Predicting sleep latency from the three-process model of alertness regulation

Area of Science:

• Chronobiology
• Sleep Science
• Mathematical Modeling

Background:

• Irregular sleep-wake patterns disrupt circadian rhythms and sleep homeostasis.
• Accurate prediction of sleep latency is crucial for managing sleep disorders and optimizing performance.
• Existing models may not fully capture the complexities of sleep regulation under erratic schedules.

Purpose of the Study:

• To modify the quantitative "three-process model of alertness regulation" for predicting sleep latency.
• To incorporate circadian and homeostatic sleep loss components for alertness prediction.
• To validate the model's predictive accuracy using empirical sleep latency data from irregular sleep studies.

Main Methods:

• Adapted the "three-process model of alertness regulation" by integrating circadian and homeostatic factors.
• Input sleep timing data from two irregular sleep studies into the modified model.
• Regressed predicted alertness at bedtime against empirical sleep latency.
• Performed cross-validation using independent datasets from irregular sleep and shift work studies.

Main Results:

• The modified model achieved a maximum R-squared (R2) of 0.88 for predicting sleep latency using an exponential function.
• Cross-validation on a separate irregular sleep study dataset yielded a maximum R2 of 0.65.
• Model predictions explained more variance in sleep latency than self-rated alertness.
• Successful cross-validation was also demonstrated with published shift work study data.

Conclusions:

• Sleep latency on irregular schedules can be accurately predicted using knowledge of the preceding sleep/wake pattern.
• The modified three-process model offers a robust tool for forecasting sleep onset under non-standard sleep conditions.
• These findings have potential practical applications in optimizing rest and activity management for individuals with irregular schedules.