The utilisation of operant delayed matching and non-matching to position for probing cognitive flexibility and working memory in mouse models of Huntington's disease

•We compared delayed matching and non-matching to position (DMTP and DNMTP) tasks in two different operant apparatus, the 9-hole operant apparatus configuration and the Skinner-like operant apparatus configuration.•We determined that the DMTP and DNMTP operant tasks produce more efficient, robust and reliable results in the Skinner-like operant apparatus configuration.•We therefore used the Skinner-like operant apparatus configuration to test DMTP and DNMTP tasks in the HdhQ111 mouse model of HD.•We tested the DMTP and DNMTP tasks in the HdhQ111 knock-in mouse model of HD which revealed significant deficits in task acquisition and reversal learning in comparison to wildtype animals. Operant behavioural testing provides a highly sensitive and automated method of exploring the behavioural deficits seen in rodent models of neurodegenerative diseases, including Huntington's disease (HD). The delayed matching to position (DMTP) and delayed non-matching to position (DNMTP) tasks probe spatial learning and working memory and when applied serially they can be used to measure reversal learning, which has been shown to be an early symptom of executive dysfunction in HD. The DMTP and DNMTP tasks were conducted in two configurations of operant apparatus; the conventional 9-hole operant apparatus, and a Skinner-like operant apparatus, to compare, contrast and optimise the DMTP and DNMTP operant protocols for use in mice. The optimised tasks were then tested in the HdhQ111 mouse model of HD. Optimisation of the operant apparatus demonstrated that the mice learned the DMTP and DNMTP tasks more rapidly and effectively in the Skinner-like apparatus configuration in comparison to the conventional 9-hole apparatus configuration. When tested in the HdhQ111 mouse model of HD, the DMTP and DNMTP tasks revealed significant deficits in reversal learning. We found that mice were capable of performing the DMTP and DNMTP tasks in both apparatus configurations, but in comparison to the 9-hole configuration, the Skinner-like configuration produced more efficient, robust and reliable results. The results presented here suggest that DMTP and DNMTP tasks, incorporating a reversal learning manipulation, are valid and robust methods for probing selected cognitive deficits in mouse models of neurodegenerative diseases.