Color and spatial frequency deferentially impact early stages of perceptual expertise training
Neuropsychologia, 2019
Hélène Devillez, Matthew V. Mollison, Simen Hagen, James W. Tanaka, Lisa S. Scott & Tim Curran
Embedded Files
The current study examined the role of color and spatial frequency on the early acquisition of perceptual expertise after one week of laboratory training with bird stimuli. Participants learned to categorize finches (or warblers) at the subordinate species level (e.g., purple finch) and categorize warblers (or finches) at the more general family level. Training images were presented in their natural colors across 6 sessions. Participants completed a subordinate level species matching task prior to training, one day after training and one week after training while event-related potentials (ERPs) were recorded. Bird images were presented in either their natural congruent color, incongruent color, grayscale, low spatial frequency (LSF < 8 cycles per image) or high spatial frequency (HSF > 8 cycles per image). Replicating previous training studies, performance benefited more from subordinate- than basic-level training. Before training, any color helped performance, but color congruence effects (congruent > incongruent) only emerged after subordinate-level training. Spatial frequency manipulations did not interact with training. The N170 ERP component was sensitive to spatial frequency manipulations, but not color. N170 spatial frequency effects did not interact with training, and training effects generalized to all manipulations except the LSF images. Like performance, color congruence effects on the N250 were only observed after subordinate level training. These results are consistent with previous reports suggesting that effects of perceptual expertise training on performance are more clearly indexed by N250 than N170 effects. Taken together, our behavioral and ERP results show that color plays an important role in both low- and high- level visual processing, supporting surface-plus-edge–based theories for object processing and recognition.
Page updated
Google Sites
Report abuse