Carl L. Thornton

COGNITIVE STYLE DIFFERENCES BETWEEN ENGINEERS AND COLLEGE STUDENTS

The description of Witkins field-independent individuals as those who tend to be analytical, logical, and able to extract subtle aspects from problems for analysis bears a close resemblance to the description of engineering job functions. It was therefore hypothesized that engineers would be more field-independent than Witkins standardization sample. It was determined that 46 engineers and technicians were significantly more field independent, as measured by the rod-and-frame test, than Witkins standardization sample. Alternate explanations for the obtained results are discussed, including intelligence, sampling, and shift to adult status.

Human Factors Evaluation of a Computer Based Information Storage and Retrieval System

The human factors aspects of a computer-based information storage and retrieval system were evaluated in three government intelligence services using questionnaire and interview techniques. It was found that many factors entered into satisfaction with the information system, including familiarity, ease of use, and importance. Other relevant factors included training in the use of the system, amount and type of information to meet needs in the system, and tolerances of individuals for irrelevant material in the output of searches. The interaction of these factors is discussed in relation to satisfaction with the system.

Quantitative Methods of Calculating the d' of Signal Detection Theory:

GREEN and Swets (1966) noted that 1,000 to 5,000 observations requiring 10 to 12 hours of testing per subject are necessary to generate an ROC curve. Despite this large expenditure of time and effort for the sake of accuracy, the determination of the important parameter d’ is made by visual comparison to an approximately fitted theoretical curve. As Green and Swets (1966, p. 404) stated, &dquo;It is common practice to fit curves to ROC data by selecting from a family of theoretical curves the one that appears upon visual inspection to give the best fit. This procedure is not very satisfactory but, apparently a more objective one is not available.&dquo; To overcome this problem, formulas were derived which should be useful with various types of data.

Field Dependence and Target Identification

Target identification has usually been studied with respect to environmental parameters which affect the performance of the “human operator.” Rather than use this method, which treats individual differences as sources of error, the investigators studied individual differences and related them to Witkins concept of perceptual style. Witkin and his associates were able to differentiate subjects on their ability to pull a visual item from an embedding context. Two experiments were conducted using the Embedded Figures Test (EFT) as a measure of perceptual style. Significant correlations between perceptual style and the ability to correctly identify targets in aerial photographs were found. Since a great deal of research has been conducted in the area of perceptual style, it is possible to utilize this research in connection with problems associated with target identification. Implications in the areas of experimental design, selection, and training were discussed.

Relation Between Hidden Figures Test and Rod and Frame Test Measures of Perceptual Style

or dependent (of the visual field) on the basis of the RFT. Found to be related to the RFT was an embedded figures test (EFT), in which S identified a simple geometric figure visually embedded in a complex design. The EFT was the basis for Witkin, Dyk, Faterson, Goodenough, and Karp’s (1962) modification of the definition of perceptual style, now simply the ability to pull an item from an embedding context. The results (Witkin et al., 1954, p. 85), however, indicate a possible paradox. Although both tests supposedly measured the

TWO METHODS OF DETERMINING BODY SENSITIVITY: A COMPARISON AND EVALUATION

Field independence, as measured by the magnitude of error in judging true vertical on the Rod and Frame test, has been considered to be related to the utilization of body cues by Witkin, et al. (1954). Benfari and Vitale (1965) have used one portion of the Rod and Frame test, analyzed in terms of direction of error from the vertical, to determine if Ss were “body-oriented.” If field-independence and body-orientation measures tap the same attribute, then they should be highly correlated. Body orientation was not significantly related to field independence when data from 46 adult males were analyzed using both procedures.

VISUAL FUNCTIONING AND EMBEDDED FIGURES TEST PERFORMANCE

A previous report (Barrett, Cabe, & Thornton, 1967) showed no significant relationship between perceptual style as measured by the Rod-and-frame Tesc (RFT) and measures of visual functioning. Witkin, Dyk, Faterson, Goodenough, and Karp (1962) postulate that perceptual style in its most general form is the ability to disembed a simple item from a more complex context, and the Embedded Figures Tesc (EFT) is, therefore, a valid measure of perceptual style as well as the RFT. Since Witkin, Lewis, Hertzman, Machover, Meissner, and Wapner ( 1954) found that the EFT and RFT correlate only between .)O and .60 (for men), the two tests may be measuring different attributes. They never, however, tested the possibility that perceptual style might contain a component of visual functioning. The present study, then, tested the relationship between EFT scores and measures of both near lateral phoria and mac~~lar stereopsis, both of which were considered elemental in efficient visual functioning. Of a sample of 44 male engineering and technical personnel, previo~isly tested for near lateral phoria and macular stereopsis by use of a Keystone Orthoscope and for perceptual style using the RFT (as reported by Barrett, er al., 1967), 37 Ss were retested approximately 6 mo. later using a form of EFT supplied by the Educational Testing Service. Remaining Ss were not available for retesting due to transfers, heavy work loads, etc. Due to the nature of the distributions, Spearman rhos were computed among the measures. Rhos of -.07 ( 9 > .05, non-corrected) were obtained between both EFT and phoria and EFT and stereopsis. Lack of significant relationship suggests that visual functioning is not a factor in performance on the EFT. Furthermore, if the EFT may be considered a measure of perceptual style, the present results support the previous finding that visual functioning is not a factor in perceptual style.

PERCEPTION OF DEPTH AS MEASURED BY MAGNITUDE ESTIMATION

11 Ss used the magnitude estimation technique to judge depth in 3 three-dimensional scenes of varying complexity. Also Ss perceptual style, as measured by the rod-and-frame test, was determined to test the hypothesis that perception of depth is significantly related to Ss perceptual style. Each S was given 30 trials under 5 eye conditions, i.e., left eye occluded, right eye occluded (both aware and unaware of occlusion), and neither eye occluded. No significant relationship was found between various eye conditions and judgment of depth or between Ss perceptual style and judgments. A significant relationship was found among scenes, with the more complex scene judged as having greater depth.

The Use of Reciprocals in Comparing Group and Individually Administered Tests

THERE are many instances where individual and group forms of a test may be correlated. The group form score is the number of items completed in a specified time period while the individual form score may be the time taken to compete a fixed number of items. At first glance these two scoring techniques appear identical. However, even if two tests were to vary identically it would be impossible to obtain a 1.00 correlation because of the nature of the scoring units.

RELATION OF PERCEPTUAL STYLE TO MEASURES OF VISUAL FUNCTIONING

Measures of Witkins concept of perceptual style were obtained with a Rod and Frame apparatus for 44 Ss and correlated with individual measures of macular stereopsis and phoria obtained from a Keystone Orthoscope. As none of the Spearman rbos were significant, individual differences in perceptual style appear not to be a function of the measured individual visual characteristics.