Conditioning Attention As a Potential Reinforcer in Shelter Dogs

ABSTRACT

CONDITIONING ATTENTION AS A POTENTIAL REINFORCER IN SHELTER DOGS

There are currently millions of dogs residing in animal shelters. One issue contributing to dogs not being adopted is poor interactions with potential adopters. Research suggests that for some dogs, human attention may not be a reinforcer for behavior. This study evaluated and attempted to condition attention as a potential reinforcer for shelter dog behavior. Five shelter dogs were assessed as subjects within this study. This study consisted of two phases: identifying the reinforcing efficacy of human attention as a potential reinforcer, and assessing a conditioning technique to condition human attention as a reinforcer. For each appropriate response of nose-to-wand touching, reinforcement was delivered in the form of attention, food, or simultaneous presentation of both attention and food dependent upon the condition assessed. Results of this study demonstrated that human attention was not a strong reinforcer for most of the dogs studied. Furthermore, results demonstrated that human attention could be conditioned to be a reinforcer when paired with a primary reinforcer such as food. Limitations and future areas of research are discussed.

Alyssa Kristine Salazar May 2018

CONDITIONING ATTENTION AS A POTENTIAL REINFORCER IN SHELTER DOGS

by Alyssa Kristine Salazar

A thesis submitted in partial fulfillment of the requirements for the degree of Master of Arts in Psychology in the College of Science and Mathematics California State University, Fresno May 2018

APPROVED

For the Department of Psychology:

We, the undersigned, certify that the thesis of the following student meets the required standards of scholarship, format, and style of the university and the student's graduate degree program for the awarding of the master's degree.

Alyssa Kristine Salazar Thesis Author

Steven W. Payne (Chair) Psychology

Constance Jones Psychology

Sharlet Rafacz Psychology

For the University Graduate Committee:

Dean, Division of Graduate Studies

AUTHORIZATION FOR REPRODUCTION OF MASTER’S THESIS

X I grant permission for the reproduction of this thesis in part or in its entirety without further authorization from me, on the condition that the person or agency requesting reproduction absorbs the cost and provides proper acknowledgment of authorship.

Permission to reproduce this thesis in part or in its entirety must be obtained from me.

Signature of thesis author:

ACKNOWLEDGMENTS I would like to acknowledge certain individuals that have helped me through this process. To begin, I would like to give a heartfelt ‘thank you’ to Dr. Steven Payne for serving as the chair of my thesis. I appreciate all of your hard work and guidance that you have provided to me not only on my thesis but throughout my entire graduate stay. I could not have asked for a better mentor to help lead me through this crazy process and for that, I thank you. Without him, I would not have had the opportunity to work with or get slobbered on by my new four-legged friends! I would also like to thank Dr. Sharlet Rafacz and Dr. Martin Shapiro for taking the time to serve on my thesis committee as well as providing me with all of their feedback. Lastly, I would like to thank Dr. Constance Jones for taking the time to step in and take part in my thesis defense. A giant ‘thank you’ goes out to my amazing research assistants Fuchi Ly and Sandra Ruby for all of their unforgettable help. They were the ones who endured all of the dog saliva, early mornings, and never-ending barking so I could collect my data. I would also like to thank the Valley Animal Center for providing me with a location to conduct my research. Thank you to their staff for being so kind and allowing me to feed and pet their adorable puppies! I would also like to take the time to thank my funder the Association of Professional Dog Trainers Foundation for providing me with the finances I needed to cover all of my research costs. Finally, I would like to thank my parents, my brother, and Seth for being my amazing support system. Thank you for your love, encouragement, and strength you all provided to me when I needed it the most. To mom and dad, this v is for you! Jesse, I hope I made you proud big brother. To Seth, I am forever thankful for your love you continue to give throughout this journey. Thank you to everyone!

TABLE OF CONTENTS Page

LIST OF FIGURES ...... vii

CHAPTER 1: INTRODUCTION ...... 1

Introduction ...... 1

CHAPTER 2: LITERATURE REVIEW ...... 3

Attention as a Reinforcer for Human Behavior ...... 3

Attention as a Reinforcer for Dog Behavior ...... 4

Conditioning Attention as a Reinforcer for Human Behavior ...... 11

Conditioning Reinforcers for Non-Human Animal Behavior ...... 12

Purpose ...... 18

CHAPTER 3: METHODS ...... 19

Subjects ...... 19

Setting ...... 19

Materials ...... 20

Response Measurement and Interobserver Agreement ...... 21

Study 1 ...... 22

Study 2 ...... 24

Social Validity ...... 25

CHAPTER 4: RESULTS AND DISCUSSION ...... 26

Results ...... 26

Discussion ...... 36

REFERENCES ...... 42

APPENDIX: IACUC ...... 46

LIST OF FIGURES

Page

Figure 1. Data are depicted for Nutty for study 1...... 29

Figure 2. Data are depicted for Squishy for study 1...... 29

Figure 3. Data are depicted for Bailey for study 1...... 30

Figure 4. Data are depicted for Sascha for study 1...... 30

Figure 5. Data are depicted for Fudge for study 1...... 31

Figure 6. Data are depicted for Squishy, Bailey, and Sascha for study 2...... 34

Figure 7. Data are depicted for Squishy for the open-field test for study 2...... 35

Figure 8. Data are depicted for Bailey for the open-field test for study 2...... 35

Figure 9. Data are depicted for Sascha for the open-field test for study 2...... 35

CHAPTER 1: INTRODUCTION

Introduction According to the American Society for the Prevention of Cruelty to Animals (ASPCA, 2017), there are nearly 3.3 million dogs that enter animal shelters every year. This high statistic is the result of strays and previously owned dogs being brought into shelters with approximately 620,000 returned as owner surrenders (ASPCA, 2017). With the high numbers of dogs entering shelters compared to the low numbers animal shelters can house, nearly 670,000 dogs are euthanized each year (ASPCA, 2017). These euthanizations are often the result of dogs having a difficult time being adopted by shelter visitors. While there are multiple criteria that are a part of the adoption process, one important aspect that greatly influences adoption is the social interactions that dogs engage in with potential adopters. Protopopova and Wynne (2014) suggested that dogs who are overlooked for adoption were found to have ignored the initiation of play by potential adopters at a rate twice as much as dogs who engaged in play. Ignoring potential adopters comes to be a behavioral issue when shelter overcrowding leads to high euthanization rates to make room for new animal intake. Protopopova, Mehrkam, Boggess, and Wynne (2014) found that when a dog spent more time at the front of the enclosure near human interaction, potential adopters found this behavior to be desirable. Also, dogs that spent time simply standing in the kennel away from the front of the enclosure was found to be undesirable. Weiss, Miller, Mohan- Gibbons, and Vela (2012) described in-kennel behaviors such as sitting, gazing, not barking, and jumping on potential adopters as behaviors that promote adoption. They additionally reported that the initial interaction with potential 2 adopters was the most important variable in determining whether the dog would be adopted. With behaviors that involve human interaction being reported as highly important influencing factors contributing to adoption rates, if human attention is not a reinforcer for dog behavior, this may greatly decrease their chances of being adopted. Rather, the dog is more likely to be passed over for another dog that engages in behaviors such as tail wagging, approaching, greeting, and jumping on potential adopters (Weiss et al., 2012). As Protopopova and Wynne (2014) found, adopters are more sensitive to undesirable behaviors such as ignoring potential adopters than desirable behaviors such as greeting and approaching. Thus, ignoring attention will demonstrate a greater impact upon the chances of adoption more than desired behaviors. This supports the notion that dogs who make no attempt at responding to human attention will remain in a shelter for a longer length of time rather than dogs who appropriately respond. While there is extensive research conducted on animal behavior, it is important to assess the behavior of shelter dogs and their responses towards human attention as a reinforcer in order to increase the chances of adoption and decrease shelter euthanization rates.

CHAPTER 2: LITERATURE REVIEW

Attention as a Reinforcer for Human Behavior There have been few studies conducted within research literature that have assessed the relative efficacy of attention as a reinforcer for dog behavior. However, there have been multiple studies that have assessed attention as a reinforcer for human behavior. Though it has been suggested that social reinforcers may be difficult to establish for individuals with Autism Spectrum Disorder (ASD), researchers have identified attention as a frequently delivered consequence for both problem and appropriate behavior (Kelly, Roseoe, Hanley, & Schlichenmeyer, 2014; McKerchar & Thompson, 2004; Thompson & Iwata, 2001). Furthermore, attention has been found to be effective as a reinforcer for problem behavior and appropriate communication by individuals with ASD (Hanely, Iwata, & McCord, 2003; Piazza et al., 1999). Piazza et al. (1999) evaluated and compared the reinforcing effectiveness of attention in the forms of praise and reprimands for the maintenance of destructive behavior. The authors conducted a reinforcer assessment to compare both forms of attention for two individuals whose problem behavior was found to be maintained by attention following a formal functional analysis. The reinforcer assessment consisted of contingent interaction with specific colored toys that resulted in either reprimands or praise. Following the assessment, results suggested that attention in the form of reprimands was the higher quality reinforcer in maintaining problem behavior for both participants given their engagement with pre-determined colored toys that resulted in reprimands. With this, it is possible to identify social reinforcement as an effective reinforcer for maintaining human behavior in the form of destructive behavior. 4

Beavers, Iwata, and Lerman (2013) published a review of functional analyses of problem behavior. The purpose of this article was to provide a statistical analyses of functional analysis results with the use of published literature from multiple databases from January 2001 through May of 2012. Of 158 published functional analysis studies, 149 included a test condition that assessed social-positive reinforcement for problem behavior of human participants with 146 assessing the effects of attention in particular. The authors of the total studies reported 91.7% of differentiated results. Of the results, social-positive reinforcement was the second highest reported finding, at 29.2% for maintaining problem behavior, and 17.2% of functional analyses found attention to be a maintaining variable. These results suggest that social-positive reinforcement in the form of attention does indeed serve as a reinforcer for human behavior.

