Poster Session 1 - Thursday Evening
Erica Hoy Kennedy (Frostburg State University) & Dorothy M. Fragaszy (University of Georgia)
Evidence for Analogical Problem Solving in a Capuchin Monkey (Cebus apella)
Analogical reasoning involves recognizing similarities among object relations when the objects themselves are dissimilar.  It has been argued that apes, but not monkeys, are capable of analogical reasoning.  This study investigated whether capuchin monkeys can use analogical reasoning in order to solve a three-dimensional search task.  The task involved hiding food under one of two (and later three) opaque cups of different sizes and allowing the subject to search for food under the cup of analogous size in their own stimulus set. Four monkeys were first trained to criterion on basic MTS. Next they were exposed to two pairs of cups and required to search for food based on physical or relational similarity. One of the four monkeys reached criterion on all variations of the relational matching task, suggesting that analogical reasoning is within the capacity of a member of New World monkey species.
Leyre Castro & Edward A. Wasserman (The University of Iowa)
Can pigeons learn to complete an analogy?
Analogical reasoning is considered to be characteristic of humans and to be beyond the abilities of non-language trained apes.  However, from an evolutionary point of view, rudiments of analogical reasoning may be evident in other animals as well.  The present study tried to see whether pigeons could learn to complete an analogy.  The birds were first shown two pictures that could be the same as or different from one another (A-A or A-B).  Then, along with a third picture, two choice alternatives appeared (C-C/D); pigeons had to select the picture that matched the relation displayed in the first place.  After extensive training, pigeons reached 60% accuracy—not very high, but significantly above chance.  Transfer tests suggested that the level of accuracy reached in training was, in part, due to learning of specific stimulus combinations rather than to learning of the relations between the displays.
Kuroshima H. (University of Georgia, JSPS), Leighty A. Katherine (Disney’s Animal Kingdom), Fragaszy M. Dorothy (University of Georgia) & Fujita Kazuo (Kyoto University)
Capuchin monkeys can understand the context of other’s action.
We investigated whether capuchin monkeys can discriminate between situations in which a person is unable versus to do and in which she is unwilling to perform a particular action. We compared the reactions of 6 subjects under four conditions; (1) Experimenter 1 (E1) tried to deliver a raisin through an opening of the test cage but failed because the opening was too small (physical-unable), or (2) because a second experimenter 2 (E2) blocked the opening (social-unable). (3) E1 began to give the raisin to the subject but retracted it despite no physical interference (physical-unwilling) or (4) no social interference (social-unwilling). The subjects protested against E1 and E2 significantly more often in the social situation than in the physical situation. Moreover, they protested against E2 significantly more often in the social-unable condition than in the other conditions. These results suggested that the monkeys interpreted the other’s action according to its context.
Jennifer Vonk & Francys Subiaul (University of Southern Mississippi, The George Washington University)
What Chimpanzees (Pan Troglodytes) Understand About Others’ Capabilities
We tested chimpanzees’ ability to predict which of two human experimenters could deliver a food tray. In the “floor” condition, legs were needed to push the tray toward the subject. In the “lap” condition, arms were needed to hand the tray to the subject. Initially, chimpanzees preferred to beg to an experimenter whose arms were not visible rather than one whose legs were not visible regardless of the task. We manipulated factors that might have affected the chimpanzees’ preferences, such as distance between experimenter and subject, amount of occlusion of experimenters’ body, contact with the tray and positioning of constraints that either impeded movement of the limbs or did not. The chimpanzees’ performance was best explained by attention to irrelevant cues such as distance and contact. When we eliminated the discriminative role of such cues, performance fell to chance levels, indicating that chimpanzees do not reason about capability as humans might.
Megan L. Hoffman, Michael J. Beran, & David A. Washburn (Georgia State University)
Working Memory for What-Where-When Information in Rhesus Monkeys (Macaca mulatta)
In recent years there has been an increased interest in whether nonhuman animals possess elements of episodic memory, including the integration of what, where, and when information in memory. In the present study, a computerized task was used to assess working memory for what-where-when information in joystick-trained rhesus macaques (Macaca mulatta).  In Experiment 1, the macaques were trained on three separate tasks: an identity DMTS task, a spatial DMTS task, and a symbolic-temporal DMTS task.  The macaques were then presented with a task in which they were required to report what, where, and when the event took place and another task in which they were required to report only one randomly selected component of the event. The results indicated that the macaques were above chance at remembering what, where, and when information, but that accuracy was affected by variables also observed in the human literature.
Miranda C. Feeney, William A. Roberts, & David F. Sherry (University of Western Ontario)
Tests of Episodic-like Memory in the Black-capped Chickadee
Episodic-like memory in scrub jays has been described by Clayton and Dickinson (1998). We conducted similar a test for episodic-like memory in another species of food-storing bird, the black-capped chickadee.  Chickadees found concealed sunflower seeds and mealworms in selected sites in a testing board within their home cage, and then searched these sites again after either short (3 hr) or long (123 hr) retention intervals. After the long retention interval, their preferred mealworms were degraded and inedible.  Chickadees showed some memory for both what kind of food they had previously encountered and where, but showed no memory for when food had previously been found.  In a second experiment, currently in progress, chickadees search for hidden sunflower seeds and mealworms in artificial trees set in an indoor aviary, increasing both the spatial naturalism of the task and the effort required to find food.  The results of this study will also be presented.
Allison L. Foote & Jonathon D. Crystal (University of Georgia)
Metacognition in the Rat
A fundamental question in comparative cognition is whether animals have knowledge of their own cognitive states, termed metacognition.  Recent evidence suggests that people and non-human primates, but not less “cognitively sophisticated” species, are capable of metacognition.  Before taking a duration test, rats were given the opportunity to decline a test if they did not know the answer.  On other trials, they were not given the option to decline.  Accurate performance on the duration test yielded a large reward whereas inaccurate performance resulted in no reward.  Declining a test yielded a small, but guaranteed, reward.  If a rat possesses knowledge about whether it knows the answer to the duration test, it would be expected to decline most frequently on difficult tests and show lowest accuracy on difficult tests that cannot be declined.  Our data provide evidence for both predictions and demonstrate that a non-primate has knowledge of its own cognitive state. 
