| |
Abstract:
We recently reported that activation for subordinate-level
recognition of common objects falls in the group-averaged fusiform
"face area". We replicated this finding with the face area defined
functionally in individuals. Imaging parameters were identical to
the original study. Subjects (11) performed two different tasks:
1-Passive viewing of faces vs. objects (F-O), to localize the face
area and 2- Categorizing whether a picture of a familiar object
matched a basic or subordinate label presented auditorily (S-B).
5x5 pixel regions of interest (ROI) were defined according to the
average or each individual face area. Percent signal change was
summed in each ROI. An ANOVA performed on the group-averaged ROI
indicated greater activation in the face area for F-O than for S-B,
as well as greater activation in the face area than surrounding
ROIs, only for F-O. However, a similar analysis performed on
individually-defined ROIs showed that the face area was more active
than ROIs surrounding it for both tasks, with no difference between
them in the face area. Even when the face area is defined
fucntionally, categorization level can account for a significant
part of the activation found when comparing faces to objects. Other
recent findings indicate that expertise with non-face objects can
also lead to the recruitment of the face area. Together, these two
factors may explain most of the specialization in the middle
fusiform area, widely considered to be selective for face
processing.
|
