The effects of stimulus salience and cue validity in the overshadowing of geometric features of an enclosed arena by discrete landmarks were investigated in rats using the water maze paradigm. Experiment 1 established that in a rhomboid-shaped arena, the acute corner was more salient than the obtuse corner. In Experiment 2, rats were trained to find a submerged platform either in one of the acute, or obtuse, corners. In addition to the information provided by corner angle, the platform was also signaled by the presence of a spherical landmark suspended above the platform for rats in the experimental group. The landmark was a more valid cue for predicting the location of the platform than the angle of the corner. This training resulted in overshadowing of learning about the angle of the corner by the presence of the landmark. The final experiment extended this result by showing that when the predictive validities of the angle and the landmark were matched in the experimental group, learning about geometry was still overshadowed by the presence of landmarks, but only in animals that were trained with the platform at an obtuse, but not acute, corner. These results uniquely demonstrate that learning about geometry can be overshadowed by discrete landmarks, and also that whether overshadowing is observed depends on the stimulus salience and the relative validity of the competing cues. These findings imply that learning based on geometric cues follows the same basic rules that apply to a wide range of other learning paradigms.

Blocking of learning about a conditioned stimulus (the "blocked" cue) occurs when it is trained alongside an additional stimulus (the "blocking" cue) that has been previously presented with the outcome. A number of theories (e.g., N. J. Mackintosh. 1975a. A Theory of Attention: Variations in the Associability of Stimuli With Reinforcement. Psychological Review, 82, 276-298; J. M. Pearce & G. Hall. 1980. A Model for Pavlovian Learning: Variation in the Effectiveness of Conditioned But Not Unconditioned Stimuli. Psychological Review, 87, 532-552) account for this attenuation in learning by proposing that attention paid to the blocked cue is restricted. In three experiments, we examined the associability of both blocked and blocking cues. In Experiment 1, rats were trained with a blocking protocol before being given a test discrimination composed of two components; one of these components required the use of the previously blocked cue as a discriminative stimulus, and the other component was soluble by using the blocking cue. To our surprise, the component that depended on the blocked cue was more readily solved than the component dependent on the blocking cue. The results of Experiments 2 and 3 suggest that this is due to the quantity of exposure that each stimulus received during initial training. Implications for theories of blocking, and more widely associative learning, are discussed.

Conditioned taste aversion (CTA) is traditionally viewed as an instance of pavlovian conditioning. This interpretation rests on the lack of an instrumental contingency between the tastant and the gastric malaise in a standard procedure of CTA. To investigate a role for instrumental punishment in CTA, we present 2 tastants sequentially ("sucrose then NaCl" or "NaCl then sucrose") in a daily alternating and counterbalanced order to rats with an explicit positive contingency between the dosage of the lithium chloride (LiCl) administered and the amount of 1 tastant drunk on that trial. In the beginning of experiment, rats suppressed their intake of both tastants. With the increase of conditioning trials, rats gradually learned to resume the intake of noncontingent solution while selectively suppressing the intake of LiCl-contingent solution. This selective suppression in CTA is the first report indicating that rats are sensitive to the subtle cues related to the covariations between the magnitude of stimulus and the magnitude of responses in a punishment paradigm involving a long delay between the gustatory stimulus of tastant ingestion and the aversive effect of LiCl injection.

Participants in 2 experiments classified face stimuli into 2 categories determined by the gender of the faces. Although the category rule was simple, the stimuli, created by morphing male and female faces, made the explicit identification of the rule difficult. Participants were classified as rule users or nonusers depending on whether they explicitly identified the gender rule on a postexperiment questionnaire. Nonusers displayed a generalization decrement to new category exemplars that were more obviously male and female but less similar to the trained exemplars. Rule users, instead, generalized their category judgments perfectly to new category members. However, when an exception to the rule was introduced, generalization decrement for stimuli related to the exception was evident in both groups. All participants displayed a near-perfect ability to reverse their category judgments, regardless of whether learning occurred in the presence or absence of the explicit gender rule. The results highlight a distinction between rule- and feature-based category learning but show that both processes are subject to cognitive control.

Studies of spontaneous behavior to assess memory are widespread, but often the relationships of objects to contexts and spatial locations are poorly defined. We examined whether object-location memory was maintained following global, but not local, changes to the geometric shape of an arena. Rats explored two trial-unique objects in a distinctively shaped arena before being exposed to two identical copies of one of these objects in a different shape in a different physical location. Rats preferentially explored objects that were novel in relation to their local geometric context rather than identifying both locations as novel in the global geometric context.

Strong interest exists in developing a rodent model of visual cognition to conduct research into the neural mechanisms of visual categorization. Yet, doubt remains as to whether rats perform visual categorization tasks as do humans and nonhuman primates. Here, we trained eight rats on two visual categorization tasks using photographs of eight objects from each of four basic-level categories: chairs, flowers, cars, and humans. In Experiment 1, rats learned to categorize chairs versus flowers; in Experiment 2, rats learned to categorize cars versus humans. After rats learned each discrimination, we tested them with eight novel pictures from each of the categories. The rats performed at reliably above-chance levels during these generalization tests. To determine which dimension(s) of the stimuli controlled the rats' behavior, we conducted regression analyses using several image dimensions. The chair versus flower discrimination was mainly controlled by the convexity of the stimuli, whereas the car versus human discrimination was mainly controlled by the aspect ratio of the stimuli. These results demonstrate that rats can categorize complex visual objects using shape-based properties of photographs.

The most common associative explanation of interference is based on a retrieval failure. Retrieval, in turn, is considered as the result of an associative activation mechanism that is thought to be fast and automatic. However, up-to-date, there is no evidence of interference based on dependent measures specifically related to this kind of low level processes. The objective of the present study was to test whether interference phenomena can be observed by using a cued response task designed to detect low level retrieval processes. Experiment 1 evaluated whether the cued response task served to show a priming effect. Such effect allowed us to interpret the results found in the remaining experiments of the series. Experiment 2 aimed to find the interference effect by using the cued response task. Experiments 3 and 4 were conducted to assess whether spontaneous recovery and context-change effects could also be observed. The results showed that interference and recovery from interference phenomena can be attributable to fast retrieval processes, which is consistent with associative accounts of interference.

When pigeons are given a choice between two alternatives, one leading to a stimulus 20% of the time that always signals reinforcement (S+) or another stimulus 80% of the time that signals the absence of reinforcement (S-) and the other alternative leading to one of two stimuli each signaling reinforcement 50% of the time, the 20% reinforcement alternative is preferred although it provides only 40% as much reinforcement. In Phase 1 of the present experiment, we tested the hypothesis that pigeons compare the S+ associated with each alternative and ignore the S- by giving them a choice between two pairs of discriminative stimuli (20% S+, 80% S- and 50% S+, 50% S-). Reinforcement theory suggests that the alternative associated with more reinforcement should be preferred but the pigeons showed indifference. In Phase 2, the pigeons were divided into two groups. For one group, the discriminative function was removed from the 50% reinforcement alternative and a strong preference for the 20% reinforcement alternative was found. For the other group, the discriminative function was removed from both alternatives and a strong preference was found for the 50% reinforcement alternative. Thus, the indifference found in Phase 1 was not due to the absence of discriminability of the differential reinforcement associated with the two alternatives (20% vs. 50% reinforcement); rather, the indifference can be attributed to the pigeons' insensitivity to the differential frequency of the two S+ and two S- stimuli. The relevance to human gambling behavior is discussed.