Research
The over-arching theme of the research is to combine mathematical and computational modelling of decision-making with experimental research into how people actually make decisions in order to better understand this process.
We have identified a set of
research projects in which the cross-disciplinary approach will provide crucial new insights. Each project draws on the specialist knowledge of the
core collaborators. The overall project will be focused on a linked set of research themes which we describe here.
- In an unstable world it is rare that sensory information is unambiguous: humans successfully cope with this stimulus uncertainty, but satisfactory inclusion of this key component into a complete model is currently an open problem.
- Many environments change over time (e.g., climate, financial, social): computationally tractable ways to deal with such changes are critical for robust artificial systems and these solutions may be inspired by human decision-making.
- Humans can rapidly switch goals: such flexibility is not normally a characteristic of theoretical models, but would significantly enhance the capabilities of artificial systems.
- Decision-makers face a dilemma over whether to exploit current knowledge or to explore the environment further: how should these be balanced optimally, in a robust and tractable manner, and how do individuals actually balance them?
- Observed behaviours often appear to be sub-optimal: this sub-optimality may be explained by optimality in a larger life strategy or to ensure robustness to a changing environment with limited data.
- Humans and artificial systems rarely act in isolation: understanding how social groupings make decisions may provide significant insights to the behaviour of individual decision-makers.
Investigations of these topics will provide a platform on which to build successful cross-disciplinary working practices, and open the door to the next generation of questions to address at this important interface between science, engineering, and social science.