Attention as a Reinforcer for Dog Behavior In terms of the reinforcing efficacy of attention as a reinforcer for dog behavior, Stanley, Morris, and Trattner (1967) assessed a passive person as a potential reinforcer with Shetland sheepdog puppies who were exposed to extinction and separated into either a P-NP-P group or NP-P-NP group. This was done by evaluating eighteen Shetland puppies and their mean speed in which each group made their way across an enclosure within two conditions: person (P) and no person (NP). The purpose of the person and no person conditions were to represent conditioning with a person present and extinction when a person is not present. To begin, the study developed an enclosure with one corner designated as the start area and a square that was placed under three meters away from the start which was determined to be the goal. The experimenters placed each puppy into a condition that contained a person within the enclosure and a separate condition 5 without a person. In the person present condition, an experimenter passively sat in the goal square, while the no person condition left the puppy alone in the enclosure. The authors recorded the running time in seconds that it took each puppy to run from the start area to the goal square once they were released inside of the enclosure. Results of this study were taken from sessions that were ran across trial blocks. Each trial block occurred every 3 days and results indicated a significant increase in running speed for both groups of puppies across both conditions. However, across three daily trial sessions, there was a temporary decrease in performance from the first trial to the third. The authors suggested that this was due to satiation over time from one trial to the next. Overall, the authors determined that given the increase in speed each puppy revealed within the person condition, a passive person can serve as a reinforcer for dog behavior. McIntire and Colley (1967) conducted a study that continued to examine the effectiveness of social reinforcement with dog behavior. This study was conducted with six German Shepherds who were provided positive verbal praise statements accompanied by physical contact in the form of rubbing on the ears and neck for approximately 5 to 10 seconds contingent upon following commands. Training sessions were conducted with a panel that consisted of seven lights in which five were associated with commands, one that signaled the end of the session, and the last light indicating latency. The purpose of these training sessions was to provide a positive consequence contingent on appropriate responses across 45 days. For days 1 through 8 and 25 through 35, no petting was provided for appropriate responses, while on days 9 through 24 and 36 through 45 petting was provided. Results of this study indicate that there was a significant decrease in latency to responding when petting was provided as the consequence in comparison to an increase in latency to responding when no petting was provided 6 as a consequence. This suggests that when providing social reinforcement to a dog, petting is an important variable to pair with praise to decrease latency of responding in dogs. Furthermore, given that there were no additional primary reinforcers such as food delivered, the evidence of this study support the idea that human attention may be a primary reinforcer for dog’s behavior. However, it is also possible that human attention had previously been conditioned as a reinforcer for the dogs’ behavior. In order to distinguish between the types of human interaction that dogs prefer, Feuerbacher and Wynne (2015) assessed dogs’ preference between physical petting from humans versus vocal praise. This was done with a mixture of both shelter and owned dogs and separated into three groups of fourteen dogs each. Each group consisted of shelter dogs, owned dogs that were exposed to strangers, and owned dogs that were exposed to their owners. The experiment took place within a room located at the animal shelter for shelter dogs, or at a daycare lobby in which the owned dogs were left for the day. Within each room, there was an exercise pen that had two chairs placed on the same side of the room and located one and a half meters apart. Around each chair was a taped off circumference that would indicate if the dog was within proximity of the individual seated in the chair. For the response to count as proximity, it was required that the dog have two paws in or on the tape, any part of its body minus the tail maintaining contact with the experimenter, and a requirement of at least 50% of the body within the perimeter that was set even if the dog was sitting, laying down, or standing. The purpose of the two chairs was to differentiate between one assistant that provided petting while the other provided vocal praise. As long as the dog met the criteria of proximity, petting or vocal praise using a high tone of voice were provided as consequences. 7

Each session was split into two, 5-minute periods, and during the first period either petting or vocal praise was provided for proximity. During the second period, the assistants changed the provided consequence, so if the assistant provided praise in the first period they would now be providing petting in the second period. This was done in order to demonstrate a preference that would continue to occur from the first period into the second period. For example, if a dog preferred petting in the first period and the experimenters changed consequences, the dog would be expected to go to the other experimenter that is now providing petting. Within each session, measurement was taken on each 5- minute period, the initial choice that the dog went to, the duration spent during each alternative, and the number of alterations that occurred between each alternative consequence. Results of this study indicated that the time allocated within each alternative across all four groups was spent with petting rather than vocal praise. This demonstrates that across all groups, dogs demonstrated a higher preference for petting when provided with a concurrent choice. Furthermore, when a stranger was providing petting, and the owner of the owned dogs was providing vocal praise, the owned dogs allocated their time towards the stranger providing petting. Feuerbacher and Wynne (2015) conducted a second experiment that assessed whether or not dogs would continue to remain social and within proximity to an assistant when only provided with vocal praise as a source of social interaction with humans. This was done with shelter and owned dogs and was conducted within the same settings as experiment one. However, only one chair was placed within each session rather than two. There were three conditions the dogs were exposed to as well as a separation into two groups of 12 shelter dogs and 12 owned dogs. The first condition consisted of shelter dogs being provided 8 with petting versus vocal praise. The second condition was vocal praise versus no interaction, and the final condition was a single alternative of petting only. These conditions were also carried across for the owned dogs. The sessions were 3 minutes each, and when the dogs were within the taped off proximity of the assistant, the designated consequence was provided as long as the dog met the same criteria of proximity set in experiment one. Each session was measured using the duration spent within proximity of the experimenter, and the frequency of bouts where the dog would enter the proximity of the assistant. Results of the first condition (petting versus vocal praise) indicated that across both shelter and owned dogs, there was a significant amount of time spent in proximity to the experimenter during the petting consequence as compared to vocal praise. Furthermore, across both shelter and owned dogs, there were significantly fewer bouts of interactions during the petting condition as compared to vocal praise. During vocal praise versus no interaction, there were no significant effects found across shelter and owned dogs, and there was no preferred choice between vocal or no interaction. Finally, during the petting condition, there was no main effect that was found for the type of dog, owned or shelter, or the session number in which petting was provided for eight sessions. Furthermore, when compared to an additional two no interaction sessions that were conducted for seven dogs across both the types of dogs, results indicated that there was a significantly less amount of time that was spent within proximity to the experimenter when compared to petting. With these results, the experimenters were able to determine that petting was the preferred consequence provided to both owned and shelter dogs in comparison to vocal praise even when provided by the owner. 9

Feuerbacher and Wynne (2012) attempted to identify the relationship between human social interaction as a reinforcer for dog behavior in comparison to a primary reinforcer: food. This study hypothesized that if attention functioned as an easily generalized primary or conditioned reinforcer, there would be an increase in shelter dogs seeking human interaction given the high amount of time deprived of human social reinforcement. The authors used shelter dogs, owned dogs, and hand raised wolves as subjects who were reinforced for an appropriate response of nose-to-hand touches with humans. Three separate conditions across five experiments were assessed: social interaction, food delivered by hand, and extinction. The first experiment consisted of shelter dogs and evaluated social interaction versus food as a delivered reinforcer for nose-to-hand responding. The sequence of conditions was counterbalanced across nine dogs and results indicated that for most dogs, social interaction was a weak reinforcer. Furthermore, the latency for responses within the food condition was shorter than when social interaction was the delivered consequence. However, one subject did show higher rates of responding across both the food and social interaction conditions. This suggests that that social interaction may be an effective reinforcer for some dogs, but this may be idiosyncratic across dogs. The second experiment aimed to compare three conditions across shelter dogs: social interaction, food, and extinction. Results indicated that once again, social interaction conditions and extinction conditions produced low rates of responding as compared to the food conditions. The authors also concluded that if social interaction functioned as a reinforcer, it was either low in value or rapidly decreased in value due to satiation. The third experiment was a replication of experiment two and utilized owned dogs rather than shelter dogs. Similar to experiment two, results indicated that the food 10 condition demonstrated high responding while social and extinction conditions provided low responding. The fourth experiment used shelter dogs and provided them with social exposure prior to a food condition to assess if social interaction would strengthen in value. However, results indicated that social interaction functioned at a lower value than food as a reinforcer. The last experiment evaluated the same conditions as experiment two and three with captive wolves. As compared to domestic dogs, wolves also demonstrated similar results of low responding within social and extinction conditions as compared to higher rates of responding within food conditions. Overall, the study suggests that for both dogs and wolves, human social interaction was not a powerful reinforcer for responding as compared with food. Although the study by Feuerbacher and Wynne (2012) was able to effectively evaluate the reinforcing efficacy of social interaction between dogs and humans, there were some limitations to this study. First, only three sessions were assessed within each condition which may not be able to account for successful discrimination between each condition. Furthermore, the authors did not attempt to condition attention as a potential reinforcer and evaluate whether it affected shelter dogs and their interactions with potential adopters. Rather, rates of responding were merely compared between conditions with regards to social, food, and extinction consequences. Lastly, the interaction between humans and dogs during the food condition could serve as a potential limitation. By hand-delivering morsels of food to each subject, physical contact was made between the person delivering food and the dog receiving food. It is possible that this additional interaction outside of food delivery may have influenced the results of the study by increasing the quality or magnitude of reinforcement in the food condition. 11 Conditioning Attention as a Reinforcer for Human Behavior Though there is little research conducted on conditioning attention as a potential reinforcer with dogs, there have been multiple studies done regarding conditioning attention as a reinforcer with human subjects. Conditioning techniques have been utilized and provided effective evidence in modifying behaviors towards the preferred direction with children (Walker & Buckley, 1968). For example, attention in the form of praise has demonstrated its reinforcing effects as a consequence to acquire and maintain appropriate behavior such as social interaction (Barton, 1981; Dozier et al. 2012; Strain & Timm, 1974). With human behavior, Dozier, Iwata, Thomason-Sassi, Worsdell, and Wilson (2012) compared two conditioning procedures in an attempt to condition praise as a potential reinforcer: stimulus-stimulus pairing and response-stimulus pairing. This study by Dozier et al. (2012) was separated into two separate studies that each individually evaluated one of the two conditioning procedures. The first study consisted of the stimulus-stimulus pairing procedure across four individuals where a primary reinforcer (edible) was paired with 10 novel statements of praise. This was done by pairing both stimuli on a fixed-time (FT) 15-second reinforcement schedule. To begin, there were five conditions assessed that included baseline, praise, pairing condition of food plus praise, praise as a test, and praise as a test with food present. The first condition consisted of baseline where no consequences were provided for each participant engaging in a target response when task materials were present. The praise condition consisted of each response resulting in the delivery of one out of 10 predetermined praise statements without repetition of praise statements. The third condition consisted of providing a praise statement prior to the immediate delivery of an edible item contingent on the 12 participant engaging in the target response on the FT 15-second schedule. Following five sessions of the third condition, the praise condition was reinstated. Finally, the last condition of praise as a test with food present followed the praise condition. This condition was assessed in order to evaluate whether the edible item present served as a discriminative stimulus for each participant to engage in the target response while only praise was delivered as the consequence following each response. Results indicated that for all four participants, praise when delivered alone did not function as an effective reinforcer. Furthermore, the stimulus pairing procedure was not successful in conditioning praise as a reinforcer. The second part of the study by Dozier et al. (2012) assessed conditioning response-stimulus pairing across eight individuals in an attempt to condition attention (praise) as a reinforcer. This was done by the simultaneous delivery of both an edible item and praise as a consequence in one condition and was compared against delivering praise alone in a separate condition. Baseline was conducted and was identical to baseline during study 1. The praise condition consisted of providing each participant 1 out of 10 statements of praise on a fixed- one ratio (FR1) following the target response. The last condition was the food plus praise condition in which 1 out of 10 praise statements was delivered as a consequence simultaneously with food on an FR1 schedule. Results of study 2 indicated conditioning of praise as a potential reinforcer was effective for four subjects when delivered alone for a target behavior. In addition, it was also effective in establishing an increase in responses for additional target behaviors.

