The Effects of 43 Hours of Sleep Deprivation on Executive Control Functions: Event-Related Potentials in a Visual Go/no Go Task

Article excerpt

In the adult population there is a high prevalence of sleep loss which can affect physical, emotional, and cognitive functions (Falkenstein, Hoormann, & Hohnsbein, 1999; Fallgatter & Strik, 1999; Filipovic, Jahanshahi, & Rothwell, 2000). For example, workplace accidents can result from sleep loss. In order to investigate how cognitive functions are affected by sleep loss, sleep deprivation is usually used as a means to induce the state of sleep loss. Results of a number of studies on sleep have indicated that higher cognitive functions are susceptible to sleep deprivation (Diekelmann, Landolt, Lahl, Born, & Wagner, 2008; Horne, 1993; Kumar, 2008; Kuriyama, Mishima, Suzuki, Aritake, & Uchiyama, 2008; Muzur, Pace-Schott, & Hobson, 2002). It has been found that performance of tasks that rely on the prefrontal cortex are greatly affected by sleep deprivation, while performances of tasks that do not rely on frontal functions are affected less (Harrison & Horne, 2000; Horne, 1993). Such findings provide evidence for the hypothesis that the prefrontal cortex is particularly susceptible to sleep deprivation. In terms of functions rather than anatomical substrates, researchers have suggested that sleep deprivation influences frontal executive control functions in particular and the sensitivity of the prefrontal cortex to sleep has also been emphasized (Jones & Harrison, 2001; Muzur et al., 2002). In research concerning psychological cognition it has been indicated that response inhibitions, information updating, and monitoring are core aspects of executive control functions (Baddeley, 1992; Botvinick, Braver, Barch, Carter, & Cohen, 2001; D'Esposito & Grossman, 1996). Therefore, it is essential to investigate whether or not total sleep deprivation (TSD) affects executive control functions. To our knowledge, to date no event-related potentials (ERP) study has been carried out in which the aim was to assess whether or not executive control functions are impaired by 43 hours of TSD.

The Go/No-Go task (Falkenstein et al., 1999; Fallgatter & Strik, 1999; Filipovic et al., 2000) is frequently used to investigate inhibition of a response that requires activation of the executive control system. Participants are asked to produce fast responses to one kind of stimulus called Go, but to refrain from responding to another kind of stimulus called No go. In most of the Go/Nogo ERP studies it has been found that there is an N2 component for No go stimuli relative to Go stimuli, particularly at frontocentral sites (Bekker, Kenemans, & Verbaten, 2004; Fallgatter & Strik, 1999; Filipovic, Jahanshahi, & Rothwell, 1999, 2000; Nieuwenhuis, Yeung, & Cohen, 2004; Nieuwenhuis, Yeung, van den Wildenberg, & Ridderinkhof, 2004; Oddy, Barry, Johnstone, & Clarke, 2005; Scisco, Leynes, & Kang, 2008). In auditory tasks, the No go-N2 at the frontal sites close to the midline is often found to be smaller compared with the No go-N2 that is elicited by visual stimulation (Falkenstein et al., 1999; Falkenstein et al., 2002; Tekok-Kilic, Shucard, & Shucard, 2001). A P300 was also found at frontal and central sites which was larger for No go than that for Go (Falkenstein et al., 1999; Falkenstein et al., 2002; Fallgatter & Strik, 1999; Filipovic et al., 2000; Kok, 1986; Roberts, Rau, Lutzenberger, & Birbaumer, 1994; Shucard, McCabe, & Szymanski, 2008; Smith, Johnstone, & Barry, 2008). The No go-N2 and No go-P3 have traditionally been associated with inhibition control, which is a component of executive control and relates to the ability to deliberately suppress dominant and automatic proponent responses. However, recently several researchers (Botvinick et al., 2001; Bruin & Wijers, 2002; Bruin, Wijers, & van Staveren, 2001; Donkers & van Boxtel, 2004; Miller & Cohen, 2001; Troy et al., 2000; van Boxtel, van der Molen, Jennings, & Brunia, 2001; Yeung, Cohen, & Botvinick, 2004) have argued that the amplitude of No go-N2 mainly reflected the presence of conflict, and seemed to be unrelated to response inhibition, or at least related only to a limited extent. …