Vestigial preference functions in neural networks and túngara frogs.

Abstract

Although there is a growing interest in understanding how perceptual mechanisms influence behavioral evolution, few studies have addressed how perception itself is shaped by evolutionary forces. We used a combination of artificial neural network models and behavioral experiments to investigate how evolutionary history influenced the perceptual processes used in mate choice by female tú ngara frogs. We manipulated the evolutionary history of artificial neural network models and observed an emergent bias toward calls resembling known ancestral states. We then probed female tú ngara frogs for similar preferences, finding strong biases toward stimuli that resemble a call hypothesized for a recent ancestor. The data strongly suggest that female tú ngara frogs exhibit vestigial preferences for ancestral calls, and provide a general strategy for exploring the role of historical contingency in perceptual biases.

A Test of the Critical Assumption of the Sensory Bias Model for the Evolution of Female Mating Preference Using Neural Networks

“…Phelps and colleagues have used neural networks to examine the effects of species recognition on general mate choice and have verified their results by comparing the predicted behavior with that of real animals (Phelps and Ryan 1998). They have also found that the past history of selection on neural networks has strong effects in predicting present day behaviors (Phelps and Ryan 2000; Phelps et al 2001). Pfennig and Ryan (2006, 2007) have used neural networks to examine the extent to which reproductive character displacement among species promotes behavioral isolation among populations within species.…”

Reproductive character displacement generates reproductive isolation among conspecific populations: an artificial neural network study

“…The networks were presented pulsatile calls in a frequency by time matrix in which the values within the matrix ranged from 0 to 1 and represented amplitude of the signal at a given frequency and time (analogous to a sonogram; see also Phelps & Ryan 1998, 2000Phelps et al 2001). We synthesized the calls using a program written in Matlab that generated each call by combining randomly chosen values (see below) of four parameters: call duration (the length of the call in terms of matrix columns); call dominant frequency (the frequency in the call with the greatest energy, measured in terms of matrix rows); pulse rate (measured as number of pulses per matrix column) and inter-call interval (the number of matrix columns between the last column of the first call and the first column of the second call).…”

The evolution of behaviour, and integrating it towards a complete and correct understanding of behavioural biology

“…In a previous study we had trained populations of artificial neural networks to recognise túngara frog calls Ryan, 1998, 2000;Phelps et al, 2001). We showed that the response of the networks to novel stimuli predicted the response of female túngara frogs to the same novel stimuli, suggesting that there might be some similarities in how the two systems decode the target stimulus and generalise to other stimuli.…”