A temperature-based, stochastic model of mate-choice with incomplete information and evolving representations

This thesis is about mate-choice, specifically, about how population-level mate-choice behavior can emerge from a number of underlying preference mechanisms possessed by the individuals in the population. Modeling the decision-making processes involved in mate-choice is interesting because it involves the interplay of a number of competing pressures and constraints. Specifically, individuals must attempt: i) to find the best mate possible, ii) with only partial knowledge of the individuals in the pool of potential candidates, and iii) in a limited amount of time. In this thesis we develop a model of mate-choice that differs with current mate-choice models in five major respects - namely, it incorporates computational temperature as a measure of choosiness; it uses, for each individual, a multi-dimensional vector of mate-value for a variety of characteristics, instead of a single, overall mate-value describing the individual; it employs a fluid representational structure for potential mates that evolves over time as new information about that person becomes available; it uses subjective mate-values, since mate-value is largely, although certainly not completely, subjective ("beauty is in the eye of the beholder"); and it incorporates a self-esteem parameter that acts as an internal gauge measuring the mate value of a particular individual with respect to potential mates. With the context of these constraints, we show that this model (the "Standard" model) qualitatively reproduces empirical data on first-marriage rates at various ages, male-female marriage age shifts (women initially marry earlier than men), mate-value correlations among married couples, and changes in marriage-rate curves when pressure to marry early is decreased. Simulations are also run in which the Standard Model is compared to a number of parameter variations to study its performance under conditions which would be possible in a real environment.