Study Of Animal Learning Research Articles

Defining the stimulus--a memoir.

The eminent psychophysicist, S.S. Stevens, once remarked that, "the basic problem of psychology was the definition of the stimulus" (Stevens, 1951, p. 46). By expanding the traditional definition of the stimulus, the study of animal learning has metamorphosed into animal cognition. The main impetus for that change was the recognition that it is often necessary to postulate a representation between the traditional S and R of learning theory. Representations allow a subject to represent a stimulus it learned previously that is currently absent. Thus, in delayed matching-to-sample, one has to assume that a subject responds to a representation of the sample during test if it responds correctly. Other examples, to name but a few, include concept formation, spatial memory, serial memory, learning a numerical rule, imitation and metacognition. Whereas a representation used to be regarded as a mentalistic phenomenon that was unworthy of scientific inquiry, it can now be operationally defined. To accommodate representations, the traditional discriminative stimulus has to be expanded to allow for the role of representations. The resulting composite can account for a significantly larger portion of the variance of performance measures than the exteroceptive stimulus could by itself.

Controversies over the role of comparison in animal psychology and a new perspective for species comparison

The purpose of this article was to address issues on what the role of species comparison is for the studies of animal psychology. Whereas some people think that the comparison has a critical role in their approaches, others do not. We classify these two groups of people as evolutionists and generalists, respectively. In the evolutionary approach, different species are compared to examine correlations between specific selective pressures (such as caching food) and cognitive abilities (such as spatial learning) in animals. In contrast, different species of animals are compared to examine universal principles of learning in the generalist approach. Although the generalists' ideas of the “universal” principles of learning provide parsimonious explanations of animals' behavior, they tend to underestimate influences of selective pressures on learning and cognition. We therefore propose a new framework for the study of animal learning and cognition; it is based on the evolutionary approach and examines ultimate causes of diversities of learning and cognition but includes ideas of generalists' approaches to explain proximate causes.

Review of The Behavior of Animals: Mechanisms, Function and Evolution.

JOHAN J. BOLHUIS and LUC-ALAIN GIRALDEAU The Behavior of Animals: Mechanisms, Function and Evolution Oxford: Blackwell Publishing, 2005, 536 pages (ISBN 0-631-23125-0, US$69.95 Paperback) Interest in the behaviour of animals dates at least to the ancient Greeks (and probably long before), for humans have long appreciated the potential insights to be gleaned about their own habits from those of animals. Until relatively recently, however, the scientific study of animal behaviour was comparatively limited in scope. Until Darwin, and for some time after, research on animal behaviour was conducted primarily by amateur naturalists and seldom explicitly and thoughtfully linked to human behaviour and psychology, the exception being the rigorous studies of animal learning that began toward the end of the 19th century and continued throughout the 20th, which was, by and large, where the field of comparative psychology started and finished. All of that changed in the middle 20th-century with the landmark works of ethologists Nikko Tinbergen, Konrad Lorenz, and Karl von Frisch, who shared a Nobel Prize for their innovative formal research programs on, and considerable insights into, the naturalistic behaviour of animals. The scientific study of animal behaviour grew precipitously after that, and today encompasses field and laboratory studies of every major animal group and the truly dizzying array of behavioural phenomena that they manifest, all with a dual focus on better understanding the natural diversity of animal behaviour and on using it to better illuminate our own. The Behavior of Animals: Mechanisms, Function and Evolution is a new textbook by Johan Bolhuis and LucAlain Giraldeau aimed at covering this increasingly dynamic and broad field. Given the vastness of the phenomena now to be covered, a compelling theoretical framework is absolutely essential if a comprehensive understanding of animal behaviour is to emerge. Fortunately, Tinbergen himself provided just such a framework for organizing research on animal behaviour, and Bolhuis and Giraldeau wisely follow it in their organization of the material in this text. Thus, a short preface rehearsing the history of animal behaviour research by Robert Hinde (one of Tinbergen's students who himself became one of the foremost contemporary ethologists) is followed by 16 chapters by 23 authors. The chapters are organized into three major sections, two of which flow directly from Tinbergen's four whys of behaviour. The first section is concerned with the first two of Tinbergen's four questions, namely, what are the immediate proximate and longer developmental causes of behaviour? This section includes chapters on perception, motivation, biological rhythms, learning and memory, and cognition. All are well written and comprehensive. David Sherry's chapter provides an especially nice balance of current cognitive theory and basic empirical research by partitioning brain effects on behaviour into three main processes: obtaining information, manipulating that information, and directing movement and behaviour in response. To illustrate these processes, Sherry uses clear and compelling behavioural examples, such as the well-researched auditory processes that barn owls use to detect and capture rodent prey. Similarly, the animal cognition chapter by Nathan Emery and Nicola Clayton offers a concise account of the mechanistic basis of spatial memory and social relationships that are central to challenges in the lives of so many species. The second section concentrates on Tinbergen's second set of questions: What is the ultimate adaptive function of a behaviour, and how did it evolve in the history of the species? These questions are explored in several chapters with different themes, including communication, sociality, mating systems, mate choice, polyandry, sperm competition, and sexual conflict. …

