Search

We introduce the Category Abstraction Learning (CAL) model, a cognitive framework formally describing category learning built on similarity-based generalization, dissimilarity-based abstraction, two attention learning mechanisms, error-driven knowledge structuring, and stimulus memorization. Our hypotheses draw on an array of empirical and theoretical insights connecting reinforcement and category learning. The key novelty of the model is its explanation of how rules are learned from scratch based on three central assumptions. (a) Category rules emerge from two processes of stimulus generalization (similarity) and its direct inverse (category contrast) on independent dimensions. (b) Two attention mechanisms guide learning by focusing on rules, or on the contexts in which they produce errors. (c) Knowing about these contexts inhibits executing the rule, without correcting it, and consequently leads to applying partial rules in different situations. The model is designed to capture both systematic and individual differences in a broad range of learning paradigms. We illustrate the model's explanatory scope by simulating several benchmarks, including the classic Six Problems, the 5-4 problem, and linear separability. Beyond the common approach of predicting average response probabilities, we also propose explanations for more recently studied phenomena that challenge existing learning accounts, regarding task instructions, individual differences in rule extrapolation in three different tasks, individual attention shifts to stimulus features during learning, and other phenomena. We discuss CAL's relation to different models, and its potential to measure the cognitive processes regarding attention, abstraction, error detection, and memorization from multiple psychological perspectives. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Polymorphous concepts are hard to learn, and this is perhaps surprising because they, like many natural concepts, have an overall similarity structure. However, the dimensional summation hypothesis (Milton and Wills Journal of Experimental Psychology: Learning, Memory and Cognition, 30, 407–415 2004) predicts this difficulty. It also makes a number of other predictions about polymorphous concept formation, which are tested here. In Experiment 4, we confirm the theory’s prediction that polymorphous concept formation should be facilitated by deterministic pretraining on the constituent features of the stimulus. This facilitation is relative to an equivalent amount of training on the polymorphous concept itself. In further experiments, we compare the predictions of the dimensional summation hypothesis with a more general strategic account (Experiment 2), a seriality of training account (Experiment 3), a stimulus decomposition account (also Experiment 3), and an error-based account (Experiment 4). The dimensional summation hypothesis provides the best account of these data. In Experiment 5, a further prediction is confirmed—the single feature pretraining effect is eliminated by a concurrent counting task. The current experiments suggest the hypothesis that natural concepts might be acquired by the deliberate serial summation of evidence. This idea has testable implications for classroom learning.

The current research examined the effects of errorful generation on memory, focusing particularly on the roles of motivation and surprise. In two experiments, participants were first presented with photographs of faces and were asked to associate four facts with each photograph. On Generate trials, the participants guessed two of the facts (Guess targets) before those correct facts, and another two correct facts (Study targets), were revealed. On the remaining Read trials, all four facts were presented without a guessing stage. In Experiment 1, participants also ranked their motivation to know the answers before they were revealed, or their surprise on learning the true answers. Guess targets were subsequently better recognised than the concurrently presented, non-guessed Study targets. Guess targets were also better recognised than Read targets, and recognition of Study and Read targets did not differ. Errorful generation also increased self-reported motivation, but not surprise. Experiment 2 showed that the results of Experiment 1 can outlive a 20-minute delay, and that they generalise to a more challenging recognition test. Together, the results suggest that errorful generation improves memory specifically for the guessed fact, and this may be linked to an increase in motivation to learn that fact.

publisher: Elsevier articletitle: In Defence of Effect-Centric Research journaltitle: Journal of Applied Research in Memory and Cognition articlelink: http://dx.doi.org/10.1016/j.jarmac.2016.10.005 content_type: simple-article copyright: © 2016 Society for Applied Research in Memory and Cognition. Published by Elsevier Inc. All rights reserved.

