The physical factor structure of random forms and their judged complexity1

This study relates the perceived complexity of 2fl random forms to their physical factor structure. Ten principal axes, accounting for 94 per cent of the total variance of 24 physical measures, were rotated using the Varimax criterion. Factor scores for each form were correlated with the complexity ratings of the forms by each of 11 Ss. A single factor accounted for most of the variance in the complexity ratings. This factor was best described by four physical measures: the number of turns in the form, the length of the perimeter, the perimeter squared to area ratio, and the variance of the internal angles of the form.     

Innovation Motivation and Artistic Creativity

Innovation motivation is a social learning model of originality comprising two variables: the need to be different and innovation expectancy. This study examined their contribution to artistic creativity in a sample of undergraduates. Participants completed measures of both innovation motivation variables as well as intelligence, adjustment, and artistic training. Participants completed three drawings using colored pencils: a human artifact, a non‐human life form, and a person. Choice of main subject and details were tabulated for each drawing and originality scores were derived based on the infrequency of each thematic element. The drawings were also evaluated for technical proficiency and creativity by two master's level art therapists. The need to be different and innovation expectancy each predicted the originality of each drawing; together, they predicted 42% of the variance in the composite originality score. The need to be different and innovation expectancy also predicted proficiency and creativity ratings, especially the latter. The other variables yielded weaker, less consistent predictions. In general, both innovation motivation variables still predicted significant proportions of the variance in originality scores and creativity ratings even after controlling for other variables.     

Confounds in pictorial sets: The role of complexity and familiarity in basic-level picture processing

Complexity is conventionally defined as the level of detail or intricacy contained within a picture. The study of complexity has received relatively little attention-in part, because of the absence of an acceptable metric. Traditionally, normative ratings of complexity have been based on human judgments. However, this study demonstrates that published norms for visual complexity are biased. Familiarity and learning influence the subjective complexity scores for nonsense shapes, with a significant training x familiarity interaction [F(1,52) = 17.53, p < .05]. Several image-processing techniques were explored as alternative measures of picture and image complexity. A perimeter detection measure correlates strongly with human judgments of the complexity of line drawings of real-world objects and nonsense shapes and captures some of the processes important in judgments of subjective complexity, while removing the bias due to familiarity effects.     

The span of apprehension: Form Identification as a function of amount of information displayed

Observers were asked to identify each member of a brief display of geometrie forms. The information content of the displays was manipulated by varying the number of forms displayed and the number of relevant dimensions of each form. The main findings were: (a) Performance was approximately invariant with the number of forms displayed and depended to a minor degree upon the relevant information in each form. (b) Serial-position curves were obtained that had a strong primacy effect, (c) The number of correct identifications per trial was approximately normally distributed, with greater variance than if the forms were independentlyidentified.These findings are discussed in relation to both “coding rate” and “memory capacity” models of immediate visual memory. Neither model accounts for the data. Undetermined organizational factors are suggested to play a role in immediate visual memory. This task appears to depend upon form Identification processes different from those indicated by RT measures.     

Imagery: A multidimensional construct

Two sets of words were independently rated for imagery. Each set of words contained concrete nouns, e.g., bell, and adjectives, e.g., yellow, as well as nouns and adjectives describing affects, e.g., angry, and physical states, e.g., hungry. Factor analysis of the ratings indicated two factors or sources of variance in each set of words. These were interpreted as a Visual factor, and an Experienced States factor. These results indicated that imagery ratings are best conceptualized as reflections of multiple modes of imagining rather than ease of visualization alone.     

MULTIFACETED CONCEPTIONS OF SELF-OTHER RATINGS DISAGREEMENT

Self-other ratings disagreement (S-ORD) is the extent to which one's self-rating of performance differs from ratings made by Other people (e.g., supervisors). As increasing number of organizations adopt self-assessment as a tool for managing performance, S-ORD is becoming a central concern in performance appraisal research. Most past research, however, only examines S-ORD in terms of the levels of ratings, and ignores other forms of disagreement. This paper identifies 7 different forms of S-ORD that are detectable through the use of confirmatory factor analysis (CFA). The causes for each form of S-ORD are discussed, as well as its conceptual and practical implications. The procedure is illustrated by comparing 332 managers' self-evaluations with their supervisors' evaluations. An extension of the CFA model to a full structural equation model is used to illustrate and discuss the fragmentary nature of existing approaches to S-ORD. It is shown that valid inferences concerning S-ORD cannot be made unless various forms of S-ORD, and not merely the differences in overall ratings, are detected and dealt with.     

Book Review

Performance-based measures and ratings of executive functions were examined in a sample of adolescents with attention deficit/hyperactivity disorder (ADHD) and comparison controls. Performance-based measures of executive function included inhibition, working memory, set shifting, and planning, and ratings of these same executive functions were completed by parents and teachers. Adolescents with ADHD demonstrated lower executive function performance than controls and displayed elevated ratings on the executive function ratings by parents and teachers. Significant associations were obtained between the performance-based measures and the parent and teacher ratings, but each measure was not uniquely associated with its respective scale on the rating scales. When performance-based measures and ratings were examined as predictors of ADHD status, the parent and teacher ratings entered as significant predictors of ADHD status. Further commonality analyses indicated that performance-based measures accounted for little unique variance in predicting ADHD status and also displayed little overlap with the behavioral ratings. These findings highlight the diagnostic utility of behavioral ratings of executive function in predicting ADHD status; however, behavioral ratings should not be assumed to be a proxy for performance on measures of executive function in clinical practice.     