Measures of temporal discrimination in fixed-interval performance: A case study in archiving data

The primary data of many experimental studies of animal learning and performance consist of the times at which stimuli and reinforcers were delivered, and the times at which responses occurred. The articles based on most of these studies report selected data, either from some sessions or some animals, or summary measures of the animals' behavior. The primary data are sufficient to produce any of the selected and summary measures, but the selected and summarized data cannot produce many of the measures used in other experimental reports. It is now feasible to archive the primary data from animal behavior experiments so that they are accessible for others to perform secondary analysis. The value of such secondary analysis of archived data is described with a case study in which rats were trained on three fixed-interval schedules of reinforcement. The full data set may be downloaded from www.psychonomic.org/archive/.

Training Tammar Wallabies (Macropus eugenii) to Respond to Predators: A Review Linking Experimental Psychology to Conservation

Animals bred in captivity often suffer high levels of predation after release into the wild. Prerelease predator avoidance training has been undertaken to try to improve antipredator skills. Applied research on predator avoidance learning in birds and mammals has not benefited from the empirical findings of extensive basic research, as it has been the case for fish. Consequently, this field has progressed slowly and the utility of prerelease antipredator training as a conservation strategy remains controversial. Here, I report one experiment and review two others that illustrate the way in which principles and experimental designs borrowed from classic studies of animal learning can be used to develop predator avoidance training techniques and to establish the content of learning. Results show that tammar wallabies (Macropus eugenii), an Australian macropodid marsupial, can acquire a fear response that is specific to predators, but that the likelihood of learning is dependent upon subtle details of the training protocol. Differential reinforcement of predator and non-predator stimuli has the potential to enhance the specificity of learning, if necessary. I discuss the implications of these results for the field of predator avoidance training and suggest that a controlled experimental approach, which enables the content of learning to be described, will be the most fruitful for this research area in the long term.

Rescuing The Science of Human Behavior from the Ashes of Socialism

The discredit into which the socialist ideal has fallen as a consequence of recent political events calls into question, not just the viability of a particular political and economic system but, the very idea that the social order can be improved by applying principles derived from the scientific study of human social behavior. Before the collapse of socialism, the idea of a science of human behavior, construed in biological terms as a branch of the science of the behavior of free-moving living organisms in general, had been undermined by Chomsky’s (1959) repudiation of the behaviorist project to construct a science of language (verbal behavior) based on principles derived from the study of animal learning. I contend that only by reinstating the link between linguistics and the study of animal learning can confidence be restored in the possibility of a genuine science of human behavior with application to the problem of constructing a better social order.

Generalization in Human Category Learning: A Connectionist Account of Differences in Gradient after Discriminative and Non discriminative Training

Two experiments are reported that investigate the difference in gradient of generalization observed between one-category (non-discriminative) and two-category (discriminative) training. Extrapolating from the results of a number of animal learning studies, it was predicted that the gradient should be steeper under discriminative training. The first experiment confirms this basic prediction for the stimuli used, which were novel, prototype-structured, and constructed from 12 symbols positioned on a grid. An explanation for the effect, based on the Rescorla-Wagner theory of Pavlovian conditioning (Rescorla & Wagner, 1972), is that under non-discriminative training “incidental stimuli” have significant control over responding, whereas under discriminative training they do not. Incidental stimuli are those aspects of the stimulus, or the surrounding context, that are not differentially reinforced under discriminative training. This explanation leads to the prediction that a comparable effect of blocked versus intermixed discriminative training should also be found. This prediction is disconfirmed by the second experiment. An alternative model, still based on the Rescorla Wagner theory but with the addition of a decision mechanism comprising a threshold unit and a competitive network system, is proposed, and its ability to predict both the choice probabilities and the pattern of response times found is evaluated via simulation.

