Acquisition and Use of ‘Priors’ in Autism: Typical in Deciding Where to Look, Atypical in Deciding What Is There

Abstract

Individuals with Autism Spectrum Disorder (ASD) are thought to under-rely on prior knowledge in perceptual decision-making. This study examined whether this applies to decisions of attention allocation, of relevance for ‘predictive-coding’ accounts of ASD. In a visual search task, a salient but task-irrelevant distractor appeared with higher probability in one display half. Individuals with ASD learned to avoid ‘attentional capture’ by distractors in the probable region as effectively as control participants—indicating typical priors for deploying attention. However, capture by a ‘surprising’ distractor at an unlikely location led to greatly slowed identification of a subsequent target at that location—indicating that individuals with ASD attempt to control surprise (unexpected attentional capture) by over-regulating parameters in post-selective decision-making.

Beyond Prior and Volatility: The Distinct Iterative Updating Account of ASD

“…Individuals with ASD appear to set the weight of the prior lower (and that of the sensory inputs higher) than optimal when the environment shifts from a normal to an unpredictable, volatile regimen. A similar inability to adapt to process unexpected sensory stimulation has been reported in several recent studies [4,24,36,51]. For example, using audiovisual recordings of hand-clapping with unexpected omissions of sound [51] found an early negative omission response (the oN1 component of the event-related potential) -a key signature of processing unexpected sensory stimulation -to be significantly more pronounced in their ASD relative to the TD group.…”

“…This may well explain the mixed reports of predictive coding in autism. Individuals with ASD can acquire appropriate priors from one-shot learning [29], prior trials [30], or statistically global settings [31,32,34,36]. That the priors are typically intact is also evidenced by the present finding that the ASD group of participants did not differ in their local prior updating from the TD group.…”

“…For example, using audiovisual recordings of hand-clapping with unexpected omissions of sound [51] found an early negative omission response (the oN1 component of the event-related potential) -a key signature of processing unexpected sensory stimulation -to be significantly more pronounced in their ASD relative to the TD group. In our previous study on probability learning of distractor locations in visual search [36], individuals with ASD showed comparable learning of the spatial distractor distribution, but an atypically strong reaction to a prediction error only when the distractor appeared at an unlikely (i.e., surprising) location.…”

“…Thus, while there is a consensus that individuals with ASD display atypical sensory processing, the underlying causes remain controversial: does it arise from overlearning of volatility [13,18] or reduced reliance on priors [12]? Or, alternatively, is learning intact [34], but the response to prediction errors is altered [36]? Of note in this context, while predictive-coding models of ASD [13,14,18] predict differences in predictive error handling in individuals with ASD (compared to TD individuals), the extant studies have focused primarily on differences in global priors and the consequent influences on sensory estimates -thus largely neglecting inter-trial effects (see [36] for an exception).…”

“…Or, alternatively, is learning intact [34], but the response to prediction errors is altered [36]? Of note in this context, while predictive-coding models of ASD [13,14,18] predict differences in predictive error handling in individuals with ASD (compared to TD individuals), the extant studies have focused primarily on differences in global priors and the consequent influences on sensory estimates -thus largely neglecting inter-trial effects (see [36] for an exception). In particular, examining how individuals with ASD (compared to TD individuals) handle trial-to-trial changes within high-and, respectively, low-volatility environments and switches between the two environments might provide crucial evidence for deciding between two promising accounts of abnormal predictive coding in ASD, namely: (a) do individuals with ASD form atypical priors regarding volatility; or, rather, (b) do they show atypical handling of prediction errors in response to volatility changes?…”