Start Time
3:30 PM	Welcome Reception & Light Snacks
	Wednesday Evening
6:50 PM	Opening Remarks - Tom Zentall
7:00 PM	Spatial Learning and Memory (Chair - Mike Brown )
7:00 PM	Michael F. Brown & Gary Giumetti (Villanova University)
	Spatial Pattern Learning in the Radial-arm Maze
	Two experiments will be described that provide evidence that rats can learn a spatial pattern of baited and unbaited arms on the radial-arm maze.  In particular, baited and unbaited arms occurred in a spatially alternating pattern on each trial.  However, the identity of the baited and unbaited arms was unpredictable over trials.  Thus, it was the abstracted spatial pattern itself, rather than the spatial relations among particular locations, that controlled choices.   The implications of these results for theories of spatial representation will be discussed.
7:14 PM	William Timberlake, Susan A. Sinning, & Joseph Leffel (Indiana University
	Beacon Training in a Water Maze Can Facilitate as well as Block Room Cue Learning
	In Stage 1 of a water maze blocking procedure, rats were trained with a Beacon predicting the location of a submerged escape platform.  In Stage 2 both Beacon rats and added controls were trained with a predictive compound of Beacon and Room Cues (Landmarks plus Background Cues).  The effect of Stage 1 training was evaluated relative to controls in a Room Test without the Beacon.  In Experiment 1, Stage 1 Pole Beacon training blocked Stage 2 learning of Room Cues.  In Experiment 2, Stage 1 Hanging Beacon training had no effect or facilitated Room Cue Learning in Stage 2.  In Experiment 3, Stage 1 Hanging Beacon training with irrelevant Room Cues showed no effect, and neither did an overshadowing control in Stage 2.  In Experiment 4 Displaced Beacon training showed facilitation.  Apparently Stage 1 Pole Beacon training contributes to blocking, and a stable framework to facilitation.
7:28 PM	Ken Cheng, Ajay Narendra (Macquarie University), & Rüdiger Wehner (University of Zurich)
	Navigation in the Central Australian Desert Ant Melophorus bagoti: Some Initial Results
	The highly thermophilic red honey ant Melophorus bagoti forages in the heat of the day. We provided ants a feeder sunk in the ground. Ants arriving there the first time were displaced, with their food, to a distant test field and allowed to home. Ants headed off in the feeder-to-home direction, indicating a directional sense based on sky-compass cues. These ants ran on average, however, less than half the homebound distance before initiating search behavior. The search was biased, drifting farther and farther from the start of the homebound run. With repeated runs to and from a food source, ants established stereotypic routes. Ants trained and tested in channels that blocked most landmark cues estimated the correct distance home, indicating an odometric ability. Preliminary evidence also suggests that nest-associated cues, possibly panoramic cues, are used for homing. This suite of strategies suits an environment filled with tussocks interspersed with trees.
7:42 PM	John B. Phillips, R Muheim (Virginia Tech), NM Edgar (New York Medical College), & KS Sloan (Virginia Tech)
	Auditory and Magnetic Compass Orientation in C57BL/6 mice
	We report the development of behavioral assays of magnetic and auditory compass orientation in C57BL/6 mice.  We show that C57BL/6 mice can be trained to position their nests in a learned relationship to directional auditory or magnetic cues in a visually symmetrical arena.  Initial findings from the magnetic compass assay suggest that mice may perceive the magnetic field as a complex three-dimensional “visual” pattern, consistent with the light-dependent, photoreceptor-based magnetoreception mechanism proposed by Ritz et al. (2000).  Development of these assays in C57BL/6 mice will make it possible to investigate the biophysical and molecular mechanisms involved in magnetoreception, as well as the neural pathways and processing mechanisms involved in auditory and magnetic compass orientation.  C57BL/6 mice also perform well in variety of spatial tasks in radial arm and water mazes, so this strain of mice can be used to investigate the interaction of spatial (e.g., visual landmark arrays) and magnetic compass cues in the orientation behavior of rodents.
7:56 PM	Eric L.G. Legge & Marcia L. Spetch (University of Alberta)
	Do Differential Outcomes Enhance Spatial Localization?
	Numerous studies have shown that differential outcomes (DO) enhance discrimination and retention accuracy in matching-to-sample tasks.  The present study tested whether DO would also promote faster learning and more accurate searching in a landmark-based spatial task. That is, would DO expectancies help an organism remember where a goal is hidden? Pigeons were trained to find a hidden goal that was either south-west of a green landmark or north of a red landmark in a spatial arena.  Only a single landmark was presented on each trial.  For the DO group, each landmark corresponded to a specific type of hidden food, whereas for the control group, the relation between food type and landmark was random across trials. If pigeons in the DO group show faster acquisition and more accurate searching than pigeons in the control group, this would indicate that outcome expectancies not only enhance discriminatory behavior but also spatial localization.
8:05 PM	Causal  Reasoning (Chair - Bill Whitlow)
8:05 PM	Bill Whitlow (Rutgers University)
	Causal Scenarios and Configural Learning
	Learning psychologists typically approach causal learning with a focus on stimulus conditions and ask what is learned about the causal roles of the events involved in a given problem?  Social psychologists typically adopt a focus on causal scenarios and ask what explanatory framework do people use to reach a causal interpretation of a given problem?  Our research has tried to integrate these two approaches by asking how the role of configural cues in causal reasoning depends on the causal scenario invoked by different tasks.  The present studies compare positive and negative patterning results in a social reasoning task and in a consummatory reasoning task.
8:19 PM	Kenneth J. Leising & Aaron P. Blaisdell (University of California, Los Angeles)
	Causal Reasoning from Interventions in Rats is Not Explained by Interference
	Blaisdell, Sawa, and Waldmann (in press) found evidence that rats are capable of deriving predictions of the outcomes of interventions after observational learning of a casual model.  This capacity is consistent with causal Bayes net theories, but alternatively, may be the result of interference between antecedent events that share a common outcome.  To test this competing hypothesis, rats received sensory preconditioning training consisting of A->X pairings in Phase 1 and A->sucrose pairings in Phase 2. At test, one group of rats received outcome X contingent upon lever pressing.  Nose poking into the feeding niche in this group was compared, using a yoking procedure, to two groups that merely observed outcome X, and one group that received an antecedent cue B prior to X (i.e., B->X). Cue B did not attenuate nose poking to X the way a lever press did, supporting causal Bayes nets and not an associative interference interpretation of intervention effects.
