4.1 Summarize Findings
The operation check showed that there was no statistical difference between the PVT test results before the nap and the PVT test results after the nap in the two factors of the participants’ caffeine intake or not. The estimation effect size data shows that the main effects and interactions of the two within-subject factors are not statistically significant. Without considering the influence of other factors on the PVT test, only the internal factor-caffeine / decaffeination on the PVT test is considered, whether the participants consume caffeine has no effect on the PVT test. In other words, the data in this study shows that caffeine does not help reduce sleep inertia after a nap when working in a simulated shift. This inference cannot support previous hypothetical.
Based on the analysis, the profile plots of pre / post-nap * caffeine / decaffeinated for each dependent variable show that there are interactions between the two factors. However, the interaction of pre / post-nap was not statistically significant. In other words, compare the differences between PVT tests before / after nap at different caffeine factor levels. From the analysis of the data, there is no difference between them, but from the analysis of the Profile Plots, there are differences between all dependent variables (Not parallel). Especially the Profile Plots of RTFASTMEAN and LAPSES. Pre / post-nap * caffeine / decaffeinated Profile Plots (Figure 2) of LAPSES show that under caffeine conditions, number times participants took longer to respond than 500ms of pre-nap more than post-nap; number times participants took longer to respond than 500ms of pre-nap more less post-nap under decaffeinated. And the interaction of the two factors of LAPSES is the only one with an approximate marginal significance, P = 0.117. The interaction between the two factors of RTFASTMEAN is not significant, but its pre / post-nap * caffeine / decaffeinated profile plots (Figure 3) show that caffeine is basically similar to mean 10% fastest reaction of pre-nap. The caffeine is significantly lower than decaffeinated to mean 10% fastest reaction of post-nap.
The profile Plots of each dependent variable and the P value of interaction were analyzed. Profile Plots showed interaction, but the reason why the P value was not statistically significant may be due to the small sample size and insufficient data to constitute statistical significance. So, the conclusion is that there is interaction between pre / post-nap and caffeine / decaffeinated in each dependent variable, but the interaction is not significant. In other words, the effect on the PVT test pre/post- nap varies at different levels of caffeine / decaffeinated, but it is not statistically significant.
In many experiments concerning caffeine and sleep, the total dose and single dose of caffeine are considered as variables. There is no statistical significance for the effect of caffeine on sleep inertia in this experiment. In addition to the small sample size, the dose of caffeine may also be one of the reasons. The degree of improvement of caffeine’s alertness will weaken as the dose of caffeine decreases (Paula, 2006)
Although theoretically and some previous studies on caffeine have shown that caffeine can effectively improve the arousal effect, the perceived cognition of caffeine may not be universally translated into objective performance. It depends on how much caffeine is consumed by each person. People with high consumption of caffeine show faster reaction times for simple choices and report a positive subjective effect on caffeine (Attwood et al., 2007). On the other hand, moderate to low-level users will not exhibit these enhancements (Attwood et al., 2007). In this experiment, non-high-concentration caffeine was administered only once a night, and the exact sensitivity of the participants to caffeine could not be determined, which caused the effect of caffeine to be biased.
4.4 Limitation and Recommendation
The limitation of this experiment is that caffeine was used only once. Therefore, in addition to a single dose of caffeine, the results of the study should not be generalized. Future studies using multiple doses may show a response to alertness. Compared to placebo or nap, receiving higher or lower caffeine doses may show better arousal state.
Another limitation is the time and duration of the nap. Napping may be an effective countermeasure for shift work, which can increase sleep time and reduce prolonged awake time (Takahashi 2003). Nap, which is usually taken as a preventive measure before night shifts, is called preventive siesta and can provide performance benefits over the next few hours (Garbarino et al., 2004). However, the benefits of naps during night shifts to reduce fatigue are not clear (Centofanti et al. 2018). Some people think that because of the low probability of entering the deep sleep stage, short naps, especially those with less than 20 minutes of sleep or less, are less likely to be related to sleep inertia (sleepiness and awake disorders) (Hilditch et al. 2016). Compared to daytime, night shifts increase people’s tendency to sleep and are usually accompanied by longer periods of waking up (Dijk and Czeisler 1995). Since these factors affect the timing of naps (Centofanti et al., 2016) and sleep inertia (Hilditch et al., 2017) and sleep stages (Folkard and Tucker, 2003), the findings from the daytime nap time study It may be different from a nap at night. Indeed, compared to daytime naps, a study that looked at naps of less than 30 minutes at night found that during the next few hours, this nap did not benefit the next job (Centofanti et al. 2017). In addition, the sleep inertia caused by nap at night may be greater than the sleep inertia caused by the usual nap time. Future research may hope to use different sleep lengths to test whether there is a newresponse relationship between sleep and sleep inertia.
The operation check showed that there was no statistical difference between the PVT test results before the nap and the PVT test results after the nap in the two factors of the participants’ caffeine intake or not.