Individual research project · 2026
An ACT-R model of inter-group comparison in public goods games
A declarative-memory model examining how social comparison changes contributions in a public goods game.
Conceptual illustration; not an experimental result.
Research question
Can an ACT-R memory model account for contribution patterns when participants compare their group with other groups? The study compares simulated behaviour with a previously published public goods experiment.
Methodology
The model combines base-level activation, partial matching and blending within declarative memory. Five hundred simulated games per condition examine how remembered outcomes influence subsequent contribution decisions.
Findings
The model reproduces the qualitative ordering of the experimental conditions but underestimates the size of the inter-group effects. This distinction separates a plausible behavioural pattern from a close quantitative fit.
Limitations
This is a computational modelling study using published experimental findings for comparison. It does not report a new human-participant experiment or establish that the model uniquely explains the observed behaviour.
Related research
Recursive theory of mind