Arnold Powell

Correlates of Adult Locus of Control

Correlations between scores on the Adult Locus of Control Scale and a number of personality scales supported the hypothesis that the control scores are related to need achievement, anxiety, psychopathology, socioeconomic status, ethnic group membership, and intellectual ability. There was some evidence that locus of control scores are also related to teacher effectiveness.

Adult Locus of Control and Mental Ability

Correlations between I-E and Adult Locus of Control scales, and a measure of mental ability, supported the hypothesis that these scales provide alternative measures of feelings of control for adults. Support was also obtained for the hypothesis that an internal control is associated with greater mental ability.

Semantic Clues in Comprehension of Novel Sentences

It has been suggested that semantics provide important clues to underlying linguistic structure. The present results supported this suggestion, since self-embedded sentences containing 3 relative clauses were easier to decode when semantic clues were present. Lack of transfer of learning from one set of sentences to another set of similarly constructed sentences was interpreted as supporting models of linguistic competence and language acquisition that give greater emphasis to context-specific processes.

Organization and recall of pictures and words in children

Recall and category clustering of pictures and words were investigated among Ss from the third, fourth, and sixth grades. Both recall and organization were greater among older Ss and for pictorial stimuli. Recall was correlated with scores on a standardized achievement test designed to measure paragraph comprehension (PC). The Z-score measure of clustering was found to yield more reliable and consistent results than the adjusted ratio of clustering. Some of the change in recall as a function of age was associated with both category clustering and PC scores. This finding was interpreted as supporting the view that memory does not develop as an isolated skill, but reflects the child’s developing ability to transform surface structures into abstractly organized deep structures. It was suggested that pictures are easier to recall because the transformation of surface structure representations into organized deep structures is less influenced by the context of encoding than it is in the case of words.

Memory Span, Organization, and Transformations in Recall of Pictures and Words

Immediate memory span, category clustering, and reading comprehension scores accounted for approximately 60% of the variance in average free-recall performance across three trials given 30 high school Ss in special reading classes. Immediate memory span tended to correlate more highly with recall of words than pictures, while reading comprehension tended to correlate more highly with recall of pictures. Category clustering correlated only slightly more with recall of words than pictures. The findings were interpreted as supporting a model of recall in which pictorial stimuli are considered to afford a greater opportunity for simultaneous processing in an imaginal information-processing system.

Conceptual and Associative Context Effects on Speed of Word Recognition

Both conceptual and associative relationships between clue words and target words influenced the speed of recognition of tachistoscopically presented target words. When clue and target words were associated, but in a conceptual context different from that in which the clue word occurred, presentation of the clue word actually interfered with recognition speed. This finding was interpreted as supporting the hypothesis that the information relevant to a particular lexical item in working memory is determined by the context in which it is presented. Another finding was that experience in a high school reading laboratory increased recognition speed but did not influence the effect of conceptual and associative contexts, which suggested that these contexts affect memory and memory search rather than perception directly.

Further Note on Rule Utilization in Solutions of Abstract Picture-Analogy Problems

As a further test of a rule model of analogical reasoning, 24 problems from the Raven Progressive Matrices were given 93 Ss, run in a 2 × 2 × 2 factorial design in which there were two presentation intervals (20 vs 45 sec.), two stimulus conditions (the first two rows of each matrix were either present or absent), and two rule conditions (some Ss had access to E-generated rules). Over-all there were fewer errors with the longer interval, when the stimulus items of each matrix were present and when Ss had access to the E-generated rules for solving the problems. Providing rules had a larger effect when the stimulus items were not presented than when they were presented. The results were interpreted as providing support for a rule model of analogical reasoning, though some of the limitations of this model were noted.

RULE UTILIZATION IN SOLUTIONS OF ABSTRACT PICTURE-ANALOGY PROBLEMS

Two experiments tested the hypothesis that rules are generated and applied in solving analogy problems by providing one group of Ss with the appropriate rules. Analogy problems from the Lorge-Thorndike Intelligence Test were employed in Exp. I, and problems from the Raven Progressive Matrices were employed in Exp. II. The results from both experiments did not support the hypothesis. It was suggested that the possibility, that the cognitive processes involved in comprehending E-generated rules may be as complex as those involved in generating rules from the stimulus materials, should be investigated further.

Reward Magnitude, Reward Shift, and Resistance to Extinction in the Mongolian Gerbil

A 3-phase (acquisition, shift, extinction) study of reward magnitude in the gerbil found that running speed was greater and resistance to extinction less with 4 pellets than with 1 pellet—similar effects to those previously reported for rats. However, the effects of a reward shift for gerbils were found to be different from those for rats; a shift from 4 to 1 pellets led to a gradual increase in running speeds. The results were compared to those reported for rats, turtles, and goldfish; results with gerbils were considered to be consistent with a model in which reward magnitude influences associative strength directly and where incentive motivation plays a minor role.

Mathematical models of delayed reactions in rats

Abstract Two models of delayed reactions, the Uncertain Trace and the Linear Trace-Decay models, were proposed and tested in a delayed-reaction T maze. S s trained with a short delay attained lower asymptotic performance levels than did S s trained on the same discrimination, but with no delay between stimulus and response. This was interpreted as reflecting the greater probability that S s forget the cue during a delay interval, and three additional findings eliminated an alternative explanation that the cue is not attended to in the delay condition. The Linear Trace-Decay model provided a more accurate description of quantitative aspects of the data; however, neither model provided a completely adequate fit to the data.