antagonists, thus, they have calming and mood-altering effects. In human medicine, sedation is an unwanted side effect, but in veterinary medicine the phenothiazine derivatives are used primarily for this purpose. 4 Phenothiazine derivatives affect the cardiovascular system due to their ability to block ? 1 adrenoceptors, resulting in an anti-adrenaline effect. This causes marked arterial hypotension primarily due to peripheral vasodilation, and a decrease in packed cell volume caused by splenic dilatation. Phenothiazine derivatives also exert an anti- arrhythmic effect on the heart (Muir et al., 1975; Muir, 1981). Acepromazine maleate is the most commonly used phenothiazine in veterinarian medicine today in the United Kingdom, North America, and Australia. Other phenothiazines, such as propionylpromazine, are used in Europe and their effects are similar to acepromazine (Taylor and Clarke, 1999). Acepromazine maleate induces tranquilization, muscle relaxation, and a decrease in spontaneous activity. It also depresses respiration and decreases locomotor and behavioral responses. A horse may appear well sedated with ACE but still respond when stimulated (Taylor and Clarke, 1999). Acepromazine maleate has no analgesic effect (McKelvey and Hollingshead, 2003). In horses, it is highly protein-bound with an elimination half-life of 184.8 minutes and is detectable in plasma for up to 8 hours. However, exercise usually alters the pharmacokinetic course and pharmacodynamics of a drug (Ma, 1990). Chou (1998) reported an exercise-associated reduction of urinary ACE resident time after a single dose. The recommended dose for horses is 0.044-0.088 mg/kg of body weight (Riebold et al., 1995). These doses may be given intraveneously (IV), intramuscularly (IM), or subcutaneously (SubQ). Onset of action occurs within 15-20 minutes following IV administration and lasts approximately 2 hours. Onset of action occurs within 45 minutes following IM administration. The actual duration of tranquilization with IM and IV 5 injection depends on the dose but may persist for 6-10 hours (Thurmon et al., 1996). As the dose is increased, sedation is enhanced, but the dose response curve rapidly reaches a plateau after which higher doses do not increase, but only lengthen, sedation and increase side effects (Tobin and Ballard, 1979). In practice, the dose is chosen in relation to the length of sedation required and the purpose for which it is needed. A calming effect on the behavior of excitable animals can be seen at doses below 0.03 mg/kg of body weight, making ACE a drug prone to abuse especially in the equine performance industry. Excitement reactions are rare but have been reported following IV (Tobin and Ballard, 1979) or IM (Mackenzie and Snow, 1977) injection of the drug. The most prominent sign that sedation has occurred in male horses is the extrusion of the penis from the sheath. Other signs include, but are not limited to, some drooping of the eyelids and slight protrusion of the third eyelid. Clinical doses have little effect on respiration. Even though sedated animals may breathe more slowly the minute volume of respiration usually is unchanged (Muir et al., 1975; Tobin and Ballard, 1979; Parry et al., 1982). Hypotension is prone to occur particulary after IV administration of ACE and can result in fainting and recumbency. Geldings and stallions may experience penile prolapse when ACE is administered intravenously. In all species of animals, acepromazine causes a dose-related fall in arterial blood pressure (Kerr et al., 1972; Parry et al., 1982). This property is thought to be mediated through vasodilatation brought about by peripheral ? 1 adrenoceptor block (Hall et al., 2001). The effects of clinical dose rates of ACE on heart rates generally are minimal according to most investigators (Mackenzie and Snow, 1977; Kerr et al., 1972; Parry et al., 1982), while others report no changes (Muir et al., 1979). ACE has little 6 antihistamine activity but has a powerful spasmolytic effect on smooth muscle including that of the gut (Hall et al., 2001). It is metabolized in the liver and both non-conjugated and conjugated metabolites are excreted in the urine. Equine enthusiasts need to keep in mind that ACE will not transform an aggressive horse into a docile one but will reduce the animal?s awareness and response to external stimuli. Also, increasing the dose does not ensure more pronounced effects, as approximately 30-40% of horses administered ACE do not attain the desired effects. In the equine industry, ACE is used as an aid in training to produce standing restraint and to reduce stress prior to transportation because it is long acting, inexpensive, and does not produce severe ataxia. Because it tranquilizes without appreciably affecting coordination, ACE may be used to obtain better control of excitable horses at the racetrack or show arena and is thus prohibited in competition by most racing commissions and show horse organizations (Chou et al., 2002). In 1982, the operant behavior of the horse, under the influence of ACE, was measured by the use of a modified ?Skinner? box apparatus (Ballard et al., 1982). The horses were trained, without ACE, to bob their heads over a feed bucket which resulted in the interruption of a light beam and a consequent food reward. The horses were switched to a variable-interval reinforcement schedule where they were able to adjust quickly to the test stimuli and to form a stable baseline response. The effect of ACE on this baseline response was measured by administering increasing doses of the drug or equal volumes of saline control to the horses ten minutes prior to introduction into the behavioral stalls. Ballard et al. (1982) found that when ACE was administered to horses before performing 7 operant conditioned behavioral tests, a decreased rate of responding resulted. Horses showed a decrease in rate of response with doses as low as 0.004 mg/kg. Discrimination Learning in Horses Aside from classical conditioning, simple discrimination is one of the most elementary forms of learning in animals and humans. A variety of discrimination tasks have been used as indicators of animals? abilities to learn. During a discrimination task, the subject must respond to a stimulus that has been chosen as correct from among two or more stimuli presented. To keep the subject motivated in performing the task correct choices are reinforced (rewarded). The basic principle of the discrimination task is that those subjects capable of making the greatest number of correct choices have the greatest ability for learning. One of the earliest recorded two-choice discrimination studies involving horses (Gardner, 1937) was also one of the simplest in design. Subjects were required to discriminate between a feedbox covered in black cloth and a feedbox which was not covered in cloth. This study successfully demonstrated that horses were capable of performing correctly in a two-choice discrimination task, opening the door for more advanced methods. Mader and Price (1980) used two different three-choice visual discrimination tests to evaluate the effects of breed, age, and social dominance on discrimination learning ability of horses. Quarter Horses in this particular study learned significantly faster than Thoroughbreds, and no relationship between learning abilities and social dominance was 8 detected. A significant negative correlation was found between age and learning ability on only one of the tests. Discrimination tests are utilized to determine differences in learning ability between two or more treatment groups. In 1989, McCall used the two choice discrimination method to reveal that body condition and gender had no effect on learning ability in adult horses A series of studies investigated horses? abilities to learn the concept of sameness under several different conditions (Flannery, 1997). Three horses were shaped to touch individually presented stimuli with their muzzles and then to make two responses to two matching cards from an array of three before experimentation began. The task in each experiment was to select the two stimulus cards that were the same and to avoid the non- matching stimulus card. Flannery (1997) used a discrimination task to demonstrate that horses can learn the concept of sameness, and that they are able to generalize this learning to a novel stimulus presentation situation. These results suggest that a relational discrimination test may be useful for assessing horses? learning abilities and the level of training appropriate for individual horses. The ability of the horse to discriminate between colors has been investigated (Smith and Goldman, 1999) using a series of two-choice color versus gray discrimination problems. The results of this study demonstrated that horses do in fact have color vision that is at least dichromatic. Warren (1962) found discrimination reversal learning tasks to be a feasible technique for measuring learning ability in individual horses. Fiske and Potter (1979) used a serial reversal learning discrimination combining spatial and brightness cues to 9 investigate individual learning ability for yearling horses within a herd. Fiske and Potter (1979) reported the ?learning to learn? phenomenon in horses. Animals that utilize information learned previously to reduce the number of trials in subsequent learning sets are said to have learned to learn (Bitterman, 1965). Their research provided the groundwork for spatial learning tests. Spatial Learning in Horses The use of mazes in investigating learning behavior mainly has been applied to rodents. However, in the recent past this technique has been used with larger domestic animals, e.g. cattle (Arave et al., 1992; Boivin et al., 1992) and horses (Kratzer et al., 1977; McCall et al., 1981; Marinier & Alexander, 1994). Most mazes constructed for large animals involve a simple Y or T maze. Kratzer (1977) investigated maze learning in Quarter Horses using a two-compartment maze which provided a single left- or right-side choice, and found that the horses showed learning ability based on decreases in latency and in errors as trials progressed. The reliability of maze tests in equine learning research was investigated by Heird et al. (1986). They investigated whether two different maze tasks (discrimination and spatial), alternating over a series of testing but using the same horses, would yield consistent results. Heird et al. (1986) found that subjects learned more rapidly and reached higher levels of performance as the series of tasks progressed. It was found that learning occurred at a faster rate on the discrimination task compared to the gradual improvement in performance observed in the place tasks. 10 McCall et al. (1981) used Hebb-Williams closed field mazes to investigate learning abilities in yearling horses. Prior to testing, subjects were acclimated to the test area. Once testing began, subjects were tested on a new maze problem each day during a 12-day period. The horses quickly adapted to the testing procedure, were successful in learning each new problem, and the maze was useful in ranking the learning ability of the individual horses. Arave et al. (1992) performed a study with dairy calves that involved a more complex maze. Arave et al. (1992) found that activity of calves was affected by sire, gender, trial direction and food source location. There were significant correlations between maze completion time and activity scores, between activity score and laterality and between time and laterality on some days (Arave et al., 1992). Mariner and Alexander (1994) used a similar maze to evaluate changes in learning ability and memory in horses. Two separate mazes similar in complexity and length and containing the same number of turns and the same number of blind pockets were used in their study. They found that all horses learned the first maze, but variation existed between them. The ability of the horses to remember also varied between subjects. This study demonstrated the suitability of the use of the maze to investigate learning and memory in horses. Based on the results of the studies mentioned previously