Final Activity Report Summary - CNGBFP (The Cognitive Neural and Genetic Basis of Face Perception)
The project had three specific aims:
1. To collect norms for a battery of face recognition measures that can be used to assess face recognition deficits in developmental prosopagnosia, a congenital deficit in face recognition abilities.
2. To reveal fMRI measures that reflects face processing abilities.
3. To examine whether we can reveal a genetic component of face processing by studying prosopagnosic and their family relatives.
The first goal was completed during the first year of the grant. The second aim during the second year of the grant. The third aim will be investigated during the 3rd year that is not funded by the Marie Curie.
The results of the first year of the project provided us with norms of face recognition abilities across 20 different measures that will allow us to compare to prosopagnosic and obtain an extensive profile of their face recognition abilities and the nature of the face representation that they generate.
In the second year we added another more sophisticated behavioural test that was not included in the original proposal which allows us to generate a 'face space' that reflects the similarity of faces on a multi-dimensional space. This space is relatively consistent across subjects and face manipulations and we can now examine whether this is also the case in prosopagnosics.
Towards the end of the first year and most of the second year we attempted to find a neural signature for intact face processing. We used a paradigm that allows us to assess whether a given brain region, which is highly responsive to faces, is able to discriminate between faces of different identities. In particular, given that prosopagnosics use information about hair or external features to discriminate between faces, we examined the effect of external features on the ability of this region to discriminate faces that differ in hair or glasses.
Our results interestingly show that brain regions that process faces are influenced by changes in external non-face features also in participants with normal face recognition abilities. However, this effect does not simply reflect the processing of the external feature (e.g. glasses) by a face processing region. It reflects the way external features modify the identity of the face. We all know that when someone we know changes their hair-style or glasses, or grow a beard, the identity of their face somewhat changes. Our fMRI shows exactly this effect. It is possible that prosopagnosic individuals will show a different pattern that will reflect the processing of the external features themselves, rather than their integration with the representation of the identity of the face.
One of the most significant project that we worked on included collection of EEG and fMRI data simultaneously. Such technique is hardly used in neuroimaging labs across the world and has not been used for the study of face processing. Because the novelty of the technique, our first study aimed to validate the data that we obtain when simultaneously recording EEG and fMRI. The MRI induces large artifact in the EEG data and in a paper that we published we showed that we are able to remove the noise from the data and reveal valid face-selective neural EEG measures (Sadeh et al., 2008). In addition we revealed strong correlations between ERP and fMRI measures of face-selectivity in the temporal and occipital lobe. This is the first study to reveal such correlations and this paper is currently under review (Sadeh et al, under review).
Finally, with my colleagues, Brad Duchaine, David Pitcher and Vince Walsh at Institute of Cognitive Neuroscience in University College London, I had the opportunity to be involved in the first transcortical- magnetic stimulation (TMS) of a face-selective brain region in humans. These studies reveal that this region is involved in early stages of face specific processing abilities. Following the publication of the first study, i visited their lab at UCL and we conducted several experiments, one of them has been recently presented in a conference and will be soon submitted for publication.
1. To collect norms for a battery of face recognition measures that can be used to assess face recognition deficits in developmental prosopagnosia, a congenital deficit in face recognition abilities.
2. To reveal fMRI measures that reflects face processing abilities.
3. To examine whether we can reveal a genetic component of face processing by studying prosopagnosic and their family relatives.
The first goal was completed during the first year of the grant. The second aim during the second year of the grant. The third aim will be investigated during the 3rd year that is not funded by the Marie Curie.
The results of the first year of the project provided us with norms of face recognition abilities across 20 different measures that will allow us to compare to prosopagnosic and obtain an extensive profile of their face recognition abilities and the nature of the face representation that they generate.
In the second year we added another more sophisticated behavioural test that was not included in the original proposal which allows us to generate a 'face space' that reflects the similarity of faces on a multi-dimensional space. This space is relatively consistent across subjects and face manipulations and we can now examine whether this is also the case in prosopagnosics.
Towards the end of the first year and most of the second year we attempted to find a neural signature for intact face processing. We used a paradigm that allows us to assess whether a given brain region, which is highly responsive to faces, is able to discriminate between faces of different identities. In particular, given that prosopagnosics use information about hair or external features to discriminate between faces, we examined the effect of external features on the ability of this region to discriminate faces that differ in hair or glasses.
Our results interestingly show that brain regions that process faces are influenced by changes in external non-face features also in participants with normal face recognition abilities. However, this effect does not simply reflect the processing of the external feature (e.g. glasses) by a face processing region. It reflects the way external features modify the identity of the face. We all know that when someone we know changes their hair-style or glasses, or grow a beard, the identity of their face somewhat changes. Our fMRI shows exactly this effect. It is possible that prosopagnosic individuals will show a different pattern that will reflect the processing of the external features themselves, rather than their integration with the representation of the identity of the face.
One of the most significant project that we worked on included collection of EEG and fMRI data simultaneously. Such technique is hardly used in neuroimaging labs across the world and has not been used for the study of face processing. Because the novelty of the technique, our first study aimed to validate the data that we obtain when simultaneously recording EEG and fMRI. The MRI induces large artifact in the EEG data and in a paper that we published we showed that we are able to remove the noise from the data and reveal valid face-selective neural EEG measures (Sadeh et al., 2008). In addition we revealed strong correlations between ERP and fMRI measures of face-selectivity in the temporal and occipital lobe. This is the first study to reveal such correlations and this paper is currently under review (Sadeh et al, under review).
Finally, with my colleagues, Brad Duchaine, David Pitcher and Vince Walsh at Institute of Cognitive Neuroscience in University College London, I had the opportunity to be involved in the first transcortical- magnetic stimulation (TMS) of a face-selective brain region in humans. These studies reveal that this region is involved in early stages of face specific processing abilities. Following the publication of the first study, i visited their lab at UCL and we conducted several experiments, one of them has been recently presented in a conference and will be soon submitted for publication.