Sleep enhances category learning

Abstract

The ability to categorize objects and events in the world around us is a fundamental and critical aspect of human learning. We trained healthy adults on a probabilistic category-learning task in two different training modes. The aim of this study was to see whether either form of probabilistic category learning (feedback or observational) undergoes subsequent enhancement during sleep. Our results suggest that after training, a good night of sleep can lead to improved performance the following day on such tasks.

Figure 1.

Sample screens of WPT stimulus and response screens. (Top) Observation mode. On each trial, subjects see a set of cards along with the correct weather outcome displayed above. They then press the corresponding key (“sun” or “clouds”) to move to the next trial. (Bottom) Feedback mode. On each trial, subjects see a set of cards and are asked to predict the correct weather outcome by pressing either the “sun” or “clouds” key. After pressing a key the actual outcome is displayed above the cards, along with the information on whether their prediction had been correct or incorrect.

Figure 2.

Sleep's effect on WPT performance. (A–C) Corrected improvement ± SEM is shown. Significant improvement was seen in the Sleep groups trained with either 200 trials in observational mode (A) or with 100 trials in short feedback mode (C), but not with 200 trials in feedback mode (B). Subjects in the Wake groups failed to show significant improvement in feedback mode (B,C). (D) Subjects who performed well at the initial test session had a higher percentage of REM sleep the following night. (*) P < 0.05; (***) P < 0.0001.

Figure 3.

Correlation between post-training performance and 12-h improvement. (OBS) Observational mode; (FB) feedback; (SFB) short feedback; (PSG) observational mode in sleep laboratory. Light dashed lines: ceiling limits; heavy dashed lines: regression fits to all points. Specifics regarding panels A and B in text.