Configural Learning Research Articles

The effects of neonatal choline dietary supplementation on adult spatial and configural learning and memory in rats

The facilitative effects of pre- and early postnatal choline chloride dietary supplementation on adult rat spatial and nonspatial learning and memory were examined using a delayed match-to-place and a transverse-patterning discrimination task. Animals were exposed to the choline supplementation both prenatally (through the diet of pregnant rats) and postnatally (subcutaneous injection) for 24 days. In the first experiment, 90-day-old rats were given pairs of trials in which they first found a hidden platform in a Morris water maze in a particular location (acquisition trials), and then were required to remember that position 10 min later (test trials). Those animals given neonatal choline pretreatment found the platform on test trials significantly faster than did animals in a saline-treated control group. All animals were subsequently tested in the same paradigm following atropine sulfate injections. The atropine eliminated the difference between experimental and control animals on test trials. In a second experiment, neonatally treated choline rats performed significantly better than controls in acquiring a visual transverse patterning discrimination task previously found to be sensitive to hippocampal and/or frontal damage. The present study extends the description of long-term functional enhancement produced by perinatal choline supplementation to include the ability to use and remember visual configural associations, working spatial memory, and to relate these effects to modifications in cholinergic basal forebrain systems.

The effects of neonatal choline dietary supplementation on adult spatial and configural learning and memory in rats

The effects of neonatal choline dietary supplementation on adult spatial and configural learning and memory in rats

An extinction cue reduces spontaneous recovery of a conditioned taste aversion

The effect of an auditory cue presented during extinction on spontaneous recovery of a conditioned taste aversion was investigated in three experiments. Experiment 1 demonstrated that the presence of the cue during extinction did not influence saccharin consumption during that phase, and that an aversion to saccharin in the absence of the cue was stronger at 18 days than at 1 day after extinction, representing spontaneous recovery rather than a renewal effect. Experiment 2 showed that a cue presented during extinction and testing reduced spontaneous recovery. Experiment 3 replicated that effect and showed that it depended on the cue’s correlation with extinction and not on an unconditioned effect; cues that had been presented during or prior to conditioning did not reduce spontaneous recovery when presented during testing. The cue’s potential to reduce spontaneous recovery through conditioned inhibition or configural cue learning is discussed, as is the possibility that the cue retrieves a saccharin extinction memory in a manner consistent with Bouton’s (1993) account of spontaneous recovery.

Hippocampal lesions and negative patterning in pigeons

A number of studies have shown that rats with hippocampal lesions are impaired on the negative patterning task, thereby supporting the view that the hippocampus may be important for configural learning. We trained 8 control sham-operated pigeons and 9 pigeons with bilateral hippocampal (hippocampus and area parahippocampalis) lesions on a negative patterning task to respond in the presence of a houselight or tone presented individually, and to withhold responding in the presence of the houselight and tone presented simultaneously. Pigeons with hippocampal lesions learned the task at the same rate as did control unoperated pigeons. This finding is in line with recent reports that the hippocampus is not critical for performance of the negative patterning task.

Preserved negative patterning and impaired spatial learning in pigeons (Columba livia) with lesions of the hippocampus.

Pigeons (Columba livia) with bilateral electrolytic lesions of the hippocampus and area parahippocampalis were compared with control pigeons on 2 tasks: negative patterning and delayed spatial alternation. Negative patterning demands configural stimulus representations for its successful solution. The only effect of hippocampal lesions on this task was an increased response rate to the rewarded stimuli. On the delayed spatial alternation task, hippocampal birds showed deficits relative to controls. Differences in the results of prior studies on negative patterning appear to be due to different response requirements to the nonreinforced stimuli. These results are consistent with prior work with rats and suggest that the avian hippocampus is essential for spatial memory and response inhibition but is not involved in configural learning.

A neural network solution to the transverse patterning problem depends on repetition of the input code.

