Biomedical Engineering Reference
In-Depth Information
as judged from the ISI dependent signal reduction. The over-
laps of activation between the local processing sites are not exten-
sive and/or they are fairly independent. There is some similarity
in left-eye and right-eye paired stimulation where the suppres-
sive interaction between ocular dominance columns was relieved
within short ISI
(4)
.
The following description on the locations and functional
roles in the above areas has been reported: FG1 was located
between visual area V4 and the lateral occipital area (LO), and
FG2 was located in the posterior fusiform gyrus (pFG)
(8)
.The
LO and pFG compose the lateral occipital complex (LOC) and
are known to be involved in object processing
(9)
. It has been
also reported that there exist populations of neurons that code
complex shapes in primate V4
(10)
. Reported human fMRI data
from V4 showed larger responses for circular concentric patterns
than for parallel grating patterns
(11)
, and human psychophysi-
cal data demonstrated the existence of context-dependent inter-
actions as well
(12)
. Therefore, in the present experiment, the
areas FG1 and FG2 are likely to be those sites that encoded the
shape of frames with size/location distinction. The reduction of
fMRI signal at 50-ms ISI could indicate the loss of effectiveness
of the encoding. If the poor encoding process is related to the
poor perception of visual information, the meta-contrast (back-
ward suppression) phenomenon of the reduced visibility of the
small frame at 50-ms ISI was associated with the signal reduction
at FG2.
On the other hand, the fMRI response in FG3 showed reduc-
tion of signal at 240-ms ISI although the visibility of the two
frames was not reduced at the ISI. FG3 is located in the dorsal
extension of the fusiform face area (FFA)
(13)
. The FFA has been
reported to be sensitive to circular concentric shapes as well as
faces, which may account for the activation at FG3 to the concen-
tric frames in this study
(11)
. It was also reported that FFA shows
size invariance and suppressed responses to repeated stimuli
(14)
.
Therefore, there is a good possibility that the paired inputs to
FG3 were identical and that the reduction of the fMRI response
at FG3 approached to the identical input cases of V1 (
Fig. 11.9
),
where the activation by the second stimulus of an identical pair
was severely suppressed when the ISI was 200 ms. These obser-
vations suggest that a site such as FG3, with suppression at the
240-ms ISI, is related to the processing for the size-independent
shape recognition that follows shape encoding in area FG2 as dis-
cussed earlier.
This showed that there exist two stages in neuronal process-
ing along the ventral visual pathway as indicated by the observed
suppression phenomena. One is related to pre-recognition stage
of an object shape and the other to the recognition stage and they
were related to the suppression at 50 and 240 ms respectively.
