Information availability shapes foraging strategy: A virtual foraging game

Understanding spatial exploration reveals how organisms sample, process, and act on spatial information. Foraging paradigms (multi-item search) offer a window into these exploration strategies. Here, participants foraged as first-person avatars in a virtual environment. We investigated how movement ability and information availability influenced foraging strategy using a generative model that predicts item selection sequences with three latent parameters: 'proximity' (items closeness), 'momentum' (relative direction), and 'turning' (left or right). Proximity and momentum were highly predictive of behaviour. Information availability also impacted strategy: participants with a map were more likely to double-back. Finally, we also show that target selection rules affect model performance: A `winner-takes-it-all` rule outperforms a stochastic rule and simple baseline model. We conclude that our model accurately predicts behaviour on an item-by-item basis and argue that this approach can serve as a template for understanding the cognitive underpinnings of human spatial exploration.