David L. Alderton

Reaction time variability and intelligence: A “worst performance” analysis of individual differences

Abstract Reaction time distributions from 303 subjects were partitioned into 16 fast-to-slow latency bands. Average latencies for the 16 bands were then correlated separately with various indices of mental ability. The slowest bands (or “worst trials”) were by far the best predictors of intelligence and working memory performance. The slowest bands also drove the individual differences in variability, and the variability/intelligence correlation. An error-based production model for worst trials was tried and abandoned.

Understanding spatial ability

Spatial ability is discussed in terms of psychometric factors and information processing research. Reanalysis of major psychometric studies suggests two major spatial factors — spatial relations and spatial visualization ability. Apparent differences between these factors in speed versus power and cognitive complexity are verified by process analyses of individual differences in spatial relations and visualization ability. Information processing studies suggest multiple sources of individual differences such as process execution speed, quality and capacity of representation, process coordination and strategies for problem solution. Consideration is also given to additional issues for research as well as implications for testing.

The ability to reason about movement in the visual field

Abstract In tests of visual-spatial reasoning, the examinee must reason about some aspect of a picture. Because of the limitations of paper-and-pencil tests, the picture do not contain any moving elements. These are referred to as static displays. Outside the test situation, people frequently have to reason about visual displays that contain moving elements. These will be called dynamic displays. This experiment investigated whether or not individual differences in the ability to reason about dynamic displays are predictable from tests using static displays. One hundred and seventy people were given a battery of tests, including six standard paper-and-pencil tests of visual-spatial ability, five computer-controlled tests using static displays, and six computer-controlled tests using dynamic displays. Several multivariate analyses indicated that the ability to reason about dynamic visual motion is distinct from the ability to reason about static displays.

Stimulus complexity effects in visual comparisons: the effects of practice and learning context

This study examined visual comparison performance for 6-24-point random polygon stimuli (Cooper & Podgorny, 1976). Stimulus complexity effects decreased with practice, consistent with Bethell-Fox and Shepard (1988). A difficult discrimination context produced greater complexity effects than an easy discrimination context, consistent with Folk and Luce (1987). The difficult discrimination context also led to more stimulus-specific learning and diminished stimulus complexity effects. Increased stimulus learning resulted in continued skill acquisition, better transfer, and less performance disruption when the task context was equated for all Ss. It is argued that improvements in performance in a perceptual comparison task are not solely a function of the amount of practice provided in responding to particular stimuli. The context in which responses are elicited is equally important and must be accommodated in theories of skill acquisition.

Risk Factors for Misconduct in a Navy Sample

Understanding antisocial behavior and organizational misconduct is an important objective, because these maladaptive behaviors are disruptive and costly to organizations and to society as a whole. The objective of this study was to identify psychosocial risk factors for misconduct and antisocial behavior in a sample of Navy personnel. A group of sailors (n = 158) who had engaged in significant misconduct were compared with a demographically similar group of sailors (n = 288) who had not engaged in misconduct and who were in good standing with the Navy. The psychosocial variables that emerged as the most important risk factors for antisocial behavior were alcohol use (odds ratio [OR] = 2.42), high impulsivity (OR = 2.20), high trait hostility (OR = 1.79), and antisocial behavior of friends (OR = 1.65). The implications of these results for the military and for research on antisocial behavior are discussed.

Arithmetic reasoning gender differences: Explanations found in the Armed Services Vocational Aptitude Battery (ASVAB)

Abstract This study examined the factor structure of the Armed Services Vocational Aptitude Battery (ASVAB) Form 7 to assess variables of environmental and spatial ability theories that explain a male advantage in arithmetic reasoning. No gender differences were found in an othogonalized hierarchical solution of the general factor and technical, quantitative, verbal, and spatial factors that would suggest differential problem-solving strategies as a function of spatial ability. However, a small technical factor loading (approximately .10) for Arithmetic Reasoning for both males and females (considered salient when compared to loadings for Mathematics Knowledge and Numerical Operations) suggested environmental or experiential interactions contribute to problem-solving ability. Consistent with these findings, an approximate .10 correlation between the sex variable and Arithmetic Reasoning for a combined group analysis was attributed to the technical factor.

Training for Optimal Strategic Skills

The present research examined the impact of discrimination difficulty on the acquisition and transfer of strategic visual discrimination skills for meaningful stimuli. Participants were trained to discriminate between airplane silhouettes that varied in similarity. Some participants were trained to make difficult discriminations between similar airplane silhouettes, whereas others were trained to make easy discriminations between dissimilar airplane silhouettes. Participants were then transferred to making discrimination judgments at all similarity levels. The results suggest that initial training difficulty influences strategic skills even when participants have a priori strategies for processing stimuli. The findings improve our understanding of strategic skill acquisition, and training suggestions are discussed.

Dynamic Administration of a General Intelligence Test.

Abstract The benefits of dynamic testing are thought to include; (a) a reduction in strategic variance, accompanied by (b) a test score increase for “disadvantaged” subjects. Sometimes forgotten, however, is that these accomplishments are illusory unless they support a specified goal (e.g., better validity). In the present study, we examine the benefits of dynamic test administration with the Ravens Advanced Progressive Matrices (APM) test of general intelligence. The results indicate that, while APM scores were significantly increased by dynamic procedures, important criteria such as reliability and construct validity were not enhanced. We conclude that the choice of dynamic procedures depends on both the ability construct and the purpose of testing, and should be justified on a case-by-case basis.

Components of Spatial Ability

Three areas of theory and research related to spatial cognition are reviewed and their integration is discussed. These include factor analysis of spatial ability, information processing theories of spatial cognition, and developmental changes in spatial processing. A primary goal is to illustrate how integration of these perspectives, both theoretically and empirically, benefits the understanding of this domain of intellectual ability and can impact test design and instructional applications.Primary consideration is given to spatial ability factors such as perceptual speed, spatial relations and spatial visualisation. The differentiation of these factors, their relationships and their relevance for technical skills training is briefly reviewed. Primary emphasis is given to cognition process approaches to analysing spatial ability and the sources of individual differences in spatial processing. Results are presented from a series of studies examining performance in perceptual comparison tasks, spatial relations and spatial visualisation tasks through componential analysis techniques. In all cases, information processing models form the basis for analysing speed, accuracy and strategy differences in the execution of mental processes. Conclusions regarding the various components of spatial ability and their manifestation in various spatial ability tests and factors are presented and related to general theories of spatial information processing such as Kosslyn’s (1981) theory of mental imagery. Extensions of this process approach to developmental analyses and analyses of the trainability of spatial components are also presented. The implications of integrating psychometric, information processing, developmental and learning approaches to the study of spatial ability are considered with regard to test development, diagnostic testing functions and adaptive instruction. Specifically it is shown how process analysis can be used to systematically design problems for presentation in computer comtrolled testing. The benefits of such an approach to test design include the ability to measure overall ability as is done by present testing procedures as well as specific cognitive components emanating from theory of performance. This type of approach also offers the possibility of using tests for diagnostic as well as predictive purposes. Finally consideration is given to extensions of this approach to analysing individual differences in the capacity to acquire greater skill in various components of spatial information processing. Implications of this assessment approach are presented with respect to the uses of tests in adaptive instruction in technical skills training areas.