Conditioning Reinforcers for Non-Human Animal Behavior Despite research regarding the conditioning of attention as a reinforcer for human behavior, little research that has been conducted assessing the efficacy of 13 attention as a reinforcer for dog behavior. However, there has been literature published that attempted to evaluate the reinforcing effect of other conditioned reinforcers (i.e., clickers). Clickers are a mechanical device that when pressed, emit an auditory “click-clack” sound (Chiandetti, Avella, Fongaro, & Cerri, 2016). They were originally created by Marian Breland and her husband in order to train animals from a distance (Gillaspy & Bihm, 2002). According to Chiandetti et al. (2016), the clicker is used to reinforce a response that is emitted in operant conditioning and serves as a conditioned stimulus or secondary reinforcer when reliably anticipating a reward. Furthermore, the auditory sound will also be used as a filler within the time delay following the appropriate response and prior to the primary reinforcer being delivered (Pryor, 1999). Feng, Howell, and Bennett (2016) conducted a review of three proposed mechanisms of clicker training: The Reinforcing Hypothesis, Marking Hypothesis, and Bridging Hypothesis to begin to understand the means in which clickers operate as a potential conditioned reinforcer in dogs. For this paper, we will examine The Reinforcing Hypothesis that is associated with the use of secondary reinforcement as a consequence to increase the likelihood of a response. The clicking sound of the clicker, also known as a SIGNAL, is a previously neutral stimulus that upon repeated pairings with a primary reinforcer such as food acquires the reinforcing value of said primary reinforcer. Once established, the SIGNAL becomes a PREDICTOR SIGNAL as suggested by Skinner (1938), will therefore increase the rate of responses, resistance to extinction, and reinforce novel responses. Egger and Miller (1962) tested lever pressing in rats when presented with a PREDICTOR SIGNAL that had a history of pairing with food versus a neutral tone plus food. Results indicated that in the presence of the PREDICTOR SIGNAL, rate of responding was significantly higher than the 14 neutral tone, strengthening the Reinforcing Hypothesis. With respect to resistance to extinction, Bugelski (1938) placed rats’ lever pressing on extinction and compared the results to a group that was not provided auditory feedback following a response. Results of this study indicated that the rats who were subjected to the PREDICTOR SIGNAL took longer for their responding to decrease than those provided with no feedback. Finally, Skinner (1938) assessed a PREDICTOR SIGNAL and its ability to train a novel behavior. This was done with rats who were provided an auditory SIGNAL that was paired with food as a consequence of 60 presentations. Following pairing, a novel task was presented to the rats, and the results indicated the rats responding was similar to extinction in which a high rate of responses occurred followed by nearly zero responses following 30-50 trials. With this, a PREDICTOR SIGNAL is suggested to be a secondary reinforcer which over time lost its reinforcing efficacy due to a lack of pairing with a primary reinforcer. Furthering research on clicker training, Smith and Davis (2008) attempted to determine whether clicker training would decrease the time it took to train a novel behavior as well as increase resistance to extinction. Smith and Davis (2008) used 35 purebred basenji dogs with no prior experience with clicker training who were deprived of food for a minimum of three hours prior to the start of each session. All sessions were conducted within the same setting of each dog’s home and the dogs were separated into two groups: clicker and control. The clicker group was conditioned using a clicker while the control group was not. Within each session, one trainer and one assistant were present with a traffic cone placed in between the trainer and the bowl as well as the assistant in the opposite area of the room behind the dog. The conditions the dogs were exposed to included 15 conditioning, training, strengthening, and extinction with each condition set at a specific criterion for the dog progress into the following condition. Prior to the start of the conditioning sessions, Smith and Davis (2008) provided each dog with pieces of food that were placed in the bowl by the trainer. Once the sessions began, food was provided into the dog bowl approximately 1- second following the click of the clicker. To meet mastery criteria, each dog was required to eat food from the bowl within 5 seconds of the clicker sound for 10 consecutive trials. The training sessions were conducted following conditioning, and the purpose of this phase was to teach each dog a novel behavior of touching their nose to the traffic cone. This phase had a limit of 40 trials per day and began on the same day as the conditioning criterion was mastered. Each appropriate response of nose touching to the traffic cone was reinforced with a click sound which was immediately followed by food within the dog bowl while the control group was only provided with food. The training phase was conducted using a 4- step shaping procedure: looking at or touching the cone, latency to the cone after being lured and reinforced with food, the trainer pointing to the cone and saying “touch,” and finally the trainer only providing the demand to “touch” the cone without pointing. Following training, the strengthening phase was conducted. Smith and Davis (2008) assessed the strengthening phase to increase the resistance to extinction for the novel nose-touching behavior that was shaped in the training phase (Pierce & Cheney, 2004). This phase was comprised of 40 trials and began on the same day as training was mastered. Each dog’s first emittance of nose- touches was reinforced on a FR-1while the following nose-touches were placed on a VR-2 schedule. Criterion for mastery was set for 2 consecutive days of responding at less than 5 seconds latency for 9 out of 10 trials. 16

The final phase that was implemented was extinction to assess whether the dogs continued to emit the nose touching behavior in the presence of the clicker and the absence of primary reinforcers (food). Dogs within the clicker group were provided with only the clicker sound while the control group was not reinforced following responding. Criterion was set in the absence of responding within thirty second trials for 10 consecutive sessions. Results of the Smith and Davis (2008) study indicated that during all phases, dogs took approximately 2 to 6 days to meet all mastery criterion. Furthermore, it was found that during the conditioning phase, there was no difference between both subject groups, control and clicker, for criterion to be met. For the training phase, there was also no difference between both groups and the total duration of trials to meet criterion. The strengthening condition also found similar results between both groups as well as within the extinction criterion. However, it should be noted that the clicker group did take longer as well as more trials to reach criterion; suggesting that the clicker sound may have served as a conditioned reinforcer. With this, Smith and Davis (2008) concluded that clicker training did not significantly decrease the amount of time required to train dogs. However, clicker training might be useful in maintaining behavior when primary reinforcers are not readily available. A second study that evaluated the efficacy of clicker training was conducted by Chiandetti et al. (2016). The purpose of this study was to assess if clicker training reduced the time required to train dogs to engage in novel behaviors and increases generalization of the learned behavior compared to a shaping procedure utilizing other reinforcers. This study used 51 dogs with no prior experience in clicker training that were separated into three equal groups: clicker, bravo, and reward only. The same trainer was assigned to all three groups 17 and would provide either the clicker sound, the auditory “bravo” praise, or food as reinforcement. The novel behavior that was trained for all dogs was opening a bread-box apparatus by pushing the wooden black rubberized handle to open it. The second apparatus used to test for complex generalization was a larger box with plastic inserts for a guillotine opening. The conditions evaluated were: preparatory, training, and test. The preparatory phase was conducted in order for the trainer to build rapport with each dog as well as present two tasks: touching the nose and placing a paw on a box. The dogs within the clicker group were provided with a click as a food predictor while the dogs in the bravo group were provided with a spoken word “bravo” in place of the click. Following the preparatory condition, training was provided for each dog with food being delivered as reinforcement for the behavior of opening the bread box. There were three sessions conducted per day for each dog and the experimenters would end one session with the dog responding appropriately ensuring all dogs gained reinforcement for the correct response prior to starting the sequential session. Criterion was set at 8 of 10 consecutive attempts of opening the bread box for each dog to move on to testing. The testing condition was conducted 1 week following the training condition and was comprised of a recall phase where each dog was required to perform the learned behavior for three out of five consecutive attempts. Following recall, each dog was presented with two unreinforced trials 5 hours later where they were presented with a complex apparatus to demonstrate generalization. Each trial was set at 5 minutes in duration and was scored by time and frequency of attempts to complete the first motor action, time and frequency of attempts from the first action to the first trial in the sequence to criterion, and the time and frequency of attempts to solve the simple and complex test. Results of Chiandetti et al. (2016) 18 found that there was no significant effect found for the presentation of an auditory reward predictor such as a clicker or spoken word. Furthermore, there were no significant differences found between the performances of each group during training or the generalization test. There are several studies in the literature that have successfully evaluated conditioned reinforcers and their use with both humans and animals. While there are studies that have successfully assessed human attention as a conditioned reinforcer for humans, there is a lack in research assessing attention as a potential conditioned reinforcer for dog behavior.

