Beyond nonutilization: Irrelevant cues can gate learning in probabilistic categorization.

In probabilistic categorization, also known as multiple cue probability learning (MCPL), people learn to predict a discrete outcome on the basis of imperfectly valid cues. In MCPL, normatively irrelevant cues are usually ignored, which stands in apparent conflict with recent research in deterministic categorization that has shown that people sometimes use irrelevant cues to gate access to partial knowledge encapsulated in independent partitions. The authors report 2 experiments that sought support for the existence of such knowledge partitioning in probabilistic categorization. The results indicate that, as in other areas of concept acquisition (such as function learning and deterministic categorization), a significant proportion of participants partitioned their knowledge on the basis of an irrelevant cue. The authors show by computational modeling that knowledge partitioning cannot be accommodated by 2 exemplar models (Generalized Context Model and Rapid Attention Shifts 'N Learning), whereas a rule-based model (General Recognition Theory) can capture partitioned performance. The authors conclude by pointing to the necessity of a mixture-of-experts approach to capture performance in MCPL and by identifying reduction of complexity as a possible explanation for partitioning. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Irrelevant sound disrupts order information in free recall as in serial recall

The claim that the sensitivity of free recall to disruption by irrelevant sound is a function of the extent to which rote rehearsal is employed as a mnemonic strategy was investigated in two experiments. The degree of disruption by irrelevant sound in terms of both item and order information was contrasted under serial and free recall instructions. Irrelevant sound was found to disrupt order and item information equally in serial and free recall tasks (Experiment 1). Contrary to previous reports, an effect of irrelevant sound was also demonstrated on free recall of particularly long lists, and the interaction between list length and retention interval in the irrelevant sound effect was examined (Experiment 2). Generally, the results support the view that irrelevant sound disrupts the use of order cues.     

Irrelevant autocorrelation in least-squares intervention models.

Autocorrelation coefficients computed on the entire series of observations obtained from interrupted time-series designs are generally irrelevant to the general linear model (GLM) independence assumption. Consequently, the argument that GLM solutions are invalidated when large coefficients of this type are encountered is incorrect. Several decompositions of the terms involved in such coefficients are provided to show how deterministic components in the correct model contaminate these coefficients. Example data sets from articles in the methodological literature that were written to promote the use of complex time-series methods are used to illustrate relevant and irrelevant autocorrelations and to demonstrate the application and viability of GLM time-series intervention models. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Children's use of frequency information for trait categorization and behavioral prediction.

Two experiments examined young children's use of behavioral frequency information to make behavioral predictions and global personality attributions. In Experiment 1, participants heard about an actor who behaved positively or negatively toward 1 or several recipients. Generally, children did not differentiate their judgments of the actor on the basis of the amount of information provided. In Experiment 2, the actor behaved positively or negatively toward a single recipient once or repeatedly. Participants were more likely to make appropriate predictions and attributions after exposure to multiple target behaviors and with increasing age. Overall, children's performance was influenced by age-related positivity and negativity biases. These findings indicate that frequency information is important for personality judgments but that its use is affected by contextual complexity and information-processing biases. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Pigeons use item-specific and category-level information in the identification and categorization of human faces.

Three experiments examined the pigeon's ability to adapt to the specific tasks of identification and categorization of complex visual stimuli by attending to information at different levels of abstraction. In Experiment 1, pigeons were successively trained to discriminate first between pairs of human faces from the same sex and then between male and female faces. A final test revealed that the associations formed in the first training phase did not survive categorization training, indicating that the pigeons used item-specific information to discriminate faces of the same sex, but shifted their attention to category-level information for discriminating between the male and female categories. Two further experiments disproved alternative explanations. The results support the feature-learning account of open-ended categorization in animals. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Preintegrative categorization of social information: The role of persons as organizing categories.

Two studies with 57 college students investigated the factors that cause persons to emerge as organizing categories during the initial classification of social information about individuals in groups. Preintegration categorization processes were studied using a speeded classification procedure for 4 social-information ensembles (unfamiliar and familiar names with high and low association values). Results suggest that person discriminability and strength of the person–feature association are important determinants of the categorization process. These findings emerged for both generic knowledge structures and newly developed person categories. (16 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Emotional response categorization.

Traditional theories of categorization in which categories are assumed to be grounded in perceptual similarity or theories ignore an important basis of conceptual structure: the emotion that a stimulus elicits in a perceiver. This article discusses the nature of, constraints on, and conditions of use of emotional response categories. Experiments in which participants sorted triads of concepts that shared both emotional and nonemotional relations indicate that individuals use emotional response categories when they are experiencing emotional states. Multidimensional scaling of similarity judgments by emotional and nonemotional perceivers supports a selective attention mechanism of these effects. Participants induced to feel happy or sad emotional states weighted the emotional responses associated with stimuli more heavily than people in relatively neutral states. The triad and multidimensional scaling findings, along with functional considerations, suggest that emotional response categorization is not only tenable, but necessary for a complete account of categorization. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Context-gated knowledge partitioning in categorization.

