No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
Stimulus validity and stimulus selection in associative learning
... At the time, Wagner (1969) regarded these findings as evidence that "relative informativeness," or "cue validity," was critically important for determining the extent to which each of two cues gain control over responding when they are presented together. He then went on to suggest that the increment in the signal value of a cue, B, is "a function of the degree to which reinforcement is predictable on the basis of the entire configuration of cues among which B is included" (p. ...
... This neglect is surprising because, as we show shortly, the eventual fate of these cues has important theoretical implications. In Wagner's (1969) blocking condition, Aϩ AXϩ, Cue A reliably signals when the outcome will occur, thus rendering X redundant as a cue for providing information about the trial outcome. Turning to the simple discrimination of the relative validity design, AYϩ BY-, Cues A and B provide accurate information about when the outcome will occur and thus render Y redundant. ...
... Turning to the simple discrimination of the relative validity design, AYϩ BY-, Cues A and B provide accurate information about when the outcome will occur and thus render Y redundant. Given the lack of informativeness, or low cue validity of X and Y, according to the speculations of Wagner (1969) these cues might be treated in the same way, and elicit responses of similar magnitude after the different training schedules. A rather different prediction is made by the more formal analysis offered by the theory. ...
In 2 experiments, participants received a predictive learning task in which the presence of 1 or 2 food items signaled the onset or absence of stomachache in a hypothetical patient. Their task was to identify the cues that signaled the occurrence, or nonoccurrence of this ailment. The 2 groups in Experiment 1 and the single group in Experiment 2 received a blocking treatment, where Cue A and a combination of Cues A and X both signaled stomachache, A+ AX+. These groups also received a simple discrimination where the outcome was signaled by one compound but not another, BY+ CY-. Subsequent test trials revealed the so-called redundancy effect, where X was regarded as a more reliable predictor of the outcome than Y. This result occurred when the trials with A+ preceded those with AX+ (Group E, Experiment 1 and Experiment 2), and when the trials with A+ and AX+ were intermixed (Group C, Experiment 1). The results challenge theories based on the assumption that cues presented together must compete for a limited pool of associative strength. Rather, they are said to support theories that assume changes in attention determine what is learned when two or more cues are presented together. (PsycInfo Database Record (c) 2020 APA, all rights reserved).
... 29 This model, although quite simple, is still one of the most efficient attempts to account for many observed learning-related behaviours : First, the logarithmic acquisition curve fits very well with the observed acquisition of conditioned response, shown for instance in rats and rabbits [28] . Second, the model also accounts for the extinction phenomenon first described by Pavlov [6] : indeed, should the CS presentation not be followed by the US anymore, it would constitute a discrepancy with previously learnt information and lead to gradual extinction of the acquired learning. ...
... After training, flies were transferred to a T-maze [28] where they could choose between the CS + and the CS − in the absence of shock during 1 min. In the case of NP, where flies faced two CS + , half of the tests were performed using one of the CS + while the other half was performed using the other CS + , both presented against the CS − . ...
Animals' survival heavily relies on their ability to establish causal relationships within their environment. That is made possible through learning experiences during which animals build associative links between the events they are exposed to. Most of the encountered stimuli are actually compounds, the constituents of which may have been reinforced (i.e., associated with a pleasant or unpleasant stimulus) in a different, sometimes opposed way. How compounds are perceived and processed is a central topic in the field of associative learning. In theory, a given compound AB may be learnt as the sum of its components (A+B), which is referred to as "Elemental learning", but it may also be learnt as a distinct stimulus (which Is called "Configural learning"). Finally, AB may bear both constituent-related and compound-specific features called "Unique Cues" (AB = A+B+u). Configural and unique cue processing enable the resolution of ambiguous tasks such as Negative Patterning (NP), during which A and B are reinforced when presented alone but not in a compound AB. Although neural correlates of simple associative learning are well described, those involved in non-elemental learning remain unclear. In this project, we rework a typical olfactory conditioning protocol based on semi-automated olfactory/electric shocks association, allowing us to demonstrate for the first time that Drosophila is able to solve NP tasks. Behavioural study of NP solving shows that its resolution relies on training repetition leading to a gradual change in the compound AB representation, shifting away from its constituents and thus becoming easier to distinguish. Next, we develop a computational model of olfactory associative learning in drosophila based on structural and functional in vivo data. Exploratory simulations of the model allow us to identify a theoretical mechanism enabling NP acquisition, the validity of which can be tested in vivo using neurogenetical tools only available in Drosophila. We propose that during a NP training, flies first acquire associative links between A, B and their reinforcement, which induces an ambiguity as the compound AB is presented without reinforcement. However, over the course of training cycles, non-reinforced stimuli representation is inhibited while the reinforced stimuli representation is consolidated. This differential modulation eventually leads to a shift in odours representation allowing flies to better distinguish between the constituents and their compound thus facilitating NP resolution. We identify APL (Anterior Paired Lateral) neurons as a plausible implementation of this theoretical mechanism, as APL inhibitory activity is specifically engaged during the non-reinforced stimulus presentation, which is necessary for NP acquisition but dispensable for non-ambiguous forms of learning. Lastly, we explore APL role in a broader context of ambiguity resolution. In conclusion, our work validates Drosophila as a robust model to investigate non-elementary learning, and present a promising model of the underlying neural mechanisms using a combination of behaviour, modelling and neurogenetical tools. We believe this opens the way to numerous interesting projects focused on understanding how animals extract robust associations in a complex world.
... Specifically, 193 X was presented in compound with A on the 75% of trials in which it was followed by 194 reward [3AX(1), X(0); where 3 indicates the proportion of trials]. This allowed A to 195 compete with X as a predictor of reward and steal its credit (e.g., Wagner, 1969). A's 196 ability to serve as a competitor was further bolstered by continuing to present it by itself 197 . ...
... If credit assignment is 205 noncompetitive (e.g., Bush & Mosteller, 1951), X should be expected to evoke more 206 responding than Y given its higher probability of reward. Conversely, to the extent credit 207 assignment is competitive, Y should be expected to evoke more responding than X 208 (Wagner et al., 1968;Wagner, 1969). To examine the role of agency, we focused our 209 analysis on responding on X(0) and Y(1) trials. ...
A fundamental assumption of learning theories is that the credit assigned to predictive cues is not simply determined by their probability of reinforcement, but by their ability to compete with other cues present during learning. This assumption has guided behavioral and neural science research for decades, and tremendous empirical and theoretical advances have been made identifying the mechanisms of cue competition. However, when learning conditions are not optimal (e.g., when training is massed), credit assignment is no longer competitive. This is a catastrophic failure of the learning system that exposes the individual's vulnerability to form spurious associations in the real world. Here, we uncover that cue competition can be rescued when conditions are suboptimal provided that the individual has agency over the learning experience. Our findings reveal a new connection between agency over learning and credit assignment to cues, and open new avenues of investigation into the underlying mechanisms.
... Therefore, the present study sought to explore whether stimuli would compete in a learning task involving a choice reaction time task and response conflict. The experiments used a procedure in which several different trial types were interspersed during a single phase of training (akin to the "relative validity" procedure used by Wagner [10]) to examine cue competition. There were two goals of the present study. ...
... It was expected that training with the A cue alone as a predictor of a compatible trial (with 80% probability) would reduce, through cue competition [9,10], the degree in which C would be learned about as a stimulus that predicts a compatible trial. Similarly, it was expected that the B-alone training trials would reduce the degree in which cue D would be learned about as a predictor of incompatible trials. ...
