Theory-based Bayesian models of inductive learning and reasoning

Inductive inference allows humans to make powerful generalizations from sparse data when learning about word meanings, unobserved properties, causal relationships, and many other aspects of the world. Traditional accounts of induction emphasize either the power of statistical learning, or the importance of strong constraints from structured domain knowledge, intuitive theories or schemas. We argue that both components are necessary to explain the nature, use and acquisition of human knowledge, and we introduce a theory-based Bayesian framework for modeling inductive learning and reasoning as statistical inferences over structured knowledge representations.     

Probabilistic models of cognition: conceptual foundations

Remarkable progress in the mathematics and computer science of probability has led to a revolution in the scope of probabilistic models. In particular, 'sophisticated' probabilistic methods apply to structured relational systems such as graphs and grammars, of immediate relevance to the cognitive sciences. This Special Issue outlines progress in this rapidly developing field, which provides a potentially unifying perspective across a wide range of domains and levels of explanation. Here, we introduce the historical and conceptual foundations of the approach, explore how the approach relates to studies of explicit probabilistic reasoning, and give a brief overview of the field as it stands today.     

Perception of speech and nonspeech stimuli by children with and without reading disability and attention deficit hyperactivity disorder

The auditory temporal deficit hypothesis predicts that children with reading disability (RD) will exhibit deficits in the perception of speech and nonspeech acoustic stimuli in discrimination and temporal ordering tasks when the interstimulus interval (ISI) is short. Initial studies testing this hypothesis did not account for the potential presence of attention deficit hyperactivity disorder (ADHD). Temporal order judgment and discrimination tasks were administered to children with (1) RD/no-ADHD (n=38), (2) ADHD (n=29), (3) RD and ADHD (RD/ADHD; n=32), and (4) no impairment (NI; n=43). Contrary to predictions, children with RD showed no specific sensitivity to ISI and performed worse relative to children without RD on speech but not nonspeech tasks. Relationships between perceptual tasks and phonological processing measures were stronger and more consistent for speech than nonspeech stimuli. These results were independent of the presence of ADHD and suggest that children with RD have a deficit in phoneme perception that correlates with reading and phonological processing ability. (c) 2002 Elsevier Science (USA).     

The impact of anticipated action effects on action planning

Three experiments with a total of 72 participants investigated the assumption that motor actions are planned in terms of their sensorial effects. Participants had to prepare a certain action A that consistently led to a sensorial effect (a tone of certain pitch). Instead of (in Experiment 1) or before (in Experiments 2 and 3) the execution of the prepared action, another response B had to be carried out, which either resulted in the same or in a different auditory effect (a tone of same or different pitch). It was found that a to-be-executed response B was in general initiated more quickly when it resulted in the same effect as a concurrently prepared response A. The results are considered as evidence for the basic notion that the preparation and initiation even of very simple actions is mediated by an anticipation of their reafferences.     

The case for coherence in scientific explanations: quantitative details can hurt qualitative understanding

In Experiments 1A and 1B, students read a concise booklet containing 653 words and 6 illustrations describing the formation, propagation, and dispersion of ocean waves (concise group) or an expanded booklet containing 327 additional words and 5 additional illustrations describing relevant mathematical formulas and computations interspersed throughout the lesson (expanded group). In Experiment 2, students viewed a multimedia presentation of narrated animations based on the concise or expanded booklet. In both studies, the expanded group performed more poorly than did the concise group on problem-solving transfer tests. The added quantitative details may have distracted the learner from constructing a qualitative model of the process of ocean waves.     

A dynamic multilevel model of demographic diversity and misfit effects

In this study, the authors proposed and evaluated the linkages of a dynamic multilevel model of demographic diversity and misfit effects in a large sample of quick-service restaurants. Using a cross-level approach, the authors studied an employee's demographic misfit in relation to coworkers' demographics as a predictor of turnover risk over time. At the business-unit level, they studied changes in restaurant demographic diversity in relation to changes in profitability over time and unit turnover rates in relation to profitability. The authors also studied the impact of the match between the racial compositions of the restaurants and their communities on profitability. The results supported linkages between demographic misfit and turnover and partially supported a negative association between racial diversity and changes in profitability.     

Attachment, self-esteem, worldviews, and terror management: evidence for a tripartite security system

On the basis of prior work integrating attachment theory and terror management theory, the authors propose a model of a tripartite security system consisting of dynamically interrelated attachment, self-esteem, and worldview processes. Four studies are presented that, combined with existing evidence, support the prediction derived from the model that threats to one component of the security system result in compensatory defensive activation of other components. Further, the authors predicted and found that individual differences in attachment style moderate the defenses. In Studies 1 and 2, attachment threats motivated worldview defense among anxiously attached participants and motivated self-enhancement (especially among avoidant participants), effects similar to those caused by mortality salience. In Studies 3 and 4, a worldview threat and a self-esteem threat caused attachment-related proximity seeking among fearful participants and avoidance of proximity among dismissing participants. The authors' model provides an overarching framework within which to study attachment, self-esteem, and worldviews.     

A Functional Role Analysis of Reasons

One strategy for providing an analysis of practical rationality is to start with the notion of a practical reason as primitive. Then it will be quite tempting to think that the rationality of an action can be defined rather simply in terms of ‘the balance of reasons’. But just as, for many philosophical purposes, it is extremely useful to identify the meaning of a word in terms of the systematic contribution the word makes to the meanings of whole sentences, this paper argues that it is extremely useful to explain the nature of practical reasons in terms of the systematic contributions that such reasons make to the wholesale rational statuses of actions. This strategy gives us a clear view of two logically distinct normative roles for practical reasons – justifying and requiring – that are often conflated, and it allows us to give clear definitions of what ‘the strength of a reason’ means within each of these roles. The final section of the paper explores some implications of the resulting view for the internalism/externalism debate about practical reasons, and for the practical significance of moral theory.     

The role of target salience in crowding

We studied 'crowding' in the parafovea using orientation identification of a Gabor target as the task, and flanking Gabors on an isoeccentric circle as distractors. Orientation-discrimination thresholds were raised by nearby flanking distractors. This crowding effect was increased by the number of distractors and decreased by the spatial separation between target and distractors. Crowding was greatest when the target was in the centre of the distractor array and smallest when the target was on the edge of the array. A cue indicating the position of the target improved performance when the position was otherwise unknown and the spatial separation between target and distractors was large, but the cue had no significant effect when separation was small. Increasing the contrast of the target relative to the distractors reduced crowding, but targets of smaller contrast than that of the distractors are even harder to identify than those of the same contrast. Putting the target and distractors in different depth planes decreased crowding for some observers, but there were qualitative individual differences. A large (say, 45 degrees) difference in orientation between target and distractors caused the target to 'pop out' in a presence/absence task, despite the evidence from other studies that crowding is still found in these conditions. We conclude that salience has, at best, modest effects on crowding.