Conference on Comparative Cognition 2005 - Poster Presentations
Babb Stephanie J. Babb & Jonathon D. Crystal (University of Georgia)
Discrimination of What, When, and Where in Rats is Not Based on Time of Day
We investigated discrimination of what, when, and where in rats (n=6) on the radial maze, and controlled for time of day. Phase 1 consisted of four choices, one of which contained chocolate. In Phase 2, all eight arms were available. After a short (1 hour) retention interval (RI), the four arms not available in Phase 1 provided food. After a long (25-h) RI, the four arms, plus the chocolate arm, provided food. The rats visited chocolate more after the long RI than after the short RI. Chocolate was then paired with lithium chloride after Phase 1, during the 25-h RI. Following the taste-aversion manipulation, the rats visited chocolate after the long RI less often than before LiCl. These data demonstrate knowledge of what, when, and where that cannot be based on time of day. The data also suggest flexibility to update memory based on information acquired in a new situation.
Batson Michael C. Hendrix & John Batson (Furman University)
Restrained Honeybees Can Use Odors to Predict the Location of Reward
The classically conditioned proboscis extension reflex (PER) of honeybees typically is studied by pairing an odor (CS) with sucrose (US) reinforcement.  Bees are usually restrained firmly and are able to move only antennae and proboscis, resulting in a conditioned response that is scored as a simple digital event.  In this study, honeybees were loosely restrained, allowing for extensive head movements, and were presented two odors from the same location, each of which was followed by sucrose either to the right or left of the animal.  The animals learned to anticipate the direction where sucrose would appear, as evidenced by head turns (with PER) in the correct direction.  This additional component of the PER may enhance the usefulness of this procedure in the study of invertebrate learning.
Beckmann Joshua Beckmann, Michael Young, & Olga Nikonova (Southern Illinois University @ Carbondale)
Stimulus Dynamics and Duration Judgments
The purpose of the present study was to observe the functional relationship between stimulus dynamics and stimulus duration judgments in humans. Stimulus dynamics were defined by how quickly the sphere rotated on its Y-axis. A bisection task was used to divide stimulus durations into two categories, short and long. Sphere rotation involved three levels: slow, medium, and fast.  Participants' duration judgments were longer the faster the sphere was rotated.  In the second experiment, durations were created using a logarithmic scale. Sphere rotation involved four levels: static, slow, medium, and fast. Participants' duration judgments were again longer the faster the sphere was rotated with the exception of the static condition; participants' duration judgments under the static condition were more similar to judgments under the medium condition. The results indicate that stimulus dynamics influence the point of subjective equality and the slope of the discrimination function.
Boisvert Michael J. Boisvert & David F. Sherry (University of Western Ontario)
Interval Timing in an Invertebrate
Learning to adjust behavior to the temporal structure of the environment is expected to be an important and phylogenetically widespread capacity.  However, interval timing - responding to elapsed temporal durations - is known to occur only in vertebrates.  In two experiments bumble bees (Bombus impatiens) were trained to extend their proboscis through a hole in a chamber wall, with reward available on fixed interval (FI) 12 and 24 second schedules.  Bumble bees showed pauses in responding immediately following reward, with longer pauses occurring on the longer schedule.  Maximum rates of responding occurred at or near the end of the interval.  These results suggest that bumble bees timed the intervals.  To our knowledge these data are the first indication of interval timing ability in an invertebrate.   
Chapman Kate M. Chapman (New College of Florida, Lemur Conservation Foundation) & Heidi E. Harley (New College of Florida)
Multiple Measures of Handedness & Laterality in Three Species of Lemur: Lemur catta, Eulemur mongoz and Eulemur fulvus rufus
Handedness has been considered an indicator of lateralization of hemispheric function in many primate species. Semi-free-ranging lemurs were assessed for hand/limb preference over multiple behavioral measures: (1) reaching for food, (2) foraging, (3) walking/climbing and (4) in a sequential food extraction task. In discrete food presentation, about half of the lemurs preferentially used their right hands, and the other half, their left. In food-related tasks, dominant hand preference increased as posture shifted from a quadrupedal to a bipedal stance.  While foraging, lemurs tended to use the hand closest to the food item; therefore, hand preferences were lacking in this measure. A population-level weak left preference was expressed for leading limb in walking. For the sequential tube task, all subjects exhibited strong right preferences, indicating a left-hemisphere specialization for sequential hand movements. Stability was not found across tasks; individuals did not always express the same lateral bias for each behavioral measure. 
Chenoweth Amber Chenoweth, Melissa D. Muller, Jessica Owens, Zach Moore, Denise P. A. Smith, & Stephen B. Fountain  (Kent State University)
Contextual Coding in Rat Serial Pattern Learning: Serial Position as Context for Temporal Phrasing Cues
In serial behavior, a common finding is that the transitions to new chunks of a serial pattern are more difficult to anticipate than elements within chunks.  Providing “phrasing cues” at chunk boundaries facilitates learning about these transitions in humans and rats.  Past studies have indicated that temporal phrasing cues chunk sequences by overshadowing interitem associations, but that rats also learn about the serial position of chunk boundaries.  To further examine the mechanisms underlying this effect, rats learned a pattern of responses on a circular array of 8 levers with temporal phrasing cues at chunk boundaries.  Two phrasing cue removal manipulations revealed that serial position cues served as occasion setters for the use of temporal cues at chunk boundaries.  These results help clarify how phrasing and serial position cues chunk patterns and suggest an explanation for the nature of sequential learning deficits following hippocampal lesions in rats.
Dally Joanna Dally, Nathan Emery & Nicola Clayton (University of Cambridge)
Cache protection strategies by western scrub-jays (Aphelocoma californica): who is watching and what have they seen?
