Poster E110, Monday, March 26, 2:30-4:30 pm, Exhibit Hall C
Quality of perceptual categories predict speeded, but not non-speeded, cognitive ability
Emily Fritzson1, F. Sayako Earle1; 1University of Delaware
The representation of perceptual categories is fundamental to the capacity for human thought. In the language domain, the quality of speech category representations predicts higher-order skills, such as spoken language processing (Marslen-Wilson & Tyler, 1980) and efficient reading (Mody, Studdert-Kennedy & Brady, 1997). Moreover, poor speech representations are linked to disorders of spoken (Joanisse & Seidenberg, 2003) and written language (Serniclaes et al, 2004). However, relatively little is known about the relationship between individual differences in the quality of perceptual categories and general (nonverbal) cognitive ability. In order to address this question, we obtained measures of categorical perception on a synthetic vowel continuum /a/-/e/, and performances on timed (Block Design) and untimed (Matrix reasoning) measures of nonverbal cognition (WASI-II, Weschler, 2011). We conducted a series of linear regressions on a preliminary dataset of 27 participants aged 18-35 (average 21.59, SD 4.66; 4 male), in which Block Design and Matrix Reasoning were the dependent variables, with goodness-of-fit indexes for categorization and discrimination as the two predictors. We found this model to significantly account for differences in Block Design (F2,24=10.42, p<.001, r2=.464), in which categorization and discrimination were both independently predictive of Block Design scores (t24=-2.83, p=.011 and t24=3.58, p<.001, respectively). In Matrix reasoning, perceptual abilities did not significantly account for the variance in task performance (F2,24=1.35, p=.278, r2=.10). Based on these findings, we argue that individual differences in the representational quality of perceptual categories are intimately intertwined with speeded problem-solving abilities.
Topic Area: THINKING: Problem solving