The present study examined language comprehension in Autism Spectrum Disorder (ASD) in light of monitoring. It was studied whether individuals with ASD monitor their language perception, and whether monitoring during language perception could be modulated with instructions. We presented higher-level (semantic) linguistic violations and lower-level (orthographic) linguistic violations in a free reading condition and in an instructed condition, recording event-related potentials. For control participants, a monitoring response as tapped by the P600 effect was found to semantically and orthographically incorrect input in both conditions. For participants with ASD, however, a monitoring response to semantically implausible input, tapped by the P600, was found only in the instructed condition. For orthographic errors monitoring was observed both in the free reading and in the instructed condition. This suggests that people with ASD are less inclined than typical individuals to monitor their perception of higher-level linguistic input, but that this can be enhanced with instructions.
Autism Spectrum Disorders (ASD) are characterized by a range of language and communication deficits (DSM-IV-TR, American Psychiatric Association, 2000). Although individuals with ASD vary greatly in their language abilities, some being completely mute and others being verbally fluent, even the most verbal individuals on the spectrum experience difficulties with more complex aspects of language (e.g., Huemer and Mann, 2010, O’Connor and Klein, 2004 and Tager-Flusberg, 2001). The present study focuses on language comprehension in high-functioning adults with ASD. In this subgroup, basic language abilities such as phonological and lexical processing are shown to be relatively intact. However, the processing of linguistic aspects that require integration of verbal information for comprehension, such as complex semantics and pragmatics, is often found to be impaired (for an overview see Kelley, 2011). Although a lot of research has been conducted on language comprehension impairments in ASD, their cognitive correlates are still poorly understood. Researchers have proposed some form of cognitive control to be involved in the impairments, but the exact mechanism has not yet been specified and warrants further examination.
Evidence from recent psycholinguistic research suggests that the quality of our understanding of language depends largely upon cognitive control processes (for a review see Ye & Zhou, 2009). Cognitive control refers to the control over attention, thoughts and behaviors in order to ensure goal-directed behavior (e.g., Botvinick, Braver, Barch, Carter, & Cohen, 2001). An aspect of cognitive control that has been found to be of specific relevance for the optimization of language comprehension is monitoring (for an overview see Van de Meerendonk, Kolk, Chwilla, & Vissers, 2009). Monitoring entails the evaluation of the demands for control, a process that can vary as a function of for instance context and individual abilities (Stuss & Knight, 2002). The role of cognitive control processes in language comprehension indicates that language (and problems associated with it) should not be studied in isolation, but in light of its interaction with monitoring. As such, the present study used the monitoring perspective on language as a starting point to study language comprehension in individuals with ASD. Below we will provide a brief overview of the literature on monitoring and language, where after we will describe how we studied language comprehension in ASD from this perspective.
In the present study, we examined language comprehension difficulties in people with ASD from a monitoring account. It was found that individuals with ASD are less inclined than individuals without ASD to monitor their perception of higher-level, semantic input, whereas they do exert cognitive control during lower-level, orthographic processing. Based on the present findings, we propose that people with ASD form weaker expectancy-based representations of linguistic input and rely more strongly on a systematic, word-by-word analysis of the sentence than typical individuals, as a result of which conflicting, unexpected, or complex information is not or incorrectly integrated into the ongoing linguistic interpretation. Furthermore, the finding that monitoring of semantic perception can be enhanced with instructions supports the idea that comprehension problems cannot be accounted for by an absolute deficit in the ability to access semantic information to process language for meaning and/or use contextual information and relevant knowledge. Instead, importantly for neuropsychological assessment and intervention, differences in the inclination to monitor semantic perception appear to play a role in the reduced processing of the meaning of language. The finding that language comprehension can be modulated with instructions strengthens the importance of theories that take an interactive, dynamic approach to cognitive functioning in ASD. Future research should examine whether the atypical monitoring tendency found in relation to semantic perception also holds for other aspects of language.