Using computer simulations, this paper investigates how input codes affect a minimal computational model of the hippocampal region CA3. Because encoding context seems to be a function of the hippocampus, we have studied problems that require learning context for their solution. Here we study a hippocampally dependent, configural learning problem called transverse patterning. Previously, we showed that the network does not produce long local context codings when the sequential input patterns are orthogonal, and it fails to solve many context-dependent problems in such situations. Here we show that this need not be the case if we assume that the input changes more slowly than a processing interval. Stuttering, i.e., repeating inputs, allows the network to create long local context firings even for orthogonal inputs. With these long local context firings, the network is able to solve the transverse patterning problem. Without stuttering, transverse patterning is not learned. Because stuttering is so useful, we investigate the relationship between the stuttering repetition length and relative context length in a simple, idealized sequence prediction problem. The relative context length, defined as the average length of the local context codes divided by the stuttering length, interacts with activity levels and has an optimal stuttering repetition length. Moreover, the increase in average context length can reach this maximum without loss of relative capacity. Finally, we note that stuttering is an example of maintained or introduced redundancy that can improve neural computations.

Contextual learning and cue association in fear conditioning in mice: a strain comparison and a lesion study

Fear conditioning with electric shock (unconditioned stimulus, US) paired with tone cue (conditioned stimulus, CS) has been extensively applied in recent molecular neurobiological analysis of hippocampal dysfunction in mice because the context-dependent test phase of this learning paradigm is claimed to detect hippocampal impairment in a specific manner, whereas the cue-dependent test serves as a control situation independent of hippocampal function. These claims are based on hippocampal lesion studies performed with rats and have not been conclusively confirmed with mice with specific hippocampal lesion. Therefore, I investigated how hippocampal ibotenic acid lesion affects conditioned fear in mice. I confirm that extensive lesions localized to the hippocampus impair context-dependent learning but also show that, unlike in the original rat studies, the behavioral impairment is only partial. Furthermore, studying two inbred strains of mice (C57BL/6 and DBA/2) with highly different hippocampal function, I show that the presence or absence of CS during training may influence the mouse's ability to learn complex multiple contextual stimuli in a genotype-dependent manner. I conclude that performance at the `context' test may be based on complex configural (hippocampal) learning but it can also be based on a more simple elemental (non-hippocampal) learning thus leading to potentially false-negative findings in the analysis of hippocampal dysfunction.

Null Mutation of the Prolactin Receptor Gene Produces a Defect in Maternal Behavior

We have studied pup-directed maternal behavior in mice carrying a germ line null mutation of the PRL receptor (PRLR) gene. Homozygous mutant and heterozygous mutant nulliparous females show a deficiency in pup-induced maternal behavior. Moreover, primiparous heterozygous females exhibit a profound deficit in maternal care when challenged with foster pups. Morris maze studies revealed normal configural learning in the heterozygous and homozygous animals. Eating, locomotor activity, sexual behavior, and exploration (all processes regulated by the hypothalamus) are normal in PRLR mutant mice. Olfactory function was tested in an aversive conditioning paradigm, results indicating that heterozygous and homozygous PRLR mutant mice are not anosmic. These studies clearly establish the PRLR as a regulator of maternal behavior.

Neither Extended Sequential nor Simultaneous Feature Positive Training Result in Modulation of Evaluative Flavor–Flavor Conditioning in Humans

In previous attempts to induce Pavlovian modulation of human evaluative flavor–flavor associations, the recurrent finding was that exposure to a Feature Positive (FP) schedule (XA+/A−), in which flavor A was reinforced by the aversive flavor Tween20 if and only if it was accompanied by a feature stimulus X, did not result in the expected X-modulated dislike for target A, but in simple unconditional A-US associations. The first experiment reported here investigated if more extended training results in a shift from simple A-US learning to the development of X-modulated A-US associations. Participants were exposed to a 32-trial, 4-session sequential FP schedule using flavors both for feature and target stimuli. The modulatory and/or excitatory power of the stimuli was assessed after each training session. After initial training, participants again acquired a simple unmodulated target A-US association. Contrary to expectations, the additional training sessions did not result in participants gradually learning the XA+/A− discrimination, but rather seemed to corroborate the already established A-US association. This was true whether or not participants acquired valid explicit knowledge about the conditional X→(A-US) relation. The second experiment tested the hypothesis that what can be described as conditional flavor preferences may actually be based on a configural learning process. Participants were exposed to a single-session, eight-trial XA+/A− simultaneous discrimination schedule, the parameters of which were selected so as to enhance the probability of configuring XA, while simultaneously making it likely to observe a pattern of evaluations similar to an X-modulated dislike for A (X was less salient than A). Even though there was good evidence that participants noticed flavor X, they again failed to solve the XA+/A− discrimination and most probably acquired an association between the more salient flavor A and the US. The possibility is discussed that evaluative learning, unlike expectancy learning, might ultimately prove not to be subject to modulation.