Purpose The purpose of this study was to identify and assess the reinforcement value of human attention shelter dog behavior. Furthermore, there was an examination into how human attention could be conditioned as a reinforcer for shelter dogs. This was done across two studies. The first study compared the reinforcing efficacy of attention and food for the behavior of shelter dogs. The second study was to determine if one could condition and increase the reinforcing efficacy of human attention with shelter dogs for whom attention was not a strong reinforcer. Finally, social-validity measures were taken in order to determine the effect of conditioning attention as a reinforcer on more naturalistic human interaction.

CHAPTER 3: METHODS

Subjects Five dogs (Canis lupus familiaris) were chosen to participate for each part of the study. Dogs were excluded from this study if they had a history of engaging in aggression towards people or were less than 1-year-old. All subjects that were chosen and assessed during this experiment were a part of a population of dogs that were available for adoption at the no-kill shelter Valley Animal Center. Furthermore, only subjects that demonstrated little to zero responding in the attention conditions of study 1 were used for study 2. Prior to selection, all subjects were vaccinated, spayed/neutered, and were treated for any medical issues. All subjects were identified by the names provided by the Valley Animal Center. Of the five dogs used, Bailey (3 y.o., female), Squishy (5 y.o., male), and Sascha (6 y.o., female) were American Staffordshire Terriers. Nutty (1 y.o., male) was a Jack Russel Terrier mix, and Fudge (1 y.o., male) was a Miniature-Pinscher Chihuahua mix. Sascha and Squishy were individually housed in their own kennels, and Bailey, Nutty and Fudge were in a shared housing with two or more dogs. Sascha spent 4 years in the animal shelter, Bailey spent 1 year, Squishy spent 2 years, and both Nutty and Fudge had spent 2 months in the Valley Animal Center.

Setting All sessions were conducted within the Dr. James W. Thornton Dog Park facility that was accessible by the Valley Animal Center. Within the outdoor facility, an area that was closed off to the use of other visitors was used for each session. This area consisted of a metal-gated pen that was closed off and was approximately 2-meters by 3-meters in length. The dogs were allowed to freely 20 roam this area with the experimenter present during sessions. In-between sessions, the dogs were released into an adjacent fenced off grass arena and were allowed to freely roam until the start of the next session. In-kennel sessions were conducted for Fudge (study 1 and 2), Nutty (study 1 and 2), and Squishy (study 2) following an outbreak of canine influenza that led to a temporary closing of the dog park. The kennels were approximately 1.5 meters by 1.5 meters and were separated by an indoor and outdoor fenced off kennel area. All in-kennel sessions were conducted on the outside section of each dogs’ kennel with access to the indoor area closed off by a guillotine door that separated the inside and outside of the kennel. The dogs were allowed to freely roam the outdoor kennel with the experimenter standing inside of the kennel during each session. An open field-test was conducted prior to and following study 2. This was conducted inside of the ‘meet-and-greet room’ located inside of the Valley Animal Center. This room was approximately 2-meters by 3-meters and was used by the shelter to present potential adopters with the shelter dog they were interested in adopting. Three of the walls were opaque concrete, while the fourth wall was transparent glass that faced the parking lot area of the shelter. The dogs were allowed to individually enter the room and freely roam with the experimenter present.

Materials For both parts of this study, the food that was provided was Purina Moist & Meaty food pouches. An application called Countee was also downloaded onto the experimenters’ cell phones to collect data within each session. This app was preset to the appropriate amount of time required per session to track frequency of 21 appropriate responding. Furthermore, an automated food dispenser called Treat & Train was used to reinforce appropriate responding. The Treat & Train contents included one food dispenser, one upright standing wand, and one remote control available to the experimenter. It was only present in the food conditions of study 1.

Response Measurement and Interobserver Agreement

Dependent Variable The dependent variable was the frequency of responding via the dog touching its nose to the upright standing wand. One instance of appropriate responding counted as the dog touching its nose, mouth, or any area of the muzzle to any area of the wand with a 2-second inter-response time. Each instance was counted as one occurrence and was summarized using the Countee app. During the social validity assessment, the duration of time the dog spent in the taped-off area was measured using a timer that was also available and preset to a total session time of 300 seconds on the Countee app.

IOA Interval-by-interval interobserver agreement (IOA) was evaluated in order to assess interrater reliability across a minimum of 33% of all sessions for studies 1 and 2. This was done by calculating each interval between two observers’ agreement and disagreement on the occurrence or non-occurrence of behavior. The total number of agreed intervals were divided by the sum of the total number of agreed and disagreed intervals and multiplied by 100. For Sascha, IOA results for 33% of all sessions for study 1 were 99.2% (range 93%-100%) and 86.1% (range73%-100%) for study 2. For Bailey 33% of all sessions were calculated for IOA, results were 100% for study 1 and 93.1% 22

(range 73%-100%) for study 2. For Fudge, IOA was also calculated across 33% of all sessions and was 94.3% (range 60%-100%) for study 1. For Squishy, 33% of all sessions were calculated and IOA was 99.7% (range 97%-100%) for study 1 and 96.2% (range 87%-100%) for study 2. Lastly, for Nutty, IOA was 100% for study 1 across 33% of all sessions.

Research Design The first study consisted of a reversal design counterbalanced across all dogs. The second study consisted of a non-concurrent multiple baseline across subjects with an embedded ABA reversal design in order to assess for experimental control. A pre- and post-test design was used for the open-field test that was conducted prior to and following study 2. All sessions of study 1 and 2, as well as the pre-post design lasted 5 minutes and contained no less than two experimenters present at all times.

Study 1

Purpose The purpose of study 1 was to identify and evaluate the reinforcing efficacy of human attention versus the reinforcing efficacy of food with shelter dogs. This was done by comparing the rate of responding when attention was delivered as a consequence to the rate of responding when food was delivered as a consequence for responding.

Procedures

General procedures. Each session within study 1 lasted a total of 5 minutes and the same experimenter conducted every session across the study. Each session 23 began with a pre-exposure to the contingencies as follows: prior to each session each dog was prompted to engage in the correct response through gestural procedures and provided the consequence dependent upon the condition that was being ran. Sessions were conducted across several days and no effort was made to control the number of sessions or conditions conducted each day. Sessions were conducted at least 3 hours following scheduled meal times to attempt to control for food deprivation. However, no attempts were made to control for the immediate history of attention from people given the shelter environment. During each session, the experimenter stood in the corner of the session area.

Baseline. During baseline, there were no programmed consequences provided for responding of any kind by the dog.

Attention. During the attention condition, attention was provided in the form of physical and vocal attention for approximately 5 seconds contingent on each response (FR1). An example of physical attention was petting the dog on either the side of the head and left or right sides of their body. Vocal attention was given in the form of “good job” or “you’re doing so good.”

Food. During the food condition, the experimenter provided one single piece of food delivered via an automated food dispenser contingent on each response (FR1). The experimenter held a remote that controlled the automated food dispenser and, upon appropriate responding, pressed the button labeled ‘dispense’ on the remote to release one morsel of food. A singular beeping sound accompanied the delivery of food that was automatically programmed in the food dispenser when the button to deliver food was pressed. No attention was delivered during this condition. 24 Study 2

Purpose The purpose of study 2 was to attempt to condition attention as a reinforcer to increase its reinforcing value in shelter dogs. Only dogs for whom attention was not shown to be a reinforcer in study 1 participated in study 2.

Procedures

General procedures. Each session within study 1 lasted a total of 5 minutes and the same experimenter conducted each session across the study. Each session began with a pre-exposure to the contingencies as follows: prior to each session each dog was prompted to engage in the correct response through gestural procedures and provided the consequence dependent upon the condition that was being ran. Sessions were conducted across several days and no effort was made to control the number of sessions or conditions conducted each day. Sessions were conducted at least 3 hours following scheduled meal times in order to attempt to control for food deprivation. However, no attempts were made to control for attention from people given the environment. If there was no responding during a session, the experimenter stood in the corner of the setting and provided no predetermined consequence. All sessions in study 2 were conducted within a single day for all dogs with the exception of Sascha whose post-test was run across 2 days.

Baseline. The procedures for baseline were identical to those of the attention condition of study 1.

Conditioning. During the conditioning phase, attention and food were delivered simultaneously contingent upon each appropriate response (FR1). 25

Attention was delivered in the same form as in baseline, and food was delivered by hand to the animal.

Post-conditioning test. The procedures in this condition were identical to those in the baseline condition.

Social Validity A social validity test was conducted prior to and after study 2. The assessment lasted 5 minutes. During this test, the dog was brought into a room where all of the Valley Animal Center shelter dogs were brought to meet with potential adopters. A 1-meter by 1-meter square area was taped off using duct tape, and a novel experimenter stood in the center of the taped-off area. The session began with the dog being taken into the room and being released from his or her leash. Contingent on the dog entering the taped-off area, the experimenter provided continuous vocal and physical attention. When the dog was outside of the taped-off area, no attention was provided. Furthermore, there were separate novel experimenters present during the pre-test and the post-test.

