Associative learning and representation: Introduction

The papers published in this Special Issue are based upon presentations at a workshop on “Associative Learning and Representation”, which was sponsored by the Experimental Psychology Society and hosted by Emmanuel College, Cambridge. The workshop celebrated the contribution of Professor Nicholas Mackintosh to animal learning and conditioning in particular and experimental psychology in general in the year of his retirement from the Chair of Psychology at the University of Cambridge after 21 years in post. The date of the workshop, 9 July 2002, was particularly auspicious as it was the day of Professor Mackintosh's birth 67 years ago. Moreover, it is particularly fitting that this tribute is published in the Comparative and Physiological Psychology Section (B) of Quarterly Journal of Experimental Psychology, which he founded in 1981 during his editorship of the Journal between 1977 and 1984.     

Associative learning of discrimination with masked stimuli

Great controversy exists on whether associative learning occurs without awareness. In Experiment 1, 31 participants received discrimination training by repeated presentations of two stimulus sequences (S1_A → S2_A, and S1_B → S2_B), S1 being a masked stimulus. S2 were imperative stimuli for a reaction time (RT) task. After the acquisition phase, all participants were tested with 20 presentations of congruent (S1_A → S2_A and S1_B → S2_B) and incongruent (S1_A → S2_B and S1_B → S2_A) stimulus sequences. The RT in the testing phase was faster in congruent than in incongruent stimulus sequences. These results are considered strong evidence of associative learning without awareness of the contingency between the stimuli. A second experiment was designed with SOA varied between three groups (23, 58, and 117 ms). The results showed that the participants responded more quickly to congruent stimulus sequences and that the SOA did not affect RT. The SOA did not modify the effect of congruence either, although the interaction was near significance.     

Generalization gradients of inhibition following auditory discrimination learning

A more direct method than the usual ones for obtaining inhibitory gradients requires that the dimension of the nonreinforced stimulus selected for testing be orthogonal to the dimensions of the reinforced stimulus. In that case, the test points along the inhibitory gradient are equally distant from the reinforced stimulus. An attempt was made to realize this condition by obtaining inhibitory gradients along the frequency dimension of a pure tone after discrimination training in which the nonreinforced stimulus was a pure tone (or tones), and the reinforced stimulus was either white noise or the absence of a tone. The results showed that some degree of specific inhibitory control was exerted by the frequency of the tone, although the gradients were broad and shallow in slope.

A further experiment was conducted to see whether the modification of an excitatory gradient resulting from training to discriminate neighboring tones could arise from a simple interaction of inhibitory and excitatory gradients. The results indicated that it could not, since discrimination training produced a concentration of responding in the vicinity of the reinforced stimulus which cannot be derived from any plausible gradient of inhibition.

     

Associative interference with taste aversions after contextual discrimination learning

In two experiments, rats were first given discriminative training with two distinctive contexts, such that a flavor was paired with lithium chloride (LiCl) in one context, alternating with presentations of the flavor alone in another context. Contextual control of the fluid ingestion was observed in that rats reduced the fluid intake in the LiCl-paired context but drank the solution in the context never paired with lithium. Having learned this discrimination, the rats were now given a second flavor in their home cage before being injected with LiCl and transferred to the previously lithium-paired context. In Experiment 1, the acquisition of an aversion to the novel flavor was blocked when this flavor and the contextual cues are conditioned as a compound. In Experiment 2, the blocking effect and the conditional control over fluid consumption were abolished when the associative strength of the LiCl-paired context had been extinguished by exposing the animals to water in the contexts after discriminative training. These results are interpreted as evidence that context dependency of conditioned taste aversions is mediated by a summative effect of the context–LiCl and flavor–LiCl associations.     

Generalization gradients in human predictive learning: Effects of discrimination training and within-subjects testing

Generalization gradients have been investigated widely in animal conditioning experiments, but much less so in human predictive learning tasks. Here, we apply the experimental design of a recent study on conditioned fear generalization in humans (Lissek et al., 2008) to a predictive learning task, and examine the effects of a number of relevant procedural parameters drawn from the generalization literature in animal conditioning. Experiment 1 shows that prior discrimination learning and steady-state testing procedures sharpen the gradient; Experiment 2 shows that within-subjects testing of the range of generalization stimuli also sharpens the gradient. In addition, Experiment 2 shows that, in case of very flat initial generalization, an orderly gradient can reveal itself through differential rates of extinction learning. Finally, Experiment 2 also evidenced an orderly gradient of generalization-of-extinction. These results suggest that discrimination processes have an important effect on the generalization of predictive learning in humans, and highlight behavioral analogies between animal conditioning and human predictive learning.     

Pattern structure and relational discrimination learning.

McGonigle and Jones take exception to Dodwell's explanation of anomalous transfer (AT) as the outcome of relational discriminations among a set of stimulus patterns which vary along a single dimension of orientational salience, from 'horizontal' to 'vertical'. In particular they do not think that the continuum is generated by units with Hubel-and-Wiesel type retinal receptive fields. Instead, they invoke Garner's notions about stimulus structure to explain Dodwell's results, as well as their own finding of two situations where AT fails to occur. It seems that McGonigle and Jones missed the point of the relational discrimination explanation of AT. In fact, it is shown that AT is not predicted by this model for the conditions in which they failed to obtain it. The relational model makes definite predictions about conditions under which AT will occur; as this is not true of their invocation of Garner's ideas, the former is to be preferred. Whether or not outputs are coded by Hubel-and Wiesel type units is not particularly relevant to the main point, that AT is a result of relational learning. In fact, the model is an instance of how structure, in Garner's sense, can be generated.     