The idea that there are multiple learning systems has become increasingly influential in recent years, with many studies providing evidence that there is both a quick, similarity-based or feature-based system and a more effortful rule-based system. A smaller number of imaging studies have also examined whether neurally dissociable learning systems are detectable. We further investigate this by employing for the first time in an imaging study a combined positive and negative patterning procedure originally developed by Shanks and Darby [Shanks, D. R., & Darby, R. J. Feature- and rule-based generalization in human associative learning. Journal of Experimental Psychology: Animal Behavior Processes, 24, 405–415, 1998]. Unlike previous related studies employing other procedures, rule generalization in the Shanks–Darby task is beyond any simple non-rule-based (e.g., associative) account. We found that rule- and similarity-based generalization evoked common activation in diverse regions including the pFC and the bilateral parietal and occipital lobes indicating that both strategies likely share a range of common processes. No differences between strategies were identified in whole-brain comparisons, but exploratory analyses indicated that rule-based generalization led to greater activation in the right middle frontal cortex than similarity-based generalization. Conversely, the similarity group activated the anterior medial frontal lobe and right inferior parietal lobes more than the rule group did. The implications of these results are discussed.

This article presents a comprehensive survey of research concerning interactions between associative learning and attention in humans. Four main findings are described. First, attention is biased toward stimuli that predict their consequences reliably (learned predictiveness). This finding is consistent with the approach taken by Mackintosh (1975) in his attentional model of associative learning in nonhuman animals. Second, the strength of this attentional bias is modulated by the value of the outcome (learned value). That is, predictors of high-value outcomes receive especially high levels of attention. Third, the related but opposing idea that uncertainty may result in increased attention to stimuli (Pearce & Hall, 1980), receives less support. This suggests that hybrid models of associative learning, incorporating the mechanisms of both the Mackintosh and Pearce-Hall theories, may not be required to explain data from human participants. Rather, a simpler model, in which attention to stimuli is determined by how strongly they are associated with significant outcomes, goes a long way to account for the data on human attentional learning. The last main finding, and an exciting area for future research and theorizing, is that learned predictiveness and learned value modulate both deliberate attentional focus, and more automatic attentional capture. The automatic influence of learning on attention does not appear to fit the traditional view of attention as being either goal-directed or stimulus-driven. Rather, it suggests a new kind of “derived” attention.

The influential competition between verbal and implicit systems (COVIS) model proposes that category learning is driven by two competing neural systems-an explicit, verbal, system, and a procedural-based, implicit, system. In the current fMRI study, participants learned either a conjunctive, rule-based (RB), category structure that is believed to engage the explicit system, or an information-integration category structure that is thought to preferentially recruit the implicit system. The RB and information-integration category structures were matched for participant error rate, the number of relevant stimulus dimensions, and category separation. Under these conditions, considerable overlap in brain activation, including the prefrontal cortex, basal ganglia, and the hippocampus, was found between the RB and information-integration category structures. Contrary to the predictions of COVIS, the medial temporal lobes and in particular the hippocampus, key regions for explicit memory, were found to be more active in the information-integration condition than in the RB condition. No regions were more activated in RB than information-integration category learning. The implications of these results for theories of category learning are discussed. Hum Brain Mapp 37:3557-3574, 2016. © 2016 Wiley Periodicals, Inc.

Does cognition begin with an undifferentiated stimulus whole, which can be divided into distinct attributes if time and cognitive resources allow (Differentiation Theory)? Or does it begin with the attributes, which are combined if time and cognitive resources allow (Combination Theory)? Across psychology, use of the terms analytic and non-analytic imply that Differentiation Theory is correct—if cognition begins with the attributes, then synthesis, rather than analysis, is the more appropriate chemical analogy. We re-examined four classic studies of the effects of time pressure, incidental training, and concurrent load on classification and category learning (Kemler Nelson, 1984; Smith & Kemler Nelson, 1984; Smith & Shapiro, 1989; Ward, 1983). These studies are typically interpreted as supporting Differentiation Theory over Combination Theory, while more recent work in classification (Milton et al., 2008, et seq.) supports the opposite conclusion. Across seven experiments, replication and re-analysis of the four classic studies revealed that they do not support Differentiation Theory over Combination Theory—two experiments support Combination Theory over Differentiation Theory, and the remainder are compatible with both accounts. We conclude that Combination Theory provides a parsimonious account of both classic and more recent work in this area. The presented data do not require Differentiation Theory, nor a Combination–Differentiation hybrid account.