CHAPTER 4: RESULTS AND DISCUSSION

Results Results for study 1 are presented in Figures 1-5. For Nutty (Figure 1), during baseline responding steadily decreased from four instances of responding to zero levels. Following baseline, Nutty was moved to the first attention condition. During the attention condition there was a slight increase in the first session as compared to baseline, but responding decreased to low levels over time. We then moved to the food condition where responding increased in initial variability and sharply increased in level during the last two sessions. We returned back to the attention condition where there was a sharp decrease in responding following the first session and maintained low levels of responding. We then returned back to the food condition where a sharp increase in responding was demonstrated for the entire condition. Finally, we returned to the final attention condition where responding decreased to low levels. These data suggest that for Nutty, food might be a potent reinforcer for responding while attention is not a strong reinforcer. For Squishy (Figure 2), baseline demonstrated little to zero levels of responding. Following baseline, the first food condition began and responding maintained at near zero levels of responding. We then moved to the attention condition where responding slightly increased in variability during the last three sessions but remained at low levels. Next, we reversed back to the food condition where variability in responding increased across sessions and the level of responding increased during the last session. Following the food condition, we moved to the attention condition where responding steadily decreased to low levels similar to the previous attention condition. Finally, we reversed back to the last food condition where responding sharply increased and maintained at high 27 levels. Similar to Nutty, these data suggest that for Squishy’s behavior, food might be a strong reinforcer while attention is not. For Bailey (Figure 3), during the baseline period, responding remained at or near zero levels. Following baseline, we moved to the food condition where similar to baseline, responding remained at or near zero. We then moved to the attention condition where responding decreased to zero levels during the last four sessions. We next reversed to the food condition where responding maintained at low levels for the first three sessions and sharply increased to high levels during the last two sessions of this condition. Following the food condition, we went back to the attention condition where responding sharply decreased to low levels of responding. Finally, we reversed back to the food condition where responding steadily increased and maintained similar to the previous food condition. Bailey’s data suggest that her responding is influenced by food as a strong reinforcer while attention may not be a strong reinforcer to increase or maintain her responding. For Sascha (Figure 4), during the baseline condition, responding remained at or near zero levels. Following baseline, we moved next to the attention condition. During this condition, the variability of responding increased slightly, but the overall level remained low. We next moved to the food condition where, similar to the attention condition, responding maintained at low levels. We then returned to the attention condition and replicated these results. However, when we next moved to the food condition, responding increased and maintained at moderate levels, with a sharp increase in responding during the final two sessions for this condition. Finally, we returned to the attention condition and responding decreased to low levels. Overall, these data suggest that food might be a strong reinforcer for Sascha’s behavior, but attention may not be a strong reinforcer for her responding. 28

Finally, Fudge (Figure 5) during the baseline condition demonstrated high responding during the initial session; however, decreased at or near zero levels of responding in the last three sessions. Fudge was then moved to the food condition where responding maintained at or near zero levels of responding. Next, we moved to the attention condition where responding sharply increased and maintained at high levels for all three sessions within the attention condition. We then moved to the food condition where responding increased and maintained for the first three sessions. However, during this condition sessions 1-4 were ran in the same day and session four indicated a sharp decrease in responding though it remained at a high level as compared to the previous food condition. Due to a possible satiation effect experienced in the fourth session of the food condition, the last two sessions were conducted the following day in an attempt to replicate initial results within this condition where responding did increase back to high levels. Next, we reversed back to the attention condition where responding sharply decreased at or near zero levels of responding. We then reversed back to the food condition to replicate previous results and similar to the second food condition, responding increased high levels. Finally, we reversed back to the attention condition where responding initially decreased to moderate levels. However, responding increased during the last session. These data suggest that for Fudge, both food and attention may serve as a reinforcer for his behavior. However, this is likely to be idiosyncratic for some dogs. Overall, the data for study 1 showed that for four out of five dogs, attention may not serve as a powerful reinforcer for responding. These data suggest that attention is not likely a reinforcer for shelter dogs, and that conditioning might be necessary to increase the potency of human attention as a reinforcer. Therefore,

Figure 1. Data are depicted for Nutty for study 1.

Figure 2. Data are depicted for Squishy for study 1. 30

70 BL Food Attention Food Attention Food

i 50

e 40

y 30

r 20

Bailey 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Sessions

Figure 3. Data are depicted for Bailey for study 1.

Figure 4. Data are depicted for Sascha for study 1. 31

Figure 5. Data are depicted for Fudge for study 1. three of the dogs from study 1 for whom attention did not serve as a powerful reinforcer participated in study 2 (Bailey, Squishy, and Sascha). For Squishy (Figure 6) during baseline, responding quickly decreased to zero levels. We then moved Squishy to the conditioning phase where responding increased and maintained at moderate to high levels. Following the conditioning phase, we moved to the post-test condition, or return to baseline, where responding maintained at moderate though variable levels. This suggests that following the conditioning phase of simultaneous presentation of attention and food, Squishy’s behavior was able to be maintained at higher levels than initial responding during baseline when only attention was delivered, suggesting that attention may have become a conditioned reinforcer. For Bailey (Figure 6) during baseline, zero levels of responding were recorded across all sessions of this condition. Following baseline, we moved to the 32 conditioning phase where responding increased during the first three sessions and became variable during sessions 4-7 however remained at high levels. We then moved to the post-test condition which was a return to baseline. During the post- test condition, responding decreased to low levels, however, maintained above original baseline levels. These data suggest that though there was a significant decrease in responding from the conditioning phase to the post-test, Bailey’s behavior slightly increased above original baseline levels when attention was delivered as a consequence, suggesting that attention may have become a conditioned reinforcer. For Sascha (Figure 6), during baseline there were zero levels of responding across all sessions. We then moved to the conditioning phase where responding increased during the initial three sessions. During the conditioning phase responding increased and maintained at moderate to high levels. Following conditioning, Sascha was moved to the post-test where responding maintained at moderate to low levels. It should be noted that there was a decrease in responding following session 6 where Sascha became visibly sick and sessions for the day were stopped at the ninth session of the condition. Following a bill of clean health by the Valley Animal Center, Sascha continued the post-test for sessions 10-12 the following day where there was a slight increase in responding. These data suggest that for Sascha’s behavior following conditioning, responding was able to be maintained at higher levels than baseline when attention was delivered as a consequence, suggesting that attention may have become a conditioned reinforcer. Finally, Figures 7-9 show the results of the open field test conducted with all Squishy (Figure 7), Bailey (Figure 8) and Sascha (Figure 9) in study 2. Beginning with Squishy as depicted in Figure 7, the first bar on the left labeled ‘pre-test’ indicated that Squishy spent 172 seconds in proximity of the novel 33 experimenter prior to the conditioning phase. The second bar labeled ‘post-test’ indicated that Squishy spent 107 seconds in proximity to the second novel experimenter following the conditioning phase. When compared to one another, results suggest a decrease in seconds spent within proximity to the second novel experimenter as compared to the first following the conditioning phase. These results also suggest that for Squishy, the conditioning period did not have a strong effect in increasing time spent within proximity to a novel individual in this test. For Bailey (Figure 8), unlike Squishy, data demonstrated an increase in seconds spent within proximity to a novel individual following the conditioning phase. In the pre-test, Bailey spent 57 seconds in proximity to the novel experimenter. Following conditioning, Bailey was moved to the post-test where 101 seconds was spent in proximity to the novel experimenter. These data suggest that for Bailey’s behavior, the conditioning phase may have played a role in increasing her time spent with human contact in this test. For Sascha (Figure 9), there was an initial low number of seconds spent within proximity of the novel experimenter at 89 seconds in the pre-test period of the open-field test. Following the conditioning phase, Sascha was moved to the post-test period of the open field test where her results indicated a decrease to 50 seconds spent within proximity of a novel experimenter. Overall, these results indicate that similar to Squishy, conditioning did not play a strong role in increasing the number of seconds Sascha spent in proximity to a novel individual during the open-field test. Overall, although study 2 demonstrated that attention may have been conditioned as a reinforcer for responding across all three dogs, the results of the open-field test suggest that this had a limited effect on the time spent in proximity with a person in the open-field test. 34

40 Pre-Test Conditioning Post-Test

g 30

f 20

F 10

Squishy 0

g 30

f 20

F 10

Bailey 0

p 30

n 20

F 10

Sascha 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Sessions

Figure 6. Data are depicted for Squishy, Bailey, and Sascha for study 2. 35

Figure 7. Data are depicted for Squishy for the open-field test for study 2.

Figure 8. Data are depicted for Bailey for the open-field test for study 2.