The first experiment shows that western scrub-jays differentiate between the risks that different observers pose to their caches. In the presence of observers, storers cache predominantly in distant sites, reducing the transfer of information to potential thieves. If unobserved ‘near’ and ‘far’ sites were used equally. During unobserved recovery periods, after being observed by a dominant or subordinate conspecific during caching, storers re-cached items in new sites. However, if the observer was the storer’s partner, or if caching occurred ‘in private’, few items were re-cached. In a second experiment, we found that the jays appeared to be sensitive to what an observer had or had not seen. If the same bird that watched them cache was present at recovery, storers used strategies to protect cache safety. If watched by a different bird, they did not re-cache, thus withholding information as to the location of the remaining caches. 
Emery Nathan D. Emery (University of Cambridge, UK)
Visual Co-orienting With Humans and Conspecifics in Common Marmosets
Gaze following using conspecific or heterospecific cues has been demonstrated in monkeys, however most species tested have a competitive rather than cooperative social system. Visual co-orienting was investigated in 8 common marmosets. In Experiment 1, a human experimenter provided Eye, Head or Point cues to subjects, who oriented more frequently towards the attended location after Point cues and looked longer at the correct object after all cues. Subjects also used social information from their partner. In Experiment 2, a Tap cue was included. Two subjects oriented more frequently and looked longer at the correct object after Tap cues. Latency to respond was shorter to the correct object and percentage of trials where the correct object was inspected first was higher than the incorrect object after all cues. Marmosets use social cues to orient towards an object attended to by a heterospecifc, however these results suggest reflexive orienting rather than understanding.
Feeney Miranda C. Feeney & William A. Roberts, University of Western Ontario
Impulsivity and Self-control in rats During Bouts of Foraging
Traditional assessments of “impulsivity” observed in animals in self-control experiments focus on two hypotheses: either animals lack the ability to foresee future consequences of choices or they discount uncertain events rather than retain value for rewards subject to delay.  Impulsivity is not considered a strategic, beneficial choice.  Typical self-control experiments require animals to make choices about accessing food, essentially foraging, but the research most frequently involves either operant chambers, or runway mazes, neither of which can be effectively used to model environmental factors such as patch density and spatial distinction, accessibility of food sources, or availability of shelter and escape routes.  The current research will focus on the phenomenon of impulsivity, in the context of foraging choices, using the radial maze.  Data will be reported from experiments in which rats chose between small immediate rewards on some arms and large delayed rewards on other arms.
Fiset Sylvain Fiset & Nathalie Malenfant (Université de Moncton, campus d'Edmundston)
Searching in the center: Domestic dogs encode absolute distance from edges of a frame
The objective of this study was to determine whether domestic dogs encode absolute or relative distance to locate a ball hidden in the middle of a frame. In Experiment 1, the ball was surrounded by a square frame. In Experiment 2, it was surrounded by a circular frame. During training, the frame was moved about in the room from trial to trial. During testing, training and test trials were mixed. On control tests, the ball was removed and the frame remained at the same location. On expansion tests, the ball was removed and the size of the frame was double. On both experiments, data revealed that on control tests, domestic dogs searched accurately at the center of the frame. On expansion tests, however, dogs searched at the training distance from edges of the frame. This suggests that dogs encode absolute distance from edges of a frame to locate a spatial position.
Flemming Timothy M. Flemming & David A. Washburn (Georgia State University)
Same vs. Different or Uniformity vs. Chaos?: Perception of Abstract Relations by Rhesus Monkeys (Macaca mulatta)
In this study, we investigate the postulation that monkeys are incapable of abstract relational processing (Thompson & Oden, 1996) by presenting five rhesus monkeys with several computerized relational matching paradigms. After failing to meet criterion discriminating among pairs of same and different objects in studies 1-3 of our project, monkeys were presented with rows of several objects in a two-choice discrimination (between rows) paradigm, rather than pairs.  From 8-3 stimuli per row, we see a pattern of slight degrading accuracy that corresponds to the number of objects presented. Our results suggest that monkeys rely, as pigeons do, on entropy in their discrimination of these abstract relations (Wasserman, Frank & Young, 2002). So, to monkeys, it may be more appropriate to use terms such as uniformity and chaos, rather than same and different because it is clear that they do not perceive these constructs in a manner identical to humans.
Fowler Facilitating and Inhibiting “Insightful” behavior in Pigeons (Columba livia)
Catherine M. Fowler and Robert G. Cook (Tufts University)
We looked to expand the current understanding of what cognitive processes underlie the production of novel behaviors. We replicated and extended Epstein,Kirshmit, Lanza, and Rubin's (1984) pigeon experiments looking at Kohler's classic insight task involving the novel combination of boxing pushing and banana pecking. Epstein et al. outlined a successful recipe for the synthesis of these previously learned behaviors; but what remains unclear is exactly how the training procedures are processed by the pigeons. We sought to determine what specific past experiences are required to produce this apparently “insightful” behavior by testing pigeons with both functional and non-functional box alternatives as solutions to the problem.
Frank Andrea J. Frank, Edward A. Wasserman (University of Iowa), and Michael E. Young (Southern Illinois University at Carbondale)
Item and relation control in same-different discrimination
We trained four pigeons on conjoint same-different and item discriminations in which collections of 16 visual icons were either all identical or all nonidentical and comprised icons either from Set 1 or Set 2.  The birds had to peck one of four buttons, each corresponding to one of the four combinations of these two independent factors:  same-different relation and item set.  We tracked control by same-different relation and item set during initial acquisition and during a later phase, in which the number of items was decreased from 16 to 2.  Both same-different relation and item set discriminations were learned to high levels, but control by item set developed faster.  Control by the same-different relation fell as the number of displayed items was decreased, whereas control by item set did not.  This interplay suggests joint control by relation and item information in pigeons’ same-different discrimination behavior.