The effects of two different types of training on rule-based and information-integration category learning were investigated in two experiments. In observational training, a category label is presented, followed by an example of that category and the participant's response. In feedback training, the stimulus is presented, and the participant assigns it to a category and then receives feedback about the accuracy of that decision. Ashby, Maddox, and Bohil (2002. Observational versus feedback training in rule-based and information-integration category learning. Memory & Cognition, 30, 666–677) reported that feedback training was superior to observational training when learning information-integration category structures, but that training type had little effect on the acquisition of rule-based category structures. These results were argued to support the COVIS (competition between verbal and implicit systems) dual-process account of category learning. However, a number of nonessential differences between their rule-based and information-integration conditions complicate interpretation of these findings. Experiment 1 controlled between-category structures for participant error rates, category separation, and the number of stimulus dimensions relevant to the categorization. Under these more controlled conditions, rule-based and information-integration category structures both benefited from feedback training to a similar degree. Experiment 2 maintained this difference in training type when learning a rule-based category that had otherwise been matched, in terms of category overlap and overall performance, with the rule-based categories used in Ashby et al. These results indicate that differences in dimensionality between the category structures in Ashby et al. is a more likely explanation for the interaction between training type and category structure than the dual-system explanation that they offered.

Background and Objectives Control theory predicts that the detection of goal discrepancies initiates ruminative self-focus ( Martin & Tesser, 1996 ). Despite the breadth of applications and interest in control theory, there is a lack of experimental evidence evaluating this prediction. The present study provided the first experimental test of this prediction. Methods We examined uninstructed state rumination in response to the cueing of resolved and unresolved goals in a non-clinical population using a novel measure of online rumination. Results Consistent with control theory, cueing an unresolved goal resulted in significantly greater recurrent intrusive ruminative thoughts than cueing a resolved goal. Individual differences in trait rumination moderated the impact of the goal cueing task on the extent of state rumination: individuals who had a stronger tendency to habitually ruminate were more susceptible to the effects of cueing goal discrepancies. Limitations The findings await replication in a clinically depressed sample where there is greater variability and higher levels of trait rumination. Conclusions These results indicate that control theories of goal pursuit provide a valuable framework for understanding the circumstances that trigger state rumination. Additionally, our measure of uninstructed online state rumination was found to be a valid and sensitive index of the extent and temporal course of state rumination, indicating its value for further investigating the proximal causes of state rumination.

It is sometimes argued that the implementation of an overall similarity classification is less effortful than the implementation of a single-dimension classification. In the current article, we argue that the evidence securely in support of this view is limited, and report additional evidence in support of the opposite proposition—overall similarity classification is more effortful than single-dimension classification. Using a match-to-standards procedure, Experiments 1A, 1B and 2 demonstrate that concurrent load reduces the prevalence of overall similarity classification, and that this effect is robust to changes in the concurrent load task employed, the level of time pressure experienced, and the short-term memory requirements of the classification task. Experiment 3 demonstrates that participants who produced overall similarity classifications from the outset have larger working memory capacities than those who produced single-dimension classifications initially, and Experiment 4 demonstrates that instructions to respond meticulously increase the prevalence of overall similarity classification.

