Article Figures & Data
Figures
- Figure 1.
Study protocol. Twenty-nine subjects were allocated into three experimental groups (A–C). Group A was trained at 8:00 A.M. and retested at 3:00 P.M. and 10:00 P.M. across wakefulness. Group B was trained at 12:00 P.M. (midday) and retested at 10:00 P.M. (after 10 h of wakefulness) and at 8:00 A.M. (10 h later) after a night of sleep. Group C was trained at 10:00 P.M. and retested at 8:00 A.M. (10 h later) after a night of sleep and at 6:00 P.M. (after 10 h of wakefulness).
- Figure 2.
The spatial variant of the n-back working memory task configurations. Four large white dots aligned in a single row were displayed on the screen, indicating the four possible places where a stimulus could appear. The stimulus consisted of one of the four dots changing from white to red. Subjects had to respond to the stimulus by pushing one of four buttons, which were arranged in lines corresponding spatially to the four possible stimulus positions, with the right fingers as quickly and accurately as possible when the next stimulus appeared. Responses had to be made after a delay of n (n-back) stimuli. The different load levels (1–9) were run in blocks of 20 + n stimuli; thus, 20 responses were obtained at each load level. The interstimulus interval was set at 500 ms, and a stimulus was displayed for 1500 ms. Arrangement of the stimuli was randomized in each test session.
- Figure 3.
Initial training performances in all groups. Accuracy (top) and RT (bottom) in the initial training session for each load level are plotted. Filled circles with error bars represent mean and SEM values in each panel. Significant interload level difference in accuracy was observed between level 5 and level 6; the accuracy linearly decreased as the task difficulty increased up to level 5 before rapidly dropping at level 6 and remaining low at <50% thereafter. We observed no significant interload level difference in RT. *p < 0.0001.
- Figure 4.
Time courses of improvement in n-back task performance. Time courses of NL (top) and RT (bottom) are displayed. A–C, Bars and error bars represent mean and SEM values in groups A–C, respectively. In the sessions after posttraining sleep, we observed significant improvement in NL compared with those before sleep (groups B and C; black bars), but RT showed no significant benefit from sleep. *p < 0.0167; **p < 0.001.
- Figure 5.
Intersession differences in improvement in n-back accuracy. Bars and error bars represent mean and SEM values of intersession differences in NL in each group. Left and right bars in each experimental group show intersession differences in NL between initial training (IT) and retest 1 (R1) sessions, and between R1 and R2 sessions, respectively. Filled bars represent NL improvements during the posttraining sleep in groups B and C. Post hoc test revealed a significant intergroup difference in NL between groups A and C (**p = 0.0109) and a trend toward intergroup difference in NL between groups A and B (*p = 0.0491).
