Dissociations between reaction times and temporal order judgments: a diffusion model approach

A diffusion model for simple reaction time (RT) and temporal order judgment (TOJ) tasks was developed to account for a commonly observed dissociation between these 2 tasks: Most stimulus manipulations (e.g., intensity) have larger effects in RT tasks than in TOJ tasks. The model assumes that a detection criterion determines the level of sensory evidence needed to conclude that a stimulus has been presented. Analysis of the performance that would be achieved with different possible criterion settings revealed that performance was optimal with a lower criterion setting for the TOJ task than for the RT task. In addition, the model predicts that effects of stimulus manipulations should increase with the size of the detection criterion. Thus, the model suggests that commonly observed dissociations between RT and TOJ tasks may simply be due to performance optimization in the face of conflicting task demands.     

Natural categorization through multiple feature learning in pigeons

Recently (Troje, Huber, Loidolt, Aust, & Fieder 1999), we found that pigeons discriminated between large sets of photorealistic frontal images of human faces on the basis of sex. This ability was predominantly based on information contained in the visual texture of those images rather than in their configural properties. The pigeons could learn the distinction even when differences of shape and average intensity were completely removed. Here, we proved more specifically the pigeons' flexibility and efficiency to utilize the class-distinguishing information contained in complex natural classes. First, we used principal component as well as discriminant function analysis in order to determine which aspects of the male and female images could support successful categorization. We then conducted various tests involving systematic transformations and reduction of the feature content to examine whether or not the pigeons' categorization behaviour comes under the control of categorylevel feature dimensions-that is, those stimulus aspects that most accurately divide the stimulus classes into the experimenter-defined categories of 'Male' and 'Female'. Enhanced classification ability in the presence of impoverished test faces that varied only along one of the first three principal components provided evidence that the pigeons used these class-distinguishing stimulus aspects as a basis for generalization to new instances.     

Sequences of similarity judgments as time series

Sequences of similarity judgments made to pairs of geometrical patterns with three variable elements were obtained from 22 subjects. Each subject made 100 judgments to a series which embodied random input, step transitions and impulse shifts. Three different instructions concerning the use of information from previous trials were used. Time series analyses gave linear transfer function coefficients whose ratios reflected the strategy employed by a subject. Evidence of local nonstationarity in the response gain functions is examined.     

Predicting similarity judgments in intertemporal choice with machine learning

Similarity models of intertemporal choice are heuristics that choose based on similarity judgments of the reward amounts and time delays. Yet, we do not know how these judgments are made. Here, we use machine-learning algorithms to assess what factors predict similarity judgments and whether decision trees capture the judgment outcomes and process. We find that combining small and large values into numerical differences and ratios and arranging them in tree-like structures can predict both similarity judgments and response times. Our results suggest that we can use machine learning to not only model decision outcomes but also model how decisions are made. Revealing how people make these important judgments may be useful in developing interventions to help them make better decisions.     

Dissociations between ABA-, ABC-, and AAB-renewal of Pavlovian modulation in human sequential feature positive discrimination learning

Using a conditioned suppression preparation, we investigated extinction and aba-, abc-, and aab-renewal of Pavlovian modulation in human sequential Feature Positive (FP) discrimination learning, X → A+/A-. Extinction treatment was administered in the acquisition context a (aaa- and aab-groups) or in a new context b (aba- and abc-groups) and comprised X → A- extinction trials. Discriminative X → A/A responding was lost in all groups when tested in the extinction context. In the aba-group, the discriminative X → A/A responding totally recovered when retested in the acquisition context a. For the aaa-, the aab-, and the abc-group, discriminative X → A/A responding did not reappear when tested for renewal in, respectively, contexts a, b, and c. The demonstration of aba-renewal of extinguished modulation, but not abc- and aab-renewal, suggests that extinction in a context different from the acquisition context and a return to the original acquisition context might both be critical for renewal of Pavlovian modulation in human FP-discrimination learning.     

