Food restriction (FR) protocols are a common strategy in neuroscience research to facilitate engagement and motivation within behavioural tasks. However, behavioural tasks often lack reproducibility due to a high degree of interindividual variability, thus limiting the impact and reliability of the data being generated. We hypothesised that a potential source of variability may be due to the occurrence of a hypometabolic and hypothermic state termed fasting-induced torpor (FIT), in laboratory mice. FIT is a profoundly altered state of physiology, associated with sleep disruption and changes to synaptic morphology. As a result, FIT has may be impacting on cognition and subsequent behavioural performance. To address this, we investigated whether FR protocols used for behavioural tasks were sufficient to induce FIT. To this end, adult male C57Bl/6J mice (N=8) were fed once daily and maintained at ~85% of their free-feeding bodyweight. As FIT expression is under circadian control, we investigated whether feeding time would alter FIT characteristics. Accordingly, mice were randomly assigned to either a light-fed group (food given at ZT1) or to a dark-fed group (food given at ZT13) of equal size (n=4). Skin temperature (Tskin) was continuously measured using non-invasive thermal imaging cameras to detect FIT. FIT was operationally defined as when Tskin was reduced by >2SD from baseline levels for at least 1 hour. Bouts of FIT were consistently observed in all mice from day 8 of FR, which coincided with when mice had reached ~85% of their free-feeding bodyweight. A relationship between bodyweight and mean daily Tskin over the course of FR was confirmed using linear regression analysis (light-fed R2=0.7351; dark-fed R2=0.6703). The distribution of Tskin values shifted from unimodal to bimodal with prolonged FR, indicating an increased occurrence of hypothermia bouts, with lower body temperatures becoming more prevalent. Mean Tskin significantly dropped for all mice over FR (P<0.05) by ~1-2oC compared to ad libitum Tskin. Similarly, the minimum daily Tskin also dropped significantly during FR by ~6oC. The mean number of torpor bouts expressed by the light-fed group was 8 (SD 3.464), whereas the mean value for the dark-fed group was 7 (SD 1.00). An unpaired t-test revealed that there was no difference in the number of bouts expressed by each group (P=0.9165). However, the length of FIT bouts expressed by the light-fed group were significantly longer that the dark-fed group (P=0.0007) (light-fed: mean 5.40 hrs, SD 2.77; dark-fed: mean 3.14 hrs, SD 1.06). These results indicate that FR protocols typically used in behavioural neuroscience are sufficient to induce FIT bouts, regardless of feeding time. Further, differences in FIT characteristics depending on feeding time may be contributing to variability and the lack of reproducibility between studies. Having established that FR for behavioural tasks induces torpor, the effects of torpor on subsequent performance will be investigated. As such, mice will perform a series of tasks during and after FIT bouts. These tasks will be used to assess locomotor ability, exploratory behaviour, and cognitive processes such as learning and memory.