Modulating the homeostatic process to predict performance during chronic sleep restriction.

Abstract

BACKGROUND: In most current sleep/performance models, the homeostatic process is generally conceived as a simple reservoir in which performance capacity increases exponentially during sleep and decays either linearly or exponentially during wakefulness. Models that include this notional homeostatic process have been successful for describing sleep-performance data under conditions of irregular sleep schedules, jet lag, and short periods of total sleep loss. However, recently described data from sleep restriction studies indicate that recovery following chronically restricted sleep is considerably slower than would be predicted by these models. These findings suggest that chronic sleep restriction induces relatively long-term, slow-recovering changes in brain physiology that affect alertness and performance. METHODS: This paper describes, both conceptually and mathematically, a generic modification to sleep/performance models that facilitates the ability to predict the rate at which alertness and performance restoration occurs during recovery sleep following chronic sleep restriction. Weighted nonlinear least-squares methods were used to compare the proposed modulated homeostatic model with recent sleep/performance observations during chronic sleep restriction and recovery. RESULTS: When compared with the classical Walter Reed homeostatic model, this proposed model was found to provide a better description of sleep restriction and recovery observations. The proposed model was also found to be consistent with the data from a recent University of Pennsylvania study. CONCLUSIONS: These two models make significantly different predictions of performance during both the recovery phase and the chronic sleep restriction phase.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 37 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.