8:26 PM
	Seth R. Wilhelmsen, V. Leah Bray, W. Travis Suits, & Martha Escobar (Auburn University)
	Overestimation of Contingencies Following Predictive but not Diagnostic Training
	Contingency estimation (specifically, causality judgments) can be made as a prediction (cause-to-effect) or as a diagnosis (effect-to-cause).  Most investigations trying to determine whether the two directionalities of learning involve identical or different processes have used designs with multiple causes or multiple effects (i.e., stimulus competition), but the results have been contradictory: Although causality learning occurs in both directions, competition in diagnostic training/testing situations has not been obtained in all preparations.  To determine whether there are basic differences in contingency estimation depending on the directionality of training/testing, participants were trained with pairings of a single cause and a single effect (DP for both directions = 0.50).   Training and testing were either predictive or diagnostic (i.e., a 2x2 design).  Participants correctly estimated the contingencies when training was diagnostic, whereas they overestimated the contingency when training was predictive, both effects regardless of testing direction.
8:35 PM	Communication, Song, and Auditory Discrimination (Chair -Ron Weisman )
8:35 PM	Ronald Schusterman, Colleen Reichmuth Kastak (University of California Santa Cruz), & Debbie Quihuis (Six Flags Marine World)
	Contingency Learning Can Modify Sound Production in Pinnipeds
	Classical ideas about acoustic communication in nonhuman mammals are based on the notion that vocalizations are non-modifiable, with emotional constraints, anatomical limitations, and genetic predispositions being the most significant features in mammalian communication systems.  Recently, several studies have shown that the vocalizations of some birds can be modified by food and social reinforcement contingencies.  The current study examines whether contingency learning can also modify the context and structure of sounds emitted by mammals.  Sounds produced by one captive harbor seal and four captive walruses were modified using food reinforcement to 1) alter in contextual control, 2) change in structure through selective shaping, or 3) vary in structure through the application of novelty or “difference” reinforcement criteria.  The results of these investigations with pinnipeds reveal a degree of vocal plasticity previously unreported for non-human mammals and suggest that contingency learning is relevant to the evolution of communication systems including human speech.
8:59 PM	Henrike Hultsch (Freie Universität Berlin) & Katharina Riebel (Leiden University)
	The Serial Position Effect in the Song Acquisition of Birds
	The 'serial position effect' is well known in short-term serial item learning (U-shaped recall probability) in humans. We asked whether analogous phenomena occur in a very different, long-term process, the song learning in nightingales (Luscinia megarhynchos).  Young males heard strings of model song-types and were tested for song imitation as adults. Each song in a tutored string was a different song-type, thus imitations could be examined for the serial position of models from which they had been copied.  We found no relation between serial position and probability of acquisition, but a significant effect on performance frequency. Models heard at the beginning or the end of a string were sung more frequently than 'central items'.  Such U-shaped function implies that the beginning and the end of a string of learning stimuli signify a particular quality during perceptual learning in nightingales that does not express itself in acquisition success.
9:13 PM	Christopher B. Sturdy, Michael R.W. Dawson and Isabelle Charrier (University of Alberta)
	Artificial neural network and statistical approaches to understanding natural vocal categories
	The ‘chick-a-dee’ call of the black-capped chickadee is composed of four note types, perceived by chickadees as open-ended categories. Here we describe the spectrograms from a sample of notes as a set of 9 summary features. An artificial neural network was trained to identify note type on the basis of these features, and obtained high (> 98%) accuracy.  An internal analysis of the network revealed a distributed code in which different hidden units generated high activities to different note types. By combining these different sensitivities, the network could discriminate all three types of notes. Network performance was compared to a discriminant analysis of the same data. This latter analysis also achieved a high level of performance (95%). Comparing the two approaches revealed some similarities, but also intriguing differences. We discuss our results in terms of developing both a tool for note classification and a theory of how birds classify notes.
9:27 PM	Laurie L. Bloomfield & Christopher B. Sturdy (University of Alberta)
	Mechanisms for Species Discrimination in Sympatric and Non-Sympatric Chickadees
	Previous experiments (Bloomfield & Sturdy, in prep.) provide sound empirical evidence that both black-capped chickadees (Poecile atricapillus) and mountain chickadees (P. gambeli) treat their own and the other species’ chick-a-dee calls as separate, open-ended categories. Further, there is no species-specific advantage for individual recognition of 30 different calls and prior experience of black-capped chickadees with mountain chickadees neither hinders nor aides in classification and recognition of chick-a-dee calls. Here we explore the mechanisms used by black-capped and mountain chickadees in discriminating among the chick-a-dee calls and whether experienced and non-experienced black-capped chickadees utilize different features of the chick-a-dee calls for accurate species discrimination. Preliminary data suggest that the terminal 'dee' portion of the calls guides species’ recognition and discrimination, the ecological factors of which will be discussed.
9:34 PM	Chris Harshaw & Robert Lickliter (Florida International University)
	The Influence of Stimulus Contingency and Stimulus Enhancement on Auditory Preferences in Bobwhite Quail Neonates
	This study explored the effects of both stimulus contingency and stimulus enhancement on the development of auditory preferences in bobwhite quail neonates. Subjects were given individual exposures to one of two variants of a bobwhite maternal call (A or B) at 24 hours of age and were subsequently tested at 48 hours of age in a simultaneous choice test between the familiar and novel calls. Exposure during training was either contingent (upon chick distress vocalizations), vicarious (exposure to both conspecific distress vocalizations as well as maternal calls sequenced to sound as if contingent upon those vocalizations), or non-contingent. The only condition (both A and B data) to show significant deviations from chance responding during testing was the contingent condition. Evidence of a significant preference, likely due to stimulus enhancement, was however seen in chicks given vicarious exposure to Call A. Additional data exploring this finding will be presented as well.