Two experiments investigated competition between cues that predicted the correct target response to a target stimulus in a response conflict procedure using a flanker task. Subjects received trials with five-character arrays with a central target character and distractor flanker characters that matched (compatible) or did not match (incompatible) the central target. Subjects’ expectancies for compatible and incompatible trials were manipulated by presenting pre-trial cues that signaled the occurrence of compatible or incompatible trials. On some trials, a single cue predicted the target stimulus and the required target response. On other trials, a second redundant, predictive cue was also present on such trials. The results showed an effect of competition between cues for control over strategic responding to the target stimuli, a finding that is predicted by associative learning theories. The finding of competition between pre-trial cues that predict incompatible trials, but not cues that predict compatible trials, suggests that different strategic processes may occur during adaptation to conflict when different kinds of trials are expected.
... At each point during training, the level of responding to each stimulus is a joint function of inhibition and excitation of that stimulus (Jenkins, 1965). Because this theoretical approach assumes stimuli are made up of many elements that each undergo continuous changes in associative strength over trials this class of models are referred to as Elemental-Continuity Theories of discrimination learning (Spence, 1936;Rudy and Wagner, 1975). This type of model, whereby discrimination training between similar stimuli has the effect of reducing conditioning of common elements, and increasing conditioning of unique elements, has been supported by experimental manipulations (Rescorla, 1976). ...
... It also does not decompose stimuli into elements but rather treats them as a more integrated configuration. Thus, his theory is a Configural-Noncontinuity Theory (Rudy and Wagner, 1975). ...
Neuroscientists are concerned with neural processes or computations, but these may not be directly observable. In the field of learning, a behavioral procedure is observed to lead to performance outcomes, but differing inferences on underlying internal processes can lead to difficulties in interpreting conflicting results. An example of this challenge is how many functions have been attributed to adult-born granule cells in the dentate gyrus. Some of these functions were suggested by computational models of the properties of these neurons, while others were hypothesized after manipulations of adult-born neurons resulted in changes to behavioral metrics. This review seeks to provide a framework, based in learning theory classification of behavioral procedures, of the processes that may be underlying behavioral results after manipulating procedure and observing performance. We propose that this framework can serve to clarify experimental findings on adult-born neurons as well as other classes of neural manipulations and their effects on behavior.
... However, although necessary, this simple Hebbian mechanism is not sufficient to account for the fact that over a number of CS-US pairing, the size of the increments in CS-US association becomes smaller as the cumulated associative strength of the CS increases towards an asymptote. This fact, commonly referred to as a negatively accelerated learning curve, was managed by early theoreticians of Pavlovian conditioning (e.g., Atkinson & Estes, 1963;Hull, 1943;Bush & Moesteller, 1955) by assuming what Wagner (1969) called a "saturation principle", in which each CS can acquire only a limited amount of association with the US, which is provided by the US itself. ...
... The conclusion of blocking and of other similar findings, like overshadowing (Pavlov, 1927) and relative validity (Wagner, Logan, Haberland, & Price, 1968), is that what is learned to one of the cues on a trial appears to depend not only upon its own current associative value, but also upon the asso-ciative value of the other cues present in the trial. This fact is sometimes called stimulus selection or stimulus competition (Wagner, 1969). ...
Pavlovian conditioning is a very simple and universal form of learning that has the benefit of a long and
rich tradition of experimental work and quantitative theorization. With the development of interdisciplinary
efforts, behavioral data and quantitative theories of conditioning have become progressively more
important not just for experimental psychologists but also for broader audiences such as neurobiologists,
computational neuroscientists and artificial intelligence workers. In order to provide interdisciplinary
users with an overview of the state of affairs of theoretically oriented research in this field, this chapter
reviews a few key mechanisms that are currently deemed necessary for explaining several critical
phenomena of Pavlovian conditioning. The chapter is divided into several sections; each referring to
a particular theoretical mechanism and to the type of phenomena that it has been designed to account.
The progression of the sections reveals phenomena and mechanisms of increasing complexity, which is
an indication of the theoretical sophistication that has been reached in this domain. Since there is not a
single theory containing all mechanisms, they are described separately from their originating theories,
emphasizing thus the fact that they might be used in almost any theoretical implementation.
... There is a long-standing interest in how to describe associations that are formed when an organism is conditioned to a compound stimulus (cf. Kehoe & Gormezano, 1980;Rudy & Wagner, 1975). Elemental theorists approach this issue by assuming that the stimulus is composed of a number of independent elements and that learning influences behavior by strengthening or weakening the association each element has with the reward or response. ...
... The implication some theorists have drawn from these data is that animals must be able to construct a unique representation of the pattern, or configuration, of stimulus elements that are present on a learning trial and use this representation to control behavior (cf. Pearce, 1987;Pearce & Wilson, 1990;Rescorla, 1972Rescorla, , 1973Rudy & Wagner, 1975;Whitlow & Wagner, 1972). Sutherland and Rudy (1989) in particular have supposed that normal animals have two learning systems, a simple associative system that operates according to principles of elemental theories and a configural association system that enables animals to form a representation of the conjunction of stimulus elements to solve problems that require nonlinear solutions (see also Mishkin & Petri, 1984;Wicklegren, 1979). ...
Little is known about the conditions that encourage animals to learn to use configural associations to guide their behavior or the consequences of such learning for transfer. This study provided some information about these issues by examining how rats solve the transverse-patterning problem, which requires a configural solution (Spence, 1952). Animals had to concurrently solve 3 simultaneous visual discriminations, represented abstractly as A+ versus B-, B+ versus C-, and C+ versus A-. Experiment 1 indicated that rats use a configural solution even when the problems have an elemental solution, provided that the significance of 1 element (e.g., B) shared by 2 problems is ambiguous (e.g., A+/B-; and B+/C-). Experiments 2 and 3 suggested that, when stimulated to use a configural solution by solving the A+/B- and B+/C- problems, rats transfer the configural solution to problems that have no ambiguous elements.
... Stimuli C and D, on the other hand, can be regarded as relevant because they can be used to predict trial outcomes accurately. It has long been appreciated that when presented together, relevant stimuli acquire more associative strength than irrelevant stimuli (e.g., Wagner, 1969), and for this reason alone, it might be expected that X, after the blocking treatment, will elicit a stronger response than Y, after the simple discrimination. If this prediction should turn out to be correct, then it would pose a serious challenge to the explanation offered by the Rescorla and Wagner (1972) theory for how humans attach more importance to relevant than irrelevant stimuli. ...
... There were two additional, noteworthy findings from the experiment. One of these was the successful demonstration of the relative validity effect (e.g., Wagner, 1969), in which the common cue from the simple discrimination, Y, was given a considerably lower rating than the common cue, G, from the pseudodiscrimination, FGϩ/-HGϩ/-. The implication of this result is that the training with BY-CYϩ was effective at restricting the rating given to Y. ...
In each of three experiments, a single group of participants received a sequence of trials involving pictures of a variety of foods presented individually or in pairs. Participants were required to predict in which trials the food would lead to a hypothetical allergic reaction. The different trials involved blocking, A+ AX+, and a simple discrimination, BY- CY+, in which each letter stands for a different food. Training trials were followed by a test in which participants were asked to predict how likely each kind of food would be followed by the allergic reaction. The principal purpose of the experiments was to determine how the redundant cue from blocking, X, would be judged relative to the redundant cue from the simple discrimination, Y. In contrast to predictions from currently influential theories of associative learning, X was regarded as a better predictor for the allergic reaction than Y. (PsycINFO Database Record (c) 2013 APA, all rights reserved).
... This overshadowing of one CS by another is well documented in a variety of preparations (Balsam, 1988;Mackintosh, 1975). Furthermore, the dominance of one element over another is enhanced if one of the CSs has been paired with the US prior to compound conditioning (Kamin, 1969) or if one of the CSs is singly paired with the US and intermixed with compound trials (Wagner, 1969;Wagner, Logan, Haberlandt, & Price, 1968). Rescorla and Wagner's (1972) theory treats the interaction of contexts and CSs in a fashion identical to the treatment of two discrete CSs in compound. ...
