The evolution of neural-network movement strategies in fluctuating environments

Valdes Villarreal Boullosa, Luis Fernando (2020) The evolution of neural-network movement strategies in fluctuating environments. Master's Research Project 2, Ecology and Evolution.

Preview	Text mEE_2020_VVBoullosaLF.pdf Download (2MB) | Preview
	Text toestemming.pdf Restricted to Registered users only Download (118kB)

Abstract

Movement is an important survival mechanism among species living in changing environments. This is all the more important when environmental change can be predicted from present or past information. Here I view condition-dependent movement as a special case of phenotypic plasticity. Previous research predicts that phenotypic plasticity will only evolve when the right conditions are met: when environmental change is noticeable within the lifetime of an individual, when the direction and/or degree of change can to a certain extent be predicted on the basis of some cues. I investigate whether, and to what extent, these predictions apply to movement. In my model, the movement decisions are taken by evolvable neural networks which processes local information on resource densities to determine the direction and/or step length of movement. The networks are transmitted from parents to their offspring (subject to some mutation). Individuals that take the most adequate movement decision can gather more resources and produce more offspring. This way, the movement-determining neural networks evolve over the generations. I regularly observed quite complicated eco-evolutionary dynamics, as well as polymorphisms, such as the dynamical coexistence of sessile and motile individuals.I can conclude that the evolution of condition-dependent movement is strongly affected by the pattern of ecological heterogeneity and the mechanisms underlying movement decisions.

Item Type:	Thesis (Master's Research Project 2)
Supervisor name:	Weissing, F.J.
Degree programme:	Ecology and Evolution
Thesis type:	Master's Research Project 2
Language:	English
Date Deposited:	30 Sep 2020 12:37
Last Modified:	30 Sep 2020 12:37
URI:	https://fse.studenttheses.ub.rug.nl/id/eprint/23449

Actions (login required)

View Item