Home // COGNITIVE 2012, The Fourth International Conference on Advanced Cognitive Technologies and Applications // View article

Energy Saving Accounts for the Suppression of Sensory Detail

Authors:
Terry Bossomaier
Lionel Barnett
Vaenthan Thiruvarudchelvan
Herbert Jelinek

Keywords: energy saving; computer simulation; spiking neural network; Bayesian prior; detail suppression

Abstract:
High functioning autistic people can exhibit exceptional skills with numbers, eidetic imagery and recall of concrete detail, as brought to popular attention in the film Rain Man. However, it now transpires that these skills are to some extent latent within all of us. We do not have access under normal circumstances to this concrete detail, yet brain stimulation experiments show that it exists in all of us. This paper proposes that one of the reasons for this lies in the brain’s need to conserve energy. Computer simulations using a spiking neural network support this hypothesis. A spiking neural network was set up with a number of feature detectors feeding an output unit, which in turn generates inhibition of the input neurons. This reduces the spike activity of the input, and thus overall energy usage.

Pages: 14 to 18

Copyright: Copyright (c) IARIA, 2012

Publication date: July 22, 2012

Published in: conference

ISSN: 2308-4197

ISBN: 978-1-61208-218-9

Location: Nice, France

Dates: from July 22, 2012 to July 27, 2012