Comparison of Several Artistic Judgment Aptitude Dimensions Between Children in Chicago and Lisbon

Abstract

This research examines measurability, statistical interrelationships and association with test IQ, and cultural robustness of several artistic judgment dimensions. Symmetry, Simplicity, Uniformity, and Expressiveness dimensions had been previously validated in a comprehensive study of professional artists and studied in Chicago with children of diverse cultural background. The fifth dimension, the Visual Aesthetic Sensitivity Test (VAST), rests on much weaker validity foundations but was included in this research because of wide use among empirical arts researchers. Cross-cultural comparison of these dimensions has never been reported. The study sample is a culturally homogeneous group of native Lisbon schoolchildren (modal age = 10 years, N = 48). Results showed adequate psychometric reliability for all dimensions (>.80) except VAST (.57). Means were statistically indistinguishable between Portuguese and American children on Simplicity and Uniformity, which support their aptitude foundations. Portuguese children showed significantly higher preference for asymmetry (Symmetry dimension), which indicates consistency with professional artists. However, Portuguese children showed much lower preference for Expressiveness than did Chicago children, which is the opposite direction of artistic expectations. With exception of VAST, these dimensions show considerable promise for cross-cultural and developmental research.

MRI of an artistic judgment aptitude construct derived from Eysenck’s K factor.

“…Third, earlier research by the present team indicated that appreciation of visual aesthetics is a complex phenomenon, and there are probably multiple dimensions of talent that can be identified (Bezruczko & Schroeder, 1989; see also Eysenck, 1972). Hence, future studies of dimensions besides VD 1 and VD 2 could build on the findings presented here.…”

“…Second, the visual-preference measure here was interpreted to assess an aptitude for judging aesthetic value (AJ). This interpretation was based on validation research on the instrument, including a study of professional artists (Bezruczko & Schroeder, 1991, 1994), a cross-sectional developmental study (Bezruczko & Schroeder, 1996), and Eysenck’s research on visual preferences (e.g., Eysenck, 1971, and Eysenck & Castle, 1970). Nonetheless, it is possible that the instrument in the present study measures dispositions that are more susceptible to change than the discussion here may have implied.…”

“…Visual images presented in this research have been in clinical use providing psychometric support for career and vocational guidance in JOCRF testing offices for over 25 years. Professional artist validation during test development was based on an exceptionally large and diverse sample: 66 professional artists distributed among several American cities (Bezruczko & Schroeder, 1991). Internal consistency and construct validity studies were also conducted (Bezruczko & Schroeder, 1989, 1990).…”

“…Professional artist validation during test development was based on an exceptionally large and diverse sample: 66 professional artists distributed among several American cities (Bezruczko & Schroeder, 1991). Internal consistency and construct validity studies were also conducted (Bezruczko & Schroeder, 1989, 1990). Conventional factor analytic methods of investigating dimensionality were conducted, and Rasch model item calibrations were computed and evaluated with fit statistics (Wright & Stone, 1979).…”

“…Accomplished professional artists selected less complex random images of moderate redundancy more frequently when presented with a more complex alternative, whereas nonartists consistently preferred more complex stochastic images.The underlying perceptual mechanism activated by these test images appears quite powerful. Standardized effect sizes of group means separating professional artists and nonartists were substantial for both factors (Bezruczko & Schroeder, 1991), almost an entire standard deviation unit, respectively. Moreover, alpha reliability was greater than .90 (see also Bezruczko & Schroeder, 1989).…”