Gerrard Leslie L. Gerrard, William R. Pfleger, & James C. Denniston (Appalachian State University)
The Renewal Effect: Occasion Setting by Context?
Two experiments using human participants investigated the mechanisms underlying the renewal effect.  Experiment 1 provided negative occasion setting training in which reinforced presentations of CSs A, B, and C (in Context C) were interspersed with nonreinforced A and B presentations in Contexts A and B, respectively.  Testing with CSs A and B in Context A revealed occasion setting and transfer of occasion setting by context.  Experiment 2 provided training analogous to Experiment 1, except that all training was provided phasically (i.e., acquisition followed by extinction).  Testing with CSs A and B in Context A revealed extinction and a prevention of the renewal effect, respectively.  The results of Experiment 2 suggest that the extinction context can function as a negative occasion setter which can transfer its modulatory control to other target CSs.  Results will be discussed in terms of occasion setting and other theories of renewal.
Gerrard Leslie L. Gerrard, William R. Pfleger, & James C. Denniston (Appalachian State University)
The Renewal Effect: Occasion Setting by Context?
Two experiments using human participants investigated the mechanisms underlying the renewal effect.  Experiment 1 provided negative occasion setting training in which reinforced presentations of CSs A, B, and C (in Context C) were interspersed with nonreinforced A and B presentations in Contexts A and B, respectively.  Testing with CSs A and B in Context A revealed occasion setting and transfer of occasion setting by context.  Experiment 2 provided training analogous to Experiment 1, except that all training was provided phasically (i.e., acquisition followed by extinction).  Testing with CSs A and B in Context A revealed extinction and a prevention of the renewal effect, respectively.  The results of Experiment 2 suggest that the extinction context can function as a negative occasion setter which can transfer its modulatory control to other target CSs.  Results will be discussed in terms of occasion setting and other theories of renewal.
Gray Emily R. Gray, Laurie L. Bloomfield, Anne Ferrey, Christopher B. Sturdy & Marcia L. Spetch (University of Alberta)
Mountain Chickadees use of geometric and featural information in a spatial environment
Encoding of the global geometric shape of an enclosed environment seems to be a ubiquitous and predominant means of orienting in humans and various other animals. Twelve wild-caught mountain chickadees (Poecile gambeli) were trained to retrieve a mealworm that was hidden in one corner of rectangular enclosure.  Separate groups of chickadees were trained either with features adjacent or opposite to the correct corner, or without any featural information.  All three groups were then tested with and without the featural information available.  Both feature groups were also tested in conditions that placed the featural and geometric information in conflict. The chickadees did not learn the geometric properties of an enclosure when salient featural information was adjacent to the correct corner during training.  In contrast, chickadees that were trained in the absence of features or with distal features only were able to utilize geometric information.
Harley Heidi E. Harley (New College of FL & Disney's Living Seas), Wendi L. Fellner, Kim Odell, & Erika Putman (Disney's Epcot's Living Seas)
Representation of Acoustic Rhythms by the Bottlenose Dolphin
In a previously reported study of rhythm discrimination, a bottlenose dolphin could discriminate among six different acoustic rhythms with high accuracy (94%).  The dolphin maintained the discrimination across frequency shifts (across two octaves), but his performance across tempo shifts (at intervals from halving to doubling) dropped significantly at most new tempos.  In the current study, the dolphin was exposed to rhythms at a variety of tempos before being tested in a transfer test with unfamiliar tempo-shifted stimuli.  Performance accuracy was significantly better (66% vs. 41%; chance = 17%) after the dolphin had general experience with rhythms presented at different tempos.  To determine the information that the dolphin was using, a comparison of performance accuracy across rhythms at different tempos and an analysis of the dolphin’s reaction times in some contexts were conducted.  The results suggested that the dolphin was using a combination of absolute and relative characteristics of the rhythms.  
Harshaw Chris Harshaw & Robert Lickliter (Florida International University)
Stimulus Contingency and Intersensory Redundancy: Effects on Perinatal Learning 
Previous research on early perceptual learning has generally utilized non-contingent, passive presentation of stimuli to infants. The learning obtained in such studies does not necessarily mirror the course of most learning during early development. The present study presents results demonstrating highly amplified auditory learning following a short (5 minute) contingent presentation of naturalistic auditory stimuli (an individual bobwhite maternal call) in bobwhite quail chicks. Chicks receiving contingent presentation of the maternal call based on the production of their own vocalizations at 24 hr following hatching remembered and preferred this familiar call over a novel maternal call when tested at 48 hr of age, whereas chicks receiving 5 min of non-contingent presentation did not. Data on the interaction of this contingency effect with the previously demonstrated effect of intersensory redundancy on early perceptual learning will also be discussed.
Herbranson Walter Herbranson (Whitman College)
Pigeons learn visual categories based on angle of movement, but not angle of orientation
Pigeons learned to categorize visual stimuli presented on a computer monitor.  When categorizing moving objects based on speed and angle of travel, pigeons divided attention across both stimulus dimensions and performed nearly optimally.  When categorizing objects based on size and angle of orientation, pigeons selectively attended to size, even when attention to orientation was required for optimal performance.  Results indicate that categories based on angle can be relatively easy or difficult for pigeons to learn, depending whether angle represents an orientation or a direction of travel.
Hoy Erica A. Hoy, Dorothy M. Fragaszy, Gene Brewer (University of Georgia), Julie Johnson-Pynne (Berry College), & Aeneas Murnane (Emory University)
A Comparison of the Ability of Chimpanzees (Pan troglodytes) and Capuchin Monkeys (Cebus apella) to Solve Two-Dimensional Detour Problems
Five chimpanzees and seven capuchin monkeys were presented with a series of 192 computer mazes.  Three chimpanzees and three capuchins were presented with mazes in order of perceived difficulty, while the remaining subjects were presented with the mazes in random order.  The number of choices within a maze varied between 1-5 and the number of “non-obvious” choices ranged from 0-3.  Non-obvious choices were those in which the incorrect choice appeared to lead more directly to the goal than the correct choice.  Performance was assessed by analyzing the frequency and type of errors subjects made while navigating through the mazes.  Results showed that chimpanzees made fewer errors, corrected their errors more often, and solved more mazes without error than capuchins.  Chimpanzees were unaffected by type of maze presentation (ordered vs. random), while capuchins performed significantly worse in the random condition.  These results suggest inherent strategy differences used by these two species.
