Do Additional Features Help or Hurt Category Learning? The Curse of Dimensionality in Human Learners

The curse of dimensionality, which has been widely studied in statistics and machine learning, occurs when additional features cause the size of the feature space to grow so quickly that learning classification rules becomes increasingly difficult. How do people overcome the curse of dimensionality when acquiring real‐world categories that have many different features? Here we investigate the possibility that the structure of categories can help. We show that when categories follow a family resemblance structure, people are unaffected by the presence of additional features in learning. However, when categories are based on a single feature, they fall prey to the curse, and having additional irrelevant features hurts performance. We compare and contrast these results to three different computational models to show that a model with limited computational capacity best captures human performance across almost all of the conditions in both experiments.     

The dark side of normal: A psychopathy-linked pattern called aberrant self-promotion

These studies investigate a personality and behavior pattern called aberrant self-promotion, conceptualized as a subclinical form of psychopathy. Aberrant self-promoters (ASPs) are theoretically defined as individuals characterized by a narcissistic personality configuration in combination with antisocial behavior. The first study verifies the existence of persons who manifest the ASP pattern. The second study validates the pattern, using as criteria the Revised Psychopathy Checklist (PCL-R) interview and records of antisocial behavior. In the first study a 179-item questionnaire, composed of five personality instruments, was administered to two separate samples of normal subjects (N = 214 and 367). ASPs were targeted by three methods: cluster analysis, item factor analysis, and person factor analysis. In Sample 1 the three methods demonstrated a convergence of 92 per cent in identifying the same individuals as ASPs; in Sample 2 the convergence rate was 94 per cent. In the second study 32 ASPs targeted in Study 1 were compared to 30 non-ASPS. The ASPs had significantly higher scores on the PCL-R and had committed significantly more antisocial acts than the non-ASPS. The general discussion focuses on the practical, theoretical, and measurement implications of considering aberrant self-promotion a distinct psychological motif.     

Religious Affiliation Influences on the Health Status and Behaviours of Students Attending Seventh-Day Adventist Schools in Australia

Students attending Seventh-day Adventist (Adventist) schools in Australia have been shown to have better health status and behaviours compared to secular norms, yet these schools cater for a high percentage of non-Adventist students. The purpose of this study was to investigate the influence of religious affiliation (Adventist/non-Adventist) on the health status and behaviours of students attending Adventist secondary schools in Australia. The sample included 1734 students who responded to a health and lifestyle survey that captured demographic details, self-reported height and weight, self-reported health status, mental health and select health behaviours. Students who identified themselves as Adventist reported significantly better health behaviours than the non-Adventist students in several behavioural domains, especially among the male students. However, this did not translate to a difference in health status. Further research is needed to understand the causal mechanisms responsible for the potential health advantage of Adventist students, which may include family or church religious influences.     

Correcting false information in memory: Manipulating the strength of misinformation encoding and its retraction

Information that is presumed to be true at encoding but later on turns out to be false (i.e., misinformation) often continues to influence memory and reasoning. In the present study, we investigated how the strength of encoding and the strength of a later retraction of the misinformation affect this continued influence effect. Participants read an event report containing misinformation and a subsequent correction. Encoding strength of the misinformation and correction were orthogonally manipulated either via repetition (Experiment 1) or by imposing a cognitive load during reading (Experiment 2). Results suggest that stronger retractions are effective in reducing the continued influence effects associated with strong misinformation encoding, but that even strong retractions fail to eliminate continued influence effects associated with relatively weak encoding. We present a simple computational model based on random sampling that captures this effect pattern, and conclude that the continued influence effect seems to defy most attempts to eliminate it.     

Word list generation performance in Alzheimer's disease and vascular dementia

Word list generation (WLG) was examined among clinical samples of individuals with Alzheimer's disease (AD) (n = 73) or ischemic vascular dementia (IVD) (n = 85), equivalent in age, education, current and estimated premorbid intellectual functioning, and proportion of men and women. The AD group performed significantly better than did the IVD group on lexical WLG, and a trend was observed indicating superior performance among the IVD group on categorical WLG. Within-groups, comparisons of group means, and profile analyses of individual performance patterns all indicated that persons with AD demonstrated a lexical > categorical pattern significantly more often than did IVD participants. The absolute difference in average performance between the AD and IVD groups on lexical and categorical WLG was small; the findings, however, generally support the clinical utility of the lexical > categorical WLG pattern in the differential diagnosis of AD and IVD. The patterns of performance support the presence of relative impairment in semantic processing among the individuals with AD and global deficits in retrieval and processing speed in individuals with IVD.     

GENDER-RELATED PATTERNS OF HELPING AMONG FRIENDS

In this study, the social role theory of gender and helping (Eagly & Crowley, 1986) was applied to understand gender differences in helping behavior. Relationships among criterion variables of time spent helping and help quality; and key predictors of problem severity, empathic tendency, anger, sympathy, closeness, causal controllability, coping, and self-efficacy were applied. Participants from a large community sample ( N = 1,004) described situations in which they helped a friend and completed questionnaires describing factors that influenced their actions. Recipients of the help also filled out similar questionnaires. It was found that across many problem settings women spend more time helping, give higher quality help, and feel more empathy and sympathy in response to their friends' problems. Further, the presence of anger toward a friend is associated with more time spent helping but a lower quality of help. In contrast, men rate their friends' problems as more controllable/blameworthy and experience more anger. Further, controllability has a greater influence on a number of help-related variables. For both men and women, self-efficacy and perception of problem severity are the greatest direct predictors of helping.