Neurocomputational models of selective attention based on alpha rhythm modulation

Selective attention enables conscious subjects to focus on behaviourally relevant stimuli while filtering out distracting information from the environment. The neural mechanisms underlying selective attention are not yet fully understood. Evidence from the literature suggests that selective attention involves not only prefrontal cortical regions, but also thalamic components interconnected through cortico-thalamic loops. Moreover, over the last two decades, the alpha rhythm (8–14 Hz) has emerged as an important mechanism in the modulation of sensory processing and attentional selection. In this work, we investigate its role in selective attention through a computational modelling approach. We extended a previously developed computational model of selective attention by implementing three neural networks inspired by mechanisms described in the literature. Alpha oscillations were used as the main mechanism for modulating attentional processing. The three networks are implemented using progressively more complex architectures to investigate temporal gating of sensory transmission, spatial attentional selection, and distractor suppression. The simulations reproduced the expected attentional behaviours described in previous studies, providing support for the consistency of the proposed implementations. The three models provide a framework for investigating how alpha modulation can regulate the flow of sensory information and contribute to different mechanisms of selective attention. Overall, our results support the use of the proposed computational framework to investigate the role of alpha oscillations in selective attention and provide a basis for deriving and testing new predictions on the underlying mechanisms.