Spatial cognitive performance is linked to thigmotaxis in field crickets

Information is acquired through cognitive processes and is essential for enhanced fitness in many animal species. However, little is known about intraspecific variation in cognitive performance, or about phenotypic correlates of this variation, especially in invertebrates. Here we quantified individual variation in spatial learning, exploration, boldness and thigmotaxis (wall hugging) in female Texas field crickets, Gryllus texensis. We trained females to locate a food reward (dish baited with food) located in one arm of a radial arm maze, followed by a cognitive test (probe) trial where no food reward was present. Our results suggest that female crickets could learn and remember the food location, as latency to find the reward dish, distance travelled, number of complete arm entries and number of errors made prior to finding the reward dish all declined with training trials and during the cognitive test trial. Furthermore, females performed equally well on all cognitive measures during the cognitive test trial and the last training trial, suggesting they did not rely on olfactory cues to locate the correct dish during the cognitive test trial. Individual variation in cognitive measures during the test trial were significantly negatively correlated with thigmotaxis during the trial, but were not correlated with body size, latency to leave the novel maze's acclimation zone or the percentage of the novel maze explored during initial exploration of the maze. Our findings suggest individuals with reduced thigmotaxis either (1) are better at learning a spatial task and/or (2) rely more heavily on vision than olfaction. These results contribute to the growing literature examining links between personality and cognition, and highlight the importance of quantifying correlated aspects of the phenotype when assessing individual variation in cognitive performance in a cognitive test. (C) 2019 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.