Distinguishing learned categorical perception from selective attention to a dimension: Preliminary evidence from a new method

AbstractA novel experimental method is motivated and applied in an effort to test for effects of category learning on perceptual discrimination so as to clearly distinguish category boundary effects of expansion and compression from changes in sensitivity to stimulus dimensions. The method includes a control group performing a task that, like category learning, requires attention to one systematically varying stimulus dimension rather than another. Discrimination accuracy is tracked over time and measured using a psychophysical staircase procedure tailored to individual participants that doesn’t rely on memory. Initial results suggest improvement in discrimination accuracy over time, particularly on the dimension relevant to the categorization or control task, but no evidence of category boundary effects or effects of category learning on dimension perception stronger than those of the control task. Possible reasons for this and directions for further research are briefly discussed.

Return to previous page