Hiromitsu Miyata & Kazuo Fujita (Kyoto University)
Future Planning in Pigeons On a Computerized Maze Task
Planning is the inner process of making decisions about a set of future actions in order to achieve a particular goal. We examined whether pigeons (Columba livia) plan the future steps on a maze task on the LCD monitor. In Experiment1, after training three pigeons to solve a +-shaped maze by moving a red square (the target) to a blue square (the goal), we found that they frequently moved the target toward the previous goal positions after the goal moved to another corner during task solution. In Experiment 2, using a syuriken (Japanese traditional throwing knife) -shaped maze we found that the pigeons frequently started by moving towards the previous goal directions when the goal position changed from the “preview” to the solution phase within a trial. These results suggest that pigeons do plan on the mazes, raising the possibility that future planning may be widespread among avian species.
Jing Pan, Thomas Pickering, Bettina von Ammon, Hika Kuroshima, & Dorothy Fragaszy (University of Georgia)
A Tufted Capuchin Monkey Becomes Proficient at Navigating Mazes
Capuchin monkeys are strongly predisposed to move directly toward a goal while navigating two-dimensional alley mazes.  We examined if a capuchin monkey could learn to choose a path continuing indirectly to the goal (a Non-obvious choice, or NOC).  One capuchin monkey solved multiple sets of orderly presented192 mazes containing both NOC’s and other choices followed by 24 probe mazes drawn from the sets. Following the 7th to 10th replicates of the sets, the monkey achieved on average 79% correct choices on the probe mazes, better than 40% correct on its first performance on randomly presented 192 mazes.   Subsequently the monkey achieved equivalent performances on 3 sets of probe mazes, each set composed of 8 novel, 8 familiar, and 8 rotated mazes (80% correct choices).  Thus, the monkey developed flexible control at choice points to move either toward or away from the goal, in accord with the visual property of continuation.   
Tyler Wilks & Brett M. Gibson (University of New Hampshire)
Use of Landmark Cues and Self-motion Cues in the Clark's Nutcracker
Numerous studies have demonstrated that the Clark’s Nutcracker (Nucifraga columbiana) can locate a hidden goal in a laboratory setting using visual landmarks to establish its sense of direction.  The role of self-motion cues has been largely unexplored in this species.  In the current study the birds were trained to actively move from a position near the perimeter of an arena into a four-sided enclosure where a hidden goal was located. During training both self-motion cues and distinct visual landmarks inside the enclosure could be used to locate the goal.  Subsequent tests were administered to determine the extent to which nutcrackers were using each type of cue.  The results from these tests appear to indicate that the nutcrackers were able to use both landmarks and self-motion cues to return to the location of the hidden goal. 
Use of Landmark Cues and Self-motion Cues in the Clark's Nutcracker
Numerous studies have demonstrated that the Clark’s Nutcracker (Nucifraga columbiana) can locate a hidden goal in a laboratory setting using visual landmarks to establish its sense of direction.  The role of self-motion cues has been largely unexplored in this species.  In the current study the birds were trained to actively move from a position near the perimeter of an arena into a four-sided enclosure where a hidden goal was located. During training both self-motion cues and distinct visual landmarks inside the enclosure could be used to locate the goal.  Subsequent tests were administered to determine the extent to which nutcrackers were using each type of cue.  The results from these tests appear to indicate that the nutcrackers were able to use both landmarks and self-motion cues to return to the location of the hidden goal. 
Heidi L. Marsh & Suzanne E. MacDonald (York University)
The Role of Perceptual Features in Categorization by Orangutans
Two experiments were conducted to identify the perceptual features used by orangutans when categorizing pictures, in terms of category relevance and feature specificity.  In Experiment 1, subjects were trained and tested for transfer on a concrete discrimination (gorillas or orangutans vs. other primates).  Irrelevant features that did not define the target category did not affect performance.  Increased performance was noted on photos with faces, particularly close-ups.  In Experiment 2, photos were systematically modified to test the effects of various features.  Colour modifications caused decreased performance.  Features seemed to be processed both locally and holistically, depending on the task.  Error trials did not generally reflect mistaken categorization, but rather, photos that may have been attractive for other reasons, such as interesting colouring or novel morphology.  Thus, the importance of analyzing performance on both the negative and target stimuli, and using direct manipulation to test the effects of features, was emphasized.
Pete Otovic (University of South Florida)
Reconciliation in mandrills?
One consequence inherent to group life involves conflicts of interest between the individuals in a group over valuable resources such as food and mates.  In order to reduce the likelihood of an individual animal’s potential loss of benefits associated with group living, conflict management strategies have evolved both to prevent conflicts and to repair damage after a conflict.  Reconciliation is one type of post-conflict strategy, and is typically operationalilzed by affiliative contact, vocalizations, or proximity between former opponents shortly after a conflict.  Reconciliation ostensibly functions to restore relationships between former opponents to baseline levels.  However, observational data from captive mandrills (Mandrillus sphinx) are not consistent with this idea.   
One consequence inherent to group life involves conflicts of interest between the individuals in a group over valuable resources such as food and mates.  In order to reduce the likelihood of an individual animal’s potential loss of benefits associated with group living, conflict management strategies have evolved both to prevent conflicts and to repair damage after a conflict.  Reconciliation is one type of post-conflict strategy, and is typically operationalilzed by affiliative contact, vocalizations, or proximity between former opponents shortly after a conflict.  Reconciliation ostensibly functions to restore relationships between former opponents to baseline levels.  However, observational data from captive mandrills (Mandrillus sphinx) are not consistent with this idea.   
Ikuma Adachi(Emory University), Wendy A. Suzuki(New York University), Benjamin M. Basile, Regina Paxton, & Robert R. Hampton (Emory University)
Assessment of social dominance concept formation using videos of artificial social interactions in rhesus macaque monkeys (Macaca mulatta)
Although it is widely thought that animals living in complex social groups garner knowledge of the relative dominance of group members, little is known about how they acquire this knowledge. We have attempted to develop the laboratory techniques needed to answer the question. Using video editing software, we created a set of video clips of artificial social interactions in which one monkey demonstrated dominance over another. Together the set of videos represented an artificial linear dominance hierarchy consisting of five stimulus monkeys. Subjects were trained to pick the dominant monkey in each video clip. After they mastered the task, we gave them probe trials to assess both the degree to which they had formed a dominance concept, and the extent to which they remembered the dominance relations of each pair of stimulus monkeys. Some monkeys performed significantly above chance in these probe tests.
