Research
My research has two broad strands: computational cognitive science, particularly the development of explicit process models within cognitive architectures, and collaborative work applying artificial intelligence and machine-learning methods to problems in health.
Memory and cognitive architecture
My current computational modelling research focuses on working memory and its relationship with longer-term memory. I use the ACT-R cognitive architecture to develop mechanistic models that specify how information is encoded, prioritised, maintained, retrieved and reactivated over the course of a task.
A particular focus is the prioritisation of information in visual working memory. The aim is to explain experimental effects in terms of identifiable cognitive mechanisms rather than adding task-specific processes solely to reproduce individual findings. Current work also examines how processing during a working-memory episode leaves traces that can influence later recognition and recall.
This work continues a longer programme of research using ACT-R to model cognition. Earlier models have addressed graph comprehension and diagrammatic reasoning, dynamic decision making, object-location memory, visual mental imagery, vigilance, learning and retention, and interaction with complex systems.
Machine learning and dementia
I collaborate on research applying machine-learning methods to dementia detection and progression. A central concern is how cognitive and functional measures can be used to identify informative markers while retaining models that are interpretable enough to support scientific and clinical scrutiny.
Recent work has examined feature selection in ADAS-Cog data, explainable approaches to dementia detection, and the relationship between cognitive measures and disease progression. I am also a co-investigator on a Dubai Future Foundation project investigating prodromal and preclinical markers that may help predict dementia diagnosis and progression.
Recent outputs include work on cognitive feature evaluation, a systematic review connecting explainable AI with the Research Domain Criteria framework, and an interpretable decision-tree approach to dementia detection. See the publications page for details.
AI, digital inclusion and mental health
Since 2026 I have contributed to the Centre for Equity in Mental Health, funded through the National Institute for Health and Care Research Mental Health Research Group programme. The Centre focuses on reducing mental-health inequalities in Calderdale, Kirklees and Wakefield through applied and community-based research.
My work is principally within Work Package 3, concerned with equitable big data, artificial intelligence and digital solutions for the future of mental health. Current work includes mapping digital inclusion and access to technology and examining the opportunities, limitations and equity implications of data-driven and digital approaches to mental-health services.
This strand is newer than the dementia work and differs from it in emphasis: the central questions concern access, equity, implementation and the responsible use of digital technologies rather than prediction alone.
Previous and related research
Previous research has included diagrammatic reasoning and graph comprehension, spatial cognition and map-based navigation, visual mental imagery, decision making, vigilance, learning, object-location memory, human factors, autonomous systems and human interaction with complex socio-technical systems.
Although the empirical topics have varied, much of this work shares a concern with explaining behaviour in terms of explicit cognitive representations and processes, and with testing those explanations against behavioural or eye-movement data.