Domain familiarity as a cue for judgments of learning

Students differ in how much they already know about topics within and across their courses. Few studies, however, have examined the relationship between participants’ levels of knowledge across topics (i.e., their “domain familiarity”) and their learning of information from those topics, their study choices related to those topics, and their subjective self-assessments of their learning about the topics. As such, in two studies we had participants (Study 1, college students; Study 2, Mturk workers) rank their domain familiarity for several to-be-studied domains (e.g., chemistry, history), rate their efficacy and interest in those domains, study and make judgments of learning (JOLs) for facts from each domain, and finally complete a short-answer test over those facts. Participants’ efficacy and interest ratings for the topics were linearly related to their topic rankings, as were their recall of and JOLs for facts from those domains. Although the JOLs were consistently overconfident, they were more overconfident for better-known than for lesser-known topics. Participants’ study times were not related to their topic rankings (Studies 1 and 2), but participants did use domain familiarity to strategically decide which domains to restudy before the test (Study 2). Participants typically chose to restudy their least-familiar topics, but chose to restudy their best-known topic under extremely limited restudy conditions. As a whole, the results suggest that participants effectively use their domain familiarity as a basis for their JOLs and restudy choices, but to some extent overuse this factor to assess their learning, and underuse it to guide initial study.     

Generalization and similarity in exemplar models of categorization: Insights from machine learning

Exemplar theories of categorization depend on similarity for explaining subjects’ ability to generalize to new stimuli. A major criticism of exemplar theories concerns their lack of abstraction mechanisms and thus, seemingly, of generalization ability. Here, we use insights from machine learning to demonstrate that exemplar models can actually generalize very well. Kernel methods in machine learning are akin to exemplar models and are very successful in real-world applications. Their generalization performance depends crucially on the chosen similarity measure. Although similarity plays an important role in describing generalization behavior, it is not the only factor that controls generalization performance. In machine learning, kernel methods are often combined with regularization techniques in order to ensure good generalization. These same techniques are easily incorporated in exemplar models. We show that the generalized context model (Nosofsky, 1986) and ALCOVE (Kruschke, 1992) are closely related to a statistical model called kernel logistic regression. We argue that generalization is central to the enterprise of understanding categorization behavior, and we suggest some ways in which insights from machine learning can offer guidance.     

Abstract categorization ability as a predictor of learning disability classification

Based on a theoretical analysis of the type of cognitive processing that should be sensitive to population differences, this study evaluated the diagnostic validity of a task measuring abstract categorization ability in six-, seven-, and eight-year-old learning disabled (LD) and non-LD peers. This research is part of a project, the major goal of which is the development of a cognitive-based preschool screening test for the early detection of children who may subsequently fail in school. Diagnostic validity is being evaluated within the context of the research strategy we have adopted. Data are presented that demonstrate that the component of abstract category knowledge that best discriminates LD children from non-LD peers, is knowledge of how members of abstract categories differ from each other. This is consistent with a priori predictions from theory.     

Relational similarity and the nonindependence of features in similarity judgments

Four experiments examined the hypothesis that simple attributional features and relational features operate differently in the determination of similarity judgments. Forced choice similarity judgments ("Is X or Y more similar to Z?") and similarity rating tasks demonstrate that making the same featural change in two geometric stimuli unequally affects their judged similarity to a third stimulus (the comparison stimulus). More specifically, a featural change that causes stimuli to be more superficially similar and less relationally similar increases judged similarity if it occurs in stimuli that already share many superficial attributes, and decreases similarity if it occurs in stimuli that do not share as many superficial attributes. These results argue against an assumption of feature independence which asserts that the degree to which a feature shared by two objects affects similarity is independent of the other features shared by the objects. The MAX hypothesis is introduced, in which attributional and relational similarities are separately pooled, and shared features affect similarity more if the pool they are in is already relatively large. The results support claims that relations and attributes are psychologically distinct and that formal measures of similarity should not treat all types of matching features equally.     

Social categorization and similarity in intergroup behaviour

The present study is one of a series exploring the role of social categorization in intergroup behaviour. It has been found in our previous studies that in ?minimal' situations, in which the subjects were categorized into groups on the basis of visual judgments they had made or of their esthetic preferences, they clearly discriminated against members of an outgroup although this gave them no personal advantage. However, in these previous studies division into groups was still made on the basis of certain criteria of ?real' similarity between subjects who were assigned to the same category. Therefore, the present study established social categories on an explicitly random basis without any reference to any such real similarity. It was found that, as soon as the notion of ?group' was introduced into the situation, the subjects still discriminated against those assigned to another random category. This discrimination was considerably more marked than the one based on a division of subjects in terms of interindividual similarities in which the notion of ?group' was never explicitly introduced. In addition, it was found that fairness was also a determinant of the subjects' decisions. The results are discussed from the point of view of their relevance to a social-cognitive theory of intergroup behaviour.     

