Lack of Embodied Effects On Stimulus Encoding in High-Functioning Autism

Background: In traditional approaches to cognition, experiences are encoded as abstract symbols, stripped of their perceptual and bases. More recent theories associated with "embodied cognition" approaches suggest that initial sensory, motor, and emotional aspects of experiences have an ongoing impact on cognitive processing. Because a) data from numerous labs suggest subtle difficulties in motor coordination in autism spectrum disorders (ASD), and b) embodied cognition may relate to deficits in emotional contagion and other social processes in ASD, this study examined motor system influences in encoding novel stimuli.

Objectives: To determine whether individuals with ASD display embodiment effects.

Methods: Drawing on the specialized physical systems of approach and avoidance, which allow us efficiently process emotional valence (Darwin, 1872), we contrasted responses to novel visual stimuli that were presented in the context of an approach posture versus an avoidance posture. Fifteen verbal individuals with high-functioning ASD, ages 11-29, age-matched to 15 typically developing (TD) individuals, completed an "embodied" stimulus encoding task.  In an encoding phase, subjects made "like/don't-like" judgments about neutral Japanese "kanji" characters while maintaining an approach or an avoidance posture.  In the response phase, subjects chose one of two pictures that best matched the kanji meaning, for the "approach" kanji, the "avoid" kanji, and novel kanji. Image pairs were drawn from the IAPS; for each of 36 pairs, one image was more positive (M = 7.18, on a scale of 1 to 9) than the other (M = 4.99).

Results: There was a a trend for a main effect of condition, p = 0.09, suggesting that, across groups, the ratings for kanji initially presented in the Approach condition were significantly higher than ratings for Control kanji, p = .046.  Repeated-measures ANOVA also  indicated a significant condition by group interaction on stimulus encoding, p = 0.04, which reflected the fact that the TD group's ratings for Approach kanji were significantly higher than for Avoid kanji, p = .01; their Approach ratings were also significantly higher than their Control kanji ratings, p = .01. The ASD group showed no such effect, and indeed tended to evaluate the Avoid kanji more positively than the Approach kanji, p= .09.

Conclusions: Consistent with prior research, TD individuals showed a significant impact of initial body postures on subsequent, non-postured, evaluations of visual kanji stimuli, such that they evaluated kanji more positively when those kanji were initially encountered during an Approach posture. In contrast, the ASD group did not show an effect of posture; if anything, they showed the opposite pattern in their responses. The current data are consistent with the hypothesis that there is a reduced influence of body posture on the encoding of novel stimuli, putting aside affective or emotional considerations. Because affective responding reflects the influence of motor systems, a failure of quick, efficient, multiple-modality stimulus mappings could lead to the failure of embodiment effects seen here, and may also be central to more general high-level social and communicative difficulties in ASD.