The prediction of averages

Criterion measures are frequently obtained by averaging ratings, but the number and kind of ratings available may differ from individual to individual. This raises issues as to the appropriateness of any single regression equation, about the relation of variance about regression to number and kind of criterion observations, and about the preferred estimate of regression parameters. It is shown that if criterion ratings all have the same true score the regression equation for predicting the average is independent of the number and kind of criterion scores averaged.Two cases are distinguished, one where criterion measures are assumed to have the same true score, and the other where criterion measures have the same magnitude of error of measurement as well. It is further shown that the variance about regression is a function of the number and kind of criterion ratings averaged, generally decreasing as the number of measures averaged increases. Maximum likelihood estimates for the regression parameters are derived for the two cases, assuming a joint normal distribution for predictors and criterion average within each subpopulation of persons for whom the same type of criterion average is available.     

Form recognition: Individual strategies in multivariate cue utilization

Twenty-five Ss were each presented 1,250 trials in a form-recognition task requiring S to find a form among 25 other forms on each trial. Fifty different eight-sided randomly derived polygons were used as stimuli, each presented in 25 different problems. Factors based on physical measures of the stimuli were correlated with observed latencies using multiple regression to assess cue-utilization strategies, while beta-weight profiles and intercorrelations of the Ss’ predicted latencies were used as indices of individual differences. Multiple correlations of physical-factor scores with observed latencies accounted for a significant proportion of the variance for all Ss, while between-S correlations of predicted latencies were never below .90. By all criteria employed, use of stimulus information was highly consistent among Ss. Replication of the experiment for one S revealed good reliability and nearly identical strategies.     

Mental rotation of possible and impossible four-cornered toruses

Possible and impossible four-cornered toruses were used as stimuli in a mental rotation paradigm. Rotation rates varied with number of “net changes in depth,” a variable that closely corresponds with ratings of impossibility, rather than with formal distinctions between the possible and the impossible. Rotation rates were also affected by perimeter shapes but not in accord with perimeter complexity as defined by number of vertices. Experiment 2 attempted to separate net changes in depth from figure symmetry and found symmetry to be the more reliable predictor of rotation rates, suggesting that symmetry may underlie the influence of net changes of depth.     

Amount of stimulus exploration and preference as differential functions of stimulus complexity

Two sets of photographic slides, one made up of scenes from the geographic environment, the other of works of non-representational modern art, were scaled for complexity by obtaining judges’ ratings of amount of variation present on several specified stimulus attributes. Fourteen slides defining a sevenpoint scale of complexity were selected from each set and given to college students to obtain measures of (a) amount of exploratory behavior (number of times S chose to expose each slide briefly), and (b) preference (evaluative ratings on a seven-point scale). In accordance with prediction, the former measure emerged as a linearly increasing function of complexity, while the relationship between complexity and preference was curvilinear, reaching a maximum at an intermediate level of complexity. The results are related to Berlyne’s distinction between specific and diversive stimulus exploration, and implications for the study of aesthetics are discussed.     

Complexity, extremity, and affect in male and female judgments.

Two studies were conducted to elucidate aspects of the interpersonal judement process as it is applied to and by males and females. Measurement of judgments by means of the Role Construct Repertory Test allowed consideration of cognitive complexity scores, as well as the use of extreme ratings and a total affective rating. Consistent findings from both studies suggest that males and females differ in their use of extreme scores and total affective ratings, but not in complexity. Characteristics of the stimulus person affect each of the three dependent measures. Interrelationships among the three judgmental measures provide information about the methodological and conceptual interplay of various factors in interpersonal judgment.     

Predictors of physical aggression in children with attention-deficit/hyperactivity disorder

The present investigation examined factors that predict physical aggression in children with attention-deficit/hyperactivity disorder (ADHD). Stepwise, multiple regression-analyses were used to examine predictors of children's physical aggression as rated by parents at a 1-year follow-up point and by teachers at both 1- and 2-year follow-up points. Early parent and teacher ratings of verbal aggression (ie, cursing, teasing, and threatening) accounted for the greatest proportion of the variance in physical aggression ratings obtained at follow-up. None of the other predictor variables, including early ratings of physical aggression and ADHD behaviors, contributed significant additional variance beyond that accounted for by early verbal aggression ratings. Temporal and cross-informant analyses revealed that the relationship between verbal aggression and later physical aggression was situation-specific for teacher ratings but not parent ratings. Although physical aggression may emerge early in development, these data suggest that verbal aggression represents a stable, temperamental characteristic that may be of greater value than early physical aggression for predicting later physically aggressive acts.     

