Biomedical subjects
Publications and source records attributed to R M Sibly.
Computer simulation was used to relax assumptions of analytical life-cycle theory about the eventual outcome of evolution in a constant environment. The computer simulation models, of diploid one-locus genetic systems, are described in detail. Good agreement was obtained between the analytical and simulation outcomes, except in some cases of discrepancy between the male and female life cycles.
Analytical and computer models were used to reexamine bet-hedging, the reduction in fecundity that is evolutionarily advantageous in conditions of greater variation in juvenile survivorship or less variation in adult survivorship. The computer simulation models represent diploid one-locus genetic systems with semidominance. Schaffer's (1974) predictions proved remarkably robust when variations were symmetrical, and a simple modification allowed successful prediction for the asymmetries that occurred in the computer simulations when variations were large.
In this paper it is argued that any model of the motivational (i.e. reversible) processes governing the behaviour of an animal can be represented by means of isoclines in a multidimensional 'causal-factor space'. The argument is axiomatic, based upon the two prime assumptions: that (1) it is always possible to classify the behavioural repertoire of a species in such a way that the classes are mutually exclusive in the sense that the members of different classes cannot occur simultaneously, and (2) these incompatible actions are uniquely determined by a particular set of causal factors. The isoclines join all points in the space which present a given 'degree of competitiveness' of a particular 'candidate' for overt behavioural expression. The competition between candidates is an inevitable consequence of the fact that animals cannot 'do more than one thing at a time', and is envisaged as taking place in the behavioural final common path. An empirical method of determining the motivational state (i.e. point in causal-factor space) is outlined. This is a 'relative' method, independent of the arbitrary calibration of the axes of the causal-factor space. It is shown that an arbitrary scale of measurement along any two axes of the causal-factor space is all that is necessary for empirical determination of the shape of a motivational isocline. Experiments in which this method has been applied to the measurement of hunger and thirst in doves are outlined, and the results are discussed in terms of their implications for motivation theory in general.