Interpreting gaze in Turner syndrome: impaired sensitivity to intention and emotion, but preservation of social cueing.

Women with Turner's syndrome (TS), who lack a complete X-chromosome, show an impairment in remembering faces and in classifying "fear" in face images. Could their difficulties extend to the processing of gaze? Three tasks, all of which rely on the ability to make use of the eye-region of a pictured face, are reported. Women with TS were impaired at judging mental state from images of the upper face ("reading the mind in the eyes"). They were also specifically impaired at interpreting "fear" from displays of the eye-region of the face. However, they showed normal susceptibility to direction of gaze as an attentional cue (social cueing), since they were as sensitive as controls to the validity of the cue, under conditions where it should be ignored. In this task, unlike those of reading the upper face for intention or expression, PIQ accounted for a significant amount of individual variance in task performance. The processing of displays of the eye region affording social and affective information is specifically affected in TS. We speculate that amygdala dysfunction is likely to be implicated in this anomalous behaviour. The presence in the female karyotype of two complete X-chromosomes is protective for some socio-cognitive abilities related to the modulation of behaviour by the interpretation of gaze.

The classification of 'fear' from faces is associated with face recognition skill in women.

Two experiments were conducted to explore the relationship between the discrimination of the facial expression of 'fear' in faces and facial recognition. On the basis of the reported role of the amygdala in both processes in patients, we hypothesised that the two skills would be correlated in normal adults. In Experiment 1, a series of tests of facial expression categorisation, of face matching and of familiar and unfamiliar face recognition was conducted on normal young women, for whom psychometric scores were also obtained (n=23). Accuracy of categorisation of fear from faces predicted variance in face recognition accuracy-especially in tasks of unfamiliar face recognition (immediate old-new discrimination). No other correlations between face processing and expression classification were significant. Experiment 2 repeated the expression classification tests and an unfamiliar face recognition test on a new sample of men (n=13) and women (n=16). While there were no sex differences in face recognition, the correlation between 'fear' and face recognition was replicated only for women. These data indicate that the amygdala supports both the specific apprehension of fear in faces and face recognition in adult human females, but that the association may not hold for men. Sex differences in the structure of the amygdala-hippocampal complex suggest a likely cortical substrate for the observed differences. We speculate that social learning, which involves identifying the faces of potentially salient others, and also their attitude to the observer, engages the amygdala more readily in women than in men.