Lisa M. Pytka, Heidi E. Harley (New College of Florida) & Rebecca L. Curtiss-Floyd (Florida State University)
O grape, where are thou? A study of spatial cognition in two lemur species (Lemur catta and Eulemur fulvus rufus) in a laboratory foraging setting
Non-human primates comprise an especially interesting taxon for spatial cognition study; they occupy diverse foraging niches in a wide range of habitats despite their close phylogenetic relatedness. Though still understudied in comparison with monkeys and apes, prosimians represent especially interesting opportunities for studying cognition due to their rapid speciation in small but geographically varied Madagascar.  In this study, two species of lemurs (Lemur catta and Eulemur fulvus rufus) foraged for thirty-six hidden grapes in a three-dimensional apparatus similar to ones used in previous work with squirrel monkeys. An analysis of their search strategies revealed that the lemurs demonstrated use of spatial learning in the basic search task (i.e., they extracted all grapes when all holes were baited) and decreased in search efficiency and spatial search efficiency following a five-minute delay in a working memory task.  A control study demonstrated that lemurs were not solving the task through 
olfactory cueing.
Bradley R. Sturz & Jeffrey S. Katz (Auburn University)
Learning of relative distance between discrete visual landmarks by pigeons (Columba livia)
In an open-field search task devoid of orienting cues and informative geometry, pigeons were trained to find a goal located at the midpoint of the hypothetical line connecting two discrete landmarks positioned in a linear array. Pigeons searched in substrate for an initially visible but eventually invisible goal location containing food. The distance between the landmarks (interlandmark distance) was fixed throughout training. Pigeons learned to locate the goal and continued to search at this location in the absence of food. After reaching training criteria, food-absent trials were conducted in which the interlandmark distance either remained the same as in training or was manipulated by contraction or expansion. Search error and location on novel interlandmark distances were identical to those obtained on training trials. Results suggest pigeons learned to search at a relative distance between the landmarks. Implications of a stable frame of reference as critical in spatial learning are discussed. 
Roger K. R. Thompson, Carl E. Hagmann, Dobromir G. Dotov, & Victoria L. Templer (Franklin & Marshall College)
Can Capuchin monkeys (Cebus apella), like Humans, Discriminate Relations-Between- Relations? Maybe…Maybe not.
Spinozzi, Lubranno, and Truppa (2004) reported that capuchin monkeys (Cebus apella) rapidly learned to match relations (same/different) between spatial relations (above/below) in 128 – 352 trials. A replication seemed worthwhile given the contrast between their results and those from prior studies with monkeys and birds. We tested the ability of capuchin monkeys (C. apella) and humans to discriminate either ‘above/below’ or ‘variability/uniformity’ relations in a non-matching to sample task. The near errorless performances of the humans on their initial discrimination tasks transferred to the alternative task. The monkeys’ performances on both relational tasks remained at chance after 624 trials, but increased to greater than 80% correct when tested in full view of the experimenter, as was apparently the case in Spinozzi et al. (2004). Performances subsequently dropped to chance again in the experimenter’s absence indicating that successes were prompted by either inadvertent experimenter 
cuing or perhaps, more interestingly, social modeling.
Timothy M. Flemming (Georgia State University) and Roger K. R. Thompson (Franklin & Marshall College)
Deficits in Tool-Using Behaviors of Squirrel Monkeys (Saimiri sciureus sciureus)
In a comparative study of tool-using behaviors of squirrel monkeys and tufted capuchins (Cebus apella), several deficits in the manipulation of probe tools to obtain a food reward yielded an overall failure by squirrel monkeys. Within the first session of testing, several individuals in a socially-housed colony of capuchin monkeys successfully used probing tools to obtain honey from an enclosure. By comparison, squirrel monkeys not only touched the tools significantly fewer times, but never successfully obtained honey from the apparatus after six months of testing and exposure. In a second condition, both monkey species were presented with probing tools of differing lengths; some tools were not long enough to obtain honey from the enclosure. Squirrel monkeys manipulated the tools equally regardless of their affordance to success. Capuchins, by comparison, manipulated and correctly used longer tools at significantly greater rates.
Monique A. Rashid and Clive Wynne (University of Florida)
Modes of Interaction and Communication Between Humans and Domestic Dogs
Several studies demonstrate that domestic dogs have developed the ability to
use human social gestures and cues in their home environment to guide their behavior in beneficial ways. The current studies were designed to identify what it is about such gestures that allows the dog to identify and respond to the stimuli in a functional way. In addition, the limitations of visual acuity in domestic dogs were measured using a choice paradigm to address other possible factors that may explain differences between individual responding to particular social stimuli.
Kelly A. Schmidtke, Bradley R. Sturz, Jeffrey S. Katz (Auburn University) & Anthony A. Wright (University of Texas Health Science Center at Houston)
Observing Response, Set-Size, and Abstract-Concept Formation by Pigeons
For pigeons learning a matching-to-sample task, abstract-concept learning can depend on the observing response requirement (Wright, 1997); but, its role in the two-item same/different procedure is unclear. In the present experiment, groups of pigeons with different observing response requirements (FR1, 10, and 20) to the sample item acquired a two-item same/different task with increasingly larger training set-sizes (8, 16, 32, 64, 128, 256, 1024). Transfer tests with novel items were conducted after each set-size expansion was acquired. Observing response influenced the rate of acquisition across set-size expansion: FR20 < FR10 < FR1. However, the observing response requirement did not affect transfer performance between groups and all pigeons fully learned the abstract concept (evidence by transfer being equivalent to baseline performance). These results suggest that the effect of set-size is independent from the observing response requirement for abstract-concept learning within the present
parameters used in the two-item same/different procedure.
Jessica Crast (University of Georgia), Ingrid Kaiser (University of Georgia), Monique Dase (Morehouse School of Medicine), Dorothy Fragaszy (University of Georgia), Catherine Wallez (University of Rennes)
Capuchin monkeys (Cebus apella) learn to cope with task irregularities while using tools.
