DPhil Studentship – Hormones, Prediction and Psychosis: How Reproductive Transitions Shape Belief Updating in the BrainUniversity

Hormones, Prediction and Psychosis: How Reproductive Transitions Shape Belief Updating in the Brain

DPhil in Psychiatry, University of Oxford · October 2027

Wolfson Marriott Graduate Scholarship.

Why does vulnerability to psychosis change across the menstrual cycle and menopause? This project investigates whether fluctuations in sex hormones alter how the brain balances prior expectations against new sensory evidence, a fundamental computation implicated in hallucinations and delusions.

The Group

You will join the research groups of Rob McCutcheon (oxfordpsychosis.com) and Laurence Hunt (huntlab.co.uk), with additional supervision from Dr Bodyl Brand. McCutcheon studies the neurobiology and treatment of psychosis, including cognition and sex differences; Hunt combines computational modelling and neuroimaging to understand learning and decision-making; and Brand investigates the effects of reproductive hormones on schizophrenia. The project forms part of Oxford Psychiatry's growing focus on women's mental health. You will be a member of Wolfson College and its interdisciplinary Mind, Brain and Behaviour Research Cluster.

The DPhil

Predictive inference concerns how the brain forms beliefs and updates them as circumstances change. Disruptions to this process may contribute to hallucinations and delusions. Psychosis symptoms can worsen during low-oestrogen phases of the menstrual cycle, while women experience a second peak in psychosis onset around menopause. Whether hormones influence the neural computations underlying these patterns remains unknown.

You will use a novel continuous EEG paradigm that yields reliable measures of belief updating from just minutes of recording. This enables comparisons within individuals across the menstrual cycle and between premenopausal and postmenopausal participants. The study will include people with schizophrenia-spectrum disorders and healthy controls, combining computational modelling, EEG, hormone measures, cognitive testing and clinical assessment.

The aim is to establish how hormonal state affects belief updating and its relationship with symptoms, potentially informing more personalised, hormone-informed treatment. You will help refine research questions and analyses, with opportunities to develop independent projects in computational neuroscience and reproductive psychiatry.

Training/supervision

You will receive training in clinical research with people experiencing psychosis; EEG acquisition and analysis; experimental study delivery; computational modelling of learning and decision-making; reproductive endocrinology and sex-informed research design; and reproducible data science in Python.

Candidate profile

Applicants should have at least an upper second-class honours degree (or equivalent) and/or a relevant Master's degree in neuroscience, psychology, medicine, physics, computer science, engineering or a related discipline, alongside experience of independent research. Programming and quantitative skills (Python, MATLAB or R), strong communication skills and willingness to work with clinical populations are essential. EEG, computational modelling or clinical research experience is desirable but not required.

Funding/application

The scholarship covers course fees up to the home-fee rate, a tax-free stipend at the UKRI rate (at least £22,460 annually), and support for research and conference attendance. Overseas-fee applicants must fund the difference in tuition fees.

Applications must be submitted via the University’s online graduate application form  www.graduate.ox.ac.uk/apply. The application form, all supporting materials (including references), and £20 application fee must be received by the submission deadline.

For informal enquiries contact: [email protected]

Submission deadline: 12.00 noon (UK time), Friday 27th November 2026 - NOTE: this is earlier than the University’s December deadline

Interview date: w/c Monday 14th December 2026