Figure 9. Data are depicted for Sascha for the open-field test for study 2. 36 Discussion Overall, this study demonstrated that attention was not likely a strong reinforcer for the behavior of most shelter dogs, but that following a pairing with food, attention may have been conditioned as a reinforcer for responding. However, this did not necessarily increase dog-initiated interactions in the open- field test. Were an animal to demonstrate attention as a highly-valued reinforcer, the chances of being adopted potentially increase. By increasing these positive interactions, there is a possible decrease in overcrowding within a shelter as more dogs are adopted. Furthermore, by subjecting the dogs to human interaction and attention, there is a possibility that shelter dogs who did not engage in interactions might begin to initiate them. This coincides with the results suggested by Weiss et al. (2012), that potential adopters see greeting and approaching them as desirable behaviors by shelter dogs making it more likely they will find a permanent home. The results for study 1 indicated that there was a difference in the value of attention and food as a reinforcer for four out of the five dogs when food and attention were alternated between conditions as the consequence for responding. When food was provided as the consequence for behavior, there was an increase in responding for all five dogs. In comparison when attention was provided as the sole consequence for responding, there were lower levels of responding for all five dogs as compared to responding in the food conditions. Overall, study 1 suggests that attention may not serve as a powerful reinforcer for dog behavior which replicated the results of Feuerbacher and Wynne (2012). Our results demonstrated higher levels of responding during the food condition as compared to the attention condition for shelter dogs, which coincide with experiment two of the Feuerbacher and Wynne (2012) study that found similar results. Also similar to Feuerbacher 37 and Wynn, Fudge’s data suggest that attention as a reinforcer might be idiosyncratic for some dogs based on their histories of reinforcement. In addition to what was previously stated, while it’s possible that the results to study 1 suggest that attention is not a powerful reinforcer for shelter dogs there are a few possible explanations for the patterns of responding. One explanation may be during the study while we conducted fewer sessions than Feurbacher and Wynne (2012) it is possible that attention may have proven to be a reinforcer had we conducted additional sessions. For Squishy, Bailey, Sascha, and Fudge responding in the food occurred within the second reversal to the food condition which suggests that additional sessions may have been needed in order to responding to occur. However, although this is a possibility, our data still suggest that food is a stronger reinforcer than attention because responding occurred under those conditions during the limited number of sessions we ran. A second explanation is behavioral contrast where given the differential quality of food and attention as reinforcers it is possible that the dogs responded less for attention due to a history of responding for food than they would have if attention were assessed in the absence of food (Reynolds, 1961). However, we did counterbalance conditions across dogs such that some dogs experienced the attention condition prior to the condition with food and none of those dogs responded for attention. The results from study 2 suggest that using a simple conditioning technique may increase the value of reinforcement of human attention for shelter dog behavior. In terms of conditioning attention as a potential reinforcer, our study is the only study in animal behavior literature who attempted to use a response- stimulus pairing as suggested by Dozier et al. (2012). In the Dozier et al. (2012) study, food was provided with a praise statement contingent upon human responding. Results initially suggested that attention may not serve as a primary 38 reinforcer for human behavior. However, a response-stimulus pairing procedure may have conditioned attention as a reinforcer following a conditioning procedure. We demonstrated similar results where attention alone did not function as a strong reinforcer for behavior as suggested in study 1. Following a response-stimulus pairing procedure where food was delivered simultaneously with physical and vocal attention, there was an increase in responding for our three dogs used in study 2. This is important in finding a method where human social interactions can be conditioned as a potential reinforcer for non-human behavior. As suggested by Dozier et al. (2012), a response-stimulus pairing procedure may be used to increase attention as a reinforcer and our results have replicated their findings by using non-human behavior when using similar pairing procedures. Though there was an increase in responding during the conditioning phase for all three dogs there was a decrease in responding during the post-test condition. One possible explanation for this decrease might be an extinction effect where following the removal of food, there was a decrease in responding while attention was provided as the sole consequence. Human attention, or the presence of a human, may have served as a discriminative stimulus for responding to access food following the conditioning phase, which might explain why responding still maintained at low levels. Finally, during the open-field test, there was a decrease in seconds spent within proximity to the novel experimenter for two of the three dogs. This suggests that there was no significant effect upon the open-field test from initial pre-test following conditioning. There are several possible reasons why the responding during the open-field test did not correspond with the findings of our post-conditioning test. First, during the post-conditioning condition, there was little stimulation available outside of human attention. However, the room in 39 which the open-field test was conducted contained many novel stimuli which may have competed with the efficacy of attention as a reinforcer. Second, during the post-conditioning test condition, it is possible that stimulus control of the environment signaled the availability of attention as a reinforcer due to the number of times the dogs contacted the contingency. However, during the open-field test, the experimenter did not engage in behavior to signal the availability of attention as a consequence for behavior (e.g., calling the dog over), and the dogs had a limited (10 minutes total) chance to experience the consequences for responding. It is possible that had the experimenter attempted more initiating of interactions with the dogs (as would likely be the case in a real adoption scenario), dogs may have spent more time in proximity to the experimenter. Though results indicated an increase in the reinforcing value of attention for all shelter dogs in study 2, there are a few limitations to the current study. First, we did not assess what types of attention were reinforcing to individual dogs. It is possible that for some dogs, vocal attention in the absence of petting would be preferred to vocal attention plus petting. However, previous research by McIntire and Colley (1967) had shown that physical and vocal attention can be used as a reinforcer (physical attention was preferred), so we decided to use a combination of these to potentially increase the potency of the attention being delivered. Second, the response used in this study is arbitrary and low-effort, and it is unlikely that the dogs would encounter such a response in their natural environment. It is possible that if attention serves as a reinforcer for this response, it may not serve as a reinforcer for more effortful responses, such as those involved in interaction with humans. This is a possible explanation for the results of the open-field test, as dog-initiated interactions might be more effortful than the arbitrary responses used in studies 1 and 2. A third limitation is the setting that 40 was used. While running sessions with the dogs, the fence that surrounded the closed-off area in which all sessions were run allowed the dogs to view out into the dog park where other people and their pets were utilizing the facility. These additional environmental stimuli may have influenced responding during the study. However, we attempted to address this issue by wrapping a 2-meter by 3- meter tarp around the outside of the fence in order to block off any opportunities the dogs may have had at viewing other people or their pets. Furthermore, this issue was addressed when the dog park was closed due to the canine influenza outbreak that led to us holding sessions within the outside kennels of the animal shelter where access to other people and animals was limited. Future researchers should consider conducting sessions in more controlled settings if possible to further isolate the effects of these reinforcers on responding. Finally, we did not conduct maintenance probes to determine if attention continued to serve as a reinforcer over a long period of time. Future researchers should include maintenance probes in order to assess the long-term effects of conditioning. With these results, it is important to note the other areas of research that should be further explored regarding conditioning attention in shelter dogs. One area that would be interesting to assess is the difference between novel versus familiar individuals. Our study used the same experimenter within the attention conditions of studies 1 and 2 across all shelter dogs used. With the assumption that every potential adopter is novel to the shelter dog, using different individuals during each conditioning session might provide a more naturalistic approach to how shelter dogs are provided attention. By using the same experimenter in the conditioning phase of study 2, the experimenter may become familiar as suggested by McIntire and Colley (1967) who predicted that satiation might have played a role into a decrease in responding for human attention over time. Another idea for 41 future research might be evaluating a different type of response that the shelter dogs would encounter in their natural environment. Our study used materials such as a wand and nose-to-wand touching responses as the contingency for reinforcement. However, it is highly unlikely that the shelter dogs will encounter this type of responding to garner human attention as a consequence. Looking at more natural responses such as approaching and sitting in front of the kennel door as a response for human interaction may provide shelter dogs with a low-effort response that potential adopters might find as a desirable behavior as suggested by Weiss et al. (2012). Finally, future researchers should continue to determine how conditioning attention as a reinforcer affects the likelihood that dogs will actually be adopted from a shelter. Though conditioning attention as a reinforcer is not a novel idea, it would be beneficial for shelter dogs to be subjected to this type of research to find a better method in which to increase interactions with potential adopters and decrease the length of stay in shelters. Overall, this study was successful in demonstrating that even when attention did not serve as a reinforcer, it may be conditioned as a reinforcer in shelter dogs.

REFERENCES

American Society for the Prevention of Cruelty to Animals (2017). Pet statistics. Retrieved from: https://www.aspca.org/animal-homelessness/shelter-intake- and-surrender/pet-statistics.

Barton, E. J. (1981). Developing sharing: An analysis of modeling and other behavioral techniques. Behavior Modification, 5, 386-398.

Beavers, G. A., Iwata, B.A., & Lerman, D.C. (2013). Thirty years of research on the functional analysis of problem behavior. Journal of Applied Behavior Analysis, 46, 1-21.

Bugelski, R. (1938). Extinction with and without sub-goal reinforcement. Journal of Comparative Psychology, 26, 121-134

Chiandetti, C., Avella, S., Fongaro, E., & Cerri, F. (2016). Can clicker training facilitate conditioning in dogs? Applied Animal Behaviour Science, 184, 109- 116.

Egger, M. D. & Miller, N.E. (1962). Secondary reinforcement in rats as a function of information value and reliability of the stimulus. Journal of Experimental Psychology, 64, 97-104

Feng, L. C., Howell, T. J., & Bennett, P. C. (2016). How clicker training works: Comparing reinforcing, marking, and bridging hypotheses. Applied Animal Behaviour Science, 181, 31-40

Feuerbacher, E. N. & Wynne, C. D. L. (2012). Relative efficacy of human social interaction and food as reinforcers for domestic dogs and hand-reared wolves. Journal of the Experimental Analysis of Behavior, 98, 105-129.

Feuerbacher, E. N. & Wynne, C. D. L. (2015). Shut up and pet me! Domestic dogs (Canis lupus familiaris) prefer petting to vocal praise in concurrent and single-alternative choice procedures. Behavioural Processes, 110, 47-59.

Gillaspy, J. A. & Bihm, E. M., (2002). Obituary: Marian Breland Bailey (1920- 2001). American Psychology, 57, 292-293.

Hanley, G. P., Iwata, B. A., & McCord, B. E. (2003). Functional analysis of problem behavior: A review. Journal of Applied Behavior Analysis, 36, 147- 185. 44

Kaplan, P. S., Hearst, E. (1982). Bridging temporal gaps between CS and US in autoshaping: Insertion of other stimuli before, during, and after CS. Journal of Experimental Psychology and Animal Behavior, 8, 187-203

Kelly, M. A., Roscoe, E. M, Hanley, G. P., & Schlichenmeyer, K. (2014). Evaluation of assessment methods for identifying social reinforcers. Journal of Applied Behavior Analysis, 47, 113-135.

Lieberman, D. A., McIntosh, D. C., & Thomas, G.V. (1979). Learning when reward is delayed: A marking hypothesis. Journal of Experimental Psychology and Animal Behavior, 5, 224-242.

McIntire, R.W. & Colley, T.A. (1967). Social reinforcement in the dog. Psychological Reports, 20, 843-846.

McKerchar, P. M., & Thompson, R. H. (2004). A descriptive analysis of potential reinforcement contingencies in the preschool classroom. Journal of Applied Behavior Analysis, 37, 431-444.

Piazza, C. C., Bowman, L. G., Contrucci, S. A., Delia, M. D., Adelinis, J. D., & Goh, H. L. (1999). An evaluation of the properties of attention as reinforcement for destructive and appropriate behavior. Journal of Applied Behavior Analysis, 32, 437-449.

Pierce, W. D., & Cheney, C. D., (2004). Behavior analysis and learning (3rd. ed.). Mahwah, NJ: Lawrence Erlbaum.

Protopopova, A., & Wynne, C. D. L. (2014). Adopter-dog interactions at the shelter: behavioral and contextual predictors of adoption. Applied Animal Behaviour Science, 157, 109-116.

Protopopova, A., Mehrkam, L. R., Boggess, M. M. & Wynne, C. D. L. (2014). In- kennel behavior predicts length of stay in shelter dogs. PLoS ONE, 9, 1-21.

Pryor, K. (1999). Don't shoot the dog! The new art of teaching and training. New York, NY: Bantam Books

Reynolds, G.S. (1961). Behavioral contrast. Journal of the Experimental Analysis of Behavior, 4, 57-71.

Skinner, B. F. (1938). The behavior of organisms: An experimental analysis. Oxford, England: Appleton-Century. 45

Smith, S. M. & Davis, E. S. (2008). Clicker increases resistance to extinction but does not decrease training time of a simple operant task in domestic dogs (Canis familiaris). Applied Animal Behaviour Science, 110, 318-329.

Stanley, W. C., Morris, D. D., & Trattner, A. (1967). Conditioning with a passive person reinforcer and extinction in Shetland sheep dog puppies. National Institute of Mental Health, 2, 19-20.

