Alexithymia in Autism: Psychophysiological Correlates and a Possible Route to Anxiety

Background: Autism Spectrum Disorder (ASD) is frequently associated with Alexithymia, which is characterised by difficulties in identifying and describing one’s own emotions (Hill et al., 2004). These difficulties have been shown to account for the problems individuals with ASD often experience in recognising and identifying emotional expressions in others (Cook et al., 2013) and there are reasons to believe that Alexithymia may also lie at the root of the high prevalence of Anxiety in ASD. A major challenge for research on Alexithymia, however, remains that the neurocognitive basis of the phenomenon remains poorly understood and that the construct is measured exclusively through self-report questionnaire measures. In the context of ASD, this hinders the formulation of accounts that specify the neurocognitive mechanisms that lie at the root of the close association between difficulties in emotion recognition, Alexithymia and Anxiety. 

Objectives: To examine the psychophysiological basis of Alexithymia in ASD and to explore possible associations between Anxiety and Alexithymia.

Methods: In experiment 1 seventy-six adults (33 ASD; 43 TD) completed standardised Alexithymia questionnaires (BVAQ & TAS-20) as well as the Becks Anxiety Inventory (BAI). In experiment 2, 13 ASD and 13 TD participants were asked to rate a series of images that ranged widely in emotional salience (valence and arousal) while their Skin Conductance Responses were measured. These smaller groups were carefully selected to match as closely as possible in terms of age, IQ as well as Alexithymia whilst ensuring that Alexithymia scores varied sufficiently to allow for meaningful correlation analyses.

Results: In Experiment 1, ASD participants were more Alexithymic on both the BVAQ (p<.05) and TAS-20 (p<.001) and the two measures correlated significantly with one another for both ASD (r = 0.42) and TD (r = 0.45) participants.  Although groups did not differ on the BAI, Anxiety and TAS-20 correlated significantly in ASD (r = .52) but not TD (r = .19) participants. In Experiment 2 subjective ratings of arousal were significantly correlated with participant’s GSR responses, again in both ASD (r = .51) and TD (r = .55) groups. By computing correlation coefficients for this association for each individual, we were able to ask to what extent the degree of coupling between subjective and psychophysiological emotional responses is in turn associated with participants’ Alexithymia and Anxiety. In both groups, reduced coupling was associated with greater Alexithymia (r < -0.54) and increased coupling with greater Anxiety (r > 0.46)

Conclusions: Our date reveal a complex pattern of associations between Alexithymia, Anxiety and the interplay between subjective and psychophysiological facets of emotional experiences that seem to suggest that abnormalities in conscious awareness of one’s state of arousal contribute to Anxiety as well as Alexithymia in ASD.