Jaime Mark Jaime & Robert Lickliter (Florida International University)
Prenatal Exposure to Temporal Synchrony Affects Postnatal Responsiveness to Spatial Contiguity in Bobwhite Quail Chicks
Evidence derived from neural and behavioral studies of animal infants suggests that temporal synchrony and spatial colocation play a key role in early intersensory development. Little is known about how sensory experience during prenatal development can contribute to postnatal responsiveness to the temporal or spatial contiguity of multisensory stimulation. This study manipulated late prenatal and early postnatal audio-visual experience available to bobwhite quail embryos and hatchlings. Results revealed that embryos exposed to temporally synchronous and spatially colocated bimodal stimulation prior to hatching subsequently preferred spatially colocated audio-visual maternal cues following hatching, despite being denied postnatal visual experience. In contrast, embryos not receiving prenatal synchronous and spatially colocated audio/visual experience failed to show a preference for the spatial contiguity of maternal cues following hatching. These results suggest that prenatal experience with amodal stimulus properties (such as synchrony) can sensitize chicks to other amodal properties in the days following hatching.
Kundey Shannon M.A. Kundey & Laurie R. Santos (Yale University)
Episodic-like memory in capuchins (Cebus apella)
Human memory combines what, when, and where information about personal events, a capacity referred to as episodic memory.  Many have hypothesized that this capacity is unique humans. Here, we explore whether capuchin monkeys can represent episodes in memory.  Capuchins were taught that a preferred degrading food hidden for 10 sec, but not 3 min, is edible and that a less preferred non-degrading food was always edible.  Capuchins then watched as the preferred and less-preferred foods were hidden for 3 min or 10 sec.  Capuchins failed to combine what, when, and where information successfully to choose the non-degrading food after 3 min and the degrading food after 10 sec.  While capuchins failed to combine what, when, and where information, they successfully combined dyads (what/when, when/where, and when/what) of these components.  Capuchins’ limited ability to use this information suggests they may lack episodic memory.
Leighty Katherine A. Leighty (University of Georgia)
A Comparative Analysis of 2D Object Perception in Chimpanzees (Pan troglodytes) and Young Children
I examined human and nonhuman primates’ ability to form complex object representations from 2D images and utilize these representations to work in 3D space.  Adult chimpanzees along with 3 and 4 year-old children served as subjects.  Subjects were presented with a live televised image of a series of hiding tasks.  After viewing an event in 2D, subjects were allowed to search the 3D object for the hidden item.  The titrated testing series required subjects to discriminate between objects using color, form, local features, and relative position, and to discriminate between locations within a single object without using distinctive local features.  Adult chimpanzees and 4 year-old children successfully completed all titrations of the task, whereas three year-old children only made cross-dimensional discriminations using color, form, and local features.  Results imply that adult chimpanzees and 4 year-old children form complex object representations from 2D images and use these representations to guide action.
Markham Rebecca Markham, Robert Lickliter, & Lorraine E. Bahrick (Florida International University)
Intersensory Redundancy Guides Selective Attention During Prenatal Development
Bobwhite quail embryos were exposed to a bobwhite maternal call under several conditions to assess the salience of intersensory redundancy on prenatal learning. An experimental group received redundant bimodal (audio-visual) exposure to the temporal features of a maternal call followed by unimodal (auditory) exposure to the same call. Three control groups received either the same exposure but in the reverse sequence, only unimodal exposure, or only bimodal exposure. Chicks from all groups were tested at 48 hr following hatching for their preference between the familiarized call and a novel bobwhite maternal call. The experimental group showed a significant preference for the familiar call over the novel call, whereas none of the control groups showed a preference. These results suggest that intersensory redundancy can direct attention to temporal properties of bimodal stimulation and this redundancy can educate attention to these temporal properties in subsequent unimodal stimulation where no intersensory redundancy is available.
McClure Erin A. McClure, K. Saulsgiver, & Clive D.L. Wynne (University of Florida)
Effects of d-amphetamine on stimulus control in pigeons exposed to duration discriminations
Two experiments used a matching to sample of durations procedure to examine changes in temporal discrimination evoked by d-amphetamine in pigeons.  The sigmoid functions relating percent of choices of the key reinforced after long duration stimuli to the duration of stimulus presented were fit by a cumulative normal function. In addition to a parameter indexing the left/right position of the sigmoid curves (which indexes temporal perception), this function also provided estimates of the range and slope of the curve (measures of stimulus control).  Results showed that, contrary to many published reports, amphetamine had no effect on the left/right positioning of the sigmoid curve.  There was, however, an effect of amphetamine on stimulus control, as shown by the general flattening of the psychophysical function.
McKenzie Tammy L.B. McKenzie & William A. Roberts (The University of Western Ontario)
What strategies do pigeons use to form categories?
Pigeons received training with two classes of stimuli, houses and dogs. They were reinforced for pecking at one category and not reinforced for pecking at the other and were tested with novel exemplars from each training category and a novel category (flowers). Pigeons transferred the appropriate pecking behavior to the novel exemplars from each training category. However, behavior to the novel category was similar to that displayed to exemplars of dogs. Thus, pigeons responded in one way to exemplars of houses and responded in the opposite manner to all other items by exclusion. In Exp. 2, pigeons received training with a highly variable category and a low variability category. It was predicted that pigeons would categorize exemplars from a novel category into the highly variable category. Pigeon’s did not use this strategy. They treated the novel category like the S- category, regardless of whether it was the high or low variability category.
