Metacognitive monitoring and control processes in children with autism spectrum disorder: Diminished judgement of confidence accuracy

Highlights

• Children with ASD showed diminished accuracy in their judgments of confidence.

• Metacognitive monitoring processes, but not control processes, were impaired in ASD.

• Impaired metacognitive monitoring in ASD has important theoretical/educational implications.

Abstract

Metacognition consists of monitoring processes (the ability to accurately represent one’s own mental states) and control processes (the ability to control one’s cognitive processes effectively). Both processes play vital roles in self-regulated learning. However, currently it is unclear whether these processes are impaired in individuals with autism spectrum disorders (ASDs). This study aimed to assess metacognition in thirty-two children with ASD, and 30 IQ-/age-matched neurotypical children, using a judgment of confidence task. It was found that children with ASD showed diminished accuracy in their judgments of confidence, indicating metacognitive monitoring impairments in ASD. Children with ASD also used monitoring to influence control processes significantly less than neurotypical children, despite little evidence of impairments in overall control ability.

Introduction

Metacognition refers to an individual’s beliefs and knowledge about cognition (often referred to as metacognitive knowledge), as well as an individual’s ability to monitor and control their own cognitive processes (often referred to as metacognitive skill). According to the standard definition, metacognition involves forming meta-representations (second order representations) of one’s own mental states. Put more simply, metacognition involves “thinking about thinking” (Flavell, 2000). It is widely thought that accurate metacognitive skill is crucial for effective self-regulation of cognition and behaviour (e.g., Nelson & Narens, 1990). By accurately monitoring one’s own mental states, one can gain a degree of control over those mental states and, by extension, control the behaviour elicited by those states (e.g., Perner, 1991). For example, whilst revising for an exam, if an individual is able to accurately assess what information they know/do not know, they can employ more effective revision techniques, thus ultimately improving their memory for the exam topic.

Whilst metacognition clearly plays an important role in self-regulation, meta-representation is also thought to play an important role in enhancing other aspects of cognition. For example, Perner (e.g., Perner, 2000) has suggested that the ability to distinguish one’s thoughts from reality is a pre-requisite for episodic remembering (but see Williams, 2010). Additionally, there is evidence to suggest that such self-related processing is involved in imagining one’s own future, and is involved in the processes of episodic future thinking (see for e.g. Wheeler, Stuss, & Tulving, 1997). In this latter respect, metacognition may play a role not only in controlling current behaviour, but also future behaviour by facilitating efficient planning for one’s future.

However, the area of cognition most frequently discussed in relation to metacognition is mindreading (i.e., the ability to represent the mental states of other people; also known as “theory of mind”). There is a substantial debate that spans developmental, cognitive, and comparative psychology, as well as philosophy and cognitive science, about the relation between mindreading and metacognition. Whereas some argue that mindreading and metacognition rely on the same underlying (metarepresentational) mechanism/processing resources (e.g., Carruthers, 2009, Gopnik, 1993), henceforth termed “one mechanism” theories, others argue either that (a) metacognition and mindreading are unrelated (e.g., Nichols & Stich, 2003) or that (b) mindreading is dependent upon metacognition, but not vice versa (e.g., Goldman, 2006).