The reference hub for two disciplines
Behavioural Data Science combines behavioural science and data science to study, model, predict, and explain human, algorithmic, and systemic behaviour. Behavioural AI applies that science to how artificial intelligence is designed, evaluated, and governed.
Explore the Handbook Browse the libraryTwo young disciplines, one hub
Behavioural Data Science is an interdisciplinary field that combines behavioural science, data science, computer science, AI, statistics, economics, engineering, and related disciplines to understand, predict, and influence human, algorithmic, and systems behaviour using theoretical models, computational methods, and large-scale behavioural data. Behavioural Data Science studies behaviour in data-rich and algorithmically mediated environments. It uses behavioural theory and data-driven methods to analyse three interdependent forms of behaviour: human behaviour, including decisions, preferences, judgement, biases, and social interaction; algorithmic behaviour, including how models, agents, and automated systems learn, decide, adapt, and produce consequences; and systems behaviour, including the emergent outcomes that arise when humans, algorithms, institutions, platforms, and infrastructures interact.
Behavioural AI is the application of behavioural science to the design, evaluation, governance, and improvement of artificial intelligence systems, with particular attention to how AI systems influence human cognition, emotion, motivation, judgement, trust, decision-making, and behaviour over repeated interaction. Behavioural AI focuses on the behavioural consequences of AI systems in use. It asks how AI systems shape what people notice, believe, feel, choose, trust, avoid, depend on, or act upon; how AI can be designed to support rather than distort human judgement; and how behavioural risks such as over-reliance, automation bias, anthropomorphism, manipulation, reduced confidence, anxiety, unhealthy dependence, and loss of agency can be detected and mitigated across the AI lifecycle.
Flagship reference · 2026
Edited by Ganna Pogrebna and Thomas T. Hills · Cambridge University Press · 44 chapters across six parts, written by leading scholars across psychology, economics, computer science, statistics, engineering, philosophy, and law.
Chapter-by-chapter
Short AI-assisted video explainers generated with NotebookLM and reviewed by the editors for all 44 chapters — added part by part as they are produced.
The library
A curated, annotated, living collection — maintained on GitHub, contributions welcome.
From the Handbook and The Missing B in AI to the behavioural science canon.
PapersProspect theory to machine behaviour — the papers that built the field.
DegreesEvery verified university programme in BDS worldwide, mapped and annotated.
CoursesCertificates, MOOCs, and professional courses — verified list coming soon.
About the curator
Ganna Pogrebna is Professor at Queen's University Belfast and a Senior Fellow at the Australian Strategic Policy Institute (ASPI). She is a pioneer of behavioural data science and behavioural AI, co-editor of The Cambridge Handbook of Behavioural Data Science (Cambridge University Press, 2026), and author of The Missing B in AI (2025). Her work explores how emerging technologies transform industries, customer experiences, and business models, and she advises businesses and governments on AI adoption, innovation strategy, and digital trust.