According to the knowledge partitioning framework, people sometimes master complex tasks by creating multiple independent parcels of partial knowledge. Research has shown that knowledge parcels may contain mutually contradictory information, and that each parcel may be used without regard to knowledge that is demonstrably present in other parcels. This article reports 4 experiments that investigated knowledge partitioning in categorization. When component boundaries of a complex categorization were identified by a context cue, a significant proportion of participants learned partial and independent categorization strategies that were chosen on the basis of context. For those participants, a strategy used in one context was unaffected by knowledge demonstrably present in other contexts, suggesting that knowledge partitioning in categorization can be complete. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Adaptive categorization in unsupervised learning.

In 3 experiments, the authors provide evidence for a distinct category-invention process in unsupervised (discovery) learning and set forth a method for observing and investigating that process. In the 1st 2 experiments, the sequencing of unlabeled training instances strongly affected participants' ability to discover patterns (categories) across those instances. In the 3rd experiment, providing diagnostic labels helped participants discover categories and improved learning even for instance sequences that were unlearnable in the earlier experiments. These results are incompatible with models that assume that people learn by incrementally tracking correlations between individual features; instead, they suggest that learners in this study used expectation failure as a trigger to invent distinct categories to represent patterns in the stimuli. The results are explained in terms of J. R. Anderson's (1990, 1991) rational model of categorization, and extensions of this analysis for real-world learning are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Conditions Favoring Superconditioning of Irrelevant Conditioned Stimuli.

Three appetitive conditioning experiments with rats found partial learning of complex XA+ , XB+ , XAB- (+ stands for reinforced; - stands for unreinforced) negative patterning discriminations with intermixed A+ and B+ trials (Experiment 1). AB+ trials (Experiment 2), and A+ , B+ , and AB+ trials (Experiment 3). In all experiments, differential responding emerged more slowly during the learning of the negative patterning discriminations than during learning of the XA+ , XB+ , XC- control discriminations. Additionally, the negative patterning groups responded more to X than to a separately reinforced Y on unreinforced test trials: thus, X derived superexcitatory properties. This pattern was reversed in the control groups. Results are consistent with theories that allow for different activation patterns when elements are combined. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

Decision boundaries in one-dimensional categorization.

Decision-boundary theories of categorization are often difficult to distinguish from exemplar based theories of categorization. The authors developed a version of the decision-boundary theory, called the single-cutoff model, that can be distinguished from the exemplar theory. The authors present 2 experiments that test this decision-boundary model. The results of both experiments point strongly to the absence of single cutoffs in most participants, and no participant displayed use of the optimal boundary. The range of nonoptimal solutions shown by individual participants was accounted for by an exemplar-based adaptive-learning model. When combined with the results of previous research, this suggests that a comprehensive model of categorization must involve both rules and exemplars, and possibly other representations as well. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Suboptimality in human categorization and identification.

Categorization and identification decision processes were examined and compared in 4 separate experiments. In all tasks, the critical stimulus component was a line that varied across trials in length and orientation, and the optimal decision rules were always complex piecewise quadratic functions. Evidence was found that identification is mediated by separate explicit and implicit systems. In addition, a common type suboptimality was found in both categorization and identification. In particular, observers apparently approximated the piecewise quadratic functions of the optimal decision rules with simpler piecewise linear functions. A computational model, which was motivated by a recent neuropsychological theory of category learning, successfully accounted for this suboptimal performance in both categorization and identification. The model assigns a key role to the striatum and assumes the observed suboptimality was largely due to massive convergence of visual cortical cells onto single striatal units. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Racism and racial categorization.

Social identity theory predicts that perceivers strongly identified with an in-group will maximize the distinction and maintain a clear boundary between their own and other groups by categorizing others' membership accurately. Two experiments tested the prediction that racially prejudiced individuals, who presumably identify highly with their racial in-group, are more motivated to make accurate racial categorizations than nonprejudiced individuals. Results indicated that prejudiced participants not only took longer to categorize race-ambiguous targets (Experiments 1 and 2), but also made more nonverbal vocalizations when presented with them (Experiment 1), suggesting response hesitation. The results support the hypothesis that, compared to nonprejudiced individuals, prejudiced individuals concern themselves with accurate identification of in-group and out-group members and use caution when making racial categorizations. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Prototype effects in categorization by pigeons.

In 4 experiments involving a total of 78 pigeons, Ss received a categorization task involving 6 simultaneous compounds in which the elements A, B, and C were more frequently paired with food than were the elements D, E, and F. Food was delivered after compounds ABF, AEC, and DBC but not after DEC, DBF, and AEF. Subsequent testing revealed a higher rate of responding during ABC than during any of the compounds that had signaled food and a lower rate of responding during DEF than during any of the compounds that had not signaled food. Exps 2, 3, and 4 further demonstrated that the rate of responding during test trials with ABC was faster than during a compound composed of 3 elements that had individually been paired with food. Results are more consistent with a configural than an elemental analysis of discrimination and categorization. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Picture-object recognition in pigeons: Evidence of representational insight in a visual categorization task using a complementary information procedure.