Configural learning in humans: The transverse patterning problem

Most learning theorists concur that some sort of configural representation system is required for animals to solve certain compound discriminations. Transverse patterning (A+ vs. B−, B+ vs. C−, C+ vs. A−) is one example of a problem that requires a configurai solution. It has been reported in the past that adult humans are unable to solve this problem (Berch & Israel, 1971; Franks, 1976). We investigated the generality of these reports. Moreover, we examined whether a stepwise approach to training this problem would facilitate learning and at what stage in training subjects adopt a configurai strategy. We found that college-aged adults had little difficulty solving the transverse patterning problem, and that their learning was greatly facilitated by using the stepwise approach described by Alvarado and Rudy (1989). Moreover, we found that subjects seem to adopt a configural strategy when faced with ambiguous stimulus pairs even when an elemental strategy would suffice. These results provide insight into how humans solve configurai problems and also suggest some direct tests of the role of the hippocampus in configural associations.

Perirhinal cortex ablation impairs configural learning and paired–associate learning equally

Combined damage to the perirhinal and entorhinal cortex has been implicated in the formation of stimulus–stimulus associative memories. We show in this article that relative to three normal controls three cynomolgus monkeys with ablations restricted to the perirhinal cortex were impaired on a visual paired–associate learning task in which subjects had to learn which of two visual stimuli were associated with a cue stimulus. The subjects with perirhinal cortex ablations also showed an impairment of a similar magnitude on a visual configural learning task in which they had to learn which of two configurations of visual stimuli were associated with food-reward. The stimuli in both tasks were comprised of alphanumeric characters presented upon a touch-screen. Both groups made fewer errors on the configural learning task than on the paired–associate learning task. We suggest that performance on both tasks relies critically on the perirhinal cortex due to the specialization of the perirhinal cortex in processing knowledge about objects. We argue that the specializations of this system and of other memory systems such as the hippocampal-fornix spatial\\episodic memory system, are conferred by the specialization of their anatomical connections to other structures. We reject the notion that there are specific memory processes such as the hippocampal based configural associative system that was proposed to be critical for configural associative learning.

Re-examining the factor structure of the Wide Range Assessment of Memory and Learning: implications for clinical interpretation.

The intercorrelation matrices of the standardization sample of the Wide Range Assessment of Memory and Learning (WRAML), a multi-component measure of memory functioning in children ages 5 to 17 years, were submitted to a hierarchical exploratory principal factor analysis (PFA). The PFA solutions were examined and compared with the published principal components analysis (PCA) solutions with the goal of examining the validity of the clinical scale configuration (Verbal Memory, Visual Memory, and Learning) proposed by the test authors. Results of the PFA differ from the PCA and do not provide statistical support for the existing three-scale structure nor the division between memory and learning. Specific factor loadings on the majority of sub-tests are higher than the common factor loadings indicating low shared variance. The low communalities together with the poor interpretability of the factor structure suggests that the subtests should be interpreted clinically as unique entities first and secondarily as factors. Theoretical and practical implications of the findings are discussed.

Induction of Relational Schemas: Common Processes in Reasoning and Complex Learning

Five experiments were performed to test whether participants induced a coherent representation of the structure of a task, called a relational schema, from specific instances. Properties of a relational schema include: An explicit symbol for a relation, a binding that preserves the truth of a relation, potential for higher-order relations, omnidirectional access, potential for transfer between isomorphs, and ability to predict unseen items in isomorphic problems. However relational schemas are not necessarily coded in abstract form. Predictions from relational schema theory were contrasted with predictions from configural learning and other nonstructural theories in five experiments in which participants were taught a structure comprised of a set of initial-state,operator → end-state instances. The initial-state,operator pairs were presented and participants had to predict the correct end-state. Induction of a relational schema was achieved efficiently by adult participants as indicated by ability to predict items of a new isomorphic problem. The relational schemas induced showed the omnidirectional access property, there was efficient transfer to isomorphs, and structural coherence had a powerful effect on learning. The “learning to learn” effect traditionally associated with the learning set literature was observed, and the long-standing enigma of learning set acquisition is explained by a model composed of relational schema induction and structure mapping. Performance was better after reversal of operators than after shift to an alternate structure, even though the latter entailed more overlap with previously learned tasks in terms of the number of configural associations that were preserved. An explanation for the reversal shift phenomenon in terms of induction and mapping of a relational schema is proposed. The five experiments provided evidence supporting predictions from relational schema theory, and no evidence was found for configural or nonstructural learning theories.