An ecological perspective on tool use suggests that an actor learns to use the affordances of objects and surfaces in instrumental activity. We investigated the ability of capuchin monkeys to cope with irregular surfaces while using tools. We predicted that capuchins’ manipulation of a tool around aberrations (diamond and rectangular barriers and holes) at six locations in a surface would improve with practice. We assessed changes in two capuchins’ skill at this task in three phases, each increasing in difficulty. Changes in success rate, efficiency and strategy were examined. For both monkeys, holes were consistently more challenging than barriers. The number of attempts per trial decreased across phases for barrier trials only, suggesting they learned more about barriers than holes. Success rates on each aberration type did not vary significantly across phases, 
implying an ability to cope with increasing challenges. Differential improvement on different aberration-location combinations is currently being analyzed. 
Qing Liu, Ingrid Kaiser & Dorothy Fragaszy (University of Georgia)
Development in Nut Cracking Skills of Young Bearded Capuchin Monkeys (Cebus libidinosus)
Wild capuchin monkeys (Cebus libidinosus) in Piauí, Brazil crack palm nuts on hard anvil surfaces using large stones. Young monkeys have been observed manipulating and striking stones and nuts against a substrate from 6 months of age. To investigate the development of nut cracking skills, we observed seven young monkeys at four time points over a period of one year and six months. Videos of activity at nut-cracking sites were collected at approximately 6-month intervals. We are coding behaviors involved in nut/stone manipulation, observing and scrounging from others, and the outcome of cracking actions. We will focus on the perceptual and motor contributions to developmental changes in activity with nuts and stones and the timeline of developing effective nut-cracking. Monkeys in Piauí crack nuts effectively at younger ages than wild chimpanzees or capuchin monkeys in other settings.
Cynthia A. Wei, Alan B. Bond, & Alan C. Kamil (University of Nebraska, Lincoln)
List Linking in Pinyon Jays (Gymnorhinus cyanocephalus): a Test of Transitive Inference.
Treichler and Van Tilburg (1996) showed that macaques can perform transitive inference between two linked five-item lists, suggesting the use of linear representations.  Based on the social complexity hypothesis and several previous studies, we predicted that pinyon jays (Gymnorhinus cyanocephalus) should also show transitive inference across linked lists.  Following methods similar to Treichler and Van Tilburg, we trained pinyon jays on adjacent pairs in two five-item lists (e.g. list 1: A>B>C>D>E and list 2: 1>2>3>4>5).  We then trained them on the linking pair (e.g. E>1 or 5>A) and subsequently tested the birds on probes of six novel pairs (for lists linked E>1, within list pairs= B>D, 2>4; between list pairs= D>1, E>2, B>2, C>3).  Patterns of response will be discussed with respect to the nature of the cognitive representation. 
Bauer, G.B. (New College of Florida; Mote Marine Laboratory), Colbert, D.E.(University of South Florida), Gaspard, J.C. III (University of Florida; Mote Marine Laboratory), Reep, R.(University of Florida), & Mann, D.(University of South Florida; Mote Marine Laboratory)
Sensory Processes of Florida Manatees:  A Review
Manatees are tactile/auditory specialists, with limited visual acuity, a pattern consistent with the frequently turbid, underwater environment they inhabit.  Neuroanatomical data suggest their chemical senses are of secondary importance.  Using the vibrissae-rich facial area, manatees demonstrate tactile discrimination of texture gratings (Weber fraction = 0.05) at a level comparable to human index finger performance.  Preliminary tests indicate manatees detect low frequency vibrations between 5 and 50 Hz, presumably through hydrodynamic sensation involving the vibrissae that cover the postfacial portion of their bodies.  Their auditory temporal processing rate is high, exceeding that for humans by a factor of 10.  Their evoked potential audiogram indicates detection of sound frequencies up to at least 40 kHz, with peak sensitivity around 24 kHz.  
Directional hearing for broadband stimuli is excellent, but localization of tonal sounds is less accurate.  They have dichromatic color vision, but visual acuity is probably no better than 20 minutes.
Carrie R. Rosengart (California University of Pennsylvania)
A Comparison of Spatial Search Task Performance of Capuchin Monkeys and Adult Humans
Capuchin monkeys and young children demonstrate perseverative searching.  Initial spatial search trials are accurate, but performance declines on subsequent trials due proactive interference. In this experiment the spatial search patterns of adult male capuchin monkeys (n=4) was compared with that of adult humans (n=40). Each participant, regardless of species, was given five trials. On each of the first three trials, the participant watched the experimenter hide a piece of food in the same spot. There were 24 different potential hiding locations.  After a delay period (0, 10, 30 or 120 seconds) the participants were allowed to retrieve the food. The distance of the first search attempt from the actual hiding location was recorded. On the next two trials, the reward was hidden in a different location. It is predicted that, like the monkeys, human search accuracy will decrease on the second set of trials.
Amber M. Chenoweth & Stephen B. Fountain (Kent State University)
Acute Nicotine Exposure Produces Only Mild Impairments of Adult Rat Serial Pattern Performance
We examined effects of nicotine, a nicotinic cholinergic agonist, on performance of a well-learned serial pattern.  Rats were trained to press levers in a sequential pattern: 123 234 345 456 567 678 781 818, where digits represent the clockwise position of levers in a circular array, spaces indicate 3-s pauses, and other ITIs were 1 s.  Once rats reached a high criterion, they received a relatively high dose of nicotine (0.4 mg/kg i.p.) for three successive days.  On Day 1 but not thereafter, nicotine produced small but significant impairments on the first element of chunks and on the final “violation” element that was inconsistent with pattern structure.  Nicotine never affected within-chunk performance.  Because muscarinic cholinergic antagonists such as atropine and scopolamine cause severe deficits in pattern performance, these small nicotine effects add to the evidence that muscarinic cholinergic systems play a more important role in pattern performance than nicotinic systems.