9:41 PM	Daniel I. Brooks (University of Iowa) & Robert G. Cook (Tufts University)
	The Discrimination of Complex Auditory Stimuli by Pigeons
	In order to better understand the categorization of complex auditory stimuli, three pigeons were trained in a go/no-go task that examined the discrimination of a major chord (S+) from four alternative chord types (S-). The four chord types were chosen for their perceptual distance (one note) from the major chord (C Major). Two chords were chosen for one note manipulations of the third (Minor Chord, Sus4), and two were chosen for one note manipulations of the fifth (Augmented, Flat 5th). Results indicate that pigeons can discriminate these complex stimuli, and do so on the basis of both relational and absolute factors. The data also provide evidence for enhanced discrimination of manipulations of the fifth, suggesting accordance with human research on perceptual similarity within musical structure.
	Thursday Afternoon
12:00 PM	Category and Concept Learning (Chair - Roger Thompson)
12:00 PM	Michael E. Young (Southern Illinois University at Carbondale) & Edward A. Wasserman (University of Iowa) A Theory of Variability Discrimination:  Finding Differences Visual variability discrimination requires an observer to categorize visual collections of items based on the variability in the collection.  We present a theory of visual variability discrimination that aggregates localized differences between nearby items to mediate the discrimination.  This Finding Differences Model was compared to a previously posited Positional Entropy Model across various data sets involving people and pigeons.  Previously published data sets were supplemented with three new experiments, two of which involved arrays comprising items with systematic, quantitative differences.  Although both theories provide strong and similar fits for the published data sets, only the Finding Differences Model is applicable to investigations involving quantitative item differences, providing equally good fits in these new studies.
	A Theory of Variability Discrimination:  Finding Differences
	Visual variability discrimination requires an observer to categorize visual collections of items based on the variability in the collection.  We present a theory of visual variability discrimination that aggregates localized differences between nearby items to mediate the discrimination.  This Finding Differences Model was compared to a previously posited Positional Entropy Model across various data sets involving people and pigeons.  Previously published data sets were supplemented with three new experiments, two of which involved arrays comprising items with systematic, quantitative differences.  Although both theories provide strong and similar fits for the published data sets, only the Finding Differences Model is applicable to investigations involving quantitative item differences, providing equally good fits in these new studies.
12:24 PM	Yasuo Nagasaka & Edward A. Wasserman (University of Iowa)
	Concept Discrimination with the Reassignment Paradigm in Pigeons
	Lea (1984) proposed a series of empirical tests for divulging concept discrimination in animals.  We studied pigeons’ concept discrimination using Lea’s reassignment paradigm.  In the training phase, pigeons were taught to peck one of four diagonally placed keys in response to five depth orientations of four geons.  In the reassignment phase, one view of each geon was reassigned to four new keys.  In the testing phase, both reassigned and nonreassigned stimuli were shown with the keys used in the reassignment phase. The results disclosed no transfer of the discrimination to the nonreassigned stimuli.  Because this outcome may have been peculiar to our use of geometric stimuli, we are now conducting a companion experiment with more naturalistic stimuli—photos of people, flowers, cars, and chairs.  We will compare and contrast the results with both kinds of stimuli and relate these findings to Lea’s proposed methods of revealing concept discrimination in animals.
12:38 PM	Robert G. Cook (Tufts Univ.) & J. David Smith (St.Univ. of New York at Buffalo)
	Stages of Abstraction and Exemplar Memorization in Pigeon Category Learning
	We examined category learning in pigeons and humans across acquisition. Five birds and eight humans learned six-dimensional perceptual categories constructed to include prototypes, typical items, and exceptions. Early on in both species, prototype performance improved and exception performance correspondingly worsened, indicating an initial mastery of the categories’ general structure. Later on, exception performance improved selectively, indicating exception-item resolution and exemplar memorization. The results suggest a psychological transition in pigeon category learning from abstraction- to exemplar-based processing similar to that previously found in humans.
12:52 PM	Tamo Nakamura, Anthony A. Wright (University of Texas Health Science Center at Houston), & Jeffrey S. Katz (Auburn University)
	Same/Different Abstract-Concept Learning by Pigeons: Relational Learning, Item-Specific Learning, and the Generalization Hypothesis
	Pigeons were trained in a two-item same/different task with either 8 or 32 picture items. The generalization-from-item-pairs hypothesis predicts that pigeons trained with 32 items would take 16 times longer to learn the task compared to those trained with 8 items because the correct response needs to be learned for each pair of items (1024 pairs vs. 64 pairs). Contrary to this prediction, the numbers of trials required to reach the criterion for the 32-item group was the same as the 8-item group. The reason that the 32-item group learned as rapidly as the 8-item group must be that they were basing their decision (in part) on the relationship between the two items presented in each trial – the same/different abstract concept. This conclusion was supported by better transfer for the 32-item group than the 8-item group. This rapid learning and partial concept learning by the 32-item group is evidence against the generalization-from-item-pairs hypothesis.
1:06 PM	Andrea J. Frank & Edward A. Wasserman (University of Iowa)
	Pigeons Process Both Items and Relations in Multi-Element Visual Arrays
	Stimulus control can be quite specific and limited to the particular items that are presented, or, in the case of two or more items, stimulus control can be more general and involve the relations between or among the items.  Here, we gave pigeons either icon set (Set 1 versus Set 2) or relation (Same versus Different) report keys after showing them a display containing 12 visual icons that were either the same as or different from one another.  This method forced the pigeons to attend to both icon set and relation on every trial.  After the pigeons attained 80% correct on both tasks, we showed them displays containing 2, 4, 6, 8, 10, 14, 16, 18, 20, 22, and 24 icons followed by the “set” or “relation” report keys.  Pigeons’ ability to flexibly report set or relation with smaller and larger numbers of items will be discussed.
1:13 PM	Jeffrey S. Katz (Auburn University), Anthony A. Wright, (University of Texas Health Science Center at Houston), & Bradley R. Sturz (Auburn University)
	Same/Different Concept Learning: Familiarity and Emergent Features
	Pigeons initially learned a simultaneous two-item same/different concept by responding to a sample picture before selecting same or different to another picture. This procedure has been argued to have a familiarity component that controlled performance (e.g., Premack, 1983). Familiarity of the sample item was manipulated by decreasing the observing response to zero and retesting for concept learning. Pigeons again showed full abstract-concept learning (transfer = baseline). This simultaneous procedure has been argued to have low-level emergent features that controlled performance (e.g., Mackintosh, 2000). Such features were eliminated by changing the procedure to a 1-item list memory task with delays up to 30 seconds. Pigeons again showed full abstract-concept learning. Taken together, these findings indicated pigeons learned the same/different abstract concept based on relational processing between the pictures.
