Neurocognitive Mechanisms Underlying Internet/Smartphone Addiction: A Preliminary fMRI Study

Round 1

Reviewer 1 Report

 

In the presented study the authors investigate the underpinning of the ability to focus attention in a self driven manner put in relation to the risk for smartphone addiction. The study provides interesting behavioral effects in combination with differences in BOLD activations. The description of the procedures is frequently incomplete. This makes it hard to judge their soundness.

 

Major:

The methods description has to be completed. More details need to be added regarding the design description and the statistical evaluation. Please see below for details. It would help to give motivation for any analysis choices in advance.

The discussion should include inhibitory control as explanation for the findings, which might be a part of endogenous shifting of attention. Inhibitory control has e.g. recently be discussed in a very similar study on a different addiction, see: Weafer, J., Gorka, S.M., Dzemidzic, M. et al. Neural correlates of inhibitory control are associated with stimulant-like effects of alcohol. Neuropsychopharmacol. 46, 1442–1450 (2021). https://doi.org/10.1038/s41386-021-01014-5 Similar regions are described in this context. It seems plausible that in low-risk subjects inhibitory control increases within the time period in which performance on the task recovers from destraction.

 

List of comments:

Introduction:

* l.53 introduce RSVP prior to use.

* l.68 Please, give reference for the definition of the multiple demand network.

Methods:

* section 2.2 add reference pointing to Fig.1 A

* ll. 83f Where is the mentioned protocol published ?

* ll. 83-90. Please provide more details on the paradigm. Does it mean each 200 ms a number (or character) was shown? How did the subjects indicate they saw a letter? How did they characterize this letter? Do I understand correctly that subjects could adapt to a 2sec rhythm for the detection of the targets, because temporal spacing was always either 2sec or 4sec?

* l. 101. Was the order of oddball and non-oddball trials randomized?

* l. 121f: Was smoothing indeed performed prior to normalization?

* l. 125 Which double gamma function? The canonical hrf? Please give a reference.

* Methods: How were ROIs defined? Peak coordinates + sphere with which radius?

* Methods: Due to the significant difference in errors some kind of error regressor has to be included into the GLM model.

* l. 141 A section on statistics is entirely missing. I assume there was no significant difference on the voxel-wise level, between the groups. In the ROI base analysis, it is not clear how statistics was performed. Were statistical comparisons performed per time point? How was Bonferroni correction for multiple comparisons performed?

Results:

* 3.1. Please give the description of the distribution of the behavioral score across all participants.

* Please, show the basic GLM contrast.

* Please, give a list of the exact MNI coordinates at which the ROIs were placed. In a table for this could be enriched with the statistical values per coordinate.

* Did any of the differences scale with the score? Was a regression with the score performed?

* l. 144 omit ‘This’

* I recommend not using the same color scheme to distinguish ‘search’ from ‘oddball’ in Fig.1, and subsequently ‘low risk’ vs. ‘high risk’ in Fig. 2

* ll. 147ff. There should be at least one target before the onset of the distractor. Why wasn’t that split off (same for the final target). Why was the performance on T1+T2 not better, although half of the targets in that case did not distinguish between search and oddball? The offset of the video could be a distractor affecting T8 in the low-risk group.

* Fig. 2 please indicate measure and units for the axes.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

tomography-1793161 

This article tested whether smartphone addiction was associated with any changes to attention-related functional brain networks. The authors found frontal pole activation was reduced in people at high risk of smartphone addition during performance of an attentionally-demanding task. While the writing is generally good, there are several things which are not well-justified in the methods. Some image processing steps are unconventional and the authors don’t sufficiently justify their choices. The presentation of the results should be changed so that the reader can understand the spatial distribution of group differences. Overall the article has some merits but the presentation should be improved.  

 

Title:  

“mechanisms”. 

Abstract:  

“The present study investigated neurocognitive mechanisms underlying”. 

“showed significantly lower levels of activation”. 

“FPC” not defined. 

Main text: 

There are grammar issues (plural, tense) throughout. I have not addressed each mistake individually as there are too many. 

Introduction 

Define “RSVP”. 

There is a structural “attention network” which the authors should cite in their introduction in contrast to the known functional networks related to attention (https://pubmed.ncbi.nlm.nih.gov/31765057/). 

 

Methods 

Can the authors better justify their choice of a 100-second high pass filter, as this is typically used for resting-state fMRI. The highpass filter duration should be changed according to the design of the stimulus presentation. 

Line 130-135, the author writes in first-person, “I assigned...”, “I was able to...”. Use third person or passive voice. 

Can the authors provide any reference to support their approach to ROI isolation, to show any previous studies supporting their choice of approach? The authors should expand on why this was done, not just what was done. 

Results 

The first paragraphs of the results should be in methods, where they are describing the statistical method and processing of the behavioural outcomes.  

At the beginning of the results, please show the subject demographics and basic smartphone addition measures. For the smartphone addition measure, describe the distribution of the data (normal? skewed?) and relevant statistics e.g. mean. 

Figure 1: please label in the figure the high risk and low risk charts by adding a title to each one e.g. “b) low risk.” “c) high risk” 

Before going into the imaging results, please detail the ROIs that were extracted, possibly showing a figure for each one. Describe the position and size, and overlapping anatomy. 

Define all acronyms for ROI regions. 

If the ROI regions were extracted based on the activation patterns, why are the ROIs in Figure 2 shown as perfect spheres? Were the coordinates used to create these? This may be an issue because the spheres shown seem to overlap multiple functional areas. 

Can the authors show in a Figure the pattern of differential activation between the two groups resulting from the FSL FEAT GLM? The timecourse itself as shown is not very informative and the reader cannot infer the spatial pattern of differences between the groups. 

Discussion 

The discussion does not go into enough detail on the functiona neuroanatomical findings. Why was the identified region the strongest. What is its involvement with attentional networks? Why does smartphone addiction result in this specific change?

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

All my concerns have been properly addressed.

Reviewer 2 Report

The authors have addressed my comments