Sleep Loss, Daytime Sleepiness, and Neurobehavioral Performance among Adolescents: A Field Study

Round 1

Reviewer 1 Report

I have read the article  with great interest and it does not seem very

In table 1b there is an error because it is not possible to bed at 25 hours.
It is rare that the method is section 4 and the logical thing is to put it after the introduction.

It would be interesting to know what day of the week the tests apply (I think they do not follow any pattern) because the lack of sleep accumulates throughout the week.

Please clarify: when calculating the sleep parameters they have used all the nights or only those that were followed by the tests?

The reason for changes of the effect of the circadian rhythm can be discussed because probably related that on weekends the morning tests are be performed later, at a time that is more appropriate for their chronotype.

In fact, in some tests it is seen that at night they equalize 

Author Response

Reviwer1

Comments and Suggestions for Authors

1. In table 1b there is an error because it is not possible to bed at 25 hours.

Thank you, we changed in the table.

It is rare that the method is section 4 and the logical thing is to put it after the introduction.

2. The method part placed in the end of the paper is according to the journal format

3. It would be interesting to know what day of the week the tests apply (I think they do not follow any pattern) because the lack of sleep accumulates throughout the week.

We added explanation in the method page 10, line 346-347:

 "one school day (arbitrary chosen by adolescents) and one non-school day (usually on Saturday").

In addition, we added limitation (page 8; line 276-280):

Although participants were objectively monitored by actigraph through all school days, they were instructed to choose arbitrarily the day they performed the neurobehavioral assessments. 

4. Please clarify: when calculating the sleep parameters they have used all the nights or only those that were followed by the tests?

Our results included comparison between averaged weekdays and weekends scores (see table 1a,1b). Further more the main hypothesis regarded the differences in neurobehavioral performance for weekday vs. weekend (Table 2).

We doubt direct associations between sleep measurements and neurobehavioral measurements.

5. The reason for changes of the effect of the circadian rhythm can be discussed because probably related that on weekends the morning tests are be performed later, at a time that is more appropriate for their chronotype.

Thank you, its interesting question:

See page 7 lines 198-202: "Our study supports previous findings [e.g., 22, 23] showing that the performance on PVT measures was worse after sleep loss, particularly on weekdays compared to weekends. Furthermore, the results pattern corresponds with the differences emerge from the forced wake time on weekdays as opposed to weekend, in which the wake time and the test time is aligned with their chronotype".

Reviewer 2 Report

Summary

This paper from Tzischinsky Orna and Barel Efrat describes a field study about the relationship between sleep and cognitive performance among adolescent students. The authors first found that students in this study suffered from sleep loss and sleepiness during weekdays compared to weekends, using both objective and subjective sleep measurements. They further showed that students behaved relatively poor in PVT and DSST tests, suggesting a link between sleep and cognitive development. In particular, the affection of sleep loss on cognitive tasks is time dependent: on weekdays, students exhibited longer response time and less accurate rates in the PVT and DSST tests in the morning compared to other time of the day.

Overall, the research topic of this paper is timely and important. The experimental design in this paper is sensible, and the data is clearly presented. The paper is well-organized, although with a few grammar and typo mistakes.  The findings regarding sleep involved in brain development and cognitive performance of adolescent students will be of broad interest to many readers. However, there are a few issues to be addressed. I think this manuscript would be appropriate for publication if the following major and minor issues can be resolved.     

Major comments

1. Comparing data from such natural setting as shown in this paper has advantages (as discussed in paper) but may not be well controlled. Although the authors showed that sleep is necessary for cognitive performance, is the sleep loss the only (or major) reason that leads to poor neurobehavioral performance? Are there any other potential factors (for example mood, stress) also involved in? I suggest adding another group of control: participants in this study who follow the same schedule of weekday at weekends and run the same tests to rule out whether sleep is the major factor for cognitive performance.
2. In table 1, I noticed that students slept later at weekends compared to weekdays (table 1a: bedtime 23:47 vs 01:47). As a result, they woke up later. Which factor is more critical for neurobehavioral performance, sleep duration or timing of sleep?

