The problem
When is it rational to trust an expert’s testimony? How far should trust in an expert’s judgement in one domain carry over to another? What makes someone trustworthy rather than merely confident, and how can trust, once broken, be rebuilt? These questions matter for public health, for science policy and for democratic life, and philosophers working on the epistemology of trust have made real progress on them.
But the field’s standard tool is the method of cases. A theorist consults their own intuitions about an imagined scenario and builds a general theory on the result. Intuitions about such cases vary with culture, gender, age and background, they shift with framing, and they can carry prejudice. The upshot is a growing number of rival theories and little convergence between them.
Our approach
Models are simplified, idealised representations used to study things too complex to investigate directly. Scientists build them with imperfect data, generate whole families of them, and use robustness analysis to find which conclusions survive across the family. We’re bringing those techniques to the epistemology of trust.
Our starting point is a model we’ve developed ourselves of how trust transports: how a rational agent should adjust trust in an expert in one field after gaining or losing trust in that expert elsewhere. The model rests on four modest requirements on rational trust, and it already has consequences for live debates. These include when experts trespass beyond their competence, how virtuous trust differs from gullibility and from excessive scepticism, and how agents can rebuild broken trust by starting again in a nearby, lower-stakes domain. It also reveals a new failure mode, reverse trespass. Because several of its parameters can be varied, the model can be used to see which conclusions are robust.
Why now
Trust in scientists and other experts has fallen sharply in recent years. Some of that distrust is political hostility to expertise. Some of it has a basis: there is evidence that the way science is organised can distort research incentives, and that independence from political pressure can also make expert advice less accountable to the people it is meant to serve. Working out which distrust is warranted, and how to repair what isn’t, needs better tools than intuitions about cases.
What we’re doing in 2026–27
- Three thematic workshops, each pairing invited speakers with researchers and PGRs in Glasgow.
- A public talk series of about six hybrid talks, from December 2026 to June 2027, recorded and shared openly.
- A day-and-a-half flagship conference in Glasgow in summer 2027, bringing the three strands together and presenting the network’s first results.
- An advisory board, now being appointed, to guide the network’s direction.
- Co-authored working papers, and a network of collaborators across philosophy, economics and the social sciences.
See the programme for details as they’re confirmed.
Funding
Modelling Trust is funded by the Glasgow Changing Futures (GCF) Collaborative Seed Fund, in the Collaborative Futures theme, from October 2026 to July 2027. It’s based in Philosophy at the University of Glasgow and works closely with COGITO Epistemology Research Centre.