Success in tasks requiring categorization of pictorial stimuli does not prove that a subject understands what the pictures stand for. The ability to achieve representational insight is by no means a trivial one because it exceeds mere detection of 2-D features present in both the pictorial images and their referents. So far, evidence for such an ability in nonhuman species is weak and inconclusive. Here, the authors report evidence of representational insight in pigeons. After being trained on pictures of incomplete human figures, the birds responded significantly more to pictures of the previously missing parts than to nonrepresentative stimuli, which demonstrates that they actually recognized the pictures' representational content. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Feature and rule selection in categorization tasks.

The process by which stimuli are assigned to categories has been traditionally conceptualized as bottom-up. Accordingly, stimulus features are supposed to be the fundamental units of analysis, and categorization to be accomplished on the basis of feature category correlations (cue validities). Alternatively, top-down processes are invoked, in which case one begins with a given categorization rule and then assigns stimuli to categories on that basis. Bottom-up and top-down views share a fundamental weakness, namely, they are unable to specify how features or rules, respectively, are acquired. This difficulty can be overcome if it is assumed that the process starts with neither features nor rules, but with stimuli. Then, as a result of experience with stimuli belonging to different categories, the cognitive system discovers and uses locally constructed features that maximally discriminate between the categories at hand. According to this view, the relationship between a target and a contrast category is the main factor affecting what subjects learn about each. Two experiments were conducted to explore this hypothesis. Both experiments support the notion that the relationship between the target and contrast category significantly determines which critical features are extracted as being defining of either category. In particular, it determines the level of generality of these features. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Irrelevant response effects improve serial learning in serial reaction time tasks.

Tones were introduced into a serial reaction time (SRT) task to serve as redundant response effects. Experiment 1 showed that the tones improved serial learning with a 10-element stimulus sequence, but only if the tone effects were mapped onto the responses contingently. Experiment 2 demonstrated that switching to noncontingent response-effect mapping increased SRT only when participants had previously adapted to contingent response-effect mapping. In Experiment 3, the beneficial influence of contingent tone effects on serial learning occurred only when there was sufficient time between the response effects and the next imperative stimuli. The results are discussed in terms of the ideomotor principle. It is claimed that an internal representation of the to-be-produced tone effects develops and gains control over the execution of the response sequence. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Irrelevant auditory material affects counting

R. H. Logie and D. A. Baddeley (1987) suggested that event counting may be supported by a phonologically based working-memory structure referred to as the phonological loop. However, inconsistent results concerning the detrimental effects of irrelevant speech on event counting led them to propose that lexical or semantic representations are also involved. In 4 experiments, the authors showed that this extension of Logie and Baddeley's original conceptualization was unnecessary. Instead, the number of irrelevant syllables spoken per time unit, a factor that was not taken into account in previous research, can explain Logie and Baddeley's results. In addition, the present results support the hypothesis that in cases of interference from the auditory channel, correction processes that help to recover the current running total from past running totals are also involved in counting performance.     

Psychiatric diagnosis as prototype categorization.

Most critiques of the psychiatric diagnostic system seem to presuppose a "classical" view of categorization; an alternative view of categorization, frequently called the "prototype" view, is emerging in the psychological literature. The present article represents an attempt to apply this alternative view to psychiatric diagnosis. The central thesis is that while clinical diagnosis may appear "messy" and disordered from the classical perspective, it seems orderly and principled from the perspective of the prototype view. Two empirical investigations were conducted to document the utility of a change in views about psychiatric categorization systems and rules as they are used in practice, not with systems and rules that appear in the diagnostic manuals. The 1st investigation concerned the content and structure of the clinical features that trained psychiatrists commonly associate with patients from 9 different diagnostic categories. The 2nd investigation was concerned with problems of reliability and confidence in clinical diagnostic judgments. Results of both investigations provide support for the utility of the prototype view in this domain. (32 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Velocity-based motion categorization by pigeons.

To examine if animals could learn action-like categorizations in a manner similar to noun-based categories, eight pigeons were trained to categorize rates of object motion. Testing 40 different objects in a go/no-go discrimination, pigeons were first trained to discriminate between fast and slow rates of object rotation around their central y-axis. They easily learned this velocity discrimination and transferred it to novel objects and rates. This discrimination also transferred to novel types of motions including the other two axes of rotation and two new translations around the display. Comparable tests with rapid and slow changes in the objects' size, color, and shape failed to support comparable transfer. This difference in discrimination transfer between motion-based and property-based changes suggests the pigeons had learned motion concept rather than one based on change per se. The results provide evidence that pigeons can acquire an understanding of motion-based actions, at least with regard to the property of object velocity. This may be similar to our use of verbs and adverbs to categorize different classes of behavior or motion (e.g., walking, jogging, or running slow vs. fast). (PsycINFO Database Record (c) 2011 APA, all rights reserved)