Muller Melissa D. Muller, Zach Moore, Denise P.A. Smith, & Stephen B. Fountain (Kent State University)
Do Rats Use Rules or Associations in Serial Pattern Learning?
This experiment sought to determine the extent to which rats use rules versus associations to guide behavior in a serial learning task. Rats were trained on one of three highly structured patterns composed of chunks of various lengths: 123-234-345-456-567-678-781-818; 1234-3456-5678-7818; and 12345-45678-78121. The numbers identify the order of correct responses on 8 levers that were arranged in a circular array. Acquisition rates were measured for chunk boundary, within chunk, and violation elements. Rule learning theories predict faster chunk boundary and within chunk acquisition, because violations, by definition, cannot be anticipated by the pattern structure. Associative theories predict that acquisition should depend on cue factors such as discriminability and contingency. Results showed violation element acquisition was faster than within chunk acquisition for all three groups. Additionally, the violation element and chunk boundaries were learned equally fast in all three groups. These results support an associative view of serial pattern learning.
Nakata Ryuzaburo Nakata & Yoshihisa Osada (Rikkyo University)
Can monkeys utilize only their faces to identify each other?
We explored whether squirrel monkey (Saimiri sciureus) can utilize faces of their own species to identify themselves. Face stimuli were squirrel monkey faces and human faces both unfamiliar to two subject monkeys. These monkeys were trained to discriminate between two squirrel monkey faces and between two human faces. After monkeys could acquire a correction level of 80%, they were presented with a pair of new faces. Monkeys could easily discriminate between two human faces, but they had much difficulty in discriminating faces of squirrel monkeys. The results suggest that squirrel monkeys can identify individual humans more easily than in the case of identifying those of their own species. In the natural environment, monkeys may utilize olfactory or auditory cues for identification of individuals of their own species rather than faces. 
Naqshbandi Mariam Naqshbandi & William A. Roberts (The University of Western Ontario)
Cognitive Time Travel in Squirrel Monkeys
Cognitive time travel refers to the ability to remember past events distinguishable in space and time and to anticipate future events.  Previous research has suggested that squirrel monkeys (Saimiri sciureus) may be able to anticipate events at least 15 min into the future.  However, the previous work involved a future goal that was relevant to the current need of the monkeys.  Can squirrel monkeys show evidence of anticipation for the future when two different motivational states exist in a single experiment?  The present experiments tested the Bischof-Kohler hypothesis that non-human animals are bound to the present by their current motivational state.  Two experiments were conducted that involved choice between 4 and 8 peanuts with choice of 4 peanuts yielding a reward satisfying a different motivational drive (water in Experiment 1 and mealworms in Experiment 2) 30 min later.  Proportion of trials on which the smaller quantity was selected was measured.  
Nikonova Olga Nikonova, Michael E. Young, & Joshua S. Beckmann  (Southern Illinois University at Carbondale)
Contingency versus Mechanism in Causal Comparisons
Effects should be contingent on the occurrence of their causes.  It is thus not surprising that events with stronger contingency relations are typically judged to be more causal.  It is also true, however, that judges believe that a causal mechanism should be present before causality can be established.  In the present study, contingency and visible mechanism were placed in direct conflict to determine how weak must a contingency be before participants will favor a perfectly contingent causal candidate without a plausible mechanism over a less contingent causal candidate with a more plausible causal mechanism?  Participants preferred a much weaker (50% predictive) candidate with direct or mediated spatial contiguity over a stronger (100% predictive) candidate that lacked contiguity.
Nolley Eric P. Nolley, Amanda W. Willey, James D. Rowan, and Brian M. Kelley (Bridgewater College)
Assessing adolescence drug exposure on adult serial-pattern learning in rats:  Alcohol and Nicotine
Most current methods for assessing the effects of adolescent drug exposure in animal models are centered on simple behaviors. However, such methods provide no insight regarding impairments of higher cognitive functions. The purpose of this study was to investigate a serial-pattern learning as a method of evaluating drug exposure during adolescence. Rats were injected 5 days a week for 5 weeks with equivalent volumes of 0.3 mg/kg nicotine, 20 % alcohol, or saline based on body weight. After 5 weeks off, all subjects were trained on a violation trill pattern (123 234 345 456 567 678 781 818) for 28 days receiving 5 patterns a day. Although no significant differences were found, a recent study using Prozac found impairments in pattern acquisition. Overall, this methodology can serve as an effective screen for examining the pharmacological and toxicological effects of adolescent drug exposure on higher-cognitive functioning in adulthood.
Olthof Anneke Olthof & Angelo Santi (Wilfrid Laurier University)
The Association of Time Intervals with Symbols by Pigeons Using the Touchscreen: Evidence for Ordinality but not Summation
Previous research has found that pigeons can sum symbols associated with various quantities of food rewards in a Modified Wisconsin General Test Apparatus. The present experiment examines whether pigeons can sum symbols associated with various time intervals in a touchscreen apparatus. Pigeons were initially trained to choose one of two symbols from the set X = {0, 1, 2, 3, 4, and 5}, and were tested with novel pairs, as well as a choice between two sums, each composed of two symbols. In Experiment 1, each symbol was associated with an X-second delay to a fixed duration of food access, while in Experiment 2, each symbol led to X-seconds of food access. The results indicated that although pigeons formed an ordered representation of the symbols, spatial proximity of two symbols on the touchscreen did not induce summation. Instead, pigeons based their decisions on the value of the individual symbols.