Minor comments

1. Line 18: replace “following” to “followed by”.
2. Line 19: replace “assessment” to “assessment than”.
3. Line 42: replace “control” to “controlled”.
4. Line 68: replace “demonstrated” to “have demonstrated”.
5. Line 105: what does the “t()” indicate here?
6. In all the tables, a lot of units were missing (for example, what’s the unit of sleep duration in table 1?), also it would be easier for readers if the p values could be added into the tables.
7. Line119: what do M, N, E stand for? I suggest that the authors could add an abbreviation table at the end of the manuscript.
8. Line134: replace “om” to “on”.
9. Line 148: data for PVT false starts is missing.
10. Line 198: needs reference.

Author Response

Reviwer2

Comments and Suggestions for Authors

Summary

This paper from Tzischinsky Orna and Barel Efrat describes a field study about the relationship between sleep and cognitive performance among adolescent students. The authors first found that students in this study suffered from sleep loss and sleepiness during weekdays compared to weekends, using both objective and subjective sleep measurements.

They further showed that students behaved relatively poor in PVT and DSST tests, suggesting a link between sleep and cognitive development. In particular, the affection of sleep loss on cognitive tasks is time dependent: on weekdays, students exhibited longer response time and less accurate rates in the PVT and DSST tests in the morning compared to other time of the day.

Overall, the research topic of this paper is timely and important. The experimental design in this paper is sensible, and the data is clearly presented. The paper is well-organized, although with a few grammar and typo mistakes.  The findings regarding sleep involved in brain development and cognitive performance of adolescent students will be of broad interest to many readers. However, there are a few issues to be addressed. I think this manuscript would be appropriate for publication if the following major and minor issues can be resolved.     

Major comments

1. Comparing data from such natural setting as shown in this paper has advantages (as discussed in paper) but may not be well controlled. Although the authors showed that sleep is necessary for cognitive performance, is the sleep loss the only (or major) reason that leads to poor neurobehavioral performance?

Are there any other potential factors (for example mood, stress) also involved in? I suggest adding another group of control: participants in this study who follow the same schedule of weekday at weekends and run the same tests to rule out whether sleep is the major factor for cognitive performance.

Thank you for your comments and the suggestion for including control group, unfortunately under the current circumstances (Covid-19) the adolescent have unstable school schedule.

Regarding to your comments about "other potential factors" we addressed it in the limitation part.

Page 8 lines 250-251

….. it is not so experimental in nature and several intervening variables could compromise our findings (anxiety, stress etc.).

2. In table 1, I noticed that students slept later at weekends compared to weekdays (table 1a: bedtime 23:47 vs 01:47). As a result, they woke up later.

Which factor is more critical for neurobehavioral performance, sleep duration or timing of sleep?

Is it a very critical question and few studies are dealing with this question, we added new reference (40) on this topic.

See page 7 lines: 203-206

Recent study has presented [40] that some sleep characteristics, such as sleep duration, sleep onset, sleep insufficiency, and rate of oversleeping, could be a significant influence on adolescents’ academic performance. However, we cannot conclude which has more influence on cognitive performance: late sleep onset or short sleep duration.

 Minor comments

1. Line 18: replace “following” to “followed by”.
2. Line 19: replace “assessment” to “assessment than”.
3. Line 42: replace “control” to “controlled”.
4. Line 68: replace “demonstrated” to “have demonstrated”.
5. Line 105: what does the “t()” indicate here?
6. In all the tables, a lot of units were missing (for example, what’s the unit of sleep duration in table 1?), also it would be easier for readers if the p values could be added into the tables.

We added t values and  p values in table 1.

7. Line119: what do M, N, E stand for? I suggest that the authors could add an abbreviation table at the end of the manuscript

WE added the explanation in the first time we used these abbreviation, .

8. Line134: replace “om” to “on”.

9. Line 148: data for PVT false starts is missing. Thank you.

10. Line 198: needs reference. (22,23)

Thank you for your comments. We did the changes according to your suggestions