No abstract

Finding an object amongst a cluttered visual scene is an everyday task for humans but presents a fundamental challenge to computational models performing this feat. Previous attempts to model efficient visual search have focused on locating targets as swiftly as possible, but so far have not considered balancing the costs of lengthy searches against the costs of making errors. Here, we propose a neuro-inspired model of visual search that offers an attention-based control mechanism for this speed-accuracy trade-off. The model combines a goal-based fixation policy, which captures human-like behaviour on a simple visual search task, with a deep neural network that carries out the target detection step. The neural network is patched with a target-based feature attention model previously applied to standalone classification tasks. In contrast to image classification, visual search introduces a time component, which places an additional demand on the model to minimise the time cost of the search whilst also maintaining acceptable accuracy. The proposed model balances these two costs by modulating the attentional strength given to characteristic features of the target class, thereby minimising an associated cost function. The model offers a method for optimising the costs of visual search and demonstrates the value of a decision theoretic approach to modelling more complex visual tasks involving attention.

Integration-to-threshold models of two-choice perceptual decision making have guided our understanding of human and animal behavior and neural processing. Although such models seem to extend naturally to multiple-choice decision making, consensus on a normative framework has yet to emerge, and hence the implications of threshold characteristics for multiple choices have only been partially explored. Here we consider sequential Bayesian inference and a conceptualisation of decision making as a particle diffusing in n-dimensions. We show by simulation that, within a parameterised subset of time-independent boundaries, the optimal decision boundaries comprise a degenerate family of nonlinear structures that jointly depend on the state of multiple accumulators and speed-accuracy trade-offs. This degeneracy is contrary to current 2-choice results where there is a single optimal threshold. Such boundaries support both stationary and collapsing thresholds as optimal strategies for decision-making, both of which result from stationary representations of nonlinear boundaries. Our findings point towards a normative theory of multiple-choice decision making, provide a characterisation of optimal decision thresholds under this framework, and inform the debate between stationary and dynamic decision boundaries for optimal decision making.