Cross inhibition improves activity selection when switching incurs time costs
Autor: | John M. McNamara, Lutz Fromhage, James A. R. Marshall, Lianne F. S. Meah, Alasdair I. Houston, Angélique Favreau-Peigné |
---|---|
Přispěvatelé: | Department of Computer Science, University of Sheffield [Sheffield], Modélisation Systémique Appliquée aux Ruminants (MoSAR), Institut National de la Recherche Agronomique (INRA)-AgroParisTech, Department of Biological and Environmental Science [Jyväskylä Univ] (JYU), University of Jyväskylä (JYU), School of Mathematics, University of Bristol [Bristol], School of Biological Sciences, European Research Council, Department of Biological and Environmental Science, University of Jyväskylä |
Rok vydání: | 2015 |
Předmět: |
0106 biological sciences
Cross inhibition Mathematical optimization Computer science [SDV]Life Sciences [q-bio] 010603 evolutionary biology 01 natural sciences Time cost 0501 psychology and cognitive sciences Foraging 050102 behavioral science & comparative psychology Geometric framework käyttäytyminen Selection (genetic algorithm) Positive feedback Behavior Geometric Framework 05 social sciences Activity selection Deadlock (game theory) Activity Selection Coupling (computer programming) Cross Inhibition Animal Science and Zoology Decision process Neuroscience |
Zdroj: | Current zoology Current zoology, Institute of zoology, Chinese academy of sciences, 2015, 61 (2), pp.242-250 ResearcherID |
ISSN: | 2396-9814 1674-5507 |
Popis: | We consider a behavioural model of an animal choosing between two activities, based on positive feedback, and examine the effect of introducing cross inhibition between the motivations for the two activities. While cross-inhibition has previously been included in models of decision making, the question of what benefit it may provide to an animal’s activity selection behaviour has not previously been studied. In neuroscience and in collective behaviour cross-inhibition, and other equivalent means of coupling evidence-accumulating pathways, have been shown to approximate statistically-optimal decision-making and to adaptively break deadlock, thereby improving decision performance. Switching between activities is an ongoing decision process yet here we also find that cross-inhibition robustly improves its efficiency, by reducing the frequency of costly switches between behaviours. |
Databáze: | OpenAIRE |
Externí odkaz: |