Switching between Counterfactual Worlds and Reality in ASD: An ERP Study

Background: Research has previously found that children with Autism Spectrum Disorders (ASD) have difficulty reasoning counterfactually, and that this may contribute to deficits in Theory of Mind ability (Grant, Riggs & Boucher, 2004; Peterson & Bowler, 2000). Using an eye-tracking and reading paradigm, we have previously demonstrated that counterfactual understanding is undiminished in adults with ASD. However, these scenarios were limited to a single sentence, which may be too simplistic to fully test the limits of counterfactual comprehension in ASD. It has been suggested that people with ASD experience a local processing bias, and exhibit superior anomaly detection abilities during reading when the anomaly is incongruent with a local than global context (Au-Yeung et al, 2017). Furthermore, ASD is associated with difficulties in cognitive flexibility and attention switching. It is therefore possible that counterfactual information embedded within a longer context, or that requires switching between counterfactual/reality, may be problematic for people with ASD.

Objectives: We examined whether adults with ASD (compared to a sex, age, and IQ matched Typically Developing (TD) group) are able to accommodate counterfactual worlds over longer passages, or switch from counterfactual to real worlds. Ease of integration was indicated by recording event-related brain potentials (ERPs), specifically the N400 response, which is sensitive to anomaly detection during reading.

Methods: Participants read scenarios on a computer screen. The first sentence established the counterfactual context (e.g. “If it had been raining this morning, Susan would have rushed to get to work.”), and was presented in full for self-paced reading. A second sentence described a consequence of this counterfactual event, and was presented word by word. Critically, this sentence included a noun that was either consistent or inconsistent with the preceding context, and either included a modal verb to indicate reference to the counterfactual world (e.g. “In the end, Susan would have arrived at work early/late”) or switched back to the factual-world (e.g. “In the end, Susan arrived at work late/early”). Factual scenarios were used as a baseline measure of contextual integration (e.g. “Because…”). ERPs were recorded from 32 electrodes, time-locked to the onset of the critical word.

Results: Results showed that participants detected the contextual inconsistency in all context conditions (i.e. showed larger N400 for inconsistent than consistent words). Crucially, this pattern was manifest in both groups, suggesting comparable counterfactual language processing in TD adults and adults with ASD. Nevertheless, some subtle differences in the topography of the N400 effect emerged between groups; the effect was left lateralised in the ASD group but was balanced across hemisphere in the TD group.

Conclusions: Adults with ASD were able to rapidly accommodate a counterfactual context, showing that they successfully maintained access to this non-real world over a longer passage. Adults with ASD were also sensitive to contextual cues, which facilitated rapid switches back to the real world, suggesting that they maintain comparable access to reality as TD adults. These findings argue against general difficulties in global coherence and complex integration in ASD, and against a general impairment in cognitive flexibility.