Oswald Tasha Oswald and Daniel J. Povinelli (University of Louisiana at Lafayette)
Chimpanzees' Understanding of Suspension Physics 
Seven peer-reared chimpanzees were given a choice task to test their understanding of suspension physics. After training the chimpanzees to successfully hang objects with holes from a dowel, the chimpanzees were presented with 18 different object pairs (one pair, consisting of a correct and incorrect object, per trial), from which they were allowed to choose one object to attempt to hang from the dowel. Only the correct object had features, such as a hook or a hole, which would allow it to successfully suspend from the dowel. Additionally, for nine of the object pairs the incorrect object contained a distracter feature (notch cut into the object).  The subjects chose more accurately on non-distracter trials. Notably, one subject also showed transfer on both distracter and non-distracter trials. These results suggest that chimpanzees may have a rudimentary working knowledge of suspension physics that may be elaborated with experience.
Pizzo Matthew J. Pizzo & Jonathon D. Crystal (University of Georgia)
Rats do not discriminate alternate days
We investigated the ability to discriminate the daily alternation of meal availability (i.e., a 48-hr intermeal interval).  The discrimination could be based on daily alternation, interval timing, or an oscillator with a period greater than 24 hours; the discrimination could not be based on a circadian oscillator.  Rats (n=14) were tested 7 days per week at a constant time of day in operant boxes; a 1-hour meal was available 1 hour after the start of the session on alternate days.  Post-session rations were adjusted to maintain a constant amount of food per day.  Although response rate increased during the first hour of testing, there was no difference in the response rate between food and non-food sessions.  The data suggest that rats do not readily discriminate alternate days.
Rahn Elizabeth J. Rahn, Tyson L. Platt, and Martha Escobar (Auburn University)
Inhibition of delay in appetitive conditioning: Summation and retardation tests
When the delivery of an outcome consistently occurs during the final segments of its signal (i.e., conditioned stimulus, CS), conditioned responding tends to be delayed until the expected time of outcome delivery (i.e., inhibition of delay).  Rescorla (1967) reported that the initial segments of a CS signaling delayed delivery of electric shock became inhibitory, as assessed with summation and retardation tests.  However, little research has been conducted to support this assumption.  Two studies using an appetitive preparation provide summation and retardation tests for the development of conditioned inhibition during the initial segments of a CS that signals delayed delivery of an outcome.
Rattermann Mary Jo Rattermann (University of Indianapolis), Alison Benowitz, Keren Mitchell & Nidhi Suri (Franklin & Marshall College)
Two, Six or Sixteen Icons: Identity is Still Special
Using the methodology developed by Wasserman and Young (Wasserman, Fagot & Young, 2001;Young & Wasserman, 2001) a touchscreen Imac was used to present human adults and 3-year-old children with either a display of 6 identical icons or a display of 6 non-identical icons. Subjects received an equal number of identity and non-identity trials, but were only rewarded for touching the screen in the presence of identity (or non-identity, depending upon condition). As was found in previous research, the adult subjects touched the screen more for identity displays, regardless of whether they were reinforced for responding to identity or non-identity. These findings suggest a predisposition in adults to respond based on identity. Previous research using two and sixteen icon displays also resulted in more identity responses, suggesting that this predisposition is robust to differing numbers of icons. 
Rosati Alexandra G. Rosati, Jeffrey R. Stevens, & Marc D. Hauser (Harvard University)
The effect of Handling Time on Discounting in a New World Primate
Animals often discount or subjectively devalue future rewards. It is not clear, however, whether animals account for the handling time associated with consuming food rewards when making discounting decisions. We offered cotton-top tamarins, a New World primate species, choices between small, immediate rewards and larger rewards after various time delays. In addition, we experimentally manipulated the handling time of each monkey; subjects either received the entirety of the reward following the delay, or food pieces were dispensed incrementally with a small delay in between to simulate increased handling time.  Preliminary analyses indicate that tamarins prefer the small, immediate rewards more in the incremental condition at larger delays, suggesting that they are sensitive to increases in handling time. This type of evaluation allows us to distinguish between different models of discounting such as rate maximization and hyperbolic discounting.
Rosengart Carrie R. Rosengart & Dorothy M. Fragaszy (University of Georgia)
Placement and Order Errors in a Seriation Task in Capuchin Monkeys (Cebus apella)
Three combinatorial methods (pair, pot and subassembly) can be used when combining nesting cups.  Strategy preferences are based on relative hierarchical understanding of the properties of the cups. Subassembly is the method primarily used by adults. Two capuchin monkeys were trained to seriate nesting cups using the subassembly method. Both monkeys were able to seriate the cups in nearly half as many moves after subassembly training.  The subassembly training procedure may have changed the type of errors.  There can be errors due to incorrect placement (attempting to nest a large cup inside of a smaller cup) or due to an incorrect order (correctly nesting a smaller cup into a larger cup, creating a stable structure, but the cups are not adjacent to each other in sequence).  An analysis of the individual moves showed the relative rates of placement and order errors after subassembly training.
Saulsgiver Kathryn A. Saulsgiver, Erin McClure, and Clive Wynne (University of Florida)
Effects of d-amphetamine on peak interval responding in pigeons
The Peak Interval (PI) procedure is a discrete trial Fixed Interval (FI) schedule in which a subset of intervals run for several times the normal FI duration and terminate without reinforcement. Previous studies have reported that the time of peak response rate on unreinforced trials (peak time) shortens with amphetamine treatment. We found dose-dependent reduction in peak time under d-amphetamine. In addition we found that wait times (times to first response on each trial) showed a dose-dependent reduction. Analysis of response rates revealed a rate dependent effect: low response rates early in each trial were increased and high rates decreased by administration of amphetamine. We hypothesize that this rate dependent effect could be sufficient to cause the observed shifts in peak time and wait time without the need to assume effects of the drug on an underlying timing mechanism.
Seed Amanda M. Seed, Nathan J. Emery, Sabine Tebbich, & Nicola S. Clayton (University of Cambridge)
Investigating Causal Cognition in Rooks (Corvus frugilegis): A ‘Two-Trap Tube’ Task
Rooks are not reported to use tools in the wild, but in a previous study we demonstrated that they are capable of solving the trap tube task.  In this study we further investigated their understanding of physical causality.  We presented rooks with a ‘two-trap tube’, which had both a functional and non-functional trap, to avoid use of the simple rule ‘pull away from the trap’. We investigated their understanding by looking for immediate transfer to a different design.  7 of the 8 rooks solved the initial problem, taking between 30 to 140 trials.  All 7 birds showed transfer (within 10 trials) to a design with a different non-functional trap.  One transferred to two further tubes, which set the previously non-functional traps against each other, each design making one of them functional.  These results suggest that rooks appreciate the causal regularities of physical problems. 
Shapiro Nileen B. Clark, Lisa J. Winter, & Martin S. Shapiro (California State University, Fresno)
The Effect of a Delayed Reward on Choice Behavior in the American Grasshopper.
Past research on learning in vertebrates has shown that animals prefer immediate to delayed rewards.  While this has been an important parameter of associative learning in vertebrates, there has been very little work with an invertebrate model.  Grasshoppers (Schistocerca americana) were trained in a two-sided Y-maze with scented arms offering food rewards after a 0 or 3 minute delay.  In the first experiment (n=12), grasshoppers were given 14 choice trials with forced experience with both options.  While the animals did prefer the immediately rewarded odor, there was some indication that further training would improve this preference.  In the second experiment (n=12), training was extended to 20 trials, which did appear to increase choice proportions of the immediately rewarded option. This protocol should allow future research with delay of reward on learning and choice in designs such as self control, optimal foraging and risk-sensitivity.   
Smith Denise P.A. Smith & Stephen B. Fountain (Kent State University)
Medial and Lateral Caudate Putamen Lesions and Rat Serial Pattern Learning
In prior research, MK-801, an NMDA receptor antagonist, disrupted serial pattern learning in rats when the serial pattern was a sequence of 24 response elements arranged in eight 3-element chunks. The final element of the sequence violated the overall pattern structure. MK-801 rats learned within-chunk elements as fast as controls, but showed permanent inability to learn the violation response, and to a lesser degree, chunk boundary responses. Dorsal hippocampus and medial frontal cortex lesions did not produce the same deficit seen with MK-801. In the present study, rats received either medial or lateral caudate putamen excitotoxic lesions. Rats were then trained on the same pattern as in previous studies. Both medial and lateral caudate putamen lesions caused learning deficits for within-chunk and the violation elements, but the deficits were less severe than those caused by MK-801.
Stollnitz Fred Stollnitz (National Science Foundation)
Funding and Employment Opportunities at the National Science Foundation
Research and related activities in comparative cognition are supported mainly through the Behavioral Systems Cluster of the Division of Integrative Organismal Biology.  Large, multifaceted projects may also be supported through Science of Learning Centers or Frontiers in Integrative Biological Research.  Education projects may involve ethics education; course, curriculum or laboratory improvement; teacher enhancement, or informal education of the general public through zoo or museum exhibits, films or TV programs, etc.  Projects that integrate research and education are particularly welcome, as in Research in Undergraduate Institutions, Faculty Early Career Development, Research on Learning and Education, Integrative Graduate Education and Research Traineeship, Research Opportunities for Teachers, Research Assistantships for Minority High School Students, Research Opportunities for Undergraduates, and Undergraduate Mentoring in Environmental Biology (defined broadly enough to include comparative cognition!). Opportunities to serve as a science assistant, as a program officer, or as a reviewer are available in many NSF programs.
Sutton Jennifer E. Sutton (University of Western Ontario) & Sara J. Shettleworth (University of Toronto)
Sense of Direction and Landmark Piloting in Pigeons
The relative importance of a sense of direction based on inertial cues and landmark piloting for small-scale navigation by pigeons was investigated in an arena search task. Two groups of pigeons learned the location of buried food in an arena containing stable landmarks but differed in whether they had access to visual cues outside the arena. After experience with two different entrances, pigeons with access to extra-arena visual cues transferred accurate searching to novel entrances. Pigeons without visual access relied on a response strategy when first entering from a novel direction but quickly learned to search accurately. Explicit disorientation before entering did not affect accuracy.  In further manipulations, landmarks and inertial cues were put in conflict or tested one at a time. Pigeons tended to follow the landmarks in a conflict situation but could use an internal sense of direction based on inertial cues to search when landmarks were unavailable.
Thompson Dana J. Gant, Nidhi Suri, Norbelina Disla, & Roger K. R. Thompson (Franklin & Marshall College)
Unequal Pay or Violated Expectations? Capuchin Monkey (Cebus apella) Responses to Qualitatively Different Rewards
Brosnan & de Waal (2003) reported that pairs of Brown capuchin monkeys (Cebus apella) responded negatively to an unequal distribution of grape and cucumber rewards between them by a human. They interpreted their findings as evidence for, “an early evolutionary origin of inequity aversion.” Reports in the media further implied that the results showed monkeys have a sense of ‘fairness and justice’. Here, we report within- and between-session incentive contrast effects with grapes and cucumbers in capuchin monkeys. Thus far, our findings are consistent with a less anthropomorphic explanation of Brosnan & de Waal’s (2003) results. Rather than appealing to abstract concepts of inequity and fairness we suggest that a simple ‘violation of expectations” is a more parsimonious account of the animals’ behavior. 