Imitative learning in male Japanese quail (Coturnix japonica) using the two-action method.

The study of imitative learning in animals has suffered from the presence of a number of confounding motivational and attentional factors (e.g., social facilitation and stimulus enhancement). The two-action method avoids these problems by exposing observers to demonstrators performing a response (e.g., operating a treadle) using 1 of 2 distinctive topographies (e.g., by pecking or by stepping). Japanese quail (Coturnix japonica) observers exposed to conspecific demonstrators showed a high correlation between the topography of the response they observed and the response they performed. These data provide strong evidence for the existence of true imitative learning in an active, precocious bird under conditions that control for alternative accounts.

Late-training amnesic deficits in probabilistic category learning: a neurocomputational analysis.

Building upon earlier behavioral models of animal and human learning, we explore how a psychobiological model of animal conditioning can be applied to amnesic category learning. In particular, we show that the late-training deficit found in Knowlton, Squire, and Gluck's 1994 study of amnesic category learning can be understood as a natural consequence of Gluck and Myers's (1993) theory of hippocampal-region function, a theory that has heretofore been applied only to studies of animal learning. When applied to Knowlton et al.'s category learning task, Gluck and Myers's model assumes that the hippocampal region induces new stimulus representations over multiple training trials that reflect stimulus-stimulus regularities in the training set. As such, the model expects an advantage for control subjects over hippocampal-damaged amnesic patients only later in training when control subjects have developed new hippocampal-dependent stimulus representations; in contrast, both groups are expected to show equivalent performance early in training. A potentially analogous early/late distinction is described for animal studies of stimulus generalization. Our analyses suggest that careful comparisons between early and late-training differences in learning may be an important factor in understanding amnesia and the neural bases of both animal and human learning.

Cam: An animal learning model of excessive and compulsive implementassisted drug-taking in humans

Animal learning studies reveal that locating the reward cue (discriminative stimulus positively correlated with the presentation of positive reinforcement) at the site of the response manipulandum (object contacted in performing the instrumental response), an arrangement referred to as CAM (cue and manipulandum), induces excessive instrumental responding. CAM induces excessive responding even when responding is negatively related to reinforcement and serves only to delay or cancel reinforcement, revealing that excessive responding induced by CAM is unrestrainable and compulsive. In addition, the response form induced by CAM resembles patterns of consummatory behaviors. Thus, animal learning studies reveal that CAM induces excessive and compulsive appetitive-consummatory responding that is triggered by objects predictive of rewarding substances. The CAM model defines conditions under which the drug-taking implement the response manipulandum at which instrumental drug-taking behavior is directed) will contribute to the development of excessive and compulsive drug-taking in humans. Implementassisted drug-taking procedures in humans provide for CAM whenever the drug-taking implement is positively correlated with the drug's reinforcing effects. This correlation is highest when the drug-taking implement is employed only to consume the drug and the drug is consumed in no other fashion. Evidence relating drug-taking implements to drug abuse is reviewed and implications for prevention and therapy are considered.

Reinforcement learning is direct adaptive optimal control

Neural network reinforcement learning methods are described and considered as a direct approach to adaptive optimal control of nonlinear systems. These methods have their roots in studies of animal learning and in early learning control work. An emerging deeper understanding of these methods is summarized that is obtained by viewing them as a synthesis of dynamic programming and stochastic approximation methods. The focus is on Q-learning systems, which maintain estimates of utilities for all state-action pairs and make use of these estimates to select actions. The use of hybrid direct/indirect methods is briefly discussed. >

A factor analysis of the Gray-Wilson personality questionnaire

Principal components analysis of the Gray-Wilson Personality Questionnaire revealed six fairly independent factors that bore only a partial resemblance to the six animal behaviour paradigms that the test was devised to measure. Taken together with the inter-scale correlations reported previously this raises certain questions about the application of brain systems derived from animal learning studies to the analysis of individual differences at the human level.

Comparative psychology and the study of animal learning.

Comparative psychology and the study of animal learning.

The evolution of multiple memory systems.

