Individuals With Autism Have No Detectable Deficit in Neural Markers of Prediction Error When Presented With Auditory Rhythms of Varied Temporal Complexity.

The final group analysis was small, with 21 autism participants and 19 controls, and the autism group was heterogeneous in age, language ability, and symptom profile. Two recruited autism participants were excluded from group comparisons, and one withdrew because of distress with the electrode cap, so the analysed group was not the recruited group. The autism group had a higher proportion of males than the control group, and the control group had higher mean IQ because the study deliberately included people with a wide range of cognitive and language abilities. 12 autism-group participants were taking psychotropic medications, and the study could not separate medication effects from autism-related differences. The task was unattended and the measured response may include other brain components, so it is not a clean stand-alone marker of prediction error. Only three minimally verbal participants were included, and that subgroup was too small for a separate reliable analysis. The null finding applies to timing deviants in auditory rhythms, not to all auditory, multisensory, or cognitive prediction processes.

The authors report no biomedical financial interests or potential conflicts of interest. The article metadata lists non-U.S. government research support.

Do not conclude that people with autism have no predictive coding difficulties in any area. This study tested an unattended auditory rhythm task and a single neural response measure, and the authors state that it does not rule out predictive mechanism differences in other auditory, multisensory, or cognitive processing.