Strain, P. S., & Timm, M. A. (1974). An experimental analysis of social interaction between a behaviorally disordered preschool child and her classroom peers. Journal of Applied Behavior Analysis, 7, 583-590.

Thompson, R. H., & Iwata, B. A. (2001). A descriptive analysis of social consequences following problem behavior. Journal of Applied Behavior Analysis, 34, 169-178.

Walker, H.M. & Buckley, N. K. (1968). The use of positive reinforcement in conditioning attending behavior. Journal of Applied Behavior Analysis, 1, 245-250.

Weiss, E., Miller, K., Mohan-Gibbons, H., & Vela, C. (2012). Why did you choose this pet? Adopters and pet selection preferences in five animal shelters in the United States. Animals, 2, 1-17.

APPENDIX: IACUC 47

Feb. 2002

File#:

Approval Date:

Animal Use Application Institutional Animal Care and Use Committee California State University, Fresno California P R O Copy this form to your word processor. Fill in and expandP as appropriate. DO NOT DELETE ANY PART OF THE APPLICATION.O Modeled after a template provided by our licensing agencies, it assures complianceS with state and federal law. All relevant sections of the application must be completedA and submitted to the chairperson of the IACUC as an email attachment. LA hard copy with signatures (SECTIONS Q, R and S) must be submitted for inspection# by licensing agencies. This file is open to the public. : A P A. ADMINISTRATIVE INFORMATION Instructional Use Research Use P

R O Initial Submission RenewalV Modification A L ( If renewal or modification, please flag those sectionsD that differ from the original application or required further elaboration.)A T E Principal Investigator: Steven Payne, Ph.D., BCBA-D : EXPIRATION DATE:

Department: Psychology P

P Mailing Address: 2555 E. San Ramon ST #11 CaliforniaO State University Fresno CA 93740 S Telephone: (559) 278-6937 Fax: (559) 278A-7910 Email: [email protected] L # Project Title: Conditioning Attention as Reinforcer in Shelter Dogs :

R O P O S A L # 48

List the names of all individuals authorized to conduct procedures involving animals under this proposal and identify key personnel (e.g., co-investigator(s), providing their department, telephone, fax, and email:

Steven Payne Graduate Advisor, Principle Investigator, Psychology Department CSU, Fresno. Same as above.

Maria Salmeron Graduate Student, Psychology Department CSU, Fresno. Same as above.

Alyssa Salazar Graduate Student, Psychology Department CSU, Fresno. Same as above.

Lauren Valverde Graduate Student, Psychology Department CSU, Fresno. Same as above.

Martha Cisneros Graduate Student, Psychology Department CSU, Fresno. Same as above.

Submission Date: Approved: ______Chair CSUF IACUC Date

(This page, when signed, may be submitted to funding agencies.)

************************************************************************ ******** ANIMAL USE MAY NOT PROCEED WITHOUT IACUC APPROVAL. ************************************************************************ ******** B. ANIMAL REQUIREMENTS

Genus: Canis Species: lupus famili aris

Strain, subspecies, or breed: all Common. name: domestic dog

Approximate age, weight or size: Adult, > 1 year old

Sex: Male and female

Bacteriological status: N/A

Viral status: N/A

Source(s): Valley Animal Center, Fresno, California

Primary housing location(s): N/A

Location(s) where manipulation will be conducted: Valley Animal Center, Fresno, California

Number of Animals to be Used: Year 1: 8 Year 2: Year 3:

Total: 8

C. TRANSPORTATION N/A

Transportation of animals must conform to all institutional guidelines/policies and federal regulations. If animals will be transported on public roads or out of state, describe efforts to comply with USDA regulations. If animals will be transported between facilities, describe the methods and containment to be utilized. If animals will be transported within a facility, include the route and elevator(s) to be utilized.

D. STUDY OBJECTIVES

Briefly explain in language understandable to a layperson the aim of the study and why the study is important to human or animal health, the advancement of knowledge, or the good of society.

Of dogs entering shelters every year, approximately only 37% are adopted, and a staggering 41% are euthanized (ASPCA, 2016). Although various factors may contribute to adoption criteria, a dog’s ability to socially interact with humans proves to be a significant one. Research suggests that dogs that are not adopted were found to spend twice as much time ignoring play initiation by adopters than those that were adopted (Protopopova et al., 2014). One possible behavioral issue common to animals in shelters is the lack of social interaction exhibited when considered for adoption and its potential to reduce the likelihood of their adoption. The present study aims to target this issue by focusing on the reinforcement value of human attention.

The primary purpose of the present research is to compare the reinforcing efficacy of attention and food with dogs. In the first experiment, we will provide attention and food contingent on a nose-touch response.. An ABCBC reversal design will be used to assess experimental control. If attention fails to increase behavior, a second experiment will be conducted.

The purpose of the second experiment will be to increase the reinforcing efficacy of attention with dogs. In this experiment, we will pair attention with food using a response- reinforcer conditioning procedure in an attempt to make attention a conditioned reinforcer that will have similar reinforcing properties as food. We will determine this by testing the reinforcer efficacy of attention in the absence of food prior to and following the conditioning procedure.

E. RATIONALE FOR ANIMAL USE (Use additional sheets if necessary.)

1) Explain your rationale for animal use. [The rationale should include reasons why non- animal models cannot be used.] 51

Dog behavior is the target behavior the current study is examining, so non-animal models would not be appropriate.

2) Justify the appropriateness of the species selected.

Many dogs are euthanized every year due to shelter overcrowding. Dogs are more likely to get adopted if they initiate play and generally respond to the attention of potential adopters.

3) Justify the number of animals to be used. [The number of animals should be the minimum number required to obtain statistically valid results.]

Because we will be assessing different methods to help train dogs to make them more adoptable. This does not require a relatively large sample.

F. DESCRIPTION OF EXPERIMENTAL DESIGN AND ANIMAL PROCEDURES

Briefly explain the experimental design and specify all animal procedures. This description should allow the IACUC to understand the experimental course of an animal from its entry into the experiment to the endpoint of the study. Specifically address the following:

 Experimental injections or inoculations (substances, e.g., infectious agents, adjuvants, etc.; dose, sites, volume, route, and schedules).  Blood withdrawals (volume, frequency, withdrawal sites, and methodology).  Surgical procedures (provide details of survival and non-survival surgical procedures in Section G.).  Radiation (dosage and schedule).  Methods of restraint (e.g., restraint chairs, collars, vests, harnesses, slings, etc.). Include how animals are restrained for routine procedures like blood withdrawals. Prolonged restraint must be justified with appropriate oversight to ensure it is minimally distressing. Describe any sedation, acclimation or training to be utilized.

Subjects: Subjects will be the population of dogs located in the dog adoption center at Valley Animal Center. All subjects will be vaccinated, spayed/neutered, and treated for any medical issues prior to entering the study.

Location: All sessions will be conducted on-site in a small room adjacent to the kennel area.

Method: All sessions will be videotaped. Data will be collected by trained graduate and undergraduate research assistants. Responses are defined as the dog touching their nose to a designated spot on the wall. All sessions will be five minute sin length

Experiment 1: 52

Baseline During baseline, there will be no programmed consequences for responding.

Attention During the attention condition, 5 seconds of physical and vocal attention will be delivered to the dog contingent on responding. No other consequences will be delivered.

Food During the food condition, a single piece of food will be delivered via machine contingent on a correct response. No human interaction will be provided. .

Experiment 2: Baseline During baseline, attention will be provided contingent on responding

Conditioning During conditioning phase attention and food will be delivered simultaneously contingent on responding. Food will be delivered by hand..

Post-Conditioning Test The post conditioning condition will be identical to baseline..

 Animal identification methods (e.g., ear tags, tattoos, collar, cage card, implant, etc.).

Identification of samples will be by the animals’ names.

 Other procedures (e.g., survival studies, tail biopsies, etc.). None  Resultant effects, if any, that the animals are expected to experience (e.g., pain or distress, ascites production, etc.). None  Other potential stressors (e.g., food or water deprivation, noxious stimuli, environmental stress) and procedures to monitor and minimize distress. If a study is USDA Classification E, indicate any non-pharmaceutical methods to minimize pain and distress.None  Experimental endpoint criteria (e.g., tumor size, percentage body weight gain or loss, inability to eat or drink, behavioral abnormalities, clinical symptomatology, or signs of toxicity) must be specified when the administration of tumor cells, biologics, infectious agents, radiation or toxic chemicals are expected to cause significant symptomatology or are potentially lethal. List the criteria to be used to determine when euthanasia is to be performed. Death as an endpoint must always be scientifically justified.  Veterinary care (indicate desired plan of action in case of animal illness, e.g., initiate treatment, call investigator prior to initiating treatment, euthanize). Veterinary care will be provided by the animal shelter

G. SURGERY No surgery is involved in this project.

If proposed, complete the following: (Use additional sheets if necessary.)

1. Identify and describe the surgical procedure(s) to be performed. Include preoperative procedures (e.g., fasting, analgesic loading), and monitoring and supportive care during surgery. Include the aseptic methods to be utilized.

2. Who will perform surgery and what are their qualifications and/or experience?

3. Where will surgery be performed and postoperative care provided (building and rooms)?

4. If survival surgery, describe postoperative care required, frequency of observation, and identify the responsible individual(s). Include detection and management of postoperative complications during work hours, after hours, weekends and holidays.

5. If non-survival surgery, describe how humane euthanasia is enacted and how death is determined.

N/A

6. Are paralytic agents used during surgery? If yes, please describe how ventilation will be maintained and how pain will be assessed.

N/A

7. Has major survival surgery been performed on any animal prior to being placed on this study? [Major survival surgery penetrates and exposes a body cavity or produces substantial impairment of physical or physiologic functions (such as laparotomy, thoracotomy, crainotomy, joint replacement, or limb amputation).] If yes, please explain:

N/A

8. Will more than one major survival surgery be performed on an animal while on this study? If yes, please justify:

N/A

H. PAIN OR DISTRESS CLASSIFICATION AND CONSIDERATION OF ALTERNATIVES

1. Pain or Distress Classification

Species USDA Number of animals used each 3 year total (common name) Classificatio year number of n* animals B, C, D or E Year 1 Year 2 Year 3 Dogs B 8 8

Total number of animals (should equal total from Section B): 8

* USDA Classifications and Examples

Classification B: Animals being bred, conditioned, or held for use in teaching, testing, experiments, research, or surgery, but not yet used for such purposes.