it can be concluded that spatial learning tests and discrimination tests are suitable ways to investigate learning ability in horses. However, none of these studies took into account the common practice in the horse industry of using a tranquilizer as a training aid. It is not known if horses can learn effectively while tranquilized, making successive attempts at a particular training or management procedure easier, or if tranquilization simply makes the horse more tractable 11 during the initial attempt. The objective of the present study is to compare learning performance of tranquilized and non-tranquilized horses using a visual discrimination learning test and a spatial learning test. 12 MATERIALS AND METHODS Subjects Forty mature horses of varying ages, breeds, and genders housed at the Auburn University Horse Unit were selected for this study. No horses used in this study had been exposed previously to the learning tests used in this study. All horses used in this study were utilized in teaching and riding activities by the Horse Unit and were accustomed to human contact and handling. Horses were maintained on pastures which were predominantly coastal bermudagrass (Cynodon dactylon). Horses were allowed free access to pasture forage during all non-testing hours but were deprived of their daily concentrate ration until testing was completed. Fresh water was available in the pasture ad libitum. Testing Procedure Horses were assigned randomly to either the control (3ml of 0.9% saline IM; 20 horses) or to the tranquilized (0.044 mg/kg acepromazine maleate IM; 20 horses) group. Horses then were assigned randomly to a specific test day. The study took place from May 13 through May 23 of 2004. Each horse performed two different learning tests on the same day with the test order randomly determined. On test days, each horse was fitted with a heart rate monitor 13 (Polar Accurex Plus, Polar Electro Inc., Kempele, Finland) and received an injection of either saline control or acepromazine maleate IM. Following injection, a 30 minute period was allowed for the ACE to induce tranquilization before testing commenced. At the end of testing horses were returned to their respective pastures. Discrimination Learning Test The discrimination test apparatus was constructed in a rectangular test stall with a packed soil floor measuring 5.6m by 3.1m. Three sides of the test stall were covered with a dark, heavy fabric, providing 1.7m high solid walls. The entrance way to the testing area was not enclosed with fabric; it consisted of a gate that could be closed behind the horse. The stimulus area of the test stall was divided by a 1.5m by 3.6m panel, constructed of metal pipe (livestock panel). A diagram of the test stall and the discrimination learning test apparatus is shown in Figure 2. For the discrimination learning test, the stimuli consisted of a white feed bucket and a black feed bucket placed on the floor of the test area. The stimulus buckets were removable and could be switched between compartments. Between test sessions, both stimulus buckets were stored with grain in them to provide similar olfactory cues. Horses were not acclimated to the test area prior to testing and each horse was tested only on one day. On the day of testing each horse was led to the test area, faced in the direction of the stimulus and released. The stimulus bucket not selected by the subject on this initial trial was designated the ?correct? stimulus for that subject during the testing period. The subject would complete 30 selection trials on testing day, not including the initial trail, and the bucket choices were recorded for each trial. 14 Figure 2: Discrimination test stall. Solid walls and partitions are designated by unbroken lines. Entryway is designated by a dashed line. Areas of feed buckets are designated by dot-dash lines. The divider is designated by a dotted line. 15 A selection trial consisted of the handler centering the horse directly inside of the entrance to the testing area and releasing it. The horse entered the testing stall unescorted and selected between the black or white stimulus buckets by entering a stimulus compartment and approaching the bucket. The presence of food in the stimulus bucket could not be determined visually from behind the stimulus area divider. Horses were considered to have made a recordable selection when their heart girth crossed into one of the stimulus compartments. Because the subjects could not see into the buckets before this point they were allowed to switch compartments if their posterior half was not already in the compartment. If the horse selected the correct color bucket, it would be allowed to eat the food reward (54.6g Life Design Compete, 10% CP, Nutrena Corporation, Minneapolis, MN) located within the bucket. However, an incorrect selection resulted in an empty bucket and no food reward. The subject then would be removed from test area after each trial was completed. The buckets would be prepared for the next trial while the handler faced the subject away from the stimulus wall. To eliminate biased cues, the sound of food being placed into both buckets could be heard. Then the horse was led into of the test stall, centered, and released for the start of the next trial. The discrimination was based on the presence of a relevant cue, the bucket color, and an ambiguous cue, spatial position. The spatial position of the stimuli within the stimulus compartments was reversed every five trials. 