Despite similar gaze discrimination performance, persons with stable schizophrenia exhibited decreased brain activation during gaze discrimination in frontal, temporal, occipito-parietal and subcortical areas that are associated with executive functioning, emotional and visual processing. The baseline condition representing a cross-hair superimposed on a scrambled face, in conjunction with the task command, was designed to produce activation findings related to certain features of the face, specifically the eyes. Within the context of the gaze discrimination task, activation differences indicate altered integration of socially relevant information, including face processing, and aspects of social cognition and emotion experience, in schizophrenia. Fusiform areas are instrumental in face processing, in particular when faces are presented in forward condition, which is considered socially more relevant. In addition, the right fusiform gyrus may be more responsive to whole faces and the left to face details (Rossion et al 2000
). Amygdala activation may relate to aspects of emotional processing during gaze discrimination (Kawashima et al 1999
), in particular to isolated eye features (Whalen et al 2004
), or to attempts to search for and establish direct eye contact (Hooker et al 2003
). Frontal lobe activation may relate to decision-making and aspects of attribution of mental states or theory of mind mechanisms (Frith and Frith 2003
, Siegal and Varley 2002
) that play a role when faces are analyzed regarding information and meaning. Striatal and thalamic activations are less readily explained, although facial attractiveness can produce ventral striatum activation (Kampe et al 2001
). Amongst other brain regions, a large body of research has implicated both function (Onitsuka et al 2006, Gur 2007) and structure (Davatsikos 2005, Onitsuka et al 2006, Crespo Faccoro 2007) of these areas in schizophrenia. Within group comparison regarding the effects of gaze direction revealed different findings in controls and patients. Controls exhibited increased activation associated with 4° deviation in frontal and temporal regions, perhaps mirroring the increased cognitive demand of the more difficult condition. Patients, on the other hand, exhibited increased activation associated with direct gaze in temporal regions, in particular regions instrumental in processing of direct gaze and fear.
Potential limitations of our study involve choice of stimuli, sample size, clinical characteristics, and lack of exploration of possible gaze activation by group interactions and relationships between brain activation, performance and clinical symptoms. The choice of faces with direct gaze that undergo horizontal rotation was similar to the previous study by Rosse et al. (1994)
. This design is more naturalistic, compared to faces in full frontal orientation with eyes rotated, and supported by findings that frontal eye fields coordinate parallel head and eye shifts (Chen 2006
). Nevertheless, head rotation of the stimuli may have affected task performance and we attempted to limit possible attention to extraocular regions by asking probands to pay specific attention to the eyes. Based on the simplistic task design and small sample size, there was insufficient statistical power to find a possible gaze direction by group interaction. Similarly, due to sample size and overall clinical stability of patients, relationships with clinical symptoms were not further pursued in this data set and remain to be explored in a future project aimed at more specific examination of these issues.
The eyes, termed by Immanuel Kant as “windows to the soul”, communicate information about the internal state of the person, commonly involving emotional content, and the environment. More recently, eyes have been described as instrumental (Calder et al 2002
) in the concept of Theory of Mind, defined as the ability to ascribe mental states, such as thoughts, emotions and intentions, to self and others (Premack and Woodruff 1978
). Isolated eye whites of fearful faces produce amygdala activation (Whalen et al 2004
) and in a person with bilateral amygdala damage inability to visually fixate on the eyes produced impaired fear recognition (Adolphs et al 2005
), which improved with explicit instructions.
Depending on the emotional expression, direction of gaze can have a specific communicatory effect, e.g. the expression of fear and direct gaze or eye contact will signal fear from the observer, whereas gaze direction aimed away from the observer will signal external danger. Until now, studies on emotion identification have presented the whole face in full frontal view and mostly with direct gaze. For persons with schizophrenia and poor eye contact, this communicatory effect of gaze may differ from the effect in healthy persons and may contribute to misinterpretation of the target emotion or even non-emotional faces, relating to paranoia and ideas of reference. This concept is partly supported by findings in a person with Capgras syndrome, characterized by misidentification of familiar persons as imposters, where failure to identify gaze was associated with failed recognition of the face (Hirstein and Ramachandran 1997
). While our study did not examine the possible relationship between gaze discrimination and clinical symptoms, direct eye contact may relate to positive symptoms of schizophrenia, such as ideas of reference and paranoia.
As in the recent studies in Fragile X (Garrett et al 2004
) and Williams syndrome (Mobbs et al 2004
), we did not find a selective effect of direct gaze on brain activation. Future studies in schizophrenia employing more subtle degrees of gaze deviation, including non-emotional and emotional faces, larger number of stimuli and patient groups with wider range of positive and negative symptoms can better explore issues regarding potential relationship with acute symptoms of psychosis.