Predicting similarity and categorization from identification.

In this article, the relation between the identification, similarity judgment, and categorization of multidimensional perceptual stimuli is studied. The theoretical analysis focused on general recognition theory (GRT), which is a multidimensional generalization of signal detection theory. In one application, 2 Ss first identified a set of confusable stimuli and then made judgments of their pairwise similarity. The second application was to Nosofsky's (1985b, 1986) identification-categorization experiment. In both applications, a GRT model accounted for the identification data better than Luce's (1963) biased-choice model. The identification results were then used to predict performance in the similarity judgment and categorization conditions. The GRT identification model accurately predicted the similarity judgments under the assumption that Ss allocated attention to the 2 stimulus dimensions differently in the 2 tasks. The categorization data were predicted successfully without appealing to the notion of selective attention. Instead, a simpler GRT model that emphasized the different decision rules used in identification and categorization was adequate.     

Structural alignment in similarity and difference judgments

Similarity is a central construct in models of cognitive processing, and as such has been the focus of psychological inquiry. This work has revealed that similarity and difference judgments are not always inverses. One explanation for this effect is that similarity judgments focus on matching relations between the items, while difference judgments focus on the mismatching attributes. A second explanation is that both similarity and difference judgments involve a process of structural alignment, and that they use the output of this process differently. These views are contrasted by using the one-shot mapping technique that places attribute similarity and relational similarity in competition. The results suggest that similarity and difference judgments both involve structural alignment.     

Measures of similarity in models of categorization

This paper concerns the use of similarities based on geometric distance in models of categorization. Two problematic implications of such similarities are outlined. First, in a comparison between two stimuli, geometric distance implies that matching features are not taken into account. Second, missing features are assumed not to exist. Only nonmatching features enter into calculations of similarity. A new model is constructed that is based on the ALCOVE model (Kruschke, 1992), but it uses a feature-matching similarity measure (see, e.g., Tversky, 1977) rather than a geometric one. It is an on-line model in the sense that both dimensions and exemplars are constructed during the categorization process. The model accounts better than ALCOVE does for data with missing features (Experiments 1 and 2) and at least as well as ALCOVE for a data set without missing features (Nosofsky, Kruschke, & McKinley, 1992). This suggests that, at least for some stimulus materials, similarity in categorization is more akin to a feature-matching procedure than to geometric distance calculation.     

The testing effect as a function of explicit testing instructions and judgments of learning

During study, people monitor their learning; the output of this monitoring is captured in so-called judgments of learning (JOLs). JOLs predict later recall better if they are made after a slight delay, instead of immediately after study (the delayed JOL effect). According to the self-fulfilling prophecy (SFP) hypothesis delayed JOLs are based on covert retrieval attempts from long-term memory, and successful retrieval attempts in themselves enhance learning (the testing effect). We compared memory for 40 Swahili-Swedish paired associates after a week as a function of three different learning conditions, namely study plus (i) explicitly instructed self-testing, (ii) delayed JOLs, or (iii) less self-testing. We showed that repeated delayed JOLs lead to a memory improvement that does not differ significantly from a comparable condition where the participants are explicitly testing memory, and both the latter groups performed reliably better than a group that self-tested less. The results suggest that delayed JOLs improve long-term retention as efficiently as explicit memory testing and lend support to the SFP hypothesis.     

Relative attention in judgments of heterogeneous similarity

Shepard’s (1964) study on similarity of stimuli with clearly discernible dimensions was repeated with some modifications, the most important being that the Os had to make numerical similarity estimates of the stimulus pairs. The overall outcome did not deviate much from Shepard’s findings. By using quantitative estimates and choosing stimulus series so that the two dimensions were negatively correlated, data for each 0 could be analyzed separately with a partial correlation technique. It was found that the more an 0 attended to one dimension the less he attended to the other. The shifts in attention seemed to be random rather than regular. Consequently, the meaningfulness of a contention like Shepard’s as to the nonexistence of a metric is questioned.