In four experiments we investigated whether signaled and unsignaled US presentations resulted in differential context conditioning. Experiments 1 and 2 showed that the presence of a tone during grain presentation facilitated the formation of tone—food associations in pigeons. Experiment 2 also showed that the acquisition of associative value by the tone did not diminish associations between context and the unconditioned stimulus (US). Experiment 3 showed that signaled USs did not interfere with the acquisition of context—US associations, and Experiment 4 showed that even when the signal was extensively pretrained, context—US associations could not be blocked. The results of these experiments are inconsistent with conditioning models that require competition between cues and contexts for associative value.
... The conclusion suggested by blocking and similar findings,256 like overshadowing (Pavlov, 1927) and relative validity (Wag-257 ner et al., 1968), is that what is learned to one of the cues 258on a trial appears to depend not only upon its own current 259 associative value, but also upon the associative value of the 260 other cues present in the trial. This fact is sometimes called 261 stimulus selection or stimulus competition(Wagner, 1969).262 The RW model was the first formal explanation proposed for 263 this competition among CSs. ...
The influence of the Rescorla-Wagner model cannot be overestimated, despite that (1) the model does not differ much computationally from its predecessors and competitors, and (2) its shortcomings are well-known in the learning community. Here we discuss the reasons behind its widespread influence in the cognitive and neural sciences, and argue that it is the constant search for general-process theories by learning scholars which eventually produced a model whose application spans many different areas of research to this day. We focus on the theoretical and empirical background of the model, the theoretical connections that it has with later developments across Marr's levels of analysis, as well as the broad variety of research that it has guided and inspired.
... The model accounted for the known properties of inhibitory learning, including the conditions for its establishment (negative prediction error) and the outcomes of summation and retardation tests. Furthermore, it did so with exactly the same formula as was used to account for excitatory learning and recently discovered stimulus competition effects such as blocking (Kamin, 1969), correlation learning (Rescorla, 1967) and relative validity (Wagner, 1969). The Rescorla-Wagner model has become the definitive model of associative learning and has also been highly influential in the human causal and predictive learning literature (e.g., Gershman, 2015;Shanks & Dickinson, 1987;Van Hamme & Wasserman, 1994). ...
One of the many strengths of the Rescorla and Wagner (1972) model is that it accounts for both excitatory and inhibitory learning using a single error-correction mechanism. However, it makes the counterintuitive prediction that nonreinforced presentations of an inhibitory stimulus will lead to extinction of its inhibitory properties. Zimmer-Hart and Rescorla (1974) provided the first of several animal conditioning studies that contradicted this prediction. However, the human data are more mixed. Accordingly, we set out to test whether extinction of an inhibitor occurs in human causal learning after simultaneous feature negative training with a conventional unidirectional outcome. In 2 experiments with substantial sample sizes, we found no evidence of extinction after presentations of the inhibitory stimulus alone in either a summation test or causal ratings. By contrast, 2 "no-modulation" procedures that contradicted the original training contingencies successfully reversed inhibition. These results did not differ substantially as a function of participants' self-reported causal structures (configural/modulation/prevention). We hypothesize that inhibitory learning may be intrinsically modulatory, analogous to negative occasion-setting, even with simultaneous training. This hypothesis would explain why inhibition is reversed by manipulations that contradict modulation but not by simple extinction, as well as other properties of inhibitory learning such as imperfect transfer to another excitor. (PsycInfo Database Record (c) 2022 APA, all rights reserved).
... Specifically, X was presented in compound with A on the 75% of trials in which it was followed by reward [3AX(1), X(0); where 3 indicates the proportion of trials]. This allowed A to compete with X as a predictor of reward and steal its credit 33 . A's ability to serve as a competitor was further bolstered by continuing to present it by itself followed by reward [A (1)]. ...
A fundamental assumption of learning theories is that the credit assigned to predictive cues is not simply determined by their probability of reinforcement, but by their ability to compete with other cues present during learning. This assumption has guided behavioral and neural science research for decades, and tremendous empirical and theoretical advances have been made identifying the mechanisms of cue competition. However, when learning conditions are not optimal (e.g., when training is massed), cue competition is attenuated. This failure of the learning system exposes the individual’s vulnerability to form spurious associations in the real world. Here, we uncover that cue competition in rats can be rescued when conditions are suboptimal provided that the individual has agency over the learning experience. Our findings reveal a new effect of agency over learning on credit assignment among predictive cues, and open new avenues of investigation into the underlying mechanisms.
... Importantly, we gave a more extensive amount of US devaluation training prior to the devaluation tests than has been done in recent studies. In Experiments 3 and 4 we examined whether CSs that support sign-or goaltracking can mutually compete with each other in a Kamin blocking task (e.g., Kamin 1968) similar to Holland et al. (2014), but under conditions that might be expected to strengthen the blocking effect (Wagner 1969). If sign-tracking rests on an underlying learning system, distinct from goal-tracking, that is primarily subcortical and not expectancy-mediated, then we would expect to see poor US devaluation effects and, possibly, noncompetitive effects in a blocking task when the pretrained and to be blocked stimuli support different response types. ...
When discrete localizable stimuli are used during appetitive Pavlovian conditioning, "sign-tracking" and "goal-tracking" responses emerge. Sign-tracking is observed when conditioned responding is directed toward the CS, whereas goal-tracking manifests as responding directed to the site of expected reward delivery. These behaviors seem to rely on distinct, though overlapping neural circuitries, and, possibly, distinct psychological processes as well, and are thought to be related to addiction vulnerability. One currently popular view is that sign-tracking reflects an incentive motivational process, whereas goal-tracking reflects the influence of more top-down cognitive processes. To test these ideas, we used illness-induced outcome-devaluation and Kamin blocking procedures to determine whether these behaviors rely on similar or distinct underlying associative mechanisms. In Experiments 1 and 2 we showed that outcome-devaluation reduced sign-tracking responses, demonstrating that sign-tracking is controlled by reward expectancies. We also observed that post-CS goal-tracking in these animals is also devaluation sensitive. To test whether these two types of behaviors rely on similar or different prediction error mechanisms, we next tested whether Kamin blocking effects could be observed across these two classes of behaviors. In Experiment 3 we asked if sign-tracking to a lever CS could block the development of goal-tracking to a tone CS; whereas in Experiment 4, we examined whether goal-tracking to a tone CS could block sign-tracking to a lever CS. In both experiments blocking effects were observed suggesting that both sign- and goal-tracking emerge via a common prediction error mechanism. Collectively, the studies reported here suggest that the psychological mechanisms mediating sign- and goal-tracking are more similar than is commonly acknowledged.
... B is a better predictor in group II, while the information value of both stimuli is equal in group III. When A is now separately tested in the three groups, conditioned responding varies in accordance with the manipulated information value (Wagner, 1969). ...
This manuscript is part of a special issue to commemorate professor Paul Eelen, who passed away on August 21, 2016. Paul was a clinically oriented scientist, for whom learning principles (Pavlovian or operant) were more than salivary responses and lever presses. His expertise in learning psychology and his enthusiasm to translate this knowledge to clinical practice inspired many inside and outside academia. Several of his original writings were in the Dutch language. Instead of editing a special issue with contributions of colleagues and friends, we decided to translate a selection of his manuscripts to English to allow wide access to his original insights and opinions. Even though the manuscripts were written more than two decades ago, their content is surprisingly contemporary. The present manuscript was originally published as part of a Liber Amicorum for Paul Eelen’s own supervisor, prof. Joseph Nuttin. In this chapter, Paul Eelen presents a modern view on Pavlovian learning. It appeared in 1980, at the heyday of cognitive psychology which initially dismissed conditioning. Paul Eelen’s perseverance in presenting learning principles as key to study human behaviour has proven correct and ahead of time.
First published as: Eelen, P. (1980). Klassieke conditionering: Klassiek en toch modern. In Liber Amicorum, Prof. J. R. Nuttin, Gedrag, dynamische relatie en betekeniswereld (pp. 321–343). Leuven: Universitaire Pers Leuven.