Quantitative and structural factors in the judgment of pattern complexity

The processes underlying judgment of pattern complexity were investigated using patterns containing eight dots in a 4 × 4 matrix. Thirty-two subjects rated the complexity of 140 patterns that were taken from patterns generated by 12 other subjects in a preliminary experiment. Factor analysis of mean complexity ratings and 12 physical variables, which objectively described configuration of dots, revealed a “quantitative factor” and a “structural factor.” Mean complexity loaded on both factors. Multiple regression analysis showed that the mean complexity was well predicted by combination of a quantitative variable with a structural variable. Analysis of individual data also suggested that individual differences in complexity judgment could be explained by the weights on these two factors. In addition, a rating experiment was carried out in which 60 patterns were presented in limited stimulus durations (50 msec, 200 msec, 1 sec, or 4 sec). The quantitative variables were highly correlated with the complexity ratings in all conditions, whereas the structural variables affected the ratings in long durations but not in short durations. These results support the view that visual complexity is judged through two processes: One is a fast process that evaluates quantitative aspects in a stimulus pattern, and the other is a slow process that detects the structure in the pattern.     

Egoistic and Moralistic Self‐Enhancement in the Eye of the Beholder: A Cross‐Informant Study

This study was conceived to assess the agreement between self‐ and other reports of egoistic (E) and moralistic (M) self‐enhancement (SE). A self‐report scale assessing E‐SE and M‐SE was filled out by 304 participants (mean age = 39.11, SD = 10.12, 54% female). The scale, with items worded in third person, was also completed by 304 informants (one for each target participant). Data were subjected to a correlated‐traits correlated‐methods model. Confirmatory factor analysis yielded two distinct factors for both self‐ and other ratings of E‐SE and M‐SE. A significant correlation across raters was observed for each factor. Proportion of trait variance was .58 for E‐SE and .35 for M‐SE. Proportion of method variance was .24 and .41, respectively. E‐SE and M‐SE were substantially correlated within each data source. However, the two measures became orthogonal after method variance was partialed out. Measures of E‐SE and M‐SE include both trait and artifactual components of variance, which are likely to reflect individual differences in the style of self‐presentation and response bias (i.e., tendencies to exaggerate agentic and communal qualities). The trait components of E‐SE and M‐SE represent two unrelated tendencies that can be captured and distinguished not only through self‐judgments, but also by an external observer.     

The Impact of Contextual Self‐Ratings and Observer Ratings of Personality on the Personality–Performance Relationship1

This study examined 2 possible ways of increasing the predictive validity of personality measures: using observer (i.e., supervisor and coworker) ratings and work‐specific self‐ratings of Big Five personality factors. Results indicated that among general self‐ratings of Big Five personality dimensions, Conscientiousness was the best predictor of in‐role performance, and Agreeableness and Emotional Stability were the best predictors of organizational citizenship behavior (OCB). Observer ratings of personality accounted for incremental variance in job performance (in‐role performance and OCB) beyond that accounted for by general self‐ratings. However, contrary to our expectations, work‐specific (i.e., contextual) self‐ratings of personality generally did not account for incremental variance in job performance beyond that accounted for by general self‐ratings.     

BEHAVIORAL OBSERVATION SCALES FOR PERFORMANCE APPRAISAL PURPOSES

Behavioral items ( N = 78) critical to the job success of logging supervisors were developed from 1204 critical incidents, the frequency with which a supervisor ( N = 300) engaged in each behavior was rated on a 5-point Likert type scale by two sets of observers. A factor analysis reduced the items to 38 and 33, respectively, for the two sets of observers which in turn constituted 10 and 11 factors or criteria for performance evaluation purposes. Multiple regression equations based on composite scores were used to predict cost-related measures of logging crew effectiveness. The shrinkage in Rs after double cross-validation was moderately small. Moreover, the behavioral observation scales (BOS) that were developed by factor analyzing the observation ratings had moderately high reliability and accounted for more variance in the cost-related measures than did the BOS developed by traditional judgmental clustering techniques. The similarities and differences between BOS and BES procedures are discussed.     

The stability of structural measures derived from repertory grids

Although the use of repertory grids in psychological research has proliferated, studies of their psychometric properties are relatively rare. For this reason, we studied the reliability and convergence of several measures of cognitive structure derived from grids, including intensity, percentage of variance accounted for by the first component, cognitive complexity, ordination, extremity of ratings, self-ideal discrepancy, and self-other discrepancy, as well two measures of rating stability, construct consistency and factor loading consistency. Eighty-two Spanish (Catalan) and American participants completed small and large grids on each of four occasions, which enabled exploratory analyses of the impact of grid size and subject characteristics on the structural scores obtained. Results indicated that the majority of scores showed impressive test-relest reliability (modal r = .85 for periods up to 1 month), but also suggested a gradual tightening effect across subsequent administrations. In general, scores intercorrelated in theoretically anticipated ways, converging on the assessment of cognitive differentiation and discrimination of elements within construct dimensions. The results have implications for the reactivity of structural scores to variations in grid format, as well as for use of grids in the study of sex differences and cross-cultural comparisons.