Generalization and response mediation of a conditional discrimination

Fifteen pigeons were given conditional discrimination training in which a colored sample stimulus determined which of two line comparison stimuli (vertical and horizontal) was correct. As part of the conditional discrimination procedure, birds were required to make an "observing response" to the sample stimulus presented on a wide key. The location on this key of the required observing response for the two sample stimuli differed by 0, 3, or 6 in. (0, 7.6, or 15.2 cm) for three groups of birds. Accuracy of conditional discrimination performance was directly related to the amount of separation. In subsequent generalization tests with novel sample stimuli, both observing-response location and comparison responding changed within the same region of the wavelength continuum from that appropriate for one of the training samples to that appropriate for the other. A maintained generalization test (continued reinforcement for training stimuli) revealed this relation more strongly. A test in which observing-response location was the only sample stimulus of a conditional discrimination revealed stimulus control by this observing response, supporting a response mediation interpretation of the data.     

Associative learning and imagery training in children

Two experiments are reported which examine imagery as a factor in children's learning to associate object pairs. The results of the first experiment indicated that presenting the objects in an interacting fashion greatly facilitated subsequent recognition performance. The outcome of Experiment II revealed that children can be trained to form their own interaction images. These findings are discussed in terms of their relevance to contemporary views of mediation in children's learning.     

Generalization and discrimination of shape orientation in the pigeon

Pigeons learned to peck a green key on which parallelogram-shapes were projected; they then received generalization tests in which the orientation of the parallelogram was varied. Nondifferential training produced very little eventual stimulus control along the orientation dimension, but when training included S- trials (absence of the parallelogram) subjects responded consistently more to certain orientations than to others. Unlike typical results for visual generalization (e.g., line-tilt), the tilt gradients obtained for this complex stimulus were bimodal, supporting predictions on the basis of human perceptual data. However, unimodal gradients could be produced by specific discrimination training along the orientation dimension. Other forms of intradimensional training also produced relatively steep gradients, often characterized by unexpected but consistent secondary peaks. An attempt to obtain inhibitory gradients (S+: green key; S-: parallelogram on a green background) resulted in virtually zero responding all along the shape-orientation dimension; therefore, specific inhibitory control could not be evaluated. All these experiments suggest that definition of this complex stimulus dimension in terms of mere "angular orientation" is inappropriate, and alternative interpretations are discussed.     

Conditional discrimination learning: A critique and amplification

Carter and Werner recently reviewed the literature on conditional discrimination learning by pigeons, which consists of studies of matching-to-sample and oddity-from-sample. They also discussed three models of such learning: the “multiple-rule” model (learning of stimulus-specific relations), the “configuration” model, and the “single-rule” model (concept learning). Although their treatment of the multiple-rule model, which seems most applicable to the pigeon data, is generally excellent, their discussion of the other two models is incomplete and sometimes inaccurate. Potential problems of terminology are discussed in the present paper, as are additional lines of research that deserve consideration by those interested in further work in this area. The issue of response versus stimulus selection (configuration versus compound-cue learning) is discussed in connection with the configuration model. Particular attention is given to Carter and Werner's criticism of the application, in studies with other species, of the learning set procedure in testing for single-rule learning. Some of the important related issues are: the bias for improvement on new problems in a series, the adequacy of a multiple-rule model to explain learning set formation, and evidence in favor of the single-rule model, at least in primates. Consideration of these additional contributions to the study of conditional discrimination learning emphasizes the usefulness of this task in the comparative study of cognitive processes.     

Observing responses and discrimination learning

An operant response in the pigeon, whose performance results in exposure to the discriminative stimuli, is described and suggested as an experimental analogue for “observing.” Such an operant response is then used to explore the relationship between observing responses and discrimination learning in a variety of discrimination situations, of progressively increasing complexity. In general, the results support the contention that the development and maintainence of observing responses is closely related to the degree of differential behaviour manifested toward the discriminative stimuli. Certain modifications are suggested in the theoretical formulation underlying the concept of “observing responses.”     

Memory retrieval and discrimination learning

A better reinforcement schedule may be associated with one exteroceptive stimulus (S+) than with another exteroceptive stimulus (S−). While all theories agree that performance in such discrimination tasks is affected by the reinforcement schedules associated with the exteroceptive S+ and S− cues, some theories suggest that performance is also affected by the reinforcement schedule associated with interoceptive reward produced cues. The two experiments reported here were concerned with identifying the conditions under which the interoceptive cues produced by nonreward come to affect performance. According to one hypothesis, such cues will acquire control over approach responding if animals make an approach response in their presence, but not otherwise. According to the memory hypothesis, the memory of nonreinforcement will become a signal for reinforcement, thus invigorating performance, if it is retrieved on a rewarded trial. In two experiments, two groups made strong approach responses to the nonreinforced cues on S− trials, but the memory of nonreward was better retrieved on a subsequent rewarded trial in one group than in the other. Subsequently, both groups were extinguished. The extinction findings support the memory view.     

Association learning and visual discrimination

The effect of learned stimulus associations on visual discrimination was measured in four experiments. The stimuli were visual shapes which had been scaled for similarity. Two shapes were selected as discriminanda, and each S was pretested and posttested for discrimination of these briefly presented simultaneous pairs of “same” and “different” shapes. During the training, each discriminandum was paired with another simultaneously presented associated shape on a paired-associate response-learning task. The two associated shapes were very similar, intermediate in similarity, or very dissimilar. There was more improvement in posttest discrimination following training with dissimilar associates. We conclude that learned stimulus associations affect visual discrimination.