... Rescorla and Wagner (1972) supposed that contingency variation can be viewed as a situation in which the context acted as a CS, A, which is reinforced when presented in compound with the target CS, X, (AX+), but also when presented by itself (A+). Thus, for instance, as the context acquires associative tendencies, it diminishes that to the CS as a function of the relative frequency of A+ and AX+ trials, in the same fashion as in the relative validity experiments of Wagner (1969) with discrete CSs. This interpretation is further supported by data demonstrating that the contingency variation effect can be reduced by exposing the subjects to the context in the absence of the CS and US (Dweck & Wagner, 1970) and by data indicating that CS-US pairings with shorter inter-trial intervals (ITIs) result in less conditioning to the CS than pairing with longer ITIs, presumably due to the fact that longer ITIs involve more extinction of the context-US association, favouring thus the CS-US association (e.g., Rescorla & Durlach, 1987). ...
The Sometimes Opponent Processes (SOP) model in its original form was especially calculated to address how expected unconditioned stimulus (US) and conditioned stimulus (CS) are rendered less effective than their novel counterparts in Pavlovian conditioning. Its several elaborations embracing the essential notion have extended the scope of the model to integrate a much greater number of phenomena of Pavlovian conditioning. Here, we trace the development of the model and add further thoughts about its extension and refinement.
... In Pavlovian conditioning, overshadowing depends on the relative salience of both overshadowing and overshadowed stimuli (Mackintosh, 1976), on their relative temporal proximity to reinforcement (Revusky, 1971), and on their relative validity (Wagner, 1969a) -i.e., whether the reinforcer is also signalled by other events. This Pavlovian phenomenon is traditionally explained by theories (such as Rescorla & Wagner, 1972, Mackintosh, 1975, and Pearce & Hall, 1980) that appeal to some form of associative competition. ...
In three experiments, a virtual preparation for humans of the Morris water task (VMWT) was used. Experiment 1 established that four landmarks were of similar salience. Then, in Experiments 2 and 3, participants were trained to locate a hidden platform in the presence or either two or four of the previous landmarks. In Experiment 2, one pair of groups was trained with four visual landmarks spaced at equal intervals around the edge of the pool, while a second pair was trained with two landmarks only, either relatively close to or far from the hidden platform. After training, a reciprocal overshadowing effect was found: on a test without the platform with two landmarks only (either close to or far from the platform position), the participants trained with four landmarks spent less time in the platform quadrant than those trained with only two. Finally, Experiment 3 showed that at least participants trained with two landmarks relatively close to the platform and then tested with four also performed worse on test than those trained and tested with two close landmarks only. This result suggests that generalisation decrement, rather than associative competition, could provide a sufficient explanation for the overshadowing observed in Experiment 2 in the proximal groups. The present set of experiments extend, although only partially, the generalisation decrement results documented in rats to human participants.
... Thus, while the Wagner-Rescorla model deals with trial-by-trial changes in associative strengths, and not with steady-state probabilities, and introduces additional parameters reflecting the properties of SN2 and Sc, both Equations 1lb and 12 imply that the change in performance during SN2 is proportional to the difference between asymptotic performance and performance at the outset of the blocking stage. This prediction has been supported with both classical (Kamin, 1969;Rescorla, 1969;Wagner, 1969) and operant (Mackintosh and Honig, 1970;Miles, 1970) procedures. The recent application of the Rescorla-Wagner model to generalization data (Blough, 1975) is consistent with the relationships presented here. ...
A formal statement of a relational principle of reinforcement is developed that makes contact with analyses of choice, interresponse-time distributions, and stimulus control. Some implications for current theoretical and empirical work in the various areas are examined. In this paper, the relational principle of reinforcement proposed by Premack (1959, 1965) is given a somewhat more formal statement that more explicitly acknowledges the role of the stimulus. This formalization of the reinforcement principle is shown to be consistent with the theoretical analysis of a number of diverse phenomena-including choice behavior , interresponse-time distributions, and the blocking of stimulus control. Relational Principle of Reinforcement Consider the following specific illustration of operant conditioning as a means of introducing the necessary terminology: a pigeon is deprived of food and is placed in an experimental chamber containing a response key and a food hopper, from which mixed grain may be made available. In the presence of the stimulus of the key, the key-pecking response may freely occur in the absence of any contingency imposed by the experimenter. The stimulus of the key is referred to as the noncontingent stimulus (SN) and the response of key pecking as the noncontingent response (RN). Conditioning is instituted when the stimulus of the grain, which stimulus controls pecking, is made contingent on a key-pecking response. The stimulus of the grain is termed the contingent stimulus (Sc) and pecking the grain is termed the 'Preparation of this paper was supported by a grant from the U.S. Public Health Service, MH-17395. For their comments on an earlier version of the manuscript, appreciation is expressed to John contingent response (Rc). When the contingency between key pecking and the stimuli controlling eating is instituted, the frequency of key pecking is observed to increase and the key-pecking response may be said to have been reinforced. According to Premack, a general statement of the events critical to the occurrence of reinforcement is as follows: in the presence of non-contingent stimuli (SN), a noncontingent response (RN) increases in probability if RN is followed by more preferred contingent stimuli (Sc) which control a second response (Rc) and if the organism has been deprived of the contingent response (Premack, 1965). Note that within the context of Premack's formulation, reinforcement is not a property of either a stimulus or a response but of a relationship between two successive elicitation processes, i.e., SN-RN and Sc-Rc (cf. Catania, 1971; Morse and Kelleher, in press). The preference for an elicitation process is defined as the proportion of time that an organism exposes itself to the stimuli that control the response when given free access to the controlling stimuli under baseline conditions which are otherwise identical to the conditions prevailing when the contingency is present. The preference for an elici-tation process is most conveniently measured by the probability (pi) of the response controlled by the eliciting stimulus, and may be defined as m(t1) n > m(t,) 1=1 (1) where m is an appropriate measure of the time, (ti) spent engaging in Ri when there are 341 1977, 27., 341-350 NUMBER 2 (MARCH)
... Sin embargo, aún a pesar de la similitud entre ambos procedimientos en cuanto a sus resultados, el ensombrecimiento descubierto por Pavlov (1927) fue relativamente ignorado posteriormente a su descubrimiento, pero actualmente ha influenciado ampliamente el desarrollo de teorías asociativas (Dickinson, 1980). Siendo uno de los fenómenos más robustos y de mayor generalidad en el estudio del condicionamiento Pavloviano, ya que se ha demostrado en diversas especies empleando diferentes preparaciones de condicionamiento Pavloviano como la supresión condicionada (Mackintosh, 1976), la aversión condicionada al sabor (Kreamer, Lariviere & Spear, 1988), el condicionamiento palpebral en conejos (Wagner, 1969), así como en el aprendizaje espacial (Prados 2011) por mencionar algunas (ver: Vila, Bernal & Monroy, 2016). Lo que contrasta con los trabajos derivados de la aproximación operante, en la cual el estudio de Reynolds (1961) solo ha sido replicado directamente con palomas en pocas ocasiones (Farthing & Hearts, 1970;Wilkie & Mason, 1976;Kendall & Mills, 1979;Vyazovska et al., 2014) y en un solo estudio con participantes humanos empleando una respuesta de escape discriminado (Vila & Monroy, 2015). ...
A study with human participants is presented, where the effect of two retention interval values (0 and 24 h) after a discriminated escape task was evaluated. An escape response was learned using a virtual task where participants located a hidden goal within a time limit with a compound Ed, with two components presented simultaneously (geometric, CG and form, CF). At the end of training, a test with the CF alone was presented at 0 or 24h. The results showed that all participants learned the task. During the test, the group that received training with the compound Ed and was tested at 0h presents a smaller number of responses to CF, while a group training with the compound Ed and tested at 24h test showed a greater number of responses. These data showed a recovery of the escape responses to the CF after a retention interval. Results were analyzed in the theoretical framework of Pavlovian overshadowing and its similarity to the original data obtained by Reynolds (1961).