Fornix Lesions Can Facilitate Acquisition of the Transverse Patterning Task: A Challenge for “Configural” Theories of Hippocampal Function

Configural theories of hippocampal function predict that hippocampal dysfunction should impair acquisition of the transverse patterning task, which involves the concurrent solution of three discrimination problems: A+ versus B-; B+ versus C-; and C+ versus A-. The present study tested this prediction in rats using computer-graphic stimuli presented on a touchscreen. Experiment 1 assessed the effects of fornix lesions when the three problems were introduced sequentially (phase 1: A+ vs B-; phase 2: A+ vs B-, B+ vs C-; phase 3: A+ vs B-, B+ vs C-, C+ vs A-). Fornix lesions significantly facilitated acquisition of the complete transverse patterning task (phase 3) but had no effect on the number of sessions or errors required to attain criterion during phase 1 or phase 2. In experiment 2, in which all three problems were presented concurrently from the outset of training, fornix-lesioned animals outperformed control animals during the seventh block of acquisition trials and were not impaired during any stage of acquisition. Importantly, these same animals were significantly impaired on two allocentric spatial tasks: T-maze alternation (experiments 1 and 2) and the Morris Swim Task (experiment 1). These results contradict the predictions of configural theories of hippocampal function and cast doubt on the popular notion that spatial learning is a special case of configural learning.

Hippocampal lesions enhance configural learning by reducing proactive interference.

Rats were trained on an operant conditional discrimination in which an ambiguous stimulus (X) indicated both the occasions on which responding in the presence of a second cue (A) would be reinforced and the occasions on which responding in the presence of a third cue (B) would not be reinforced (X --> A+, A-, X --> B-, B+). Both rats with neurotoxic lesions of the hippocampus and control rats learned this discrimination more rapidly when the training trials were widely spaced (mean intertrial interval of 8 min) than when they were massed (mean intertrial interval of 1 min). With spaced practice, lesioned and control rats learned this discrimination equally well. But when the training trials were massed, lesioned rats learned more rapidly than controls. At the end of training, performance of all rats was tested at three different intertrial intervals, 0.5 min, 1 min, and 8 min. The control rats trained with 8-min intertrial intervals showed deficits in discrimination performance when the test intertrial interval was 0.5 min or 1 min. An analysis of sequential effects indicated that a major source of this performance deficit was the control rats' failure to withhold responding on nonreinforced trials when those trials were immediately preceded by reinforced trials within 0.5 min or 1 min. In contrast, performance of lesioned rats was not affected by either the test intertrial interval or by the nature of preceding trials. Thus, with short intertrial intervals, hippocampal lesions may have improved learning or performance on a given trial by reducing proactive interference from the previous trial.

Simple and configural association learning in rats with bilateral quisqualic acid lesions of the nucleus basalis magnocellularis

We hypothesized that bilateral quisqualic acid lesions of the nucleus basalis magnocellularis (NBM) in rats would impair configural but not simple association learning. In experiment 1, rats were tested in a negative patterning operant discrimination where they were food-reinforced for responding to a light or a tone (L +, T +) but not for responding to the configural stimulus consisting of the light and tone presented simultaneously (LT −). Consistent with our hypothesis, NBM-lesioned rats showed a transient but significant impairment, responding normally to L + and T + but responding more often to LT −, in addition to responding more often during the inter-trial interval (ITI) than controls. In experiment 2, rats were tested in a simple operant discrimination where rats were food-reinforced for responding to a light (L +) but not for responding to a tone (T −). Although NBM-lesioned rats again responded normally to L + as predicted, NBM-lesioned rats were transiently impaired, making more T − responses and more ITI responses than controls. Together, these results suggest that the NBM is involved in both configural and simple association learning but that this involvement is limited to learning to withhold responding to non-reinforced contextual or discrete stimuli. Finally, rats from experiment 2 underwent extinction trials, where results showed no difference between NBM-lesioned and control groups, suggesting that the NBM is not involved in the extinction of conditioned responding to previously reinforced stimuli.