Tremblay Joseph Tremblay & William A. Roberts (University of Western Ontario)
The Role of Exploratory Experience in Rats' Formation of Cognitive Maps
One of the more enduring debates in animal cognition has been whether animals are able to form and use cognitive maps. The use of a novel shortcut between two previsouly visited locations can be taken as evidence of the existence of a cognitive map. Chapuis, Durup, and Thinus-Blanc (1987) found that when golden hamsters were exposed to a path connecting two previously visited subspaces they chose a shortcut significantly more often than hamsters not exposed to the connecting path. We report a study in which two groups of rats were exposed to a portion of a cross maze in the first phse. In the second phase, one group was exposed to a "connecting" path. The results allowed us to test the prediction that the group exposed to the connecting path would choose a shortcut more often than the group not exposed to the connectin gpath.
Urcelay Gonzalo P. Urcelay & Ralph R. Miller (SUNY-Binghamton)
Nonadditive effects of overshadowing and degraded contingency
A series of four experiments using rats as subjects investigated the effects of combining two treatments known for their response decrementing effects: overshadowing and degraded contingency. Experiments 1 and 2 demonstrated that subjects that experienced both treatments during training did not show any impairment at the time of testing, in contrast with subjects that experienced either treatment alone. Further studies demonstrated that extinguishing either the context in which training occurred or the overshadowing cue had detrimental effects on responding in subjects that experienced both overshadowing and degraded contingency treatments.  However, extinction of the training context produced a recovery from simple degraded contingency and extinction of the overshadowing cue produced a recovery from simple overshadowing. The present results are problematic for current associative theories of learning, but are consistent with recent data showing that the basic principles of conditioning differ for cues trained together and cues trained apart.
Verbeek Eric L. Verbeek, Marcia L. Spetch (University of Alberta), Ken Cheng (Macquarie University), & Colin W.G. Clifford (University of Sydney)
Range Effects in Face Recognition: Complex Stimuli or Complex Dimensions?
The effects of test stimulus range on generalization gradients were assessed for discriminations between morphed faces (Experiment 1), or between faces that varied in brightness (Experiment 2) or orientation (Experiment 3). Consistent with a previous study, the range of stimuli presented in testing did not affect the generalization gradients for the discrimination of morphed faces. However, a significant range effect in the direction predicted by adaptation level theory occurred when faces varied along the brightness or orientation dimension. These results suggest that resistance to range effects is due to the complex dimension produced by morphing rather than to the complex nature of face stimuli.
Vlasak Anna N. Vlasak (University of Pennsylvania)
The Relative Importance of Global and Local Landmarks in Navigation by Columbian Ground Squirrels (Spermophilus columbianus)
For efficient orientation animals can use environmental features that serve as landmarks on local or global scales. Although landmark-based navigation has been explored in many species, animals that have to remember locations of many burrows for successful survival have received little attention. I investigated landmark use by free-ranging Columbian ground squirrels (Spermophilus columbianus), burrowing mammals. Experiments tested squirrel’s ability to locate burrows during escape when local or global landmarks were obstructed. Results suggest that squirrels rely on both local and global features of the environment for successful navigation. The lack of information from one type of landmarks (local or global) cannot be completely compensated by the other type. In addition, partial obstruction of global landmarks reveals that squirrel attend preferentially to the upper portion of the horizon, which potentially shows the most prominent and reliable features of the environment.
Werner Daniel C. Werner, Amanda R. Willey, Amanda C. Alexander, and James D. Rowan (Bridgewater College)
The effects of MK-801 on phrasing in rat serial-pattern learning in rats
One explanation used to explain the rats’ deficits in learning of serial patterns is that MK-801 blocks the animals’ ability to utilize the temporal phrasing cues used to parse the patterns.  Thus, this experiment examines the effects of MK-801 on phrasing.  Rats were assigned to one of three phrasing groups.  The pattern was phrased for the separate groups by inserting a 3-sec pause (at all other locations there was only 1-sec between elements).  In the Good Phrasing group, the 3-sec pauses were placed at the boundaries of the chunks determined by the formal structure of the pattern.  In the Bad Phrasing group, the phrasing cues were placed in the middle of chunks.  The No Phrasing received the pattern with no 3-sec phrasing.  Overall, MK-801 impaired learning but the shape of the error profiles were similar, indicating that the effect by the drug is not due to the inability to utilize phrasing cues.
Willey Amanda R. Willey, James D. Rowan (Bridgewater College), and Stephen B. Fountain (Kent State Universtiy)
The role of correction in double alternation learning in rats
Traditionally, rats have shown great difficulty in learning a double alternation pattern.  Recent experiments using a procedure originally developed to examine serial–pattern learning in rats failed to replicate this difficulty.  One difference between this procedure and the traditional procedures is that the serial-pattern learning task uses a correction procedure.  The goal of this experiment is to examine the importance of correction in the learning of double alternation patterns.  In this experiment, rats were divided into two groups.  Both groups were required a 24 element double alternation pattern in an octagonal operant chamber.  Rats in one group were required to make the correct response (and received reinforcement) before progressing to the next trial while rats in the other group progressed to the next trial after any response and only received reinforcement for correct responses. The results support the hypothesis that correction dramatically improves acquisition.
Wilson Patricia A. Wilson, Janet Metcalfe & Herbert S. Terrace (Columbia University & NYPI)
Transforming Humans into Monkeys:  A Memory Experiment
Two rhesus monkeys performed a Serial List Recognition (SLR) task in which 4 arbitrary photographs were presented successively and then presented along with 6-8 arbitrary distractors.  Subjects were reinforced for selecting all items from the list, regardless of presentation order, without selecting distractors. A strong recency effect was obtained but no primacy effect.  Human subjects that  performed the same task showed both primacy and recency effects.  When the arbitrary photographs were changed to black and white fractals that are more difficult to discriminate, the accuracy of human subjects decreased and output order began to mirror that of monkeys. An experiment in progress that decreases subjects’ reaction time (to match that of the monkeys) is expected to result in monkey-like performance with respect to output order and recency effects.