... In both his learning texts, The Psychology of Animal Learning (1974) and Conditioning and Associative Learning (1983), Nicholas Mackintosh argues that common associative learning processes underlie Pavlovian and instrumental conditioning on the basis of the empirical commonalties between the different forms of learning. For example, in a series of experiments, Wagner demonstrated that when a target cue is trained in compound with another stimulus, the amount of conditioning accruing to the target depends upon its validity as a predictor of the reinforcer relative to the other stimulus (Wagner, 1969;Wagner, Logan, & Haberlandt, 1968). Mackintosh then notes that instrumental conditioning of wheel running shows comparable sensitivity to the relative validity of this response as a predictor of a food reinforcer (Mackintosh & Dickinson, 1979). ...
Associative learning theories regard the probability of reinforcement as the critical factor determining responding. However, the role of this factor in instrumental conditioning is not completely clear. In fact, a wealth of evidence from instrumental free-operant experiments has shown that participants respond at a higher rate on variable ratio than on variable interval schedules even though the reinforcement probability on the interval schedule is the same as or greater than that on the ratio schedule. This difference has been attributed to the differential reinforcement of long inter-response times (IRT) by interval schedules, which acts to slow responding. In the present study, we used a novel experimental design to investigate human responding under random ratio (RR) and regulated probability interval (RPI) schedules, a type of interval schedule that sets a reinforcement probability independently of the IRT duration. We trained participants separately on each type of schedule before a final choice test in which they distributed responding between two schedules similar to those experienced during training. Although response rates on the various schedules did not differ reliably during training, the participants responded at a lower rate on the RPI schedule than on the matched RR schedule during the choice test. This preference cannot be attributed to a higher probability of reinforcement for long IRTs and questions the idea that similar associative processes underlie instrumental and classical conditioning.
... This simply requires that two discriminative clauses be added to the production. Responding to such configural cues was a problem for some of the earlier discrimination theories (see Rudy & Wagner, 1975 for a review). The power of the ACT theory over these early theories is that productions can respond to pattern of elements rather than to each element separately. ...
Abstract
Anderson (1982) undertakes highly focused research on the rationale based on cognitive theories of learning styles and designs a cyclical process of learning. Accordingly, he points out that a framework for skill acquisition is proposed in which there are two major stages in the development of a cognitive skill, a declarative stage in which facts about the skill domain are interpreted and a procedural stage in which the domain knowledge is embodied directly in procedures for performing the skill. This general framework has been instantiated in the ACT system in which facts are encoded in a propositional network and procedures are encoded as productions. Two types of interpretive procedures are described for converting facts in the declarative stage into behavior--general problem-solving procedures and analogy-forming procedures. Knowledge compilation is the process by which the skill transits from the declarative stage to the procedural stage. It consists of the subprocesses of composition which collapses sequences of productions into single productions and proceduralization which embeds factual knowledge into productions. Once proceduralized, further learning processes that operate on the skill to make the productions more selective in their range of applications. These learning processes include generalization, discrimination, and strengthening of productions. Comparisons are made to similar concepts from past learning theories. It is discussed how these learning mechanisms apply to produce the power law speed-up in processing time with practice. Much of the evidence for this theory of skill acquisition comes from work on acquisition of proof skills in geometry but other evidence is drawn from the literature on automatization, language acquisition, and category formation.
Key words: Interpretive procedures, Proof skills, Category formation, Knowledge compilation, Language acquisition, Power law, Strengthening.
... We believe that this training procedure reduced the total amount of associative strength that F could have gained (for the same result with nonhuman subjects see Rodrigo et al, 2005). This procedure was inspired by the notion of relative validity (Wagner, 1969). According to this notion, a more valid cue, one which better predicts the occurrence of reinforcement, will overshadow a less valid one. ...
In two experiments in a virtual pool the participants were trained to find a hidden platform placed in a specific position in relation to one (Experiment 1) or two (Experiment 2) objects; then, all the participants received a test trial, without the platform, and the time spent in the segment where the platform should have been was measured. In Experiment 1, groups differed in the distance between the landmark and the hidden platform. Test results showed that the control acquired by the landmark was different depending on its relative distance from the platform: Closer landmarks acquired a better control than distant ones. In Experiment 2, two objects, B and F, were simultaneously present during acquisition. Object B was just above the hidden platform (i.e., a beacon for the platform) while object F was above the edge of the pool (i.e., a frame of reference). On the test, the spatial location of B in relation to F was manipulated in the different groups and a generalization gradient was found: Participants spent more time in the segment where B was when B was in front of F (training position), and this time decreased symmetrically with distance of B from F. The two experiments provide convergent evidence of spatial learning effects in a virtual task with humans.
... Rescorla and Wagner (1972) proposed a theory of Pavlovian conditioning in which a stimulus becomes conditioned to a US according to the degree to which the US was 'unexpected' on a trial on which that stimulus was presented. This approach differs from that taken by Spence, and is able to account for other findings that are problematic for such a theory, e.g. that a stimulus that is presented in conjunction with a better predictor in a feature positive (A0 | AB+) design will, after a period of discrimination training, elicit little responding (Wagner, 1969). Blough (1975) showed that with the addition of a further assumption, namely that experimental stimuli are composed of a set of independent features, all of which are capable of entering into associations, the Rescorla-Wagner theory is able to account for increased generalisation between more similar stimuli, and for the peak shift. ...
In a series of experiments, both pigeon and human subjects were trained to categorise two groups of confusable stimuli, with each category being made up of distortions of a ‘Prototype’. Once the subjects had successfully learned to categorise the training stimuli, they were tested on their responding to a variety of previously unseen stimuli: these were distortions of the Prototypes towards (‘Closer’ exemplars), or away from (‘Further’ exemplars), the other category, and the Prototypes themselves. Pigeons responded more to positive Further exemplars that were close to the Prototype than to the Prototype itself, or to exemplars even further away from the category boundary. This result is an example of the peak shift (Hanson, 1959), and can be explained by interacting excitatory and inhibitory generalisation gradients (Spence 1937).
When the pigeons were autoshaped using stimuli from the positive category before learning the categorisation, they failed to show a peak shift; greatest response rates on test were elicited by the positive Prototype. This result could be explained by the interaction of the autoshaping producing a ‘prototype effect’, i.e. a generalisation gradient with a maximum at the Prototype of the positive category, which masks the development of the peak shift. Further experiments showed that a similar abolition of the peak shift occurred when the pigeons were given prior experience of the negative category in an extra-dimensional discrimination designed to produce an inhibitory analogue of the prototype effect.
A connectionist model of categorisation learning is presented, based on representation of the stimuli as sets of independent features. Simulations conducted using this model showed that, with few assumptions, such an analysis was capable of accounting for all the results found with pigeon subjects, some of which present a problem for alternative instance theories of categorisation (e.g. Pearce, 1984).
Human subjects also categorised the Further exemplars better than the Prototypes, but did not show a peak shift. Performance increased with greater distance from the category boundary, consistent with subjects having abstracted and applied a cognitive strategy. When trained in an incidental learning paradigm, designed to minimise the opportunity for using such a strategy, subjects showed evidence of learning without any knowledge of the categorisation ‘rule’. The performance in this ‘implicit’ task had some similarities to the results of the studies with pigeon subjects, suggestive of a peak shift. These results indicate that similar associative processes may underlie categorisation in both humans and non-humans, although higher-level ‘symbolic’ processes may control human performance in laboratory studies.
... Although this pattern of results is puzzling with respect to frequency or novelty principles, it is at least partially consistent with a conceptualization growing out of a body of research on animal conditioning (e.g., Rescorla & Wagner, 1972;Rudy & Wagner, 1975). The main idea is that component cues associated with some outcome compete to predict that outcome. ...