16 Spatial Learning Test The spatial learning test was a simple T maze that utilized a barn aisle measuring 13m by 2.7m and two stalls measuring 3.6m by 3.4m. The majority of the test area was covered with a dark, heavy fabric to prevent outside distractions. This created solid walls a minimum of 1.7 m high in all areas where visual contact with conspecifics might have been possible, such as common walls between the barn aisle and other stalls and the start of the maze and the storage stalls. The barn aisle and stall flooring was packed soil. A diagram of the spatial learning test apparatus is shown in Figure 3. Both stalls at the end of the maze contained an identical reward box that was fixed in position and could not be moved or knocked over by the subject. Between test days grain was stored in the reward boxes to provide uniform olfactory cues. To provide uniform olfactory cues during the test both reward boxes contained concentrate reward that was inaccessible by the subject. Subjects were not acclimated to the test area prior to testing. On the day of testing the subject was led to the start of the maze (leg of the ?T?) and released to travel down the maze. The side not selected by the subject on this initial trial was designated the ?correct? stimulus for that subject during the testing period. The subject would complete 30 trials on testing day, not including the initial trail, and the selections were recorded. Each subject was tested only one day in this study. A selection trial consisted of the handler centering the horse at the beginning of the maze and releasing it to travel down the maze. The horse traveled down the maze unescorted and selected between the right or left side of the maze by entering a stimulus compartment and approaching the box. The presence of food in the stimulus bucket 17 Figure 3: Spatial test apparatus. Solid walls are designated by unbroken lines. Entryway is designated by a dotted line. Areas of feed buckets are designated by dot-dash lines. 18 could not be visually determined from the entrance of the compartment. Horses were considered to have made a recordable selection when their front hooves crossed into one of the stimulus compartments. Since the subjects could not see into the buckets before this point they would have an opportunity to switch compartments if their front hooves were not already in the compartment. If the horses selected the correct compartment, it would be allowed to eat the food reward (54.6g Life Design Compete, 10% CP, Nutrena Corporation, Minneapolis, MN) located within the box. However, an incorrect selection resulted in an empty box and no food reward. The subject would be led from the stimulus compartment back to the beginning of the maze after each trial was completed. The boxes then would be prepared for the next trial while the handler faced the subject away from the stimulus compartments at the end of the maze. To eliminate auditory cues, the sound of food being placed into both boxes could be heard. The subject then would be turned around, centered, and released for the start of the next trial. All resting heart rates, prior to administration of treatment, were recorded. Heart rate then was recorded every fifteen seconds during each test period. This information was used to evaluate if either the tranquilizer or the testing procedure affected the heart rate in the tranquilized horses compared to the control horses. Statistical Analysis Heart rate data was analyzed using a GLM procedure of SAS (SAS Inst. Inc., Cary, NC). Correct responses resulting from the discrimination learning test and the spatial learning test were analyzed for variation between subject groups (Control/Tranquilized) using the SAS t-test procedure. 19 RESULTS AND DISCUSSION Three subjects were dropped from the discrimination test due to non-performance. To be classified as a non-performer a subject failed to complete ten trials, lasting five minutes each. At the end of each non-completed trial the subject was removed from the testing area. The subject then would be reintroduced to the test area at the start of the next trial. One subject was dropped from the spatial test due to a hoof abscess. Eight observations on the heart rate monitor were lost due to technical difficulty. Five of the files belonged to horses in the control group, while three of the files belonged to horses in the tranquilized group. It was difficult to establish and maintain a heart rate on some of the horses using the human electrodes. More accurate heart rate data may have been obtained using an electrode specifically designed for a horse. To improve heart data the electrode should be lubricated with electrode gel (Lectron II Heart Rate Monitor Electrode Gel) and placed in direct contact with the skin. The mean heart rate for the control and tranquilized horses in the discrimination test is shown in Figure 4. Mean heart rates for control and tranquilized groups (49.8+1.4 and 53.8+1.5 respectively), were not significantly different (P=0.06). The mean heart rates for control and tranquilized groups in the discrimination group approached significance. There are several reasons why this might have occurred, the main reason is 20 Mean Heart Rate (Beats per Minute) + S.E. of Control and Tranquilized Horses during the Discrimination LearningTest 0 10 20 30 40 50 60 70 Control (n=15) Tranquilized (n=17) Group M e an H e a r t R a te (b p m ) Figure 4. Mean heart rates for control and tranquilized groups was 49.8+1.4 and 53.8+1.5, respectively, during the discrimination test 21 that the discrimination test had an open back that allowed the horses to see con-specifics and to become easily distracted. To reduce distractions on the discrimination test it may be beneficial to put a solid barrier behind the horse, rather than the open barrier used in this study or to situate the foal area so that subjects are moving toward conspecifics when performing the test. During the spatial test no difference was detected (P=0.36) between mean heart rates for the two groups. The mean heart rate for the control and the tranquilized horses was 55.1+4.0 and 51.3+1.5, respectively, during the spatial learning test (Figure 5). Breed and gender were found to have no significant (P=0.97 and P=0.44, respectively) effect on heart rate during the discrimination test. Likewise, no significant differences were found between breed and gender (P=0.80 and P=0.17, respectively) during the spatial test. Heart rate differences between the control and tranquilized horses, were not expected. Acepromazine does not affect heart rate significantly (Kerr et al., 1972; Mackenzie and Snow, 1977; Muir et al., 1979). Mackenzie