Effect of time of day on reward circuitry: Further thoughts on methods, prompted by Steel et al 2018

The interplay between circadian and reward function is well understood in animal models, and is of growing interest as an aetiological explanation in psychopathologies. Circadian modulation of reward function has been demonstrated in human behavioural data, but understanding at the neural level is limited. In 2017, our group published results of a first step in addressing this deficit, demonstrating a diurnal rhythm in fMRI-measured reward activation. In 2018, Steel et al wrote a constructive critique of our findings, and the aim of this paper is to outline how future research could improve on our first proof-of-concept study. Key challenges include addressing divergent and convergent validity (by addressing non-reward neural variation, and testing for absence of variation in threat-related pathways), preregistration and power analysis to protect against false positives, wider range of fMRI methods (to directly test our post-hoc hypothesis of some form of reward prediction error, and multiple phases of reward), the parallel collection of behavioural data (particularly self-reported positive affect, and actigraphically measured activity) to illuminate the nature of the reward activation across the day, and some attempt to parse out circadian versus homeostatic/masking influences on any observed diurnal rhythm in neural reward activation.
Comment: 9 pages, 1 figure