This article is concerned with the use of base-rate information that is derived from experience in classifying examples of a category. The basic task involved simulated medical decision making in which participants learned to diagnose hypothetical diseases on the basis of symptom information. Alternative diseases differed in their relative frequency or base rates of occurrence. In five experiments initial learning was followed by a series of transfer tests designed to index the use of base-rate information. On these tests, patterns of symptoms were presented that suggested more than one disease and were therefore ambiguous. The alternative or candidate diseases on such tests could differ in their relative frequency of occurrence during learning. For example, a symptom might be presented that had appeared with both a relatively common and a relatively rare disease. If participants are using base-rate information appropriately (according to Bayes’ theorem), then they should be more likely to predict that the common disease is present than that the rare disease is present on such ambiguous tests. Current classification models differ in their predictions concerning the use of base-rate information. For example, most prototype models imply an insensitivity to base-rate information, whereas many exemplar-based classification models predict appropriate use of base-rate information. The results reveal a consistent but complex pattern. Depending on the category structure and the nature of the ambiguous tests, participants use base-rate information appropriately, ignore base-rate information, or use base-rate information inappropriately (predict that the rare disease is more likely to be present). To our knowledge, no current categorization model predicts this pattern of results. To account for these results, a new model is described incorporating the ideas of property or symptom competition and context-sensitive retrieval.
... Compounds as unitary events. A different modification to associative models to incorporate trial order effects is to acknowledge a compound stimulus as a unitary event (e.g., Rescoria, 1981;Rescorla & Durlach, 1981;Rudy & Wagner, 1975). Shanks and Dickinson (1987), for example, proposed a model for human contingency judgment that treated a compound (whether of two CSs or of a CS and the context) as a single event with its own associative strength, which was independent of that attached to the components. ...
The study of the mechanism that detects the contingency between events. in both humans and non-human animals, is a matter of considerable research activity. Two broad categories of explanations of the acquisition of contingency information have received extensive evaluation: rule-based models and associative models. This article assesses the two categories of models for human contingency judgments. The data reveal systematic departures in contingency judgments from the predictions of rule-based models. Recent studies indicate that a contiguity model of Pavlovian conditioning is a useful heuristic for conceptualizing human contingency judgments.
... Since A. R. Mattocks (cited in Underwood & Schulz, 1960) observed that college students often select only the first letter of nonsense trigrams for use as their effective stimuli in paired-associate (PA) learning, the "redundant-trigram" model of PA learning has enjoyed great popularity in investigations of the process of stimulus selection (e.g., for adult Ss, Berry, Detterman, & Mulhern, 1973;Davis, Brown, & Ritchie, 1968;Lovelace & Blass, 1968;Lovelace & Greenberg, 1969;Postman & Greenbloom, 1967;Rabinowitz & Witte, 1967: Richardson, 1972, and for child and/or retarded Ss, Baumeister & Berry, 1970Berry & Baumeister, 1973: Berry, Joubert, & Baumeister, 1971Rabinowitz & McClinton, 1971). It is pOSSible, however, that the redundant-trigram research model is not as useful as an analogous model based on triads of unrelated words. ...
College Ss acquired paired associates (PAs) consisting of redundant trigrams (trigrams with no repeated letters) paired with single-digit responses, whereupon they were tested for single-letter selection and quizzed as to how they had learned each PA. The purpose of the study was to determine whether or not college Ss process redundant trigrams on the basis of their initial characteristics to any significant extent. The results showed that a significant number of the trigrams were indeed processed on the basis of the initials they spelled out. It was argued that the occurrence of initial selection makes the redundant-trigram model a less useful research model than was heretofore supposed.
... The origin of this difference is not difficult to discern. In the Orme-Johnson and Yarczower study and the present work, the pigeon's peck bore a more immediate and reliable relationship to shock than the visual SD and thus the former association overshadowed (Miles & Jenkins, 1973;Wagner, 1969) the latter. But the opposite can also occur. ...
Two pigeons keypecked on a multiple schedule of discriminated punishment. A white line on a green surround was associated with variable interval reinforcement and electric shock following each response, and the green surround alone was associated only with variable-interval reinforcement. When the white line was on, the key responding was suppressed. Probe presentations of the white line alone and the punishment contingency alone, and sessions under a mixed schedule of punishment without the white line, showed that response dependent shock controlled responding to a much greater degree than its putative discriminative stimulus. The results are seen as an instance of overshadowing of a correlation between a stimulus and shock by a more immediate and reliable correlation between a response and shock.
... It has long been suggested that there are two systems for associating external stimuli with reinforcement (Rudy and Wagner, 1976;Fanselow, 1999): one system, the elemental learning system, is responsible for associating an elemental stimulus such as a single tone or light with a reinforcer (Estes, 1950;Rescorla and Wagner, 1972) and the other, the configural or contextual system, is responsible for associating a more complex stimulus, such as a visual scene in the background, with a reinforcer Fanselow, 1999). Whether the two learning systems always compete with each other or function in parallel is still debatable although it has been suggested that the existence of a contextual stimulus prevents an elemental stimulus from controlling behavior (Fanselow, 1999). ...
Learning theories categorize learning systems into elemental and contextual systems, the former being processed by non-hippocampal regions and the latter being processed in the hippocampus. A set of complex stimuli such as a visual background is often considered a contextual stimulus and simple sensory stimuli such as pure tone and light are considered elemental stimuli. However, this elemental-contextual categorization scheme has only been tested in limited behavioral paradigms and it is largely unknown whether it can be generalized across different learning situations. By requiring rats to respond differently to a common object in association with various types of sensory cues including contextual and elemental stimuli, we tested whether different types of elemental and contextual sensory stimuli depended on the hippocampus to different degrees. In most rats, a surrounding visual background and a tactile stimulus served as contextual (hippocampal dependent) and elemental (non-hippocampal dependent) stimuli, respectively. However, simple tone and light stimuli frequently used as elemental cues in traditional experiments required the hippocampus to varying degrees among rats. Specifically, one group of rats showed a normal contextual bias when both contextual and elemental cues were present. These rats effectively switched to using elemental cues when the hippocampus was inactivated. The other group showed a strong contextual bias (and hippocampal dependence) because these rats were not able to use elemental cues when the hippocampus was unavailable. It is possible that the latter group of rats might have interpreted the elemental cues (light and tone) as background stimuli and depended more on the hippocampus in associating the cues with choice responses. Although exact mechanisms underlying these individual variances are unclear, our findings recommend a caution for adopting a simple sensory stimulus as a non-hippocampal sensory cue only based on the literature.
Four experiments using 196 albino rabbits sought to determine whether in conditioning to a serial compound, CS1–CS2–UCS, there are (a) associative mechanisms operating to extend conditioning beyond the bounds of a CS–UCS contiguity gradient and (b) stimulus selection processes acting to attenuate the potency of CS–UCS contiguity. In Exps I and II the CS2–UCS interval was held at .35 sec while the CS1–UCS interval was varied across groups from .75 to 2.75 sec. CS1 test trials revealed substantial CR acquisition at all CS1–UCS intervals. Moreover, Exp II indicated that, when the contribution of cross-modal generalization from CS2 to CS1 was factored out, there still remained a substantial level of conditioning, which Exp III indicated was attributable to an associative mechanism like higher-order or sensory conditioning. The observation of CR acquisition at CS1–UCS intervals of 4.75, 8.75, and 18.75 sec in Exp IV suggested that serial compound training yields conditioning to CSs located well beyond the single CS contiguity gradient for the rabbit's nictitating membrane response. Exps I and II also indicated the presence of stimulus selection processes, because at the shorter CS1–UCS intervals, the levels of test-trial responding to CS2 fell below those observed to the less contiguous CS1. (36 ref)
The model elaborated here adapts the influential pooled error term, first described by Wagner and Rescorla (Rescorla & Wagner, 1972; Wagner & Rescorla, 1972), to govern the formation of reciprocal associations between any pair of stimuli that are presented on a given trial. In the context of Pavlovian conditioning, these stimuli include various conditioned and unconditioned stimuli. This elaboration enables the model to deal with cue competition phenomena, including the relative validity effect, and evidence implicating separate error terms and attentional processes in association formation. The model also includes a performance rule, which provides a natural basis for (individual) variation in the strength and nature of conditioned behaviors that are observed in Pavlovian conditioning procedures. The new model thereby begins to address theoretical and empirical issues that were apparent when the Rescorla–Wagner model was first described, together with research inspired by the model over the ensuing 50 years.