Shannon M. Kundey (Kent State University), James D. Rowan (Wesleyan College), & Stephen B. Fountain (Kent State University)
Some Limitations of the Sequential Pairwise Associative Memory (SPAM) Model of Rat Serial Pattern Learning
Wallace and Fountain (2002, 2003) showed that the sequential pairwise associative memory (SPAM) model, a computational model based on pairwise associations and generalization, could simulate a variety of rat serial pattern learning phenomena for sequences composed of successive food quantities.  In the present studies, we examined how well SPAM simulates earlier data from rat serial pattern learning studies in two different paradigms, namely, a stimulus anticipation paradigm in a 6-light linear array and a pattern production paradigm in which rats anticipated the successive positions of correct responses in an 8-lever circular array.  In both paradigms, patterns were highly structured and extensively “branching” sequences of positions in the arrays (i.e., the sequences were characterized by cues that signaled different events at different points in the pattern).  SPAM failed to simulate rats’ performance in both these paradigms.  We examine possible reasons for SPAM’s failures in these
paradigms and suggest possible remedies. 
Poster Session 2 - Saturday Evening
Justin S. Johnson, Whitney L. Kimble, & Martha Escobar (Auburn University)
Directionality of training, directionality of testing, and contingency estimation
Previous research from our laboratory has shown that subjects tend to overestimate cause-effect contingencies if training occurs in the predictive (cause-to-effect) but not the diagnostic (effect-to-cause) direction, regardless of testing direction (Wilhelmsen, Suits, Bray, & Escobar, CO3 2006).  Three further experiments investigated the effects of nondirectional training, nondirectional testing, and noncausal test questions on contingency estimation. Participants were trained with a single cause and a single effect, and &#61508;p was set at 0.5 for both the predictive and diagnostic relationships.  Overestimation was again observed whenever training occurred in the predictive direction, but only if the test question implied causality.  These results are consistent with theoretical approaches assuming intrinsic differences between predictive and diagnostic judgment.
W. David Stahlman (UCLA), Seth Roberts (UC-Berkeley), & Aaron P. Blaisdell (UCLA)
The Relationship Between Reward Probability and Operant Screen Pecking in Pigeons
We investigated the role of probability of reward on the rate and variability of operant responding in pigeons.  Gharib, Gade, and Roberts (2004) previously found an inverse relationship between behavioral variability and probability of reinforcement in rats' operant lever pressing.  In four experiments, we presented pigeons with a set of colored stimulus targets on a touch screen. Pecks to the stimuli were rewarded with grain. The color of a target signaled a specific likelihood that pecks would be rewarded.  As expected, operant variability increased as a function of decreasing likelihood of reinforcement.  We also report interesting response rate differences that appear to be inconsistent with contemporary theoretical accounts of behavior.
Oshri L. Hakak, Ruey K. Cheng, Christina L. Williams, & Warren H. Meck (Duke University)
Effects of sex and prenatal choline treatments on environmental and object exploration
Prenatal choline supplementation has been shown to enhance adult rats’ performance on cognitive tasks. Recent studies have associated prenatal choline supplementation with increased gamma oscillations in the adult hippocampus, which in turn are suggested as important for memory consolidation during sleep states. In this study, we explored whether prenatally choline supplemented rats of both sexes would spend more time interacting with novel objects, and subsequently engage in more REM sleep. Local field potentials were recorded in the dentate gyrus of 36 rats of both sexes and different prenatal choline backgrounds (i.e., deficient, sufficient, and supplemented) during both interaction with novel objects and subsequent sleep. Preliminary analysis indicated that environmental and object exploration did not correlate with prenatal choline treatments. A significant sex difference was observed, however, for these two behaviors, suggesting that female rats interacted more with the novel objects and
environments than did males.
Rebecca M. Rayburn-Reeves, Laura A. Bullard, L. Brooke Poerstel, Katherine E. Bruce & Mark Galizio (University of North Carolina Wilmington)
Functional Equivalence in Rats with Olfactory Stimuli
Functional equivalence was studied in rats using a repeated reversals procedure with class-specific reinforcers and olfactory stimuli.  Six rats were tested using scented sand as olfactory stimuli in a two-choice modified operant chamber and an odor arena.  Stimuli were randomly assigned to one of two sets.  Each set was baited with a specific reinforcer: sugar or grain.  Responses to one set were reinforced until criterion levels of performance were reached; at this point, the contingencies were reversed.  Contingencies were reversed each time criterion was met.  Most rats showed a reduction in the number of sessions to reach criterion, as well as an increase in the percent correct on the first exposure to the reversed stimulus sets. Thus these data illustrate functional class formation with olfactory stimuli in rats. 
Kenneth J. Leising (UCLA), Kosuke Sawa (Senshu University), & Aaron P. Blaisdell (UCLA)
Temporal integration in Pavlovian appetitive conditioning in rats
We used an appetitive sensory preconditioning procedure to investigate temporal integration in rats in two experiments. In Phase 1, rats were presented with simultaneous compound trials on which 10-s conditioned stimulus (CS) X was embedded within 60-s CS A. In Group Early, CS X occurred during the early portion of CS A, while in Group Late, CS X occurred during the latter portion of CS A. In Phase 2, CS X was paired simultaneously with sucrose. On a subsequent test with CS A, the rate of magazine entries peaked during the early portions of the stimulus in Group Early and in the latter portions of the stimulus in Group Late (Experiments 1 and 2). Similar response peaks were not observed on tests with a control stimulus that had been presented in compound with a stimulus that did not signal reward (Experiment 2).
Ikuma Adachi & Robert R. Hampton(Emory University)
Individual recognition of conspecifics in videos by rhesus macaques (Macaca mulatta)
The cognitive ability to identify group members is advantageous for animals living in social groups. While a variety of findings from the field setting demonstrate individual recognition, there are few studies of this capacity in the lab. We examined this ability in rhesus macaques using a delayed video-to-picture matching-to-sample procedure in which a 5-sec video clip of a familiar monkey was presented as a sample, followed by five still images as comparisons. Five stimulus monkeys were used for each subject: a cage mate and four monkeys living in the same room. We prepared twenty six video clips of each stimulus monkey. Subjects were trained initially with two video clips for each. Each time performance exceeded 75% in the two consecutive sessions, we added six new video clips of each of the stimulus monkeys. All subjects learned to generalize performance to new video clips on the first trial.
