Optimizing touchscreen measures of rodent cognition by eliminating image bias

Abstract

For the last twenty years, the Bussey-Saksida touchscreen-based operant conditioning platform has evolved in close parallel alongside the Cambridge Neuropsychological Test Automated Battery (CANTAB) to produce batteries of tests for studying complex cognitive functions in rodents that are increasingly analogous to human diagnostic tests and greatly narrow the translational gap in cognition research. Naturally, with this increasing usefulness comes increasing use, and with it, a need for a greater understanding of and controlling for confounding factors that may limit the system's ability to optimally detect cognitive deficits. In the present study, we show a strong image preference bias in a standard pairwise discrimination task with a widely used spider-plane image pairing. This bias greatly influenced the performance of our experimental mice, significantly affecting the length of time it took mice to complete the task, their progress over time, and several accessory measures usefully recorded by the Bussey-Saksida touchscreen system. We further show that this bias can be corrected by using more highly similar image pairings without sacrificing the animal's ability to learn to distinguish the stimuli. This eliminated all significant stimuli specific differences seen with the spider-plane pairing. We then analyze the pixel composition of the various stimuli to suggest that the bias is due to a difference in image brightness. These findings highlight the importance of carefully modulating paired touchscreen stimuli to ensure equivalence prior to learning and the need for more studies of visual perception in mice, particularly as it relates to their performance in cognitive assays.

Novel Tool Development and the Dynamics of Control: The Rodent Touchscreen Operant Chamber as a Case Study

“…Some hypothesis-driven studies aimed at diagnostic probing have sought to determine whether mice exhibit preferences for specific stimuli or stimulus locations in touchscreen-based tasks (e.g., Delotterie et al 2014) or whether they have biases for specific images or their aspects (e.g., Belarde et al 2021). As new learning tasks have been proposed for use in touchscreens, probative work also has shaped their development and optimization (e.g., Delotterie et al 2014).…”