Transfer testing after serial feature-ambiguous discrimination in Pavlovian keypeck conditioning

Four experiments were performed to explore modulatory transfer after serial feature-ambiguous (F-A) discrimination (X→A+, A−, X→B−, B+) in Pavlovian keypeck conditioning (autoshaping) with pigeons. Diffuse features were used in the first three experiments, and no modulatory transfer was found: (1) There was no modulatory transfer between two concurrently trained F-A tasks; (2) modulatory transfer to conditioned and then extinguished stimuli was not observed; and (3) responding to the targets (A and B) after their contingencies were reversed was not affected by presentation of their original feature stimulus (X). Transfer between two serial F-A tasks was obtained in the last experiment, in which keylights were used as features, but other evidence indicated that this was due to stimulus generalization between the features. Together, the results strongly suggest that specific target-food occasion setting or configural learning is the underlying mechanism of F-A discrimination.

Elemental and Configural Learning and the Perception of Odorant Mixtures by the Spiny Lobster Panulirus argus

Livermore A., M. Hutson, V. Ngo, R. Hadjisimos and C. D. Derby. Elemental and configural learning and the perception of odorant mixtures by the spiny lobster Panulirus argus. Physiol Behav 62(1) 169–174, 1997.—The present study used a conditioning assay to investigate if the type of learning task that spiny lobsters (Panulirus argus) were required to perform influenced the way that they perceived odorant mixtures. Mixtures were composed of 2 food-related compounds (adenosine-5′-monophosphate, betaine, or l-glutamate) at concentrations that produced the same duration of searching behavior in unconditioned animals. Aversive conditioning of search behavior coupled with generalization testing was used to evaluate perceptual similarity between related mixtures. When animals were conditioned to stop searching to a binary mixture AX, they did not generalize significantly from this mixture to either of its components (A or X), or to a binary mixture containing one novel component (AY). However, when lobsters were conditioned to avoid AX but to continue responding to AY, they generalized between AX and X and between AY and Y. The results support the hypothesis that altering the salience of a mixture's components by giving them different reinforcement contingencies changed the way that the mixtures were perceived. As a result of such conditioning, animals perceived the mixture's components as separate elements, rather than as a configuration, and, as a consequence, animals generalized between binary mixtures and their most salient or predictive components.

Temporo-frontal disconnection impairs visual-visual paired association learning but not configural learning in Macaca monkeys.

Cynomolgus monkeys were tested in two computer-controlled visual associative memory tasks. The monkeys chose between visual objects on a screen by touching one. In the configural learning task one correct object and one wrong object were presented in each trial. Each of these two objects was composed of two coloured alphanumeric characters abutted together. The designation of the objects as 'correct' or 'wrong' followed a configural rule: e.g. if AB and CD are correct objects then AD and CB are wrong. In the paired association learning task in each trial three spatially separate objects (single alphanumeric characters) were presented. The central object was an instruction cue and the designation of the side objects as 'correct' or 'wrong' choices followed a paired association rule: e.g. if A, C and B are presented (C in the centre) then A is correct and B is wrong; however, if A, D and C are presented then C is correct and A is wrong. Disrupting the direct cortico-cortical interaction between the inferior temporal cortex and the prefrontal cortex by uncinate fascicle transection led to a learning deficit in the paired association task but not in the configural task. These results suggest that the uncinate fascicle facilitates visual-visual associative learning only in the specific case where a visual object acts as an instruction cue to guide the conditional choice of another, spatially separate object, and they support the evidence for a specific role of the uncinate fascicle in the learning of conditional tasks with visual instruction cues.

Recoding can lead to inaccessible structures, but avoids capacity limitations

The distinction between uninformed learning (type-1) and learning based on recoding using prior information (type-2) helps to clarify some long-standing psychological problems, including misunderstanding of mathematics by children, the need for active construction of concepts in cognitive development, and the difficulty of configural learning tasks. However, an alternative to recoding some type-2 tasks is to represent the input as separate dimensions, which are processed jointly. This preserves the original structure, but is subject to processing capacity limitations.