- Repository URL:
- Cognition; Visual perception; Learning, Psychology of; Human information processing
thesis / dissertation description
An experiment was conducted to test the hypothesis that word potency, frequency, and certain graphic characteristics affect word recognition in the parafoveal field under very brief exposure conditions. The rationale for the experiment was based on information-processing theories of cognition, theories of word recognition and reading processes, and studies of visual processes in the parafoveal-peripheral field. By (1) pairing high potency words with neutral words matched on frequency, word length, and similar graphic characteristics, (2) separating them by six spaces, and (3) presenting them at exposures too brief for eye movements to occur, a test was made of the hypothesis that word potency is a significant variable. Subjects were 92 male and female college students, who were tested individually at a tachistoscope. Controls saw word-pairs of neutral words only, while Experimental subjects saw word-pairs in which potent words were paired with neutral words. The first sample, 36 males and 16 females, were exposed to durations of 30 msec. The second sample, 6 males and 14 females, were exposed to 40 msec duration. Two-way ANOVA and regression analysis were done on each sample data. Sample I showed a significant main effect of Words. Regression analysis indicated that besides Word Potency, graphic characteristics of Ascenders (b, l, h, t, k) and Word Density were significant variables. Sample II showed a significant interaction effect of Sex by Words. Regression analysis showed Sex to be the only significant variable. Conclusions were that word potency is a significant variable, but that under the experimental conditions of brief exposures and parafoveal location, graphic characteristics are also significant variables in word recognition. The mechanism of selective attention was speculated to determine further processing once the word is recognized, thus explaining the potency effect. Parafoveal recognition data also support Hochberg's (1970) model of reading, i.e., peripheral search guidance followed by cognitive search guidance.