Good vibrations switch attention: an affective function for network oscillations in evolutionary simulations

Bram T Heerebout,R Hans Phaf
DOI: https://doi.org/10.3758/CABN.10.2.217
Abstract:In the present study, a new hypothesis on the neural mechanisms linking affect to attention was brought forward by evolutionary simulations on agents navigating a virtual environment while collecting food and avoiding predation. The connection strengths between nodes in the networks controlling the agents were subjected to random variation, and the fittest agents were selected for reproduction. Unexpectedly, oscillations of node activations emerged, which drastically enhanced the agent's fitness. We analyzed the mechanisms involved in the modulation of attention and found that oscillations acted on competitive networks. Response selection depended on the connection structure, but the speed and efficacy of switching between selections was modulated by oscillation frequency. The main focus of the present study was the differential emergence of stimulus-specific oscillation frequencies. Oscillations had a higher frequency in an appetitive motivational state than in an aversive state. We suggest that oscillations in biological networks also mediate the affective modulation of attention.
What problem does this paper attempt to address?