Abstract
Perceptual categorization, the fundamental process by which sensory events are differentiated and organized, occurs extremely rapidly under natural viewing conditions. Yet, despite a wealth of research in human vision, category-selective responses to single-glanced, natural (i.e., unsegmented) images in a perceptually continuous presentation stream have not been identified and characterized in terms of magnitude and spatio-temporal dynamics. We presented 16 human observers with variable natural images of objects at a fast periodic rate of 12.5 Hz, i.e., every 80 ms. Variable natural face images were inserted every 3, 5, 7, 9, or 11 stimuli, defining face stimulus-onset-asynchrony (SOA) conditions from 240 to 880 ms, i.e., face presentation frequencies (Fs) from 4.17 to 1.14 Hz (e.g., for the 240 ms face SOA, F = 1/0.24 s = 4.17 Hz). After just a few minutes of stimulation, face-selective responses were objectively identified at F for every condition in the frequency domain of the high-density electroencephalogram (EEG, 128 channels). Additional harmonic-frequency responses (i.e., 2F, 3F, etc.) were distributed and characterized within a common frequency range across conditions, providing novel evidence that baseline-corrected harmonic responses should be summed for quantification. Thus, the magnitude of the face-selective response was revealed to be stable across conditions; however, for the lowest 240 ms face SOA, the response was significantly reduced by 25% over three maximal right occipito-temporal channels. Correspondingly, only face SOAs above 240 ms revealed four successive face-selective response components, emerging from about 100 ms post-stimulus onset and progressing from posterior to occipito-temporal electrode sites until about 550 ms. Uncovering category-selectivity in a rapid stream of single-glanced natural images and characterizing its spatio-temporal dynamics goes well beyond previous evidence obtained from spatially and temporally isolated stimuli, opening an avenue for increasing our understanding of human vision, including its neural basis and relationship to visual categorization behavior.
Meeting abstract presented at VSS 2016