and Snow (1977) evaluated chemical restraining agents in the horse using five times the recommended dose of ACE. They found that ACE caused a slight elevation in heart rate approximately 60 minutes after administration. The heart rate monitors were used in this study as a reference point for the handlers. Once treatment was administered, the heart rate was used to establish that the horse had calmed down from the injection and had returned to its resting heart rate prior to testing. The mean percent correct responses for control and tranquilized horses in the discrimination test (67.6+2.0 and 69.8+2.0, respectively) did not differ (P=0.43); (Figure 6). Mean percent correct responses ranged from 43.4 to 80.0. In 1936, Gardner, completed a feed box discrimination test using draft breeds. The average box error 22 Mean Heart Rate (Beats per Minute) + S.E. of Control and Tranquilized Horses during the Spatial Learning Test 0 10 20 30 40 50 60 70 Control (n=14) Tranquilized (n=17) Group M ean Heart Rat e (b p m ) Figure 5. Mean heart rate for the control and the tranquilized horses was 55.1+4.0 and 51.3+1.5, respectively, during the spatial test. 23 Mean % Correct Responses+ S.E. for Control and Tranquilized Horses on Discrimination Learning Test 0 20 40 60 80 100 Control (n=18) Tranquilized (n=19) Group M e a n % C o r r ect R e s pon s e s Figure 6. Mean percent correct responses for the control and tranquilized groups was 67.6+2.0 and 69.8+2.0, respectively, for the discrimination test. 24 reported was 0.43 while in the current study the average error was 0.30. Sappington and Goldman (1994) found all their subjects successfully learned a simple black verses white discrimination task, complementary to the discrimination test used in the present study. The discrimination test detected no difference in mean percent correct responses for breed and gender (P=0.20 and P=0.60, respectively). Similarly, the spatial test detected no difference (P=0.13) in mean percent correct responses for control and tranquilized horses (72.5+3.6 and 80.0+3.4, respectively). This is illustrated in Figure 7. Mean percent correct responses ranged from 36.7 to 96.7. The error rate obtained during the spatial test is similar to the error rate approximately 20% reported by Kratzer et al. (1977). Heird et al. (1981) reported a similar learning performance in intermediately handled yearlings during initial acquisition of T-maze test. Warren and Warren (1962) also showed that two horses learned successive reversal problems very quickly. The spatial test detected no difference in mean percent correct responses for breed and gender (P=0.93 and P=0.17, respectively). Acepromazine maleate had no effect on the learning ability of horses at the doses used during this study. As shown, in this study, tranquilized horses had similar learning performance on simple discrimination and spatial test as those that were not. Therefore, ACE effectively can be used when teaching a horse a new procedure. However, the question remains how well can a horse then recall the information learned at a later point in time without the ACE? Drug-induced state-dependent learning is a well established phenomenon in psychology. The term is used to describe the finding that behavior learned in one drug state is better remembered when retention is tested in 25 Mean % Correct Responses + S.E. for Control and Tranquilized Horses on Spatial LearningTest 0 20 40 60 80 100 Control (n=20) Tranquilized (n=19) Group M e an % Co rrec t R e s pon s e s Figure 7. Mean percent correct responses for the control and tranquilized groups were 72.5+3.6 and 80.0+3.4, respectively, for the spatial test. 26 the same drug state (Lowe, 1986). In Lowe?s study 24 human subjects were required to learn a simple route map after ingesting 0.66g alcohol/kg body weight and smoking two medium tar cigarettes. The subjects were tested 24 hours later under various drug states. The highest recall was observed in the alcohol and nicotine subjects while the placebo drink and nicotine subjects, along with the no drug subjects, had the lowest recall. In studies with human subjects, alcohol (Lowe, 1981), marijuana (Darley et al., 1974), barbiturate and amphetamine (Bustamente et al., 1970) and nicotine (Peters and McGee, 1982) have all been shown to produce state-dependent learning effects. Does acepromazine produce state-dependent learning? Based on the above mentioned studies it is possible that if a horse learns under the influence of ACE, their recall should also be greatest under the influence of ACE. This study did not test the ability of the horse to recall the information that it learned. The sole purpose of this study was to see if a horse could learn under the influence of acepromazine. Now that it has been established that a horse can learn a simple spatial and discrimination test it would be interesting to see how the recall ability differs between control and tranquilized subjects. 27 IMPLICATIONS All of the equine subjects in this study exhibited similar abilities to perform simple discrimination learning tasks as well as simple spatial learning tasks. Therefore, ACE at the doses used in this study does not affect the horse's ability to learn these types of tasks. A horse is capable of learning a task while sedated. However, it is not known how well a horse can recall that information for later use. It would be of interest to investigate the effects of ACE on learning routine procedures that are commonly encountered in horse management (e.g., trailer loading) One of the limitations of this study was that it only considered whether the subject could learn under the influence of a tranquilizer. In retrospect, it would have been useful to test the recall of those same horses after a withdrawal period from the tranquilizer. It also might have been beneficial to record the temperature and humidity during test periods if testing during different times of day or over an extended period of time. Environmental conditions may affect motivation. In this study temperature and humidity were not recorded