This research explored the role that associative learning may play in human sequence learning. Two-choice serial reaction time tasks were performed under incidental conditions using 2 different sequences. In both cases, an experimental group was trained on 4 subsequences: LLL, LRL, RLR, and RRR for Group "Same" and LLR, LRR, RLL, and RRL for Group "Different," with left and right counterbalanced across participants. To control for sequential effects, we assayed sequence learning by comparing their performance with that of a control group, which had been trained on a pseudorandom ordering, during a test phase in which both experimental and control groups experienced the same subsequences. Participants in both groups showed sequence learning, but the group trained on "different" learned more and more rapidly. This result is the opposite that predicted by the augmented simple recurrent network used by F. W. Jones and I. P. L. McLaren (2009, Human sequence learning under incidental and intentional conditions, Journal of Experimental Psychology: Animal Behavior Processes, Vol. 35, pp. 538-553), but can be modeled using a reparameterized version of this network that also includes a more realistic representation of the stimulus array, suggesting that the latter may be a better model of human sequence learning under incidental conditions.

The effects of picture manipulations on humans' and pigeons' performance were examined in a go/no-go discrimination of two perceptually similar categories, cat and dog faces. Four types of manipulation were used to modify the images. Mosaicization and scrambling were used to produce degraded versions of the training stimuli, while morphing and cell exchange were used to manipulate the relative contribution of positive and negative training stimuli to test stimuli. Mosaicization mainly removes information at high spatial frequencies, whereas scrambling removes information at low spatial frequencies to a greater degree. Morphing leads to complex transformations of the stimuli that are not concentrated at any particular spatial frequency band. Cell exchange preserves high spatial frequency details, but sometimes moves them into the "wrong" stimulus. The four manipulations also introduce high-frequency noise to differing degrees. Responses to test stimuli indicated that high and low spatial frequency information were both sufficient but not necessary to maintain discrimination performance in both species, but there were also species differences in relative sensitivity to higher and lower spatial frequency information.

Demonstrations of retrospective revaluation suggest that remembered stimuli undergo a reduction in association with the unconditioned stimulus (US) present during learning. Conversely, demonstrations of mediated conditioning in flavor-conditioning experiments with rats suggest that remembered stimuli undergo an increase in association with the US present during learning. In a food allergy prediction task with 23 undergraduates, we demonstrated simultaneous backward conditioned inhibition and mediated conditioning effects. These results are compatible with the hypothesis that the direction of change (decrease or increase) in associative strength depends on whether the remembered stimulus was of a different category (conditioned stimulus/antecedent) or the same category (US/outcome) as the presented US.

The effect of concurrent load on generalization performance in human contingency learning was examined in 2 experiments that employed the combined positive and negative patterning procedure of Shanks and Darby (1998). In Experiment 1, we tested 32 undergraduates and found that participants who were trained and tested under full attention showed generalization consistent with the application of an opposites rule (i.e., single cues signal the opposite outcome to their compound), whereas participants trained and tested under a concurrent cognitive load showed generalization consistent with surface similarity. In Experiment 2, we replicated the effect with 148 undergraduates and provided evidence that it was the presence of concurrent load during training, rather than during testing, that was critical. Implications for associative, inferential, and dual-process accounts of human learning are discussed.

Contemporary theories of learning typically assume that learning is driven by prediction errors—in other words, that we learn more when our predictions turn out to be incorrect than we do when our predictions are correct. Results from the recording of electrical brain activity suggest one mechanism by which this might happen; we seem to direct visual attention toward the likely causes of previous prediction errors. This can happen very rapidly—within less than 200 milliseconds of the error-causing object being presented. It is tempting to infer that if learning is driven by prediction errors, then little can be learned in the absence of feedback. Such a conclusion is unwarranted. In fact, the substantial learning that is sometimes the result of simple exposure to objects can also be explained by processes of directing attention toward the likely causes of previous prediction errors.

The processes of overall similarity sorting were investigated in 5 free classification experiments. Experiments 1 and 2 demonstrated that increasing time pressure can reduce the likelihood of overall similarity categorization. Experiment 3 showed that a concurrent load also reduced overall similarity sorting. These findings suggest that overall similarity sorting can be a time-consuming analytic process. Such results appear contrary to the idea that overall similarity is a nonanalytic process (e.g., T. B. Ward, 1983) but are in line with F. N. Milton and A. J. Wills's (2004) dimensional summation hypothesis and with the stochastic sampling assumptions of the extended generalized context model (K. Lamberts, 2000). Experiments 4 and 5 demonstrated that the relationship between stimulus presentation time and overall similarity sorting is nonmonotonic, and the shape of the function is consistent with the idea that the three aforementioned processes operate over different parts of the time course.

