Throughout this discussion we have assumed that behaviour

time of day, feeding history, sex, age, etc.). Failure to control experimentally for these sources of variability could inflate estimates of R (in the case of sex differences) or underestimate R (in the case of time of day variation). For instance, the repeatability of PD0332991 declines with time between successive measures ( White, et al., 2013), suggesting that ontogenetic or ageing effects may confound our estimates of broad sense repeatability if we do not account for time effects.