Dennis Garlick & Aaron P. Blaisdell (UCLA)
The Hitchhiker’s Guide to Data Collection and Analysis using Visual Basic
Training in psychology typically does not include extensive instruction in computer programming. Modern programming languages, however, enable the almost instantaneous completion of tasks that might take researchers thousands of hours to complete manually. Not only can this time be better spent on other academic activities, but very labor-intensive procedures increase the likelihood of errors being made, and also limit the number of alternative analyses that are performed. A number of utilities programmed in Visual Basic for Windows will be described to illustrate the usefulness of computer programming. Utilities include automated administration of Med-PC IV test sessions, writing data directly to spreadsheets rather than text files, accessing current test session status across the internet from any web browser, and automatic graphing of current experimental data. The production version of the Automated Remote Environmental Navigation Apparatus (ARENA) will also be demonstrated. The possibility of sharing these tools with other laboratories will be discussed.
Suits, W. T., Wilhelmsen, S. R., & Escobar, M. (Auburn University)
Preexposure and inhibition effects in long-delay conditioning
In two experiments, rats were trained with an appetitive preparation to determine whether the initial segment of a long-delay conditioned stimulus (CS) that predicted delivery of an unconditioned stimulus (US) in its final segment could become inhibitory (i.e., inhibition of delay). Previous research from our laboratory had determined that, with this preparation, the initial segment of the long-delay CS passed a retardation but not a summation test for conditioned inhibition. Experiment 1 assessed whether CS preexposure of the initial segment of the long-delay CS resulted in its passing the retardation but not the summation test. Consistent with this hypothesis, changing the context between the long delay and retardation trials abolished retardation. In Experiment 2, evidence of summation was obtained when the absence of the US was made explicit through the use of a discrete excitor presented during the initial segment of the long delay CS. 
Patricia A. Wilson & Herbert S. Terrace (Columbia University)
How do monkeys (Macaca mulatta) represent items in serial memory?
A variety of recent experiments, on both human and non-human primates, have reported distance effects.  In each instance, subjects were tested on 2-item subsets that were derived from one or more lists on which they were trained (D'Amato & Colombo, 1988; Terrace, Son & Brannon, 2003). The general result was that accuracy of responding increased and reaction time decreased as the ordinal distance between the test items is increased.  Such distance effects suggest that list items were represented spatially along a linear continuum.  To investigate that hypothesis, we used monkeys (Macaca mulatta) with experience on the simultaneous chaining paradigm in our current study.  On the original simultanous chaining paradigm, subjects received primary reinforcement if and only if they had responded, in a particular order, to an array of 4 photographs that were displayed simultaneously on a touch-sensitive video monitor. To insure that subjects couldn’t learn the required sequence as a motor plan, the position of the list items varied randomly from trial to trial. In the current paradigm we used the same procedure with a few modifications.  A simultaneous chain was displayed in one of four different configurations:  linear (vertical [top to bottom] and horizontal [left to right] where the
spatial position matches the list order), probe linear (horizontal or vertical but spatial position does not match order position), and simchain (random spatial position). If subjects represented list items spatially, they might be able to use the hint provided on those trials on which the required sequence could be executed by following a spatial rule.  Our results to date show that, on 4-item lists, monkeys benefited from trials on which they could respond to list in a linear manner.  Positive transfer from linear trials to trials on which the spatial location of list items varied randomly support the hypothesis that subjects do indeed organize list items along a spatial continuum. This is the first step in a series of experiments targeted towards determining if presentation of a list with matching spatial and order positioning will increase the rate of acquisition and retention of a serial list.
Erin Stromberg (Smithsonian National Zoo), Sharon A. Himmanen (Lehman College/CUNY), & Karyl B. Swartz (Great Ape Trust of Iowa).
Does ordered list presentation affect response patterns in an unordered report task?
Two orangutans developed and used a spatial response strategy when reporting items on a touch-sensitive video screen from a list that had been presented randomly.  In a second study, when list items were presented in a constant order, one animal abandoned the spatial response strategy, but did not show evidence of a serial position effect.  In a third study, when the list items from previously-ordered lists were presented in a random order, the spatial response pattern returned, suggesting that attributes of the list items, when presented in a constant order, may have contributed to response patterns.
Colin Ellard & Erica Stuart (University of Waterloo)
Anxiety, temperament, and anti-predator behaviour in the Mongolian gerbil
Some standard laboratory tests of anxiety are predicated on the idea that the construct of anxiety may be related to processes normally engaged by an animal to avoid predation (vigilance, thigmotaxis, freezing, etc.).  There have been few studies in which the relationship between an animal's temperament and its ability to avoid potential predators have been examined directly.  We carried out such a study by measuring the behaviour of Mongolian gerbils on a host of open field variables and on some standard and some custom measures of anxiety.  Though we found some evidence for enduring temperament, and individual differences in 'anxiety', there was little or no correlation between these measures and the ability of an animal to conduct brisk and adaptive responses to imminent threat of predation.  These findings suggest that anxiety and effective anti-predator behaviour may not be related.
Emily Ward (Franklin & Marshall College), Yuko Hattori (Kyoto University), Roger K. R. Thompson (Franklin & Marshall College), & Kazuo Fujita (Kyoto University)
Effects of Two-Dimensional Noise and Feature Configuration on the Recognition of Faces in Capuchin Monkeys (Cebus apella)
Face recognition in Old World monkeys is remarkably similar to that of humans, but relatively little is known about this ability in New World monkeys, such as capuchins (Cebus apella).  In order to investigate capuchins’ perceptual and cognitive ability to recognize faces, the animals were trained to spontaneously discriminate between grayscale pictures of both humans and nonhuman primates and non-face objects.  Two-dimensional visual noise was then introduced into both images as a means to decrease face recognition (cf. Harmon & Julesz, 1973). The animals were able to make the discriminations even at a high level of image degradation. In order to minimize variations in the stimuli physical properties, the facial features of the face stimuli were randomly scrambled and paired with an original intact face. In this second spontaneous discrimination task, none of the animals could discriminate between the scrambled and intact face at a level significantly above chance.