The existence of multiple memory systems has been proposed in a number of areas, including cognitive psychology, neuropsychology, and the study of animal learning and memory. We examine whether the existence of such multiple systems seems likely on evolutionary grounds. Multiple systems adapted to serve seemingly similar functions, which differ in important ways, are a common evolutionary outcome. The evolution of multiple memory systems requires memory systems to be specialized to such a degree that the functional problems each system handles cannot be handled by another system. We define this condition as functional incompatibility and show that it occurs for a number of the distinctions that have been proposed between memory systems. The distinction between memory for song and memory for spatial locations in birds, and between incremental habit formation and memory for unique episodes in humans and other primates provide examples. Not all memory systems are highly specialized in function, however, and the conditions under which memory systems could evolve to serve a wide range of functions are also discussed. Memory is a function that permits animals and people to acquire, retain, and retrieve many different kinds of information. It allows them to take advantage of previous experience to help solve the multitude of problems with which their environment confronts them, such as how to recognize the familiar, predict events, return to particular places, and assess the consequences of behavior Recently the question has arisen as to whether the

Comparative psychology and the study of animal learning.

Comparative psychology and the study of animal learning.

Tolman and modern conditioning theory

Tolman proposed that the changes we observe in studies of animal learning do not directly mirror what the animal has learned in the way stimulus‐response theory implied. Pavlovian conditioning, for example, involves the acquisition of sign‐Gestalt expectancies, which are only indirectly related to behaviour. This account of conditioning is widely accepted today, and Tolman was alert to its central problem, that of explaining the form of the conditioned response. His scattered thoughts on this question have not yet been superseded by any more systematic theory. But Tolman was convinced that Pavlovian conditioning was a relatively uninteresting form of learning to be sharply distinguished from discrimination or maze learning. His argument here is unconvincing: the theory of Pavlovian conditioning he proposed is quite powerful enough to encompass the results of research in these areas.

What Is the Study of Animal Learning About?

What Is the Study of Animal Learning About?

Reciprocal heuristics: A discussion of the relationship of the study of learned behavior in laboratory and field

At the turn of the century, the initiators of laboratory study of animal learning advocated two very different approaches to the subject matter. Willard Small favored the investigation of learning in ecological settings appropriate to individual species. E. L. Thorndike treated the process of association formation in animals as a general one, best studied in situations distant from those to which subject species were adapted. The latter view dominated laboratory study of animal learning for 80 years. The consequent absence of ecological content from laboratory investigations of animal learning, together with recent expansion in knowledge of the behavior of free-living animals, has created opportunities for ecologically sound investigations of animal learning of the type first advocated by Small. Studies of taste-aversion learning, although introducing evolutionary issues into the study of animal learning, do not exemplify such an approach. Integration of field and laboratory studies of behavioral plasticity in animals requires a change in the methods used to select phenomena for analysis. Such integration does not often lead to identification of the behavioral processes underlying the development of particular behaviors observed in nature. Rather, its results are an increase in the variety of learning processes investigated in laboratory settings and enhanced understanding of the behavioral capacities of subject species.

Neuronlike adaptive elements that can solve difficult learning control problems

It is shown how a system consisting of two neuronlike adaptive elements can solve a difficult learning control problem. The task is to balance a pole that is hinged to a movable cart by applying forces to the cart's base. It is argued that the learning problems faced by adaptive elements that are components of adaptive networks are at least as difficult as this version of the pole-balancing problem. The learning system consists of a single associative search element (ASE) and a single adaptive critic element (ACE). In the course of learning to balance the pole, the ASE constructs associations between input and output by searching under the influence of reinforcement feedback, and the ACE constructs a more informative evaluation function than reinforcement feedback alone can provide. The differences between this approach and other attempts to solve problems using neurolike elements are discussed, as is the relation of this work to classical and instrumental conditioning in animal learning studies and its possible implications for research in the neurosciences.

Biological constraints on instrumental and classical conditioning: Retrospect and prospect

The adequacy of traditional approaches to the study of animal learning to account fully for learning phenomena has been seriously questioned during the past decade. Critics of traditional analyses advocated a biological orientation to the interpretation of associative processes and introduced a variety of concepts intended to provide a new framework for the study of animal learning. This promise of a reorientation of the field has not been realized. The concepts of biological constraints, adaptive specializations, and situation specificity of learning have had a less profound influence on the general process approach to instrumental and classical conditioning than anticipated. The present paper makes explicit the conceptual bases of the original biological approaches to learning, identifies reasons why they failed to change fundamentally the study of instrumental and classical conditioning, and proposes an alternative approach to the use of ecological and evolutionary principles in studies of conditioning. We suggest a renewed comparative approach to the study of learning phenomena that avoids many of the difficulties inherent in earlier formulations by providing (1) a strategy for the discovery of adaptive specializations in learning, (2) an ecological framework for the discussion of these adaptive specializations, and (3) a renewed emphasis on the study of species differences in learning.