Mice can monitor their timing errors
The ability to accurately monitor the passage of time is pivotal for many functions such as associative learning and planning. Earlier experiments show that humans and rats can integrate their representational (endogenous) uncertainty about time intervals into decisions in a nearly normative fashion, suggesting that they can monitor their timing errors. This ability can be formalized as knowing whether and how much one has under- or over-estimated the duration of an event without any feedback, which we refer to as metric error monitoring (MEM). Although MEM has been documented in humans and very recently in rats (with two-alternative two choice procedure), whether mice can monitor their timing errors based on confidence-like measures is unknown. I tested this hypothesis in C57BL/6 male mice (N=16). Mice were trained to depress a lever for a minimum target duration in order to receive a reward in the food hopper. No reward was given when mice under-produced the minimum required target interval. During test trials, the rate of nose-pokes into the food hopper during a variable response window after releasing the lever was recorded. Mice nose-poked more vigorously (reflecting higher reward expectancy) following temporal productions around the target duration compared to when they under-produced the minimum target interval. This result suggests that mice judge whether or not their temporal production was close to meeting the task criterion to receive a reward, and thus show temporal error monitoring abilities in mice. Our findings also provide the necessary behavioural tool to study the neural basis based on correlational and manipulative methods in non-human animals.
Mice, Timing, Error Monitoring, Cognition and Behaviour