because all horses were tested within a two week period at the end of May. Testing begin each day at seven in the morning and concluded no later then noon. Thus, temperatures and humidity were not extremely variable over the test days in this study. Some of the test subjects also may have lacked motivation for performing the tests because they were allowed free access to pastures. In this study, it is not believed 28 that motivation from differential hunger was an issue. Those horses identified as non- performers were so classified due to distractions during testing or injury. The adult horses used in this study were used to receiving concentrate feed twice a day. If subjects seem to have motivation problems it might be advisable to place the horses in the stall 12 hours prior to testing. Because tranquilization with ACE makes the horse more tractable without significantly affecting learning performance, it is a tool that may be utilized effectively by less skilled handlers on fractious horses to perform mildly aversive procedures (e.g., trailer loading or clipping) while allowing the horse to learn to tolerate these procedures. However, because this tranquilizer is economical and easily administered it is often used unnecessarily in the equine industry. For example, ACE sometimes is administered prior to showing or sales inspections. When a horse is tranquilized prior to showing or sales inspections the true temperament of horses will be misrepresented. Tranquilizers should not be used in place of appropriate training. Since ACE often is used unnecessarily in the equine industry, one needs to consider animal welfare. In the case of an inadequate trainer trying to perform an aversive procedure on a fractious horse it may be to the horse?s benefit for the trainer to use ACE. An inadequate trainer can actually endanger the welfare of the horse and can cause the horse discomfort and pain. An inadequate trainer may try to force an un- tranquilized horse to perform an aversive procedure, potentially inflicting pain or causing injury to the horse, while a tranquilized horse will be more tolerant of the aversive procedure. For example, an inadequate trainer may resort to hurtful training procedures 29 or training devices with an un-tranquilized horse when tranquilization might allow the horse to respond without these aversive procedures. Future research might consider if ACE produces drug-induced state-dependent learning. That is, does the learning that takes place under ACE become less accessible to recall during the horse?s normal state? A better understanding of ACE?s impact on learning and tractability in horses would assist veterinarians, trainers, and horse owners in determining when and how ACE can be used as a tool to help horses learn to tolerate aversive procedures. This research may clarify if ACE really can be used as a tool to help horses learn to tolerate aversive procedures. Additional studies should look at the recall ability of horses under tranquilizer (ACE) compared to the recall ability of horses not under the influence of a tranquilizer, as demonstrated in human subjects by Lowe (1986). Lowe explored drug-induced state- dependent learning in humans; finding that behavior learned in one drug state is better remembered when retention is tested in the same drug state. According to Lowe, horses in the tranquilized group will recall more while under the influence of the tranquilizer. Finally, it might be beneficial to record the exact ages of the horses if all horses are not of mature age. Weanlings and yearlings with less training then mature horses can easily become distracted or lose motivation. The age of the horses used in this study were not collected because all horses were mature horses (5-23 years old) and were currently in a riding program. Drug interventions are likely to play an increasing role in horse training, care, and maintaining. Understanding their effects on learning and post- drug temperament is critical to humane and effective use of these chemical tools 30 LITERATURE CITED Arave, C. W., R. C. Lamb, M. J. Arambel, D. Purcell and J. L. Walters. 1992. Behavior and maze learning ability of dairy calves as influenced by housing, sex and sire. Appl. Anim. Behav. Sci. 33:149-163. Baer, K. L., G. D. Potter, T. H. Friend and B. V. Beaver. 1983. Observation effects on learning in horses. Appl. Anim. Ethol. 11:123-129. Baker, A. E. and B. H. Crawford. 1986. Observational learning in horses. Appl. Anim. Behav. Sci. 15:7-13. Ballard, S., T. Shults, A. A. Kownacki, J. W. Blake and T. Tobin. 1982. The pharmacokinetics, pharmacological responses and behavioral effects of acepromazine in the horse. J. Vet. Pharmacol. Therap. 5:21-31. Bitterman, M. E. 1965. Phyletic differences in learning. Amer. Psychol. 20:396-410. Boivin, X., P. Le Neindre, J. M. Chupin, J. P. Garel and G. Trillat. 1992. Influence of breed and early management on ease of handling and open-field behaviour of cattle. Appl. Anim. Behav. Sci. 32:313-323. Bustamente, J. A., A. Jordan, M. Vila, A. Gonzalez, and A. Insua. 1970. State dependent learning in humans. Physiol. Behav. 5:793-796. Chou, C. C., C. L. Chen, A. C. Asbury, A. I. Webb and T. W. Vickroy. 1998. Development and use of an enzyme-linked immunosorbent assay to monitor serum and urine acepromazine concentrations in Thoroughbreds, and possible changes associated with exercise. Amer. J. Vet. Res. 59:593-597. Chou, C. C., C. L. Chen, B. L. Rice and P. T. Colahan. 2002. Reduced resident time and pharmacodynamic effects of acepromazine after subclinical multiple dosage in exercised Thoroughbreds. J. Vet. Pharmacol. Therap. 25:379-382 Darley, C. F., J. R. Tinklenberg, W. T. Roth, and R. C. Atkinson. 1974. The nature of storage deficits and state-dependent retrieval under marijuana. Psychopharmacologia. 37:139-149. 31 Dougherty, D. M. and P. Lewis. 1991. Stimulus generalization, discrimination learning, and peak shift in horses. J. Exp. Anal. Behav. 56:97-104. Fiske, J. C. and G. D. Potter. 