1:27 PM	Anthony A. Wright, Tamo Nakamura, Jacquelyne J. Rivera (University of Texas Health Science Center at Houston), & Jeffrey S. Katz (Auburn University) The hypothesis that generalization from the training stimuli accounts for transfer and concept learning in the same/different task was tested using human similarity ratings. Similarity was rated on a five point scale in terms of shared features for each of the unique 90 transfer stimuli tested with monkeys and pigeons following successive acquisitions with sets of 8, 32, 64, and 128 stimuli. The generalization hypothesis predicts that transfer will vary as a function of the degree of similarity between the transfer and training stimuli. Contrary to the generalization hypothesis, pigeon and monkey transfer was unrelated to the degree of similarity as rated by humans.
	The hypothesis that generalization from the training stimuli accounts for transfer and concept learning in the same/different task was tested using human similarity ratings. Similarity was rated on a five point scale in terms of shared features for each of the unique 90 transfer stimuli tested with monkeys and pigeons following successive acquisitions with sets of 8, 32, 64, and 128 stimuli. The generalization hypothesis predicts that transfer will vary as a function of the degree of similarity between the transfer and training stimuli. Contrary to the generalization hypothesis, pigeon and monkey transfer was unrelated to the degree of similarity as rated by humans.
1:41 PM	NSF Grant Writing (Chair - Bob Cook)
1:41 PM	Jerry O. Wolff (National Science Foundation)
	Competitive Grant Writing for NSF
	I will provide an outline and discuss the critical content and approach for writing a competitive grant proposal for NSF.  Those areas that are the most critical in making a grant competitive are the theoretical construct, testing alternative hypotheses, providing an explanation for interpretation of results, be truly integrative, and apply to the broadest audience in the field.  Common problems encountered in proposals that preclude receiving an award are application of the model species to theory and other taxa, “unnaturalness” of the test arena, too proximate without ultimate (evolutionary) consideration, avoiding the “so what” question, and narrowness of the question with regard to appeal to the greater field of animal behavior.  Successful proposals must have a “wow” factor, be original and creative, reject alternative hypotheses, have an “endpoint” (reach a conclusion), advance the field, be integrative, and contribute significantly to the broadest field of behaviorists.
1:55 PM	Snack Break
2:40 PM	Associative and Memorial Processes (Chair - Jeff Katz)
2:40 PM	Mark E. Bouton, Ceyhun Sunsay, Ana Garcia-Gutierrez, & Michael C. Hendrix (University of Vermont)
	Understanding Intertrial Interval Effects
	A series of experiments isolated at least three mechanisms explaining why Pavlovian learning is better with spaced trials than with massed trials: (1.) recent presentations of the conditioning events “primes” them in short-term memory, making them less surprising (e.g., Wagner, 1981); (2.) contextual conditioning that blocks learning with massed trials is extinguished with longer intertrial intervals; and (3.) spaced trials engage a third process (the “X Factor”) that allows better conditioning even when the other two factors are controlled.  Recent experiments in which ITI signaled whether or not the next CS would be reinforced uncovered surprising asymmetries in how rats discriminated short from long ITIs that are consistent with the idea that time is coded as a series of hypothetical stimuli (A-B-C-D). If early cues (e.g., A) are more salient than later cues (e.g., C), then other aspects of the results, including the “X Factor,” may be explained.
3:04 PM	Nina Tarner (Sacred Heart University) & Zoe Warwick (University of Maryland, Baltimore County) Environmental Cues That Signal a High Calorie Meal Enhance Satiety An arbitrary stimulus (CS) that reliably precedes food presentation can come to elicit feeding. Arbitrary stimuli have also been shown to activate cephalic responses (e.g., gastric acid secretion, increased insulin levels, etc.) The magnitude of postprandial satiety is influenced by a variety of factors such as the macronutrient profile of the meal and the flavor-cued expectancies of the meal’s satiating effects. What is not clear is whether the satiety effect of the food is affected by the conditioned response. The present experiment investigated the impact of environmental cues on the magnitude of satiety produced by a nutritive preload. Both oral and IG preloads were evaluated.
	Environmental Cues That Signal a High Calorie Meal Enhance Satiety
	An arbitrary stimulus (CS) that reliably precedes food presentation can come to elicit feeding. Arbitrary stimuli have also been shown to activate cephalic responses (e.g., gastric acid secretion, increased insulin levels, etc.) The magnitude of postprandial satiety is influenced by a variety of factors such as the macronutrient profile of the meal and the flavor-cued expectancies of the meal’s satiating effects. What is not clear is whether the satiety effect of the food is affected by the conditioned response. The present experiment investigated the impact of environmental cues on the magnitude of satiety produced by a nutritive preload. Both oral and IG preloads were evaluated.
3:18 PM	Gonzalo P. Urcelay & Ralph R. Miller (SUNY-Binghamton)
	Spacing Extinction Trials Attenuates Renewal and Spontaneous Recovery.
	Two experiments investigated the effect of varying the intertrial interval among extinction trials while keeping the session duration constant. In the first experiment we used an ABA renewal preparation and observed that massed trials resulted in renewal of responding whereas spaced trials attenuated renewal. In the second experiment we used a similar approach but instead of changing the physical context at the time of testing, we interposed a retention interval (change in temporal context). Results showed that responding after massed extinction trials was recovered after a retention interval and this effect was reduced if extinction trials were spaced. These results suggest that extinction is more effective when the extinction trials are spaced in time rather than massed. While the benefits of spaced trials are small when there is no contextual change from extinction to testing, a change in context following extinction has detrimental consequences for extinction when the trials are massed rather than spaced.
