Effects of genetic complexity and time resolution in an individual-based model of mesopelagic fish with hunger and fear
Year: 2014
Type: dissertation
Abstract: Many studies have shown that emotion systems are in use to direct behavior, which has also been utilized in a recent mathematical model. To investigate the general validity on the results from this model for vertical migration in planktivorous, mesopelagic fish, I increased the potential for precision in the neuronal response functions (by adding more genes) and executed a sensitivity analysis of the temporal resolution. I found that many aspects of the model were sensitive to the temporal resolution, especially when this was decreased. With fewer generations, it is possible to shorten simulation time by more than a factor of 10, but higher temporal resolutions gave better results. More complex neuronal responses did not seem to improve growth or survival. The increased genetic complexity did not incur longer time to find adaptive solutions in the model. One reason for this may be that the extra genes were only partly utilized to perform more complex neuronal responses by the simulated populations. It was found that for almost all stages in life, the increased genetic complexity was not needed. The usage of more genes to make more complex response curves to their perceptions varied between populations and was not consistent for the genes involved. Complex response curves were mostly used for increased ability to have gradually increasing responses at either very low or very high perception strengths, while still maintaining a sigmoid shape of the curve. As for previous versions of the model, several solutions evolved in the same population. Adding more time steps in each diel cycle gave a significantly higher population egg production at the highest resolution, while adding more diel cycles in a generation did not yield any differences. At the lowest temporal resolution, reducing the number of diel cycles gave lower population egg production. Total amount of fear during a lifetime varied considerably among simulations in each experiment, but showed no specific pattern in regard to temporal resolution. However, the relative length of continual periods of fear were longer in simulations with low resolution in each diel cycle and decreased as the resolution increased. The levels of fear in the different experiments did not stabilize or converge between simulations, opposite to the case for all other traits. Not even after 120,000 generations did levels of fear stabilize or converge between simulations.
Author: Bjørn Andersen
Language: English
Domain: Life Sciences
SDG: Life below water
Open Access status: green
OpenAlex ID: W2274588546