1979. Discrimination reversal learning in yearling horses. J. Anim. Sci. 49(2):583-588. Flannery, B. 1997. Relational discrimination learning in horses. Appl. Anim. Behav. Sci. 54:267-280. Gardner, L. P. 1937. The response of horses in a discrimination problem. J. Comp. Psychol. 23:13-34. Haag, E. L., R. Rudman and K. A. Houpt. 1980. Avoidance, maze learning and social dominance in ponies. J. Anim. Sci. 50:329-335 Hall, L. W., K. W. Clarke and C. M. Trim. 2001. Veterinary Anaesthesia. Harcourt Publishers, London, 561 pp. Heird, J. C., A. M. Lennon and R. W. Bell. 1981. Effects of early experience on the learning ability of yearling horses. J. Anim. Sci. 53:1204-1209. Heird, J. C., C. E. Lokey and D. C. Cogan. 1986. Repeatability and comparison of two maze tests to measure learning ability in horses. Appl. Anim. Behav. Sci. 16:103-119. Houpt, K. A., M. S. Parsons and H. F. Hintz. 1982. Learning ability of orphan foals, of normal foals, and of their mothers. J. Anim. Sci. 55:1027-1032. Kerr, D. D., E. W. Jones, M. S. Holbert, and K. Huggins. 1972. Comparison of the effects of xylazine and acetylpromazine maleate in the horse. Amer. J. Vet. Res. 33:777-784. Kratzer, D. D., W. M. Netherland, R. E. Pulse and J. P. Baker. 1977. Maze learning in Quarter Horses. J. Anim. Sci. 46:896-902. Lowe, G. 1981. State-dependent recall decrements with moderate does of alcohol. Curr. Psychol. Res. 1:3-8. Lowe, G. 1986. State-dependent learning effects with a combination of alcohol and nicotine. Psychopharmacology. 89:105-107. Ma, V. B. 1990. Influence of exercise on the pharmacokinetics of drugs. Clinical Pharmacokinetics. 19:32-43. 32 Mackenzie, G. and D. H. Snow. 1977. An evaluation of chemical restraining agents. Veterinary Record. 01:30. Mader, D. R. and E. O. Price. 1980. Discrimination learning in horses: Effects of breed, age and social dominance. J. Anim. Sci. 50:962-965 Marinier, S. L. and A. J. Alexander. 1994. The use of a maze in testing learning and memory in horses. Appl. Anim. Behav. Sci. 39:177. McCall, C. A. 1989. The effects of body condition of horses on discrimination learning abilities. Appl. Anim. Behav. Sci. 22:327-334. McCall, C. A., G. D. Potter, T. H. Friend and R. S. Ingram. 1981. Learning abilities in yearling horses using the Hebb-Williams closed field maze. J. Anim. Sci. 53:928-933. Mckelvey, D. and K. W. Hollingshead. 2003. Veterinary Anesthesia and Analgesia. Mosby, St. Louis, Missouri, 448 pp. Muir, W. W. 1981. Drugs used to produce standing chemical restraint in horses. Veterinary Clinics of North America. 3:17-44. Muir, W. W., R. T. Skarda and W. C. Sheehan. 1979. Hemodynamic and respiratory effects of xylazine-acetylpromazine drug combination in horses. Amer. J. Vet. Res. 36:1299. Muir, W. W., L. L. Werner and R. L. Hamlin. 1975. Effects of xylazine and acetylpromazine upon induced ventricular fibrillation in dogs anesthetized with thiamylal and halothane. Amer. J. Vet. Res. 36:1299. Parry, B. W., G. A. Anderson and C. C. Gay. 1982. Hypotension in the horse induced by acepromazine maleate. Austr. Vet. J. 59:148-152. Peters R. and R. McGee. 1982. Cigarette smoking and state dependent memory. Psychopharmacology. 76:232-235. Riebold, T. W., D. R. Geiser and D. O. Goble. 1995. Large Animal Anesthesia Principles and Techniques. Iowa State University Press, Ames, 304 pp. Rubin, L., C. Oppegard and H. F. Hintz. 1980. The effect of varying the temporal distribution of conditioning trials on equine learning behavior. J. Anim. Sci. 50:1184- 1187. Sappington, B. F. and L. Goldman. 1994. Discrimination learning and concept formation in the Arabian horse. J. Anim. Sci. 72:3080-3087. 33 Smith, S. and L. Goldman. 1999. Color discrimination in horses. Appl. Anim. Behav. Sci. 62:13-25 Taylor, P. M. and K. W. Clarke. 1999. Handbook of Equine Anaesthesia. W. B. Saunders Company, London, 194 pp. Thurmon, J. C., W. J. Tranquilli and G. J. Benson. 1996. Lumb & Jones? Veterinary Anesthesia. Williams & Wilkins, Baltimore, 928 pp. Tobin, T. and S. Ballard. 1979. Pharmacological review the phenothiazine tranquilizers. J. Equine Med. and Surgery. 3:460-466. Warren, J. M. and H. B. Warren. 1962. Reversal learning by horse and raccoon. J. Gen. Psychol. 100:215-220. 34 APPENDICES 35 Appendix 1: GLM procedure results for heart rate in the spatial test Treatment N Mean Standard Deviation Control 15 55.1 15.6 Tranquilized 17 51.3 6.1 P=0.36 36 Appendix 2: GLM procedure results for heart rate in the discrimination test Treatment N Mean Standard Deviation Control 15 49.8 5.4 Tranquilized 17 53.8 6.2 P=0.06 37 Appendix 3: T-test procedure results for mean percent correct responses in the discrimination test Treatment N Mean Standard Deviation Minimum Maximum Control 18 67.6 8.3 43.3 76.7 Tranquilized 19 69.8 8.9 46.7 80.0 P=0.43 38 Appendix 4: T-test procedure results for mean percent correct response in the spatial test Treatment N Mean Standard Deviation Minimum Maximum Control 20 72.5 16.2 33.3 93.3 Tranquilized 19 80.0 14.7 36.7 96.7 P=0.13 39 Appendix 5. Test date, treatment group and test order for each horse Date Name Group First Test Breed Gender 05/13/04 Dave Control Maze Thoroughbred Male 05/13/04 Money Tranquilized Discrimination Quarter Horse Male 05/14/04 Scarlet Control Discrimination Quarter Horse Female 05/14/04 Andy Tranquilized Maze Paint Male 05/14/04 GoGo Control Maze Thoroughbred Female 05/14/04 Regal Tranquilized Discrimination Thoroughbred Male 05/15/04 Yankee Control Discrimination Warmblood Male 05/15/04 Vinny Tranquilized Maze Quarter Horse Male 05/15/04 Montana Control Maze Quarter Horse Male 05/15/04 Radical Tranquilized Discrimination Quarter Horse Male 05/16/04 Flashy Control Maze Thoroughbred Male 05/16/04 Billy Tranquilized Discrimination Thoroughbred Female 05/16/04 Venture Control Discrimination Thoroughbred Male 05/16/04 Bourbon Tranquilized Maze Quarter Horse Male 05/17/04 Sunkist Control Discrimination Thoroughbred Male 05/17/04 Sterling Tranquilized Maze Thoroughbred Male 05/17/04 Major Control Discrimination Thoroughbred Male 05/17/04 Noah Tranquilized Maze Quarter Horse Male 05/18/04 Nie Control Discrimination Thoroughbred Female 05/18/04 Stick Tranquilized Maze