Jul. 23, 2014

Faces comprise a unique visual category with a special evolutionary and social significance, and hence it is not surprising that face perception is probably the most developed visual perceptual skill in humans. At the neural level, there is growing evidence that the summed activity of multiple nodes of a distributed cortical network, including “core” regions located in posterior parts of the brain, as well as “extended” regions in anterior parts of the brain is necessary to support intact face recognition in humans. 

Face recognition is seemingly effortless for most people, but is extremely difficult for individuals who suffer from a disorder termed prosopagnosia. Acquired prosopagnosia (AP) occurs following brain damage to core face regions in adulthood and has provided a unique window into the psychological and neural substrate of face processing.  

Recently, attention has been given to congenital prosopagnosia (CP), which refers to an impairment in face processing that is apparent from birth in the absence of any noticeable brain damage. Such individuals thus provide Dr. Galia Avidan (pictured above) from the Department of Psychology with the prospect of pinpointing the mechanisms underlying normal face representation and perception.  

In previous studies, using behavioral experiments, Avidan documented impairments in various aspects of face processing in CP individuals but despite these dramatic behavioral impairments and contrary to existing hypotheses, previous functional magnetic resonance imaging (fMRI) studies consistently revealed normal face-related activation in core regions of the face network in these individuals. 

“In our recent study we have attempted to further explore the nature of the impairment and showed that in the brains of individuals with CP, posterior face regions are not being efficiently connected to regions of the extended face system that are involved in face identification. In contrast, connectivity of the core regions to the amygdala associated with emotion recognition and part of the extended face system was intact, consistent with normal emotion recognition in CP,” says Avidan. 

The study was published in the Oxford journal Cerebral Cortex under the title “Selective Dissociation Between Core and Extended Regions of the Face Processing Network in Congenital Prosopagnosia” by Galia Avidan, Michal Tanzer, Fadila Hadj-Bouziane, Ning Liu, Leslie G. Ungerleider, and Marlene Behrmann. 

The findings suggest that the profound impairment in CP may arise not from a dysfunction of the core face areas but from the failure to propagate signals between the intact core system and specific nodes of the extended face network that are related to identity processing. The converging evidence enables the framing of this disorder as a disconnection syndrome rather than a syndrome related to a localized brain alteration as is typically assumed for AP.  

Importantly, a similar account was also offered recently for other neurodevelopmental disorders, such as developmental dyslexia. Specifically, analogous to CP, the neural mechanism underlying this disorder has been implicated in the connectivity of a network of brain regions associated with reading abilities. 

These findings contribute to further understanding and classifying such syndromes and to the establishment of general mechanisms common to different neurodevelopmental disorders. Such information may be important in the establishment of future automated diagnostic tools which will be based on functional and structural imaging data. Moreover, these findings may also aid in the development of rehabilitation approaches which will be based on biofeedback targeted at enhancing connectivity between brain regions associated with specific cognitive abilities such as reading or face perception.