When humans process visual stimuli, global information often takes precedence over local information. In contrast, some recent studies have pointed to a local precedence effect in both pigeons and nonhuman primates. In the experiment reported here, we compared the speed of acquisition of two different categorizations of the same four geometric figures. One categorization was on the basis of a local feature, the other on the basis of a readily apparent global feature. For both humans and pigeons, the global-feature categorization was acquired more rapidly. This result reinforces the conclusion that local information does not always take precedence over global information in nonhuman animals.

It has been demonstrated that when people free classify stimuli presented simultaneously in an array, they have a preference to categorize by a single dimension. However, when people are encouraged to categorize items sequentially, they sort by “family resemblance,” grouping by overall similarity. The present studies extended this research, producing 3 main findings. First, the sequential procedure introduced by G. Regehr and L. R. Brooks (1995) does not always produce a preference for family resemblance sorts. Second, sort strategy in a sequential procedure is sensitive to subtle variations in stimulus properties. Third, spatially separable stimuli evoked more family resemblance sorts than stimuli of greater spatial integration. It is suggested that the family resemblance sorting observed is due to an analytic strategy. Categorization is a fundamental building block of everyday cognition; it is hard to imagine how people would function effectively without it. Categorization enables one to react to different objects in the same way and to make inferences about how novel objects should be treated. For instance, when a person sees a novel object, classifying it as a “dog” allows the person to deal with it in an appropriate manner. Because of the immense variety of discriminable objects that people encounter in the natural environment, it is necessary that this categorization process should be highly constrained, as there are virtually limitless numbers of ways in which objects can be partitioned. It is therefore of fundamental importance to understand the principles that underlie the categories that people have. Traditional categorization experiments give the participant itemby-item specific feedback about category membership. While such an approach is undoubtedly useful in examining category learning, it seems extremely unlikely that people receive this level of feedback anywhere other than in the laboratory. One way of addressing this issue is to examine how people spontaneously categorize a group of objects. This can be done by providing them with a group of stimuli and asking them to categorize them in the way that they think is most appropriate. No feedback, or other information, is given by the experimenter. Such an approach has variously been called category construction (e.g., Medin, Wattenmaker, & Hampson, 1987), free sorting (e.g., Bersted, Brown, & Evans, 1969), and free classification (e.g., Handel & Imai, 1972). It seems reasonable to assume that the categories we prefer to create would reflect the underlying structure of objects we encounter outside the laboratory. Over the years, there have been several

What are the effects of preexposure of stimuli on participants’ subsequent ability to categorize them accurately? An experiment employing artificial, abstract, visual stimuli confirms that, for adult humans, the effect of preexposure is dependent upon category structure. Whether preexposure has beneficial or detrimental effects is shown to be dependent on the way category examples are generated from the category base patterns. The results are predicted by salience reduction accounts of perceptual learning but may be problematic for stimulus differentiation accounts.

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

Unitization, the creation of new stimulus features by the fusion of preexisting features, is one of the hypothesized processes of perceptual learning (Goldstone Annual Review of Psychology, 49:585–612, 1998). Some argue that unitization occurs to the extent that it is required for successful task performance (e.g., Shiffrin & Lightfoot, 1997), while others argue that unitization is largely independent of functionality (e.g., McLaren & Mackintosh Animal Learning & Behavior, 30:177–200, 2000). Across three experiments, employing supervised category learning and unsupervised exposure, we investigated three predictions of the McLaren and Mackintosh (Animal Learning & Behavior, 30:177–200, 2000) model: (1) Unitization is accompanied by an initial increase in the subjective similarity of stimuli sharing a unitized component; (2) unitization of a configuration occurs through exposure to its components, even when the task does not require it; (3) as unitization approaches completion, salience of the unitized component may be reduced. Our data supported these predictions. We also found that unitization is associated with increases in overt attention to the unitized component, as measured through eye tracking.

