Multidimensional scaling of nominal data: The recovery of metric information with alscal

Multidimensional scaling has recently been enhanced so that data defined at only the nominal level of measurement can be analyzed. The efficacy of ALSCAL, an individual differences multidimensional scaling program which can analyze data defined at the nominal, ordinal, interval and ratio levels of measurement, is the subject of this paper. A Monte Carlo study is presented which indicates that (a) if we know the correct level of measurement then ALSCAL can be used to recover the metric information presumed to underlie the data; and that (b) if we do not know the correct level of measurement then ALSCAL can be used to determine the correct level and to recover the underlying metric structure. This study also indicates, however, that with nominal data ALSCAL is quite likely to obtain solutions which are not globally optimal, and that in these cases the recovery of metric structure is quite poor. A second study is presented which isolates the potential cause of these problems and forms the basis for a suggested modification of the ALSCAL algorithm which should reduce the frequency of locally optimal solutions.     

The relationship of hemispheric preference,as measured by conjugate lateral eye movements,to accuracy of emotional facial expression

This study investigated the relationship between hemispheric preference and communication accuracy of facial affect. Hemispheric preference was determined by rating conjugate lateral eye movements. Forty right-handed females were classified: 10 right-movers (left hemisphere preference), 15 left-movers (right hemisphere preference), and 15 bidirectionals (no hemispheric preference). Subjects were shown emotionally loaded slides portraying happiness, fear, and disgust and their videotaped facial expressions were independently rated for accuracy. It was predicted that negative emotions involved right hemisphere activity and negative affect involved left hemisphere activity. As expected, it was found that left-movers were significantly better than right-movers at nonverbally communicating disgust and fear; hemispheric preference was unrelated to expression of happiness. The accuracy scores of the bidirectionals generally fell in between those of the left-movers and those of the right-movers. The results were discussed in reference to the differential lateralization of positive and negative emotions.This study was conducted as a senior research project by the first author under the supervision of the second author.     

BEHAVIORAL RESEARCH IN PREVENTIVE DENTISTRY: EDUCATIONAL AND CONTINGENCY MANAGEMENT APPROACHES TO THE PROBLEM OF PATIENT COMPLIANCE

This study examined the effects of reinforcement on compliance with an oral hygiene education program. Patients 18 years of age or older who enrolled in an ongoing program at a periodontal practice received 3–5 sessions of instruction in preventive dental care. Using a between-subjects design, patients who entered the program during alternating months also had a portion of their fees refunded contingent upon improvements in their dental plaque scores. Pre- and posttreatment data showed that all subjects exhibited lower plaque levels following the program, but that greater improvements were seen in patients who were exposed to the fee reduction contingency. Plaque scores taken at a 6-month follow-up revealed some relapse for the Fee Reduction subjects. However, their scores were still substantially better than pretreatment, and better than those of the Education only subjects, whose data differed little from untreated Controls. Methodological and practical issues related to behavioral research in preventive dentistry are discussed.     

A new approach to Eysenck's theory of criminality

Studies examining Eysenck's theory of criminality by contrasting delinquent and non-delinquent groups have produced equivocal results. Some studies have shown criminals to score higher, as the theory predicts, on psychoticism, neuroticism and extraversion, whilst other studies have not. The current study examines the theory from a new approach by using cluster analysis to demonstrate the heterogeneity of personality types in delinquent sample (N =100) and in a comparison group (N = 100).Four personality types were found in each group. Two personality types which were present in the delinquent sample were not found in the comparison group. The first type consisted of individuals scoring high on neuroticism and extraversion whilst the second type consisted of individuals scoring high on psychoticism, neuroticism and extraversion. These types appeared to be consistent with Eysenck's theory of criminality.     

The role of physical appearance in managerial decisions

This study was designed to examine stereotypical beliefs about the physical appearance of employees and the effects of these beliefs on decision-making. Current stereotypes about obese and very thin people were assessed, and the effect of beliefs about obese and very thin people on subjects' evaluations of employee performance was measured. The findings indicated that subjects used information about employees' weight and body build differently, depending on whether they were responding to questions about discipline, the likelihood of recurrence of behavior, or their desire to work with certain employees. Overall, the results suggest that subjects exclude nondiagnostic aspects of employees (e.g., weight and body build) in some types of work decisions.     

Two investigations of “female modesty” in achievement situations

Two experiments examined motivations underlying the common finding that females present themselves more modestly than males in achievement situations. In Study 1, 388 first-year college students, primarily European-Americans, predicted their first semester grade point averages (GPAs) in one of 3 public and 2 private conditions, which varied the salience of modesty concerns and/or concerns about the others' feelings. In the public conditions combined, but not in the private conditions, women's predictions were lower than men's, although there were no gender differences in actual GPA. The public condition in which the others' feelings and modesty concerns were made salient accounted for this difference between men and women. In Study 2, 230 first-year college students predicted their first-semester GPAs in private, in public to a “nonvulnerable” other, or in public to a “vulnerable” other (someone who supposedly had earned a low GPA). Women's estimates were lower than men's in the latter condition only and lower than their estimates in the other conditions. These results suggest that relational motivations, rather than a simple lack of self-confidence or modesty alone, are a factor in gender differences in self-presentation of achievement.     

Reflections …

Remarks by the symposium discussant attempt to briefly and broadly summarize the panelists’ thoughtful discussions of thePast is Prologue … for the Society for Computers in Psychology. The unique role computers have played, are playing, and will play in the behavioral sciences is reviewed. Although enthusiasm and promise for the future roles of computers as tools, as models, and as sources of study remains high, a caveat is offered.     

Strategy use in question-answering: Memory strength and task constraints on fan effects

The time taken to recognize a studied fact increases as a function of the number of other previously studied facts sharing concepts with the test fact. The phenomenon, known as the “fan effect,” has been shown to disappear and sometimes even reverse itself when the set of facts are thematically related. The shift from interference toward facilitation occurs only when subjects can use a plausibility-like strategy. In this experiment, subjects learned variously sized sets of rehted facts about fictitious people. Subjects were asked to make either recognition judgments (“Did you study this fact?”) or consistency judgments (“Is this fact consistent with what you studied7”). Subjects made these judgments both the day the material was acquired and 2 days later. The research reported here supports the hypothesis that, with delay, there is a shift in tendency toward more use of the plausibility strategy, away from the careful strategy of searching for an exact match that produces the fan effect. The plausibility strategy produced either a speedup with greater fan or an increase in error rates when the strategy was inappropriate. Plausibility effects were larger at a delay, in both reaction time and error patterns, regardless of whether subjects were asked to make consistency judgments or recognition judgments. Also as predicted, response times became faster as the tendency to adopt the plausibility strategy without first trying direct retrieval increased.