Thoroughbred Male 05/18/04 Preacher Control Maze Thoroughbred Male 40 05/18/04 Tessy Tranquilized Discrimination Warmblood Female 05/19/04 Tejos Control Discrimination Thoroughbred Male 05/19/04 Bud Tranquilized Maze Grade Male 05/19/04 S. Sound Control Maze Quarter Horse Male 05/19/04 Flash Tranquilized Discrimination Quarter Horse Male 05/20/04 G. Effort Control Discrimination Quarter Horse Male 05/20/04 Blue Tranquilized Maze Quarter Horse Male 05/20/04 Mac Control Maze Grade Male 05/20/04 Gus Tranquilized Discrimination Thoroughbred Male 05/21/04 Ty Control Maze Thoroughbred Male 05/21/04 Babe Tranquilized Discrimination Quarter Horse Male 05/21/04 Dan Control Discrimination Thoroughbred Male 05/21/04 Franny Tranquilized Maze Thoroughbred Female 05/22/04 Nilla Control Discrimination Quarter Horse Male 05/22/04 Zip Tranquilized Discrimination Quarter Horse Female 05/22/04 Molly Control Discrimination Thoroughbred Female 05/22/04 Rubin Tranquilized Maze Warmblood Male 05/23/04 Dutch Control Discrimination Thoroughbred Male 05/23/04 Speck Tranquilized Maze Thoroughbred Male 41 Appendix 6: Number and percent correct responses for individual horses on the discrimination test. Date Name Group Number Correct Percent Correct 05/13/04 Dave Control 18 60.0 05/14/04 Scarlet Control 20 66.7 05/15/04 Yankee Control 19 63.3 05/15/04 Montana Control 20 66.7 05/16/04 Flashy Control 21 70.0 05/16/04 Venture Control 21 70.0 05/17/04 Sunkist Control 17 56.7 05/17/04 Major Control 21 70.0 05/18/04 Nie Control 20 66.7 05/18/04 Preacher Control 20 66.7 05/19/04 Tejos Control 20 66.7 05/19/04 S. Sound Control 22 73.3 05/20/04 G. Effort Control 23 76.7 05/20/04 Mac Control 13 43.3 05/21/04 Ty Control 21 70.0 05/21/04 Dan Control 23 76.7 05/22/04 Nilla Control 23 76.7 05/22/04 Molly Control 23 76.7 05/13/04 Money Tranquilized 22 73.3 05/14/04 Regal Tranquilized 14 46.7 05/15/04 Vinny Tranquilized 19 63.3 05/15/04 Radical Tranquilized 24 80.0 05/16/04 Billy Tranquilized 20 66.7 05/16/04 Bourbon Tranquilized 20 66.7 05/17/04 Sterling Tranquilized 24 80.0 05/17/04 Noah Tranquilized 24 80.0 05/18/04 Stick Tranquilized 16 53.3 05/18/04 Tessy Tranquilized 22 73.3 05/19/04 Bud Tranquilized 21 70.0 05/19/04 Flash Tranquilized 22 73.3 05/20/04 Blue Tranquilized 19 63.3 05/20/04 Gus Tranquilized 21 70.0 05/21/04 Babe Tranquilized 20 66.7 05/21/04 Franny Tranquilized 23 76.7 05/22/04 Zip Tranquilized 21 70.0 05/22/04 Rubin Tranquilized 23 76.7 05/23/04 Speck Tranquilized 23 76.7 42 Appendix 7: Number and percent correct responses for individual horses on the spatial test. Date Name Group Number Correct Percent Correct 05/13/04 Dave Control 21 70.0 05/14/04 Scarlet Control 26 86.7 05/14/04 GoGo Control 28 93.3 05/15/04 Yankee Control 11 36.7 05/15/04 Montana Control 22 73.3 05/16/04 Flashy Control 18 60.0 05/16/04 Venture Control 10 33.3 05/17/04 Sunkist Control 26 86.7 05/17/04 Major Control 28 93.3 05/18/04 Nie Control 25 83.3 05/18/04 Preacher Control 22 73.3 05/19/04 Tejos Control 25 83.3 05/19/04 S. Sound Control 23 76.7 05/20/04 G. Effort Control 16 53.3 05/20/04 Mac Control 20 66.7 05/21/04 Ty Control 24 80.0 05/21/04 Dan Control 23 76.7 05/22/04 Nilla Control 22 73.3 05/22/04 Molly Control 23 76.7 05/23/04 Dutch Control 22 73.3 05/13/04 Money Tranquilized 26 86.7 05/14/04 Andy Tranquilized 25 83.3 05/14/04 Regal Tranquilized 24 80.0 05/15/04 Vinny Tranquilized 22 73.3 05/15/04 Radical Tranquilized 27 90.0 05/16/04 Billy Tranquilized 20 66.7 05/16/04 Bourbon Tranquilized 11 36.7 05/17/04 Sterling Tranquilized 27 90.0 05/17/04 Noah Tranquilized 27 90.0 05/18/04 Stick Tranquilized 23 76.7 05/18/04 Tessy Tranquilized 27 90.0 05/19/04 Bud Tranquilized 22 73.3 05/19/04 Flash Tranquilized 16 53.3 05/20/04 Blue Tranquilized 27 90.0 05/21/04 Babe Tranquilized 25 83.3 05/21/04 Franny Tranquilized 25 83.3 05/22/04 Zip Tranquilized 26 86.7 05/22/04 Rubin Tranquilized 27 90.0 05/23/04 Speck Tranquilized 29 96.7 43 Appendix 8: Mean heart rate for each horse during the discrimination test. Date Name Group Mean BPM 05/13/04 Dave Control 55.5 05/14/04 Scarlet Control 54.2 05/15/04 Yankee Control 46.5 05/15/04 Montana Control 52.6 05/16/04 Flashy Control 46.8 05/17/04 Sunkist Control 43.8 05/17/04 Major Control 43.9 05/18/04 Nie Control 52.3 05/18/04 Preacher Control 42.3 05/19/04 Tejos Control 54.9 05/19/04 S. Sound Control 44.7 05/20/04 Mac Control 45.0 05/21/04 Ty Control 55.0 05/21/04 Dan Control 49.9 05/22/04 Nilla Control 59.4 05/13/04 Money Tranquilized 53.3 05/14/04 Regal Tranquilized 55.4 05/15/04 Vinny Tranquilized 43.1 05/15/04 Radical Tranquilized 59.8 05/16/04 Billy Tranquilized 45.9 05/16/04 Bourbon Tranquilized 50.8 05/17/04 Sterling Tranquilized 56.3 05/17/04 Noah Tranquilized 48.3 05/18/04 Stick Tranquilized 43.0 05/18/04 Tessy Tranquilized 50.4 05/19/04 Bud Tranquilized 59.6 05/19/04 Flash Tranquilized 50.9 05/20/04 Blue Tranquilized 59.9 05/21/04 Franny Tranquilized 63.4 05/22/04 Zip Tranquilized 59.1 05/22/04 Rubin Tranquilized 57.7 05/23/04 Speck Tranquilized 58.0 44 Appendix 9: Mean heart rate for each horse during the spatial test. Date Name Group Mean BPM 05/13/04 Dave Control 46.1 05/14/04 Scarlet Control 59.6 05/15/04 Yankee Control 104.8 05/15/04 Montana Control 56.6 05/16/04 Flashy Control 61.9 05/17/04 Sunkist Control 43.7 05/17/04 Major Control 45.3 05/18/04 Nie Control 54.6 05/18/04 Preacher Control 42.1 05/19/04 Tejos Control 51.2 05/19/04 S. Sound Control 40.0 05/20/04 Mac Control 47.4 05/21/04 Ty Control 56.0 05/21/04 Dan Control 55.7 05/22/04 Nilla Control 61.7 05/13/04 Money Tranquilized 49.2 05/14/04 Regal Tranquilized 62.3 05/15/04 Vinny Tranquilized 42.0 05/15/04 Radical Tranquilized 53.5 05/16/04 Billy Tranquilized 44.7 05/16/04 Bourbon Tranquilized 53.6 05/17/04 Sterling Tranquilized 50.5 05/17/04 Noah Tranquilized 42.8 05/18/04 Stick Tranquilized 45.4 05/18/04 Tessy Tranquilized 54.0 05/19/04 Bud Tranquilized 52.8 05/19/04 Flash Tranquilized 52.0 05/20/04 Blue Tranquilized 58.3 05/21/04 Franny Tranquilized 53.0 05/22/04 Zip Tranquilized 62.2 05/22/04 Rubin Tranquilized 44.9 05/23/04 Speck Tranquilized 51.3