3:25 PM	Ana Martins & E.J.Capaldi (Purdue University)
	Rapid Reacquisition in Pavlovian Learning: The role of memory
	Rapid reacquisition following extinction in prior Pavlovian studies has been attributed to the CS retaining its excitatory capacity. In the instrumental area, the Sequential Model attributes rapid reacquisition to memories of reward and nonreward events retaining excitatory capacity despite extinction. Two Pavlovian investigations are reported, in which rapid reacquisition occurred, that could only be attributed to reward produced memories retaining excitatory capacity in extinction. In these experiments the CS was presented in every trial and what changed was the delivery, or not, of the UCS. This is not to suggest that the original interpretation applied to the earlier Pavlovian studies is incorrect. The CS may indeed retain excitatory capacity in extinction, as suggested, but this has yet to be demonstrated under conditions that cannot be explained by the memory view.
3:37 PM	Timing (Chair - Jon Crystal)
3:37 PM	Angelo Santi, Dwayne Keough & Patrick Van Rooyen (Wilfrid Laurier University)
	Differences in the Perception of Empty and Filled Time Intervals by Pigeons
	Pigeons were trained in a within-subjects design to discriminate empty intervals (2-s and 8-s bound by two 1-s visual markers) and filled intervals (2-s and 8-s of a continuous visual signal). The two types of intervals were signaled by different visual stimuli and they required responses to different sets of comparison stimuli. Empty intervals were judged longer than filled intervals. The difference was not a result of pigeons timing the empty interval markers. This timing difference reliably occurs when there is no ambient illumination present during test sessions, but it can be eliminated if ambient illumination is present. Different explanations of the timing difference between empty and filled intervals will be discussed.
4:01 PM	Catalin V. Buhusi & Warren H. Meck (Duke University)
	Time Sharing in Rats
	The Time-Sharing and Instructional-Ambiguity hypotheses were evaluated in a peak-interval procedure with gaps and distracters. The Instructional-Ambiguity hypothesis predicts that subjects time through distracters and delay responding after gaps. The Time-Sharing hypothesis assumes that the timer shares attentional/memory resources with other processes. According to the Time-Sharing hypothesis both gaps and distracters delay timing by decreasing the attentional and memory resources allocated for timing, thus reducing the ability of the timer to maintain the pre-gap duration in working memory.  Response functions were displaced by both gaps and distracters in accord with the Time-Sharing hypothesis. Computer simulations with a Time-Sharing model successfully matched these data and related results.
4:25 PM	Marcelo Caetano (Brown University)
	Training Temporal Discriminations in Rats: Memorization or Relearning?
	Rats can be trained to discriminate between time intervals when different intervals are signaled by different stimuli. This study paired three intervals (30, 60 and 120 s) with a single stimulus or with three different stimuli. The stimulus-interval combinations were trained either simultaneously (different combinations presented in the same session) or in blocks (one combination per session). When the stimulus-interval combinations were trained simultaneously with different stimuli signaling different intervals, the rats memorized the stimulus-interval combinations; when the stimulus-interval combinations were trained in blocks, even after extensive training, the rats relearned the stimulus-interval combinations on each session. An evaluation of possible cues used for prediction of food availability indicated the possibility that daily relearning of the interval was the default strategy, and memorization of the stimulus-interval combination occurred only when the default strategy was ineffective.
4:32 PM	Paulo Guilhardi (Brown University)
	Maintenance of Learning During Extensive Extinction
	The memory of rats for interval durations is maintained after extensive extinction when the conditions of original acquisition and extinction are the same. The goal was to determine whether the memory for interval durations is also preserved when the conditions during extinction are different from those of original acquisition. Twenty-four rats were trained on fixed-intervals 30-, 60-, and 120-s signaled by noise, light, and clicker. The stimulus-interval combinations (A) during acquisition were either maintained for 12 rats or changed (B) for the other 12 rats during extensive extinction. Following extinction, half of each group had reacquisition of the combinations A, and the other half B. The rats learned new stimulus-interval combinations during extinction, but the new learning did not eliminate the memory of the original combinations shown in the reacquisition transfer test. The restoration of the original context revealed the preservation of the original learning.
4:46 PM	Linlin Yi (Brown University) Pattern and Rate of Responding: An Explicit Solution A three-phase experiment (acquisition, extinction and reacquisition) with three fixed intervals (FI 30, 60 and 120 s) was conducted to explore the effects of extinction on response pattern and response rate in rats. The response rate in extinction was much lower than the rate in acquisition and reacquisition, but the response patterns in the three phases were very similar. This suggests that extinction affects response rate, not response pattern. The separation of pattern and rate is well demonstrated by the explicit solution of packet theory, which contains two independent memories that corresponds to pattern and rate separately.
	Pattern and Rate of Responding: An Explicit Solution
	A three-phase experiment (acquisition, extinction and reacquisition) with three fixed intervals (FI 30, 60 and 120 s) was conducted to explore the effects of extinction on response pattern and response rate in rats. The response rate in extinction was much lower than the rate in acquisition and reacquisition, but the response patterns in the three phases were very similar. This suggests that extinction affects response rate, not response pattern. The separation of pattern and rate is well demonstrated by the explicit solution of packet theory, which contains two independent memories that corresponds to pattern and rate separately.
4:53 PM	Mika Macinnis (Brown University)
	Automatic Identification of Adjunctive Behaviors in Rats
	In a standard operant chamber a rat exhibits a wide range of behaviors, although only a few are automatically recorded for analysis (e.g. lever press, head entry).  Other methods of measurement, including digital video analysis, have been used to record gross measures such as distance traveled or trajectory.  Additional behaviors the rats exhibit, sometimes referred to as adjunctive behaviors, must be hand coded by an observer (e.g. exploring, grooming).  An automatic method for behavior identification would make it possible to develop a model of behavior that encompasses the full repertoire of the rats’ behavior.  This project proposes a method for the automatic identification of adjunctive behaviors in rats, in a simple operant environment.
5:00 PM	Sara Cordes (Duke University), Adam King (Fairfield Unversity), & C. R. Gallistel (Rutgers University) Subtraction in the Mouse: Time Left Revisited Evidence suggests that both rats and pigeons can respond according to the difference in two temporal durations (Time Left: Gibbon & Church, 1981) and two numerical values (Number Left: Brannon, Wusthoff, Gallistel, & Gibbon, 2001). The neural basis for these nonverbal computations is an open question although it has been suggested that genetics may provide an answer. Using a modified Time left procedure, the current study provides the first investigation into temporal subtraction in the mouse, a species in which the genetic code is available. Results of two experiments reveal that mice responded as a function of the difference between two standard latencies, suggesting that like rats and pigeons, mice can perform online subtraction. These results open the door for investigations of the cellular and molecular basis of nonverbal computations.
