Face Processing And P1 Effect

Faces are among the most informative stimuli we ever perceive:even a split-second glimpse of a person's face tells us his identity, sex, mood, age, race, and direction of attention. Studies done on face processing suggested a face specific brain region, and recent functional imaging, neuropsychological and electrophysiological studies have provided evidence for a fast, low-spatial-frequency, subcortical face-detection pathway that modulates the responses of certain cortical areas to faces and other social stimuli. The present study is mainly focused on the relationship of face processing and event-related potential P1 component. We recorded and analyzed EEG in three experiments:in experiment 1, we investigated investigate whether the ERP differences between intensities of fearful expressions are driven by the differential salience of the eyes in the fearful faces, ERPs were measured within a backward masking paradigm, where observers were asked to do a gender-decision task with male and female neutral faces. The emotional stimuli used were low-intensity (50%), prototypical (100%), and caricatured (150%) fearful eye whites that were derived from corresponding intensities of fearful faces respectively. Three groups of white squares that have the same pixels as the eye whites were created as control conditions. Analysis of the ERP data showed a linear increase in amplitudes of the parietal-occipital P120 by three intensities of fearful eye whites. These ERP effects were proved sensitive to intensities of negative emotions but not to the simple physical features as the same patterns of differences were not observed on white squares. The P120 result of experiment 1 suggested that fearful expression may modulate the activation in visual cortex, but how it works is still unclear. So experiment 2 in the present study was carried out to test the neural pathway from amygdale to visual cortex. In experiment to, inverted or upright faces with fearful or neutral expression, and high spatial frequency or low spatial frequency gabors were presented for 67ms and 33ms consequentially, with the SOA of 100ms, at the center of the monitor. Subjects'task were to detect the tilted Gabors, which were presented in 20 percent of the trials with gabors after faces. The behavior result showed a selective improvement of d-prime on low spatial frequency gabor detection preceded by upright fearful faces compared to the same stimuli preceded by neutral faces, but neither the same pattern of result was found for inverted fearful vs. neutral faces nor for the high spatial frequency stimuli. By analyzing ERP data in experiment 2, we found the same selective enhancement on P1 amplitude on low spatial frequency Gabor stimuli preceded by upright fearful faces compared to the same stimuli preceded by neutral faces. And neither the effect was found for inverted fearful vs. neutral faces nor for the high spatial frequency gabor. The result suggested that the modulation of activities in visual area by expression only affect low spatial frequency information processing. It may also suggested that visual area is modulated by expression via low spatial frequency pathway, magnocellular pathway. The result we found in experiment 1 might be due to the activation of the same pathway. Experiment 3 is to investigate whether the early categorization effect P1 which was larger for inverted faces than upright faces could be modulated by spatial attention. Two streams of stimuli were presented, one of them is of letters/numbers stream, which was always presented on the upper visual field at the frequency of 6.7 Hz. The other stream is of faces/houses, which was presented at the center of the monitor. In half of the task blocks in experiment 3, subjects were ask to attend to the faces/houses stream, and subjects were asked to attend to the letters/numbers stream in the other half of the blocks but always fix at the center of the monitor. ERPs were analyzed time-lock to the onset of faces/houses stream, larger PI amplitude was found for inverted faces than upright face, and the effect wasn't affect by spatial attention. But the similar pattern of result wasn't found for houses, neither in attended nor in unattended condition. The result suggested a fast categorization of faces as early as 100ms, and the fast categorization doesn't need attention resources or may be need very little attention. But the fast categorization could be modulated by face inversion, like face identity and expression procession.