Background: We have previously shown that the White-crowned Sparrow (WCS) decreases sleep by 60% during a period of migratory restlessness relative to a non-migratory period when housed in a 12 h light: 12 h dark cycle. Despite this sleep reduction, accuracy of operant performance was not impaired, and in fact rates of responding were elevated during the migratory period, effects opposite to those routinely observed following enforced sleep deprivation. To determine whether the previously observed increases in operant responding were due to improved performance or to the effects of migration on activity level, here we assessed operant performance using a task in which optimal performance depends on the bird's ability to withhold a response for a fixed interval of time (differential-reinforcement-of-low-rate-behavior, or DRL); elevated response rates ultimately impair performance by decreasing access to food reward. To determine the influence of seasonal changes in day length on sleep and behavioral patterns, we recorded sleep and assessed operant performance across 4 distinct seasons (winter, spring, summer and fall) under a changing photoperiod.
Results: Sleep amount changed in response to photoperiod in winter and summer, with longest sleep duration in the winter. Sleep duration in the spring and fall migratory periods were similar to what we previously reported, and were comparable to sleep duration observed in summer. The most striking difference in sleep during the migratory periods compared to non-migratory periods was the change from discrete day-night temporal organization to an almost complete temporal fragmentation of sleep. The birds' ability to perform on the DRL task was significantly impaired during both migratory periods, but optimal performance was sustained during the two non-migratory periods.
Conclusions: Birds showed dramatic changes in sleep duration across seasons, related to day length and migratory status. Migration was associated with changes in sleep amount and diurnal distribution pattern, whereas duration of sleep in the non-migratory periods was largely influenced by the light-dark cycle. Elevated response rates on the DRL task were observed during migration but not during the short sleep duration of summer, suggesting that the migratory periods may be associated with decreased inhibition/increased impulsivity. Although their daily sleep amounts and patterns may vary by season, birds are susceptible to sleep loss throughout the year, as evidenced by decreased responding rates following enforced sleep deprivation.