	Subtraction in the Mouse: Time Left Revisited
	Evidence suggests that both rats and pigeons can respond according to the difference in two temporal durations (Time Left: Gibbon & Church, 1981) and two numerical values (Number Left: Brannon, Wusthoff, Gallistel, & Gibbon, 2001). The neural basis for these nonverbal computations is an open question although it has been suggested that genetics may provide an answer. Using a modified Time left procedure, the current study provides the first investigation into temporal subtraction in the mouse, a species in which the genetic code is available. Results of two experiments reveal that mice responded as a function of the difference between two standard latencies, suggesting that like rats and pigeons, mice can perform online subtraction. These results open the door for investigations of the cellular and molecular basis of nonverbal computations.
8:00 PM	Poster Session I (8:00 - 10:00)
	See Poster Abstracts Starting on Page 22
	Poster Presenters: Please set up your posters between 7:30 and 8:00
	Friday Afternoon
12:00 PM	Visual Processing - Pigeon Models (Chair - Robert Cook)
12:00 PM	Donald Blough (Brown University)
	Do Hard Decisions Take Longer, and If So, Why?
	A red spot appeared on each trial of a go/no-go discrimination with pigeons.  A particular shade of red was the S+, and a different shade of red was the S-.  If during a block of sessions the S- was quite similar to the S+,   responding was slower to the S+ than it was when, in other blocks of sessions,  the S- was less similar to the S+.    This result is consistent with a simple signal detection model that includes plausible assumptions about response speed and response bias.  But harder to account for are the findings that responses to S+ fastest of all when the S- and the S+ were identical, and also that a difficult red discrimination also slowed responses to a green S+ that appeared in the same sessions.
12:14 PM	Kevin Beale, Angie Koban, & Robert Cook (Tufts University)
	Rate Classification by Pigeons
	This research examined whether pigeons can learn motion categories, and more specifically, whether they can classify speed of motion. Pigeons (n=5) were trained to classify forty left and right rotating objects by their fast or slow rate of movement.  Subjects easily learned this rate discrimination and transferred this property to novel objects, rotations and type of movements. These findings suggest pigeons may be able utilize abstract categorization in defining some types of motion.
12:21 PM	Angie Koban & Robert Cook (Tufts University)
	Tests of Motion and Identity Discrimination with Dynamic Object Stimuli.
	Four pigeons were tested in a motion and identity categorization task. Object stimuli were composed of 40 different objects.  In phase 1, the right versus left rotation of the objects had to be categorized by the pigeons in a go/no-go procedure. After extended training and several procedural changes, all pigeons failed to learn this motion categorization task. In phase 2, the pigeons were found to be able to learn this left/right discrimination with a single object. In phase 3, the 40 objects had to classified by their identity with half of the 40 objects being reinforced and the other half not reinforced in a go/no-go procedure. The pigeons were successful with this multiple object task.  The results showed that pigeons can discriminate rotation with a single object and can learn to recognize up to 40 objects, but the number of objects interferes with discriminating directional rotation as a class. Implications of these results for our understanding of motion perception, classification and object recognition will be discussed.
12:28 PM	Marcia Spetch & Alinda Friedman (University of Alberta)
	Pigeons See Correspondence Between Objects and Pictures
	Pictorial stimuli are used extensively in studies of avian cognition, yet attempts to show that birds recognize the correspondence between static pictures and the objects they depict have produced mixed results.  We trained pigeons to discriminate between two identically-colored but differently shaped 3-D objects seen directly or as pictures and we found clear transfer of learning in both directions between the actual objects and their pictures. Transfer was also seen when pigeons were trained with multiple views of the 3-D objects and transferred to novel views seen in pictures. Because the transfer could not be based on two-dimensional cues such as color, we believe that our results provide the clearest evidence yet that pigeons can recognize the correspondence between objects and pictures.
12:35 PM	A. Kieres, T.B. Patton, J. VandenBosche, T. Shimizu, & R.G Cook (Tufts Univ. & USF)
	Functional Segregation Within the Entopallium in Pigeons
	The avian visual pathway is thought to be organized anatomically, physiologically and functionally in a parallel fashion. Here we aimed to determine if there is functional specificity between the anterior and posterior entopallium, by employing a variety of visual discrimination tasks. Bilateral lesions showed a clear distinction between the function of these two areas. Lesions to the anterior entopallium produced selective impairment in the birds’ ability to perform a color/shape discrimination task, while posterior lesions selectively impaired performance on a motion discrimination task. These results point to possible similarities in visual processing in pigeons and primates, namely through parallel processing visual channels.
12:42 PM	#REF!
12:45 PM	Social, Symbolic, and Cognitive Processes (Chair - Karen Hollis)
12:45 PM	Jennifer Vonk (University of Southern Mississippi), Daniel J. Povinelli (University of Louisiana at Lafayette) & Joan B. Silk (University of California, Los Angeles)
	Chimpanzees do not Donate Food to Unrelated Group Members
	Humans exhibit a wide range of prosocial behaviors, even under conditions in which their kind acts can not be reciprocated. It is possible that these other-regarding preferences are unique to humans. If they are not, we might expect evidence for similar sentiments in our closest living relatives, the other great apes. Chimpanzees in particular might be expected to show concern for the welfare of others given their proclivity towards forming strong social bonds and engaging in cooperative behaviors such as hunting. In two experiments using slightly different methodologies, seven unrelated chimpanzees that have been part of a stable social group for more than 15 years did not opt to deliver food rewards to group members when they could do so at no cost to themselves. Instead, they were just as likely to choose an option that delivered food rewards only to themselves and not to their group-mates.
