Until recently, most of
the studies on face perception were conducted using still
stimuli, such as drawings or photographs. However, the faces
that we encounter in everyday life are constantly in motion
and are very rarely static. It has been shown that besides
facilitating communication, such facial movements can also
convey information about emotions, age, gender, and to a certain
extent about the identity of the individual.
The question
is then to know how exactly movement help in identifying a
face. Is it because we have a better structural representation
of the faces in motion, or is it because our facial representations
contain a specific dynamic signature? Can we recognise faces
by the way they move? How do the cues about shape and facial
movements interact with each other? The research project proposed
here will study the role of facial movements in the processing
of faces such as the recognition of facial identity and the
recognition of facial emotions.
It has been shown in the literature that although the information
obtained from faces in movement can facilitate the recognition
of familiar faces (Lander and Bruce, 2000; Lander et al.,
1999), its effects on unfamiliar faces are far from being
certain (Christie and Bruce, 1998; Pike et al., 1997). It
has been suggested that either movement does not facilitate
the recognition of unfamiliar faces, or its benefits are counterbalanced
by other factors. One of these factors could be that the social
importance of facial movements could distract the participants,
preventing them from encoding the identity of the presented
faces.
Recently, using studies on the activation of cortical areas
in response to faces stimuli, Haxby and collaborators have
proposed a distributed neural system for face perception in
which invariant representations and changing aspects of faces
are differentiated (Haxby et al., 2000). Haxby et al. propose
the involvement of two main cortical areas to illustrate their
model. The first region is the lateral fusiform gyrus (‘Fusiform
Face Area’, FFA), which has been revealed in many neuro-imaging
studies on face perception, and represents the invariant facial
information useful to identify faces.
The second region is
the posterior Superior Temporal Sulcus (pSTS), which is involved
in the processing of the changing aspects of the movement
of faces. Neurophysiological studies in non human primates
and neuro-imaging studies in humans have demonstrated that
this cortical area is involved in the detection of information
on gaze, the orientation of the head, and on expression (Allison
et al., 2000; Haxby et al., 2000; Haxby et al., 2002; Narumoto
et al., 2001).
It has also been shown that the STS could be
involved in the perception of biological movements of the
body, hands, eyes, and mouth (Allison et al., 2000). Moreover,
the STS system is extended to cortical areas that are involved
in the detection of gaze and orientation, in the perception
of speech through the movements of the mouth, and in the perception
of emotions. The infero-temporal system is extended to cortical
areas involved in the retrieval of the identity of individuals,
their names, and of biographical information (Haxby et al.,
2000).
The functional division between these high-level cortical
areas for the processing of dynamic and invariant information
about faces is in agreement with the transfer of visual information,
from the retina to the ventral stream on one hand (high resolution
and colour sensitivity, object recognition) and to the dorsal
stream on the other hand (low resolution and movement sensitivity,
spatial orientation) – (Ungerleider and Mishkin, 1982).
Therefore, the dynamic information about faces could mainly
be processed by the dorsal stream, while the static characteristics
about faces would mainly be processed by the ventral stream
(Haxby et al., 2000).
|
