Face Processing In Persons with Asperger Syndrome

Background:  

It is still unclear if face processing is deviant within the whole spectra of autism. A greater understanding of visual search strategies and their impact on face recognition and recognition of facially expressed emotions in controlled experimental conditions is thus needed. An awareness of the visual search strategies is also a prerequisite for the development of interventions. However, prior to any interventions, the transferability of findings from a laboratory setting into the “real world” needs to be investigated.

Objectives:

-to explore face identification abilities, the ability to recognise facially expressed emotions and visual search strategies, with special focus on the importance of the eye area in adults with Asperger syndrome (AS)

-whether or not these laboratory results are transferable to a “real world” situation

Methods:  

While wearing a head mounted eye tracker, 24 adults with AS and their 24 matched controls viewed 12 pairs of photos of faces. The first photo in each pair was cut up into six puzzle pieces. Six of the 12 puzzle pieced photos had the eyes bisected. The second photo showed either three faces of which one was the same person as in the first photo, or three faces showing a happy, an angry and a surprised face of the same person as in the initially shown puzzle pieces. Differences in visual search strategies between the groups were established with respect to fixation durations and number of fixations. Following the completion of the photo viewing parts, the participants kept the eye tracker on and were seated at a table in the same room, opposite to the test leader. A dialogue about the test created an interactive dynamic condition, so the visual strategies on a face during such a “real world” event could be measured.

Results:  

Persons with AS had more difficulties in identifying faces and recognizing basic emotions than controls. However, the entire face identification superiority in controls was found in the condition when the eyes were distorted, supporting that adults with AS do use the eye region to a great extent in face identification. The visual search strategies in controls were more effective and relied on the use of the ‘face information triangle’, i.e., the two eyes and the mouth, while adults with AS had more fixations on other parts of the face.

A within-group comparison showed that people with AS, and their matched controls, displayed a high degree of stability in visual search strategies when viewing faces, regardless of the facial stimuli being photos (static) or real “real world”, as in the interactive dynamic condition.

Conclusions:  

There is a difference between the overall visual search strategies for information seeking in faces between adults with AS and matched controls. In future intervention strategies, the distribution of fixations on facial details may be most important to cue. The stability of our results across conditions warrants laboratory findings to be generalized to “real world” situations.