Lapsing during sleep deprivation is associated with distributed changes in brain activation

Abstract

Lapses of attention manifest as delayed behavioral responses to salient stimuli. Although they can occur even after a normal night's sleep, they are longer in duration and more frequent after sleep deprivation (SD). To identify changes in task-associated brain activation associated with lapses during SD, we performed functional magnetic resonance imaging during a visual, selective attention task and analyzed the correct responses in a trial-by-trial manner modeling the effects of response time. Separately, we compared the fastest 10% and slowest 10% of correct responses in each state. Both analyses concurred in finding that SD-related lapses differ from lapses of equivalent duration after a normal night's sleep by (1) reduced ability of frontal and parietal control regions to raise activation in response to lapses, (2) dramatically reduced visual sensory cortex activation, and (3) reduced thalamic activation during lapses that contrasted with elevated thalamic activation during nonlapse periods. Despite these differences, the fastest responses after normal sleep and after SD elicited comparable frontoparietal activation, suggesting that performing a task while sleep deprived involves periods of apparently normal neural activation interleaved with periods of depressed cognitive control, visual perceptual functions, and arousal. These findings reveal for the first time some of the neural consequences of the interaction between efforts to maintain wakefulness and processes that initiate involuntary sleep in sleep-deprived persons.

Figure 1.

Task stimuli used in the experiment were single large, global letters (H or S) composed of several smaller local letters (H or S). The global letter and the local letters were either congruent (global H made up of local Hs and global S made up of local Ss) or incongruent (global H made up of local Ss and global Ss made up of local Hs). A red fixation dot was displayed at the center of the screen throughout each run.

Figure 2.

Illustrated are the RT histograms for correct responses and raster plots of responses associated with three representative subjects. Data were obtained during RW after a normal night's sleep or after 22–24 h of total SD. The top, middle, and bottom panels depict the distribution of responses and response patterns associated with an individual with few lapses, an individual with a moderate number of lapses, and an individual with a high number of lapses. The expanded inset illustrates the slower RT and greater trial-to-trial variation in RT when a subject was sleep deprived.

Figure 3.

Plots on the left half of the figure depict the average signal response after collapsing all correct responses regardless of RT. These reflect the mean task-related activation during RW after a normal night's sleep or SD. Plots on the right half of the figure depict the relationship between delay in RT and fMRI signal. Delayed responses were associated with a modest but state-indifferent reduction in peristimulus signal and larger, state-sensitive increases in peak signal in frontoparietal control regions.

Figure 4.

3D plots showing the results of trial-by-trial modeling of fMRI signal associated with RTs ranging from 0.2 s faster than the mean RT for a given individual, to 0.7 s slower than the mean RT. The signal time course at the mean RT is marked in green. a, Medial frontal region; b, intraparietal sulcus; c, lateral occipital (extrastriate) cortex. Note that peak signal in the frontoparietal control regions increased with slower responses, albeit to a lesser extent during SD. In contrast, response slowing was associated with decrease in extrastriate peak signal during SD.

Figure 5.

Task-related fMRI signal associated with responding at the average RT and during a lapse (modeled as a response 0.5 s longer than the average RT for that state). Lapses were associated with higher peak signal in the medial frontal cortex (top) and bilateral intraparietal sulcus (middle) in both RW and SD. In the occipital region, lapsing significantly reduced peak signal during SD (bottom) but did not significantly modulate the peak fMRI signal after RW, although there was a delay in the time-to-peak. Random effects analysis using a threshold of p < 0.001 was used to detect task-related activation. Significant differences between peak signal associated with a lapse trial and the mean response for each state are marked with an asterisk. The shaded time points indicate those contrasted to assess significant state effects. The inset shows the mean peak signal associated with the time points under consideration. Error bars represent SEM. *p < 0.05, **p < 0.005, ***p < 0.001.

Figure 6.

Differential cortical responses associated with the fastest 10% RTs and the slowest 10% RTs after RW and SD. There was higher peak fMRI signal for the slowest 10% of trials in the medial frontal cortex (top) and bilateral intraparietal sulcus (middle). Between-state differences were observed only for the slowest 10% RTs in bilateral intraparietal sulcus and the inferior occipital cortices (bottom). Peak signal was significantly lower in the occipital region even for the fastest responses across states. Random effects analysis using a threshold of p < 0.001 was to detect task-related activation. Significant differences between peak signal associated with a lapse and the average response for each state are marked with an asterisk. The shaded time points indicate those contrasted to assess significant state effects. The inset shows the mean peak signal associated with the time points under consideration. Error bars represent SEM. *p < 0.01, **p < 0.001.

Figure 7.

3D plot showing the results of trial-by-trial modeling of fMRI signal in the thalamus. fMRI signal evolution for RTs ranging from 0.2 s faster than mean RT for a given individual to 0.7 s slower than mean RT are displayed. The signal time course at the mean RT is marked in green. Lapses in SD significantly attenuated task-related thalamic activation. This contrasted with RW in which lapses were associated with significantly greater peak fMRI signal. Error bars represent SEM. *p < 0.01, **p < 0.001.