It has been assumed that the formation of S-R links, whereby stimuli previously experienced contiguously with the execution of a response come to elicit that response, is either sufficient to explain social learning, or, particularly in the case of observational learning, that it is inadequate to do so. Contemporary learning theory conceives of learning rather differently; as the formation and association of mental representation. It is argued in this thesis that the problems which affected learning theory in the behaviourist era, and those which arise in trying to understand social learning, are similar. Therefore, it is reasoned, social learning theory might escape its difficulties by understanding learning to involve the cognitive processes that contemporary associative learning theory proposes. The purpose of the empirical work in this thesis was to develop paradigms to investigate the extent to which social learning can be encompassed by contemporary associative learning theory. In particular, procedures were designed to expose observers to the conditions necessary for learning in isolated animals, and it was assumed that if social learning effects depend on these same conditions, it would suggest that they are mediated by the same associative mechanisms. In particular, evidence was sought that, like asocial learning, social learning is: 1) contingency dependent (Chapter 3); 2) subject to blocking and overshadowing (Chapter 4); 3) sensitive to both S-S and R-O relationships (Chapter 5); and 4) subject to the effects of stimulus pre-exposure (Chapter 6). In each case, the associative predictions were, to an extent, supported. However, it is concluded that in order to establish whether a particular socially learned behaviour is controlled by an associative structure involving mental representations, additional experiments are required to rule out S-R learning.
When repeatedly exposed to simultaneously presented stimuli, associations
between these stimuli are nearly always established, both within as well as
between sensory modalities. Such associations guide our subsequent actions
and may also play a role in multisensory selection. Thus, crossmodal
associations (i.e., associations between stimuli from different modalities)
learned in a multisensory interference task might affect subsequent information
processing. The aim of this study was to investigate the processing level of
multisensory stimuli in multisensory selection by means of crossmodal
aftereffects. Either feature or response associations were induced in a
multisensory flanker task while the amount of interference in a subsequent
crossmodal flanker task was measured. The results of Experiment 1 revealed
the existence of crossmodal interference after multisensory selection.
Experiments 2 and 3 then went on to demonstrate the dependence of this effect
on the perceptual associations between features themselves, rather than on the
associations between feature and response. Establishing response associations
did not lead to a subsequent crossmodal interference effect (Experiment 2),
while stimulus feature associations without response associations (obtained by
changing the response effectors) did (Experiment 3). Taken together, this
pattern of results suggests that associations in multisensory selection, and the
interference of (crossmodal) distractors, predominantly work at the perceptual,
rather than at the response, level.
2 types of correlation between reinforcement and the elements of a compound stimulus are described, as well as an experiment demonstrating their independent effects. Pigeons were trained to peck at compound line (0° vertical) and color (C1 or C2, red or green) stimuli. The Correlated (true discrimination) group was reinforced for responses to 0°C1 but not for responses to 0°C2; the Uncorrelated (pseudodiscrimination) group was reinforced for responses to both compounds; and the Control group was exposed only to 0°C1 with reinforcement. In a subsequent test the frequency of responding to the 0° element presented alone was greatest in the Uncorrelated group, intermediate in the Control group, and lowest in the Correlated group. However, the slopes of the relative line-tilt generalization gradients were not significantly different for the three groups.
The experiments described in this section are concerned with a behavioral analysis of the various permutations and combinations of classical and operant conditioning schedules. For the present purposes, a schedule describes the interrelationship between three primary events: environmental stimuli, responses, and reinforcer delivery. An operant conditioning schedule describes the stimuli in which a reinforcing event is delivered following and contingent upon the occurrence of the recorded response (Ferster and Skinner, 1957). Dependent variables in operant conditioning are the rate, duration, force, and latency of the recorded response (Skinner, 1950; Premack, 1965). A classical conditioning procedure describes the conditional relationship between an environmental stimulus (conditioned stimulus, CS) and the subsequent occurrence of an unconditionally reinforcing stimulus (unconditioned stimulus, UCS) that reliably elicits a recorded response (unconditioned response, UCR) (Pavlov, 1927). The frequency, duration, and latency of responses elicited by the CS, not the UCS, are the primary dependent variables in classical conditioning schedules.
Two experiments with 72 undergraduates investigated attraction to a stranger, which has been previously shown to depend on the social context within which that directed action is acquired. In both experiments, a confederate stranger agreed with the S 100% of the time. Attraction to the stranger, however, was blocked if the stranger agreed with the S while in the presence of someone the S already found attractive. Based on these findings, the present authors employed N. E. Miller's (1959) "extension of liberalized stimulus–response theory" in developing a reinforcement-context theory to predict acquisition and blocking effects in attraction. Additional novel predictions addressing contextual phenomena are generated from the theory. (45 ref)
Examined the conditions under which presenting a signal for reinforcement decreases or increases the rate of leverpressing in rats. Response rate on a variable-interval (VI) schedule of reinforcement was decreased by a brief signal for reinforcement. In Experiment 1, requiring 1 short IRT to occur on completion of the VI requirement attenuated this response decrement. Requiring three responses to be emitted within a minimum amount of time at the end of the VI resulted in the reinforcement signal's enhancement of response rates. Experiment 2 replicated this increase in responding and showed that the reinforcement signal attenuated response rates on a simple VI schedule that yielded the same overall rate of reinforcement. In Experiment 3, the reinforcement signal enhanced response rates when 3 responses were required at the end of a VI schedule, but the signal attenuated response rates when the 3 responses could occur at any time in relation to the VI. These results suggest that the pattern of responding emitted immediately prior to reinforcement is a critical factor in determining the effect of reinforcement signal on response rate.
Models of associative learning have proposed that cue-outcome learning critically depends on the degree of prediction error encountered during training. Two experiments examined the role of error-driven extinction learning in a human causal learning task. Target cues underwent extinction in the presence of additional cues, which differed in the degree to which they predicted the outcome, thereby manipulating outcome expectancy and, in the absence of any change in reinforcement, prediction error. These prediction error manipulations have each been shown to modulate extinction learning in aversive conditioning studies. While both manipulations resulted in increased prediction error during training, neither enhanced extinction in the present human learning task (one manipulation resulted in less extinction at test). The results are discussed with reference to the types of associations that are regulated by prediction error, the types of error terms involved in their regulation, and how these interact with parameters involved in training.
Kapitel 7 beschreibt die aus verschiedenen Quellen, aber vor allem aus der behavioristischen Schule hervorgegangenen Erkenntnisse und Theorien über den Vollzug des Lernens, des klassischen und operanten Konditionierens und des Imitationslernens sowie die Einflüsse von Kognition und Einsicht. Besprochen werden u. A. die Prinzipien von Chaping und Chaining sowie die Konditionierung autonomer Funktionen. Hinzu kommen die neueren Erkenntnisse über komplexes und kognitives Lernen, Einsichtlernen, implizites Lernen sowie latentes Lernen.
Psychologists generally accept that simple associative learning processes are among those that are fundamental in enabling organisms to extract meaning about predictive event relationships in the environment, and in controlling adaptive modes of behavior. This chapter concerns with the issue of identifying critical conditions for Pavlovian learning to take place. It then discusses studies of Pavlovian learning. The chapter explains intensity and novelty of conditioned stimulus (CS) and unconditioned stimulus (US). It explores the major variables determining Pavlovian excitatory conditioning. The chapter also discusses psychological principles of Pavlovian learning and its limitations. It reviews some of the major empirical findings and some of the major theoretical principles generally assumed to account for many of these findings. The chapter provides an overview of some of the neural mechanisms that support Pavlovian conditioning. It considers the main processing pathways involved in prediction error coding in neural circuit for Pavlovian learning.
