Information structure processing in young adults with autism spectrum disorders: an EEG study

No Thumbnail Available
Issue Date
Journal Title
Journal ISSN
Volume Title
In the present study we investigated the possible underpinnings of anomalous prosody, which often adds to the severity of communication problems in Autism Spectrum Disorders (ASD), in 16 young adults with ASD and 16 typically developing (TD) matched controls. Our study included an EEG session and a behavioural one. In the first part of the EEG session, our participants listened to short dialogues in which the prosodic cue of pitch accent was used to drive the listener's attention to the most important part of a message (the focus), which is to convey Information Structure (IS). We predicted that ERP analysis could allow us to identify the origin of anomalous prosody in ASD either in the automatic attentional orienting to accented information or in ability to predict the presence of focus based on discourse context. In the second part of the EEG session we used an auditory oddball task to test the participants' ability to perceive acoustic changes and to predict their occurrence in a non-linguistic context. Finally, in the behavioural session of our experiment, we collected audio-recordings to analyse the participants' ability to use prosody consistently with IS. The results of the production experiment showed that both ASD and TD participants used prosodic features like pitch and duration to distinguish new information from background information, and thus could correctly convey IS. ASD patients, however, used a wider pitch range than TD participants independently from IS. Results from the two EEG experiments suggest that participants in both groups could perceive pitch variations and make predictions on their occurrence, both in a linguistic and in a non-linguistic context. In the IS task, modulations of the effects of Accent based on top-down processes were present as early as in the 100-300 ms time-window for both groups. Moreover, both groups showed an N400 effect after incongruous accentuation. The most important group differences concern the main effect of Accent, which tends to appear earlier in the TD group (300-500 ms time-window) than in the ASD group. Moreover, in the 700-1000 ms time-window, Congruity violations yielded a positive effect in the TD group and a negative effect in the ASD group. In the oddball task, the intact sensitivity to pitch variation and violations of IS was confirmed in both groups. However, group differences, in the form of an increased sensitivity of ASD patients to expectancy violations was also remarked. Notably, in the ASD group, an N1 effect occurred for all unexpected tones, even when these were acoustically identical to the preceding one in the series. hi the TD group, the N1 effect only occurred for all high tones, which were acoustically different from the preceding tones, both when the change was expected and when it was not. These results suggest that no specific impairment in processing prosodic cues is present for what concerns ASD participants. However, important differences in the use and perception of prosody exist, the explanation of which should be searched in the general features that characterise ASD, like the more prominent role of top-down processing and the more effortful handling of exceptions often observed in autistic patients.
Faculteit der Sociale Wetenschappen