12:59 PM	Francys Subiaul, Jennifer Vonk, Jochen Barth, Sanae Okamoto-Barth, & Daniel Povinelli (University of Louisiana at Lafayette)
	Chimpanzees Infer the Reputation of Strangers
	Several studies in the economics of information and social psychology have demonstrated that humans habitually attribute specific characteristics to unknown individuals. Assigning reputations to others appears to be an important feature of cooperative endeavors and necessary to solve the “tragedy of the commons.” While various studies have reported how humans make reputational judgments, none have explored whether non-humans make similar judgments based entirely on observations of third party interactions. Here we present data which demonstrates that, when presented with two strangers, chimpanzees prefer to beg from an individual who gave food, versus an individual who refused to give food, to a third party [t(3) = 5.25, p < .01, one-tailed]. Though cooperation among strangers is rare in the primate order, the result reported here suggests that as with humans, reputation may be an available mechanism for mediating pro-social behaviors among chimpanzees.
1:13 PM	Thomas Bugnyar (University of Vienna) & Bernd Heinrich (University of Vermont)
	Knower-Guesser Differentiation in Ravens
	Human social behavior is influenced by attributing mental states to others. It is debated whether such skills play a role in the behavior of non-human animals. We here used the mutually antagonistic interactions occurring between food-storing ravens and conspecific pilferers to test for the possibility of knowledge attribution in birds. Since pilfer success in ravens depends on memory of observed caches, we manipulated the view of birds at caching, thereby designing competitors who were either knowledgeable or ignorant of cache location, and then tested the responses of both storers and pilferers to those competitors at recovery. We show that ravens modify their cache protection and pilfer tactics not simply in response to the behavior of competitors but on the recognition of individuals that could or could not see the caching. Our results suggest that ravens know that obstacles can obstruct the view of others and that this affects pilfering.
1:27 PM	Wendi Fellner (The Living Seas, Epcot) & Heidi E. Harley (New College of Florida & The Living Seas, Epcot)
	Dolphin Vocal Responses to Acoustic Stimuli
	Each dolphin produces a stereotyped whistle, a signature whistle, with a unique frequency contour. The frequency contour is stable, but the whistles vary in terms of absolute frequency, duration, and amplitude. It is not clear what influences those changes. In this study, the vocalizations of an adult male bottlenose dolphin at Epcot’s Living Seas were recorded during a behavioral audiogram. For the audiogram, the dolphin was trained to whistle in response to a pulsed tone and to remain silent in its absence. Although any whistle was acceptable, the dolphin produced a stereotyped whistle to the pulsed tone stimuli used in the audiogram. The mean peak frequency of the stereotyped whistle increased when the frequency of the acoustic stimuli increased. The dolphin also produced a different category of vocalization after the flat-frequency-contour whistle used as the secondary reinforcer.  These data suggest that dolphins vary their signature whistles based on their acoustic environment.
1:34 PM	Ellen E. Furlong & Sarah T. Boysen (The Ohio State University)
	The Effects of Enculturation on Chimpanzees’ Understanding of Visual Attention
	Visual attention has been the focus of many recent studies of non-human primate theory of mind.  Povinelli and colleagues (1996) conducted a task examining 7 non-enculturated chimpanzees’ understanding of the importance of posture, the face and the eyes in visual attention.  Povinelli’s chimpanzees failed most of this task, passing only posture conditions.  We found that our group of 9 enculturated chimpanzees at The Ohio State University performed as if they understood visual attention when presented with the same task, passing all three conditions (posture, face and eyes).  In order to determine whether this discrepancy was due to the effects of enculturation, we tested a third group of semi-enculturated subjects using the same paradigm.  The semi-enculturated chimpanzees performed intermediately between the two other groups, succeeding on the posture and face conditions, but failing the eyes condition.  Our results indicate that enculturation may affect the extent to which chimpanzees understand visual attention.
1:41 PM	Anna M. Yocom, Ellen E. Furlong, & Sarah T. Boysen (The Ohio State University)
	Discrimination of Relevant and Irrelevant Problem-Solving Features by Capuchin Monkeys
	We tested three capuchin monkeys on a “means-end” problem solving task first used by Hauser and colleagues (1999) to determine whether tamarin monkeys generalize a learned solution to the task across functionally irrelevant changes.  The task required the monkeys to pull one of two cloths; one with a piece of food on the surface or one with a piece of food next to the cloth surface. The capuchins showed evidence of generalizing their knowledge across such irrelevant variations as reward size, color and location, as well as cloth shape and color, indicating that the capuchins based their responses on the functionality of the cloth as the means to access the food.  These results are in contrast to a similar task used by Povinelli and colleagues (2000) with seven chimpanzees who failed most of the conditions presented to them.
1:50 PM	Spatial Geometrics and Search (Chair -Jerry Cohen )
1:50 PM	Debbie M. Kelly (University of Saskatchewan) & Walter F. Bischof (University of Alberta)
	Influence of Features and Geometry on Goal-Directed Search Behavior in Virtual Environments
	Successful navigation may be achieved using many different mechanisms including guidance based on featural and geometric information.  In this study, we investigated how the environmental shape and the featural properties influenced goal-directed search behavior by adults.  Groups of participants were presented with distinctive featural cues, as either discrete objects or as surface properties, in a virtual environment.  Furthermore, these featural cues could either be contiguous with the geometric information or not.  Transformation tests showed that contiguity of the two cues led to higher accuracy in the localization of a hidden target position.  However, for women, the featural cues could overshadow the geometric cues, if the features were presented as discrete objects. Our results suggest that adults use a flexible strategy for encoding spatial information.
2:04 PM	Sylvain Fiset (Université de Moncton in Edmundston) & Martine Perreault (University of Ottawa)
	Do Domestic Dogs Understand Invisible Displacement of Object in an Opaque Tunnel?
	We tested whether dogs understand invisible displacement of objects through an opaque tunnel. In Experiment 1 to 3, a tunnel was fixed horizontally over six opaque boxes and in Experiment 4, it was placed diagonally. On control trials, the tunnel was transparent and connected with a box. On testing trials, the tunnel was transparent at one end and opaque at the other end where it connected with a box. A ball was pushed through the tunnel and the dog’s task was to open one box to find the object. On control trials, the dogs easily succeeded the task. On testing trials, the dogs mostly failed in all four experiments and predominantly tended to search inside the box adjacent to the disappearing location of the object inside the opaque tunnel. These results suggest that dogs do not understand that an object can move through an opaque tunnel.
