"I do not see the world as others do." Diminished perceptual adaptation, hypo-priors and autism.

Autism is a complex developmental condition, most well known for the way that it affects how a person interacts and communicates with others. But autism can affect an individual's behaviour in other important and equally debilitating ways, such as in an obsessive desire for everything to be the same - like taking the exact same route to school every day - and in senses that are working too well - like being averse to florescent light - or not well enough - like being drawn to spinning objects. In this proposal, we suggest that this wide range of autistic behaviours might be caused by fundamental differences in the way that a person with autism perceives and interprets the world around them.

For all people, knowledge about the world arrives through our senses. To process this information rapidly and efficiently, the brain must continually "tune" its sensitivity to match what is in the environment. When one wakes during the night and turns on the light, at first, it is difficult to see very much at all until receptors in the eye gradually adjust to the new conditions. Such light "adaptation" is truly remarkable. It illustrates the way in which the brain constantly - and flexibly - "adapts" to changes in the outside world. This proposal asks what the world would look like, and how might be it interpreted, if these adaptation processes were disrupted? Specifically, it asks whether less flexible brain adaptation processes might be part and parcel of autism.

In the project, we will compare the perceptual functions of children with autism and typically developing children using innovative experimental techniques. We will see whether the perceptual experiences of children with autism are less "adaptable" than children without autism. We also want to know whether these differences occur generally throughout the brain or whether they might be specific to processing only certain types of sensory information. Furthermore, we will use powerful new forms of computer modelling, which have rarely been applied to the study of autism, but which can be of great assistance in pinpointing precisely which psychological processes might be different in autism.

We hypothesise that children with autism might be less adaptable because they rely too heavily on what they sense in the "here and now" rather than using prior knowledge - knowledge that they've accrued with experience - to shape and make sense of what is in the environment. The possibility that people with autism perceive the world as it "really is" rather than as imbued by prior experiences is a critical new insight: not only can it help explain the differences in the way people with autism experience the world but it might also explain some of the hallmark features of autism, especially the sensory sensitivities and their difficulties dealing with new situations and experiences.

If confirmed, our suggestion would offer a new way of thinking about, and studying, autism capable of playing a crucial role in interventions aimed at enabling people with autism to perceive and experience the world around them with less distress.

Visual perception is a highly dynamic process, continuously recalibrating neural sensitivity to match the characteristics of the current environment. Such adaptation - ubiquitous in sensory systems - represents a rapid form of experience-dependent plasticity in which our current sensory experience is intimately affected by how we viewed the world only moments before. How would the world seem if these processes showed less adaptation?

We hypothesize that diminished adaptation shapes the perceptions of people with an autism spectrum condition, and may explain core features of the autism phenotype - especially the restricted and repetitive behaviours and sensory sensitivities. These hypotheses build on our previous work which has shown both that children with autism do not adapt as readily as typical children to changes in facial cues, and that the extent of adaptation is inversely related to degree of autistic symptomatology. We will also use Bayesian methods, a principled framework for deriving inferences in the face of uncertain information, to identify further the mechanisms underpinning diminished adaptation in autism, which we propose is due to fewer prior constraints or "hypo-priors".

This proposal uniquely combines human visual psychophysics, computational methods, and developmental psychopathology to probe further the pervasiveness of diminished adaptation in autism (Objective 1) and its computational causes (Objective 2). It is well documented that altered sensations and perceptions in autism can produce adverse effects on these individuals' everyday lives. The outputs of this research will have direct implications for our understanding of these altered perceptions and of the nature of autism itself. Identifying the mechanisms that underlie altered visual perception in autism is critical for the formulation of a new theoretical account of autism and in providing the necessary empirical base from which to develop effective interventions.

Health and educational professionals and parents and carers are well aware of the debilitating impact sensory sensitivities and perceptual differences can have on the everyday lives of people with autism. Indeed, school, college, and hospital staff are trained in autistic differences in "sensory processing" and specialist therapeutic assistance is often sought from "sensory integration" experts. Yet there is consensus neither on the range and severity of these sensory sensitivities nor on the underlying causes of them. Advancing such knowledge is therefore vital, especially since these symptoms will be included for the first time in the revised diagnostic guidelines (DSM-5) for autism, due to be published in 2013.

The proposed research will crucially improve the knowledge base about these symptoms and their underlying cause(s) and the pathways to impact will bring that knowledge direct to core communities throughout the research process. The main non-academic beneficiaries of the proposed work include parents and carers of children with autism, educators, clinicians and health professionals, autistic people themselves, public, private, and third-sector organisations (including especially the charities, the National Autistic Society and Ambitious about Autism).