L. Brooke Poerstel, Laura A. Bullard, Rebecca M. Rayburn-Reeves, Kelly Weiland, Katherine E. Bruce,& Mark Galizio (University of North Carolina – Wilmington)
Evaluating Olfactory Identity Matching-To-Sample (MTS) in Rats Using an Open-field Apparatus
Emergence of generalized identity matching-to-sample (MTS) was evaluated in rats using olfactory stimuli, multiple exemplars, and an open-field apparatus. Five subjects were presented a sample stimulus followed by multiple comparison stimuli. Responses to the comparison identical to the sample were always reinforced, whereas responses to the dissimilar comparison were never reinforced. Following criterion performances for a given set of stimuli, subjects were tested for generalized matching during the first presentations of a novel set of ‘test’ stimuli. The previous set of stimuli was removed from baseline and the novel test set was then trained to criterion. Using multiple exemplar training, subjects were repeatedly tested with new sets of novel stimuli. Acquisition of training discriminations and evidence for emergence of generalized matching was observed in most rats. 
Janice M. Hassett (Emory University, Yerkes National Primate Research Center, Center for Behavioral Neuroscience), Julie Martin-Malivel (Emory University, Yerkes National Primate Research Center, Center for Behavioral Neuroscience), Henry Lange (Emory University), Andrew Fischer (Emory University, Center for Behavioral Neuroscience), & Kim Wallen (Emory University, Yerkes National Primate Research Center, Center for Behavioral Neuroscience)
Age and Rank Influences on Access to an Automated System for Cognitive Testing in Socially Housed Rhesus Monkeys
We developed an automated touch-screen system allowing rhesus monkey social groups continuous access to computerized tasks, with interactions individually identified.  Monkeys are usually individually tested thus it was unknown whether monkeys of all ages and ranks would have access to the kiosk in a social context.  Kiosk access was investigated using a task in which monkeys touched a small screen image to obtain a reward.  Monkeys were not trained thus performance reflected spontaneous acquisition.  In 3 days of testing, 94/132 monkeys accessed the kiosk and 66 completed the maximum 50 correct trials allowed.  Age significantly affected access and completion, with older animals accessing and finishing
less. For rank, lower-ranked animals accessed and reached criterion significantly less. However, individuals from all ranks and ages successfully accessed the kiosk. There were no sex differences and aggression related to access was rare.  Thus, the kiosk provides an unobtrusive method of collecting cognitive data.
MacQueen, D. A., Bullard, L. A., & Galizio, M. (UNC Wilmington)
Determinants of Olfactory Memory Span in Rats
The present study developed an incrementing non-match to sample procedure in rats designed to be analogous to memory span tasks used with humans. Rats were placed large arena with 18 food locations. In the initial trial of each session, one food cup marked with a distinct olfactory stimulus was present and responding to it was reinforced. Each subsequent trial added a new olfactory stimulus and responding to the new stimulus was always reinforced (non-matching). The dependent measures were number of stimuli that incremented without error (span) and overall percent correct responses. Spans improved with successive testing and frequently exceeded 20 stimuli. However, errors were unaffected by the number of stimuli to be remembered, raising questions about the analogy to human memory span procedures. 
Patchouly N. Banks & Eduardo Mercado III (University at Buffalo)
Memory for Shells in Hermit Crabs (Clibanarius tricolor)
Hermit crabs inhabit empty gastropod shells to protect their soft abdomens, and are thus continuously in search of the perfect fitting shell. In the current experiment, hermit crabs (Clibinarius tricolor) were tested on their ability to recognize previously encountered shells. A total of 48 hermit crabs were given three opportunities to investigate the same shell, and three opportunities to investigate three different shells. Crabs spent significantly less time investigating familiar shells than unfamiliar shells. When judging a new shell, a crab presumably compares the qualities of that shell with its current shell to determine which will provide better protection. To make this comparison, the crab must have retained some information acquired during its investigations of its current shell, possibly over an extended period. Further research should explore this idea, as well as the possible mechanisms by which hermit crabs recognize shells,including shell and self-recognition through chemical cues.
Debborah E. Colbert (University of South Florida, Sea Life Park by Dolphin Discovery), David Mann (University of South Florida), Joseph Gaspard (University of Florida), Gordon B. Bauer (New College of Florida), Kim Dziuk & Adrienne Cardwell (Sensory Biology and Behavior Program, Mote Marine Laboratory & Aquarium), & Roger Reep (University of Florida)
Sound Localization Abilities of Florida Manatees, Trichechus manatus latirostris
Two experiments measured the underwater sound localization abilities of two manatees.  In a four-choice experiment, subjects selected from four speakers in a horizontal plane at -90o, -45o, 45o and 90o relative to the subject’s head (0o). Three broad-band signals (0.2-20, 6-20, and 0.2–2kHz) were tested at four durations (3,000, 1,000, 500, and 200ms) and two tonal signals (4 and 16kHz) were tested at 3,000ms.  All signals were tested at 100dB re:1 uPa (±1.5dB).  Broad-band accuracy ranged from 93%-52%, well above the chance level of 25%.  Tonal signal accuracy ranged from 49%-32%. In an eight-choice experiment, subjects were positioned at a depth of 1.5m at the center of equally-spaced speakers in the horizontal plane. Three broad-band signals (0.2-24, 18-24 and 0.2–1.5kHz) were tested at two durations (3,000 and 200ms).  Preliminary results indicate that 
the subjects are capable of accurately localizing the 3000ms test signals presented in this circular array. 
Noam Y. Miller & Sara J. Shettleworth (University of Toronto)
An Associative Model of Geometry Learning
Cheng (1986) suggested that learning the geometry of enclosing surfaces takes place in a geometric module blind to other spatial information. Failures to find blocking or overshadowing of geometry learning by features near a goal seem consistent with this view. However, discussions of these effects have overlooked that spatial learning in an arena or watermaze is an operant task. We present an operant model in which learning spatial features competes with geometry learning, as in the Rescorla-Wagner model. Relative total associative strength of cues at a location determines choice of that location and thus frequencies of reward paired with each cue. The model shows how competitive learning of local features and geometry can appear to result in potentiation, blocking, or independence when choice is measured in a spatial task and how the results depend on enclosure shape and kind of features. It reproduces numerous findings from dry arenas and watermazes.