2:18 PM	Steven W. Badelt, Michael A. Parenteau, Kenneth J. Leising, & Aaron P. Blaisdell (UCLA)
	ARENA: Automated Remote Environmental Navigation Apparatus
	We have developed a wireless device which automates open field experiments.  The system eliminates the time required for manual setup of each trial and reduces experimenter inconsistencies between trials.  Our new open field is composed of independently-addressable, remotely-controlled units (up to 240) which can be arbitrarily arranged on any surface within a 25+ m. radius.  Each of these units may be configured to serve one of two functions. As a navigational beacon, a unit is configured to illuminate one of several shapes in any of 16 million combinations of luminance and color.  Each unit can alternatively serve as a response device by providing a key that is illuminated at one of 256 levels.  Units of either type are linked to a coordinating computer, which can establish arbitrary links between devices to test simple forms of learning or complex cognition in either pigeons or rats.
2:25 PM	Emily Gray & Marcia Spetch (University of Alberta)
	Pigeons Encode Absolute Distance but Relational Direction from Landmarks and Walls
	Many recent studies have examined animals’ abilities to use absolute or relational distances to find a hidden goal.  In general, when trained with an array of landmarks, expansion tests reveal that the default strategy of most animals is to search at an absolute distance from one or more landmarks.  In contrast, when trained in enclosures that block external cues, animals often search on the basis of the relationship between walls of the enclosure.  In the present study, pigeons were trained to find the center of either an array of landmarks or a set of short walls that did not block external cues.  Expansion tests showed that both groups of pigeons tended to use an absolute distance strategy.  However, on rotational tests, pigeons continued to search in the center of the array, suggesting that direction was learned in relation to array.
2:32 PM	W. David Stahlman (UCLA), Seth Roberts (UC-Berkeley), & Aaron P. Blaisdell (UCLA)
	Response Rate and Behavioral Variability of Operant Screen-Pecking in Pigeons Over a Continuum of Reward Probability
	Gharib, Gade, and Roberts (2004) demonstrated a direct relationship between reward probability and variability in bar press duration in an operant task in rats. We investigated the relationship between reinforcement likelihood and operant variability in the spatial domain. Pigeons were trained to peck at a set of colored discriminative stimuli; each stimulus signaled a different probability of reward.  We found that spatial variability of responses increased with decreasing probability of reward, but only for pecks outside the discriminative stimulus. Furthermore, peck rates within the stimulus did not differ across stimuli, but higher rates were observed to stimuli signaling a higher likelihood of reward. It is difficult to reconcile these results with the law of effect.
2:39 PM	Verner P. Bingman & Daniele Nardi (Bowling Green State University) The Avian Hippocampus: Space or Spatial Salience? The avian hippocampus has been proposed to play a crucial role in memory representations of space that guide goal recognition and navigation, including the geometric coding of space. However, a growing body of evidence suggests that the geometric coding of space in homing pigeons is not blocked by hippocampal lesions, but rather re-assigned a subordinate role with respect to other goal recognition strategies such as the use of feature properties. A shift in the salience of spatial cues is also seen in lateralization studies when damage to the hippocampus of one hemisphere influences the relative salience but not capacity for representing space. As such, many previous findings on impaired spatial behavior in hippocampal lesioned pigeons may actually reflect more a shift in the salience of spatial information rather than lost spatial ability.
	The Avian Hippocampus: Space or Spatial Salience?
	The avian hippocampus has been proposed to play a crucial role in memory representations of space that guide goal recognition and navigation, including the geometric coding of space. However, a growing body of evidence suggests that the geometric coding of space in homing pigeons is not blocked by hippocampal lesions, but rather re-assigned a subordinate role with respect to other goal recognition strategies such as the use of feature properties. A shift in the salience of spatial cues is also seen in lateralization studies when damage to the hippocampus of one hemisphere influences the relative salience but not capacity for representing space. As such, many previous findings on impaired spatial behavior in hippocampal lesioned pigeons may actually reflect more a shift in the salience of spatial information rather than lost spatial ability.
2:53 PM	Izabela Szelest & Debbie M. Kelly (University of Saskatchewan) Pseudo-neglect in Clark’s Nutcrackers? Humans routinely show a slight left-sided visuospatial bias, over-selecting for objects on their left in comparison to their right, when engaging in cancellation tasks.  This asymmetry has been explained by reference to the superiority of the right hemisphere for spatially guided tasks.  Recent research has suggested that birds, like humans, may show a similar left-sided visuospatial bias.  This finding has put into question the role of the corpus callosum in the emergence of spatial asymmetries.  Our study examines whether a left-sided visuospatial bias is also evident in the food-storing Clark’s nutcracker.
	Pseudo-neglect in Clark’s Nutcrackers?
	Humans routinely show a slight left-sided visuospatial bias, over-selecting for objects on their left in comparison to their right, when engaging in cancellation tasks.  This asymmetry has been explained by reference to the superiority of the right hemisphere for spatially guided tasks.  Recent research has suggested that birds, like humans, may show a similar left-sided visuospatial bias.  This finding has put into question the role of the corpus callosum in the emergence of spatial asymmetries.  Our study examines whether a left-sided visuospatial bias is also evident in the food-storing Clark’s nutcracker.
3:00 PM	Snack Break
4:15 PM	In Honor of the Contributions of Russell M. Church
4:15 PM	Introduction - Jon Crystal
4:20 PM	Kimberly Kirkpatrick & Rona Russell (University of York)
	The Role off Temporal Generalization and Categorical Learning in the Bisection Task in Rats
	Two experiments were conducted in which rats were trained to discriminate two or three stimulus durations using a bisection task. When trained to discriminate short vs. long signals, rats demonstrated a standard peak shift effect. Conditions where training involved three signals (short, medium, and long) revealed that rats: (a) discriminated three durations (short, medium, and long) if three distinct responses were available; (b) learned to respond to short and medium signals as a category vs. a long signal; and (c) were unable to learn to respond to short and long signals as a category vs. a medium signal. The results suggest that generalization plays a major role in determining discriminability, but that categorical learning may also contribute.
4:35 PM	Peter R. Killeen & Federico Sanabria (Arizona State University)