Many naturally occurring categories vary across multiple stimulus dimensions (e.g. size, color, texture). When humans categorize multidimensional stimuli on the basis of a single dimension this has been taken to indicate use of a rule that could be verbalized. Sorting on the basis of all the stimulus dimensions (‘overall similarity’ or ‘family resemblance’) has been taken to indicate a more basic, implicit, automatic, perhaps associative process. However, a review of the literature on animal discrimination learning shows that animals often discriminate on the basis of one dominant dimension. In recent experiments, situations conducive to more complex cognitive processes have increased family resemblance sorting in humans. In an effort to resolve this apparent paradox, experiments were conducted in which humans and pigeons were exposed to multidimensional category discrimination tasks under closely similar conditions. Preliminary results show no evidence that even a non-verbal rule can be said to be involved in pigeons’ choices in these conditions, despite the fact that under some conditions a single dimension may dominate their behavior.

Stimulus sets defined in terms of artificial polymorphous concepts have frequently been used in experiments to investigate the mechanisms of discrimination of natural concepts, both in humans and in other animals. However, such stimulus sets are frequently difficult for either animals or humans to discriminate. Properties of artificial polymorphous stimulus sets that might explain this difficulty include the complexity of the individual stimuli, the unreliable reinforcement of individual positive features, attentional load, difficulties in discriminating some stimulus dimensions, memory load, and a lack of the correlation between features that characterizes natural concepts. An experiment using chickens as subjects and complex artificial visual stimulus sets investigated these hypotheses by training the birds in discriminations that were not polymorphous but did have some of the properties listed above. Discriminations that involved unreliable reinforcement or high attentional load were found to approach the difficulty of polymorphous concept discriminations, and these two factors together were sufficient to account for the entire difficulty. The usual kind of artificial polymorphous concept may not be a good model for natural concepts as they are perceived and discriminated by birds. A RULEX account of natural concept learning may be preferable.

We argue for an example of “core cognition” based on Diamond and Carey's (1986) work on expertise and recognition, which is not made use of in The Origin of Concepts. This mechanism for perceptual learning seems to have all the necessary characteristics in that it is innate, domain-specific (requires stimulus sets possessing a certain structure), and demonstrably affects categorisation in a way that strongly suggests it will influence concept formation as well.

Background Two theories explain how negative mood primes smoking behaviour. The stimulus–response (S-R) account argues that in the negative mood state, smoking is experienced as more reinforcing, establishing a direct (automatic) association between the negative mood state and smoking behaviour. By contrast, the incentive learning account argues that in the negative mood state smoking is expected to be more reinforcing, which integrates with instrumental knowledge of the response required to produce that outcome. Objectives One differential prediction is that whereas the incentive learning account anticipates that negative mood induction could augment a novel tobacco-seeking response in an extinction test, the S-R account could not explain this effect because the extinction test prevents S-R learning by omitting experience of the reinforcer. Methods To test this, overnight-deprived daily smokers (n = 44) acquired two instrumental responses for tobacco and chocolate points, respectively, before smoking to satiety. Half then received negative mood induction to raise the expected value of tobacco, opposing satiety, whilst the remainder received positive mood induction. Finally, a choice between tobacco and chocolate was measured in extinction to test whether negative mood could augment tobacco choice, opposing satiety, in the absence of direct experience of tobacco reinforcement. Results Negative mood induction not only abolished the devaluation of tobacco choice, but participants with a significant increase in negative mood increased their tobacco choice in extinction, despite satiety. Conclusions These findings suggest that negative mood augments drug-seeking by raising the expected value of the drug through incentive learning, rather than through automatic S-R control.