| |
Abstract:
Abstract: Evidence from a variety of techniques suggests the
existence of face-specific mechanisms in primate occipitotemporal
cortex. fMRI studies on fusiform face area have demonstrated that
face-specific fusiform activity can be modulated by visual
attention directed toward versus away from the spatial location or
visual attributes of the face. Here we used magnetoencephalography
(MEG) to test whether attentional modulation of face processing
occurs only at a relatively late stage of processing, or instead
whether attention can modulate the early face-selective response
that occurs 170 ms after stimulus onset (Liu et al, 2000).
Face-selective occipitotemporal sensors of interest (SOI) were
first localized individually in each subject in an independent
localizer scan. Attentional modulation was then tested at these
SOIs in each subject. On each trial subjects first viewed a face or
a house cue stimulus, then decided whether the same face or house
appeared in a subsequent target stimulus containing a single face,
a single house, or a face transparently superimposed on a house.
All stimuli were presented foveally. Preliminary data suggest that
attention to faces versus houses modulated the amplitude of the
face-selective response at a latency of 160 ms for both faces
presented alone and for faces transparently superimposed on houses.
These data suggest that even when spatial attention cannot be used
to select the face stimulus, feature-based attention can modulate
the initial phase of face processing.
|