Anna Wilkinson & Kimberly Kirkpatrick (University of York, UK)
The effects of training history on pigeons’ tracking and capture of moving objects
This experiment examined the effects of training history on tracking and capture of moving objects by pigeons. Two groups of birds were given a task where they had to capture a moving object on a touch screen monitor in a customised operant chamber. The first group had previously been trained on a stimulus that appeared on the left side of the monitor and moved across the screen in a rightward linear motion. The size of the stimulus systematically decreased and speed increased across phases. The second group did not receive any prior training with moving objects. All birds were then presented with a stimulus that was 12 pixels in diameter travelling at 68.8 pixels/s, which could appear from any side of the monitor and move directly across to the opposing side. It was found that prior rightward training lead to a rightward bias in pecking errors for all motion types. Acquisition of the four-motion task was slower and the rightward bias remained despite over 150 sessions of subsequent training. The implications for the mechanisms involved in capture and tracking will be discussed.
Krista Macpherson & Bill Roberts (University of Western Ontario)                        
Some Tests of Counting in Dogs
Domestic dogs were tested for their ability to count by placing different numbers of food items, one at a time, in each of two containers while a dog watched.  The dog was then allowed to choose between the containers and to eat all of the food in the container chosen.  In two experiments, no evidence was found that dogs preferred the container into which the greater number of items had been placed.  This finding contrasts with chimpanzee experiments, in which chimps regularly took the larger number of food items (Beran & Beran, 2004).  The possibility that these results may be explained by the recent evolutionary history of dogs is discussed.
Gin Morgan, Nicholas K. DeWind, and Herbert S. Terrace (Columbia University)
An Exception to Weber’s Law in Numerical Representations by Rhesus Macaques
There is a considerable body of evidence supporting the hypothesis that, in nonhuman primates, representations of numerical stimuli obey Weber’s Law, i.e., increasing variability of responding with increasing target magnitude.  Using a delayed non-identical numerical matching-to-sample procedure with rhesus macaques (Macaca mulatta), we have found that while accuracy shows a distance effect, responses do not show scalar variability. Instead, they show strong end effects on both ends of the stimulus continuum.  In this experiment, rhesus macaques (n=2) completed a numerical matching-to-sample task with three different stimulus continua of equal length (1-11, 3-13, and 5-15).  Results compare response patterns and examine the contributions of absolute numerical value and relative position in the stimulus continuum to numerical representation in nonhuman primates.
Dobromir G. Dotov & Anthony Chemero (Franklin & Marshall College)
Detection of Entropy in the Array Allows Two-Layer Networks to Discriminate Same and Different
Measuring entropy is one way to quantify the amount of information in a set of bits (Shannon & Weaver, 1949). Animals could use entropy in the visual array for perceptual categorization tasks. Experiments with pigeons reveal strong positive correlation between entropy in the array and the ability to learn same-different discrimination (Young & Wasserman, 1997). We trained simulated networks to perform the same task in six conditions. We found that it became increasingly difficult for our networks to learn the categorization as the number of icons decreased from sixteen to two, that is, the range of possible entropy values shrunk. Furthermore, with only two layers, our networks are only capable of association, and so could not have computed the entropy values. This indicates that the networks simply detected entropy. The similarity in the performance of the networks and pigeons suggests that the pigeons might also be detecting, and not computing, entropy.
Christina Wesolek, Joseph Soltis, Anne Savage (Disney's Animal Kingdom), Kirsten Leong (Cornell University), & John D. Newman (NIH)
Emotional Arousal in the Voiced Sounds of the Rhesus Monkey and African Elephant
Emotional arousal is expressed in the mammalian voice, but there are no consistent acoustic measures used and few comparative analyses. We apply a representative set of source and filter features to rhesus macaque (Macaca mulatta) ‘coos’ and African elephant (Loxodonta africana) ‘rumbles’ produced in “low” and “high” arousal contexts. In both species, MANOVA showed that the 15 measures separated calls across arousal categories. In rhesus macaques, high arousal was associated with increased and more variable fundamental frequencies, increased amplitudes, and a shift in formant locations. In African elephants, increased arousal was associated with lower amplitudes. In addition, low ranking female elephants expressed a greater magnitude of acoustic change compared to high ranking females. These acoustic features may successfully characterize arousal state in a variety of mammals, but the specific acoustic features that reflect arousal and the specific pattern of acoustic response may vary by
species, individual and arousal context.
Faith E. Warner, Stephen E. G. Lea (University of Exeter)
Concept-Level Generalization in Monkeys Following Training on a Single Exemplar of a Familiar Category
Visual discrimination of biologically relevant concepts was tested in four non-human primates, following training on multiple copies of a single exemplar.  The subjects were one Sulawesi black macacque (Macaca nigra), two white-throated capuchins (Cebus capucinus) and one mandrill (Mandrillus sphinx).  A ‘pokeboard’ was used to test generalization to varying exemplars of a familiar category (conspecifics amongst allospecifics) and an unfamiliar category (tigers amongst other felids).  Each subject was tested on both categories, and saw each test stimulus once only.  The test stimuli varied in aspect, background, age and sex of individual depicted, but all were shown in natural orientation.  Reliable generalization occurred in all cases for the conspecific condition (median Herrnstein &#961; scores: 0.74, 0.66, 0.60, 0.88). Generalization to tigers amongst other felids was lower overall (0.51, 0.58, 0.70, 0.55).  This difference may be due to prior conceptual knowledge of conspecifics or to an innate ability to process species-specific faces.
E. Hartmann & M. Kiley-Worthington. (Eco-Etho Research & Education Centre, France. & University of California, Berkeley)
Enhancing learning of verbal cues in horses (Equus caballus) through cooperative teaching
Studies investigating linguistic skills in mammals have indicated that these species can learn verbal cues. This is facilitated  in horses by cooperative interaction between the animal subject and the human teacher. This can have considerable application for their welfare and teaching. The study tested a group 6 horses to discriminate colours by name using a cooperative teaching approach involving words & gestures, primary & secondary positive reinforcement. The tests  showed  that the horses  recognized the colours by a verbal cue &  learnt quickly. Implications for handling and teaching equines are discussed.