Phrenology, that age old study of the phrenes, continues to be the accepted premise of contemporary psychology—unfortunately. The names and numbers of phrenes have been hotly debated and frequently changed across the decades. The appropriate measuring device is still at issue. Yet, the fundamental concept of physical dimensions as definitive measures of hypothetical entities and causes has not only been unchanged, but assimilated into all current psychologies. In the beginning, mental states were measured by the shape of the skull. More recently, physical behavior has supplanted physical structure as the measuring tool. With some irony, Pavlov’s rejection of the “fantastic states” and mentalisms has been ignored, and Pavlovian conditioning is now offered as a premier technique for measuring drives, fears, helplessness, motivation, etc. Similarly, the straightforward assertion by Skinner that behavior is only behavior has also been compromised, with once “radical” behaviorists now promoting overt responses as indices of preferences, memories, inhibitions, values, general emotional states, etc. Within the phrenological systems, behavior as a dependent variable is transformed into behavior as an index or indicant, an intrinsically trivial but convenient epiphenomenon to measure the more interesting workings of the mind (Ebel, 1974).
The hypothesis of distinct memory codes proposes that entities belonging to global categories as verbal and spatial representations, or to more specific categories as faces, objects, living or non-living entities, are stored and reactivated in functionally separate memory partitions. In this chapter we will summarize evidence which supports this principle of code-specificity in memory. First, we will briefly review the empirical roots which can be found in Cognitive Psychology, Experimental Neuropsychology, and Clinical Neuropsychology. We will then outline a general neuroscientific theory which explains why code-specific memory representations do most likely exist, and finally, in the main part of the chapter, we will give an overview over recent brain findings that are highly consistent with the idea of code specific storage and retrieval within topographically distinct neural networks.
The tradition founded by Pavlov and Thorndike is motivated by the conviction that certain pervasive phenomena of learning are captured in two paradigms; the one exemplified by Pavlov’s experiment in salivary conditioning, the other by Thorndike’s experiments with cats in a puzzle box. The Pavlovian experiment arranges a relation between two events, a CS and US, independently of behavior. It is directed toward understanding what Pavlov referred to as the process of interchangeable signification which he believed rested equally on the conditioning of inhibition and excitation. The Thorndikian or operant conditioning experiment arranges a relation between behavior and an event, or a response and reinforcer. It is directed toward understanding the shaping and strengthening of action through reward, the suppression and redirection of action through punishment, and the development of stimulus control over punished and reinforced actions.
The capability of the central nervous system (CNS) to adapt its functional and structural organization to current requirements is known as neural plasticity. Such changes can be examined at different organizational levels of the CNS; changes at the molecular-, synaptic-, neural-, system-, and behavioral level are mutually dependent (Shaw & McEachern, 2001). Plastic changes are triggered by learning, e.g., perceptual and motor training and by injuries, e.g., a deafferentation of parts or of all afferents of a sensory system. Moreover, the capacity to change is a characteristic feature of the CNS throughout life although there are qualitative and quantitative differences between developmental and adult plasticity. This chapter reports major findings on training- and lesion-induced plasticity. Results from animal and human research in the somatosensory, auditory, visual and motor system are reviewed and the possibly mechanisms underlying brain plasticity are discussed. Moreover, possible differences between developmental and adult plasticity are considered.
This chapter presents an introductory-level description of the sometimes opponent process (SOP), which is an abstract model of animal learning and performance that was developed in an attempt to integrate conceptually the principal behavioral regularities of Pavlovian conditioning, emphasizing the essence of its propositions and some of the fundamental regularities that it addresses. SOP is a connectionist model—that is, the knowledge system is conceived as a set of nodes that take on activity states and are capable of influencing each other according to the excitatory and/or inhibitory connections obtained among them. Performance is attributed to a presumed tendency of activity in certain nodes to drive overt behavior, and the acquisition of conditioned responding is interpreted in terms of a presumed change in the way some nodes do this via alterations in the linkages that they have with other nodes. The chapter reviews the current knowledge concerning the essential neural circuitry involved in eyeblink conditioning in the rabbit.
Computational theories of classical conditioning are theories whose propositions are stated as mathematical relationships. From such propositions one can compute, that is to say, deduce, presumed consequences for conditioned responding in circumstances addressed by the theory. This chapter is an attempt to roughly categorize and briefly summarize the major computational theories that have been developed to understand behavior in classical conditioning. It is organized around the essential ideas about conditioning that the various theories embrace. This chapter is not intended to provide a relative evaluation of the theories mentioned; to do so is considerably beyond the scope of what can be accomplished in the space available. Data are mentioned and judgments made, but only to provide understanding of the inspiration for the different models.
Administering unsignaled USs during daily CER training sessions interfered with CER conditioning, as has frequently been reported. This effect was reduced, however, when additional daily sessions were administered during which Ss were simply exposed to the experimental environment in the absence of the CS and US. The results indicate that S’s treatment with respect to “situational” cues is important in the determination of CS-US contingency effects, and are in agreement with recent formulations of Wagner (in press, a) and Rescorla (in press) which emphasize that the degree of conditioning to a CS depends upon the associative strength of the constellation of cues in which the CS is imbedded during training.
Twelve rats were given a fixed number of inescapable electric shocks in a T-maze. Under these conditions, Ss chose to run to the end of the arm containing another rat and engaged in the stereotyped shock-induced fighting. This choice supports previous studies showing that the opportunity for aggression is a reinforcement of behavior of Ss receiving inescapable shock.
Pigeons were trained in an operant, go/no-go, discrimination by successive presentations of discrete, positive and negative trials. In separate groups, the rate of discriminaion learning based on an auditory cue, on visual cues of different discriminability, and on combined auditory and visual cues was determined. The auditory cue was tone/no-tone, the visual cue was a difference in brightness level on the key. Following discrimination training, stimulus control was tested in extinction by presenting tone or notone at each of four key-brightness levels.Learning was more rapid with two cues than with either cue singly, demonstrating summation. The contribution made by the tone cue to the learning of the discrimination decreased with increasing discriminability of the light cue. The control exerted by the tone cue also decreased (i.e., the tone cue was overshadowed) with increasing discriminability of the light cue. In certain cases, the tone cue decreased control by brightness, showing that in two-cue discriminations each cue may reduce the control exerted by the other.
Rats were trained on three different responses, and then one of the two reinforcers of each response was devalued by pairings with lithium chloride. During instrumental training, a brief stimulus consistently signaled one of the reinforcers for one of the responses and the alternate reinforcer for one of the other responses. Neither reinforcer was signaled in the case of the third response. The major finding of the experiment is that the effect on subsequent extinction responding of post-conditioning devaluation of one of two reinforcers of a response depends on whether one of the two reinforcers was signaled during training and, if so, whether the reinforcer signaled was the one subsequently devalued. Relative to the level of performance of a response where neither reinforcer was signaled during training, the level of performance of a different response, where one of the two reinforcers was signaled during training, was higher if the signaled reinforcer was the one later devalued and lower if it was the unsignaled reinforcer later devalued.
Two studies are reported which are concerned with the role of the CS and its relationship to the CR in the standard avoidance learning situation. Study I demonstrates that the decrement associated with a trace CS is greatly reduced in one-way avoidance as compared with two-way. The suggestion is made that the differences in CS effects is attributable to differences in task complexity. Study 2 varied CS duration and the CS-CR termination contingency. Rapid acquisition of the avoidance response was obtained only when the CS remained on for the entire CS-US interval unless terminated by a response. The results were discussed with respect to the theoretical aspects of maintenance and emergence of the avoidance response.
ResearchGate has not been able to resolve any references for this publication.