Examples:  Breeding colonies of any animal species (USDA does not require listing of rats, mice, birds) that are held in legal sized caging and handled in accordance with the Guide and other applicable regulations. Breeding colony includes parents and offspring.  Newly acquired animals that are held in proper caging and handled in accordance with applicable regulations.  Animals held under proper captive conditions or wild animals that are being observed.

Classification C: Animals upon which teaching, research, experiments, or tests will be conducted involving no pain, distress, or use of pain-relieving drugs. Examples: 55

 Procedures performed correctly by trained personnel such as the administration of electrolytes/fluids, administration of oral medication, blood collection from a common peripheral vein per standard veterinary practice (dog cephalic, cat jugular) or catheterization of same, standard radiography, parenteral injections of non-irritating substances.  Euthanasia performed in accordance with the recommendations of the most recent AVMA Panel on Euthanasia, utilizing procedures that produce rapid unconsciousness and subsequent humane death.  Manual restraint that is no longer than would be required for a simple exam; short period of chair restraint for an adapted nonhuman primate.

Classification D: Animals upon which experiments, teaching, research, surgery, or tests will be conducted involving accompanying pain or distress to the animals and for which appropriate anesthetic, analgesic, or tranquilizing drugs will be used.

Examples:  Surgical procedures conducted by trained personnel in accordance with standard veterinary practice such as biopsies, gonadectomy, exposure of blood vessels, chronic catheter implantation, laparotomy or laparoscopy.  Blood collection by more invasive routes such as intracardiac or periorbital collection from species without a true orbital sinus such as rats and guinea pigs.  Administration of drugs, chemicals, toxins, or organisms that would be expected to produce pain or distress but which will be alleviated by analgesics.

Classification E: Animals upon which teaching, experiments, research, surgery, or tests will be conducted involving accompanying pain or distress to the animals and for which the use of appropriate anesthetic, analgesic, or tranquilizing drugs will adversely affect the procedures, results, or interpretation of the teaching, research, experiments, surgery, or tests. Examples:  Procedures producing pain or distress unrelieved by analgesics such as toxicity studies, microbial virulence testing, radiation sickness, and research on stress, shock, or pain.  Surgical and postsurgical sequella from invasion of body cavities, orthopedic procedures, dentistry or other hard or soft tissue damage that produces unrelieved pain or distress.  Negative conditioning via electric shocks that would cause pain in humans.  Chairing of nonhuman primates not conditioned to the procedure for the time period used.

NOTE REGARDING CLASSIFICATION E: An explanation of the procedures producing pain or distress in these animals and the justification for not using appropriate anesthetic, analgesic or tranquilizing drugs must be provided on Attachment 1. This information is required to be reported to the USDA, will be available from USDA under 56 the Freedom of Information Act, and may be publicly available through the Internet via USDA’s website.

2. Consideration of Alternatives

If any procedures fall into USDA's Classification D or E, causing more than momentary or slight pain or distress to the animals, describe your consideration of alternatives and your determination that alternatives are not available. Delineate the methods and sources used in the search. Database references must include databases searched, the date of the search, period covered, and the keywords used. Alternatives include methods that (1) refine existing tests by minimizing animal distress, (2) reduce the number of animals necessary for an experiment, or (3) replace whole-animal use with in vitro or other tests. When ascites production is used to produce antibodies, justification needs to be given as to why in vitro systems cannot be used. Note that you must certify in Section Q.5. that no valid alternative was identified to any described procedures which may cause more than momentary pain or distress, whether relieved or not.

I. ANESTHESIA, ANALGESIA, TRANQUILIZATION, OTHER AGENTS

For animals indicated in Section H.1., Classification D, specify the anesthetics, analgesics, sedatives or tranquilizers that are to be used. Include the name of the agent(s), the dosage, route and schedule of administration. If information is provided in Section G. above, please cross-reference. Describe tracking and security of controlled drugs (Drug Enforcement Agency requirements).

J. METHOD OF EUTHANASIA OR DISPOSITION OF ANIMALS AT END OF STUDY

Indicate the proposed method of euthanasia. If a chemical agent is used specify the dosage and route of administration. If the method(s) of euthanasia include those not recommended by the AVMA Panel Report on Euthanasia (e.g., decapitation or cervical dislocation without anesthesia), provide scientific justification why such methods must be used. Indicate the method of carcass disposal if not described in Section K. below.

Species Method Drug Dose (mg/kg) Route

K. HAZARDOUS AGENTS Use of hazardous agents requires the approval of the institutional Biosafety Office/Committee. Attach documentation of approval for the use of recombinant DNA or potential human pathogens.

Hazardous Agent Yes No Agent Date of Biosafety Tracking # Approval Radionuclides X Biological Agents X Hazardous X Chemicals or Drugs Recombinant DNA X

Study Conducted at Animal Biosafety Level: 1 2 3 4

Describe the practices and procedures required for the safe handling and disposal of contaminated animals and material associated with this study. Also describe methods for removal of radioactive waste and, if applicable, the monitoring of radioactivity.

Additional safety considerations: - N/A

L. BIOLOGICAL MATERIAL/ANIMAL PRODUCTS FOR USE IN ANIMALS – N/A (e.g., cell lines, antiserum, etc.)

1. Specify Material:

2. Source: Material Sterile or Attenuated: Yes No If derived from rodents, has the material been MAP/RAP/HAP tested? [MAP - Mouse Antibody Production; RAP - Rat Antibody Production; HAP - Hamster Antibody Production]

Yes (Attach copy of results) or No

3. I certify that the MAP/RAP/HAP tested materials to be used have not been passed through rodent species outside of the animal facility in question and/or the material is derived from the original MAP tested sample. To the best of my knowledge the material remains uncontaminated with rodent pathogens.

SP Initials of Principal Investigator.

M. TRANSGENIC AND KNOCKOUT ANIMALS - N/A 58

Describe any phenotypic consequences of the genetic manipulations to the animals. Describe any special care or monitoring that the animals will require.

N. EXEMPTIONS FROM ENVIRONMENTAL ENHANCEMENT FOR NONHUMAN PRIMATES OR EXERCISE FOR DOGS - N/A

1. For nonhuman primates, are you seeking an exemption for scientific reasons from the institution’s plan for environment enhancement? Yes or No If yes, provide the basis of the request.

2. For dogs, are you seeking an exemption for scientific reasons from the institution’s plan to provide dogs with the opportunity for exercise? Yes or No If

yes, provide the basis of the request.

O. FIELD STUDIES

If animals in the wild will be used, describe how they will be observed, any interactions with the animals, whether the animals will be disturbed or affected, and any special procedures anticipated. Indicate if Federal permits are required and whether they have been obtained.

P. SPECIAL CONCERNS OR REQUIREMENTS OF THE STUDY

List any special housing, equipment, animal care (e.g., special caging, water, feed, or waste disposal, environmental enhancement, etc.). – N/A

Q. PRINCIPAL INVESTIGATOR CERTIFICATIONS

1. I certify that I have attended the institutionally required investigator training. Course/Program: LATA Date: HA 101 and HA 203 9/1/15 or that I have equivalent experience in the conduct of projects of this nature. ______

2. I certify that I have determined that the research proposed herein is not unnecessarily duplicative of previously reported research. 3. I certify that all individuals working on this proposal who are at risk are participating in the Institution's Occupational Health and Safety Program. 4. I certify that the individuals listed in Section A. are authorized to conduct procedures involving animals under this proposal, have attended the institutionally required 59

investigator training course, and received training in: the biology, handling, and care of this species; aseptic surgical methods and techniques (if necessary); the concept, availability, and use of research or testing methods that limit the use of animals or minimize distress; the proper use of anesthetics, analgesics, and tranquilizers (if necessary); and procedures for reporting animal welfare concerns. 5. For all USDA Classification D and E proposals (see section H.1.): I certify that I have reviewed the pertinent scientific literature and the sources and/or databases as noted in Section H.2. and have found no valid alternative to any procedures described herein which may cause more than momentary pain or distress, whether it is relieved or not. 7. I certify that I will obtain approval from the IACUC before initiating any significant changes in this study. 8. I certify that I will notify the IACUC regarding any unexpected study results that impact the animals. Any unanticipated pain or distress, morbidity or mortality will be reported to the attending veterinarian and the IACUC. 9. I certify that I am familiar with and will comply with all pertinent institutional, state, and federal rules and policies.

Principal Investigator:

Name: Steven W. Payne Signature: Date:

R. CONCURRENCES

PROPOSAL NUMBER ______(leave blank)

Supervisory concurrence as applicable:

Name: Signature: Date:

Safety Office/Committee Certification of Review and Concurrence: (Required of all studies utilizing hazardous agents.)

Name: Signature: Date:

Facility Manager/Veterinarian certification of resource capability in the indicated facility to support the proposed study:

Facility: Name: Signature: Date:

Comments:

Attending Veterinarian certification of review and consultation on proper use of anesthetics and pain relieving medications for any painful procedures:

Name: Michael Thomas Signature: Date:

S. FINAL APPROVAL:

Certification of review and approval by the Institutional Animal Care and Use Committee:

Name: Paul R. Crosbie Signature: Date:

List any attachments here: NONE

Attachment 1

Explanation for USDA Classification E (This report is required to accompany USDA Form 7023 to support any USDA Classification E listings.) This document must be typed.

Name of investigator:

Animal Study Proposal Title:

Species and number of animals listed in Classification E for each year:

Species: Number of animals: year 1 - year 2 - year 3 - Total:

Description of project including reason(s) for species selection:

Provide a scientific justification to explain why the use of anesthetics, analgesics, sedatives or tranquilizers during and/or following painful or distressing procedures is contraindicated:

Signature of investigator: Date:

Signature of IACUC Chairperson: Date: