TY - THES
AU - Cuff, Wilfred R.
PY - 1972
TI - A consideration of the concepts of generality and complexity, as used in experimental components analysis
KW - Thesis/Dissertation
LA - eng
M3 - Text
AB - Two features of experimental compoents analysis (Holling, 1966) are considered. A method of writing general functions is proposed and verified with some observations from hydra. An analytical method of studying certain features of a simulation model of attack is then proposed and tested.
The method proposed to attain generality of functions is a strategic one. Operationally, one need only determine a strategy used by many animals, uncover some opposing variables which may have been important in its development, and predict on the basis of these antagonistic forces some characteristic response of the strategy. It is suggested that this characteristic response is a broadly applicable as is the strategy.
The dynamic response of a number of modes of searching through volume is predicted to be of shorter duration the larger the hunger threshold to which the mode responds. This is found to be the case for hydra. A similar hypothesis, that the steady states of these searching modes be directly related to the hunger thresholds of the modes, does not give as clear results. It is suggested that an additional strategy must be considered to give a better picture. These results are then used to write a model for the time which an animal spends searching for food.
The sensitivity analysis of the attack simulation model attempts to formalize the conventional method: going through the block diagram and picking out what seems to be important. A Boolean expression is written to describe all possible paths through the model. This expression is used to tabulate the various forms of the equation of the dependent variable, the attack rate. These forms of the attack rate are used to illustrate how one might analyze a complex, non-linear model.
N2 - Two features of experimental compoents analysis (Holling, 1966) are considered. A method of writing general functions is proposed and verified with some observations from hydra. An analytical method of studying certain features of a simulation model of attack is then proposed and tested.
The method proposed to attain generality of functions is a strategic one. Operationally, one need only determine a strategy used by many animals, uncover some opposing variables which may have been important in its development, and predict on the basis of these antagonistic forces some characteristic response of the strategy. It is suggested that this characteristic response is a broadly applicable as is the strategy.
The dynamic response of a number of modes of searching through volume is predicted to be of shorter duration the larger the hunger threshold to which the mode responds. This is found to be the case for hydra. A similar hypothesis, that the steady states of these searching modes be directly related to the hunger thresholds of the modes, does not give as clear results. It is suggested that an additional strategy must be considered to give a better picture. These results are then used to write a model for the time which an animal spends searching for food.
The sensitivity analysis of the attack simulation model attempts to formalize the conventional method: going through the block diagram and picking out what seems to be important. A Boolean expression is written to describe all possible paths through the model. This expression is used to tabulate the various forms of the equation of the dependent variable, the attack rate. These forms of the attack rate are used to illustrate how one might analyze a complex, non-linear model.
UR - https://open.library.ubc.ca/collections/831/items/1.0101225
ER - End of Reference