Motion detection and neuropsychology (Version 1.0)

  • Research type

    Research Study

  • Full title

    Visual statistical learning and Bayesian inference in psychiatric disorders

  • IRAS ID

    204249

  • Contact name

    Stephen Lawrie

  • Contact email

    s.lawrie@ed.ac.uk

  • Sponsor organisation

    University of Edinburgh

  • Duration of Study in the UK

    5 years, 0 months, 0 days

  • Research summary

    A growing idea in neuroscience is that perception and decision-making can be well described using statistical (Bayesian) models. According to these models, visual perception is shaped by both incoming sensory information and learned expectations about the world. Interestingly, these learned expectations can have a large influence on 'what' we see, particularly in times of uncertainty. Research indicates that internal representations of likelihood (i.e. probabilities) are used to make these visual expectations.

    It has been suggested that the unusual perceptual experience in some psychiatric diagnoses could result from differences in the development and use of visual expectations and statistical learning. The current study therefore aims to look at how individuals with certain diagnoses, more specifically schizophrenia, autism, and attention deficit hyperactivity disorder, learn implicit (without conscious awareness) visual expectations, and how these go on to influence what they see. Compared to controls we anticipate that these groups will rely less upon learned expectations or will do so in a sporadic manner.

    To evaluate this, participants will complete a one-hour visual motion detection task, and demographic, clinical, and cognitive measures (e.g. problem solving tests). We will be recruiting 120 participants (30 schizophrenia; 30 autism spectrum disorder; 30 attention deficit hyperactivity disorder; 30 healthy controls) from mental health services, research register databases, universities, and third sector organizations, either through referrals from direct care teams or responses to advertisements. The research project will take approximately 60 months.

    The main benefit of the study will be the novel application of Bayesian modelling in psychiatry, which will hopefully deepen our understanding of how different people make sense of the world and consequently what might be the best way to support them.

  • REC name

    South East Scotland REC 01

  • REC reference

    16/SS/0117

  • Date of REC Opinion

    22 Aug 2016

  • REC opinion

    Further Information Favourable Opinion