Understanding Balance Control in the Context of Riding a Personal Mobility Device
Round 1
Reviewer 1 Report
This paper presents itself as providing CoP-based balance patterns that could improve the design of self-balancing personal mobility devices (SPMD). However, study design choices means that the results are not meaningful when applied to this application.
The authors present SPMD as being predominantly controlled through forward (anterior) and backwards (posterior) CoP movements with turns controlled through other means. In the study, the authors assess only anterior CoP movements (not posterior) solely and in combination with left and right CoP movements. Therefore, the collected data has minimal overlap with expected SPMD CoP movements.
Additionally, other study choices by the authors further distance the data from the intended application of understanding SPMD CoP movements:
-Using a level ground force plate to measure CoP movements. While this ensures accuracy, it is inherently more stable than a SPMD. It is also consistently horizontal (level ground) whereas (at least some) SPMD will tilt.
-Mapping CoP movements to virtual movements in the screen to hit the target. For SPMDs, CoP movement is typically mapped to velocity. Therefore, there the path data and patterns collected would not represent SPMD CoP path patterns.
Therefore, the collected data does not appear to be valuable in achieving the authors' stated purpose/goals. This seems to be demonstrated in the discussion where minimal connections are made between the results and SPMD.
Additionally, the conclusions presents largely new information most of which is not supported by the study results, discussion, etc.
Some other comparatively minor issues of note:
-How was footedness determined? How was health history determined? Were these self-reports?
-Why did session 2 rely on participant report of stability in the target location instead of 1s in the target position (similar to session 1)?
-For the lost data, was it the analysis that distorted the data? Or was the raw data distorted? This is unclear.
-For variable names, subscripts are used in Table 1 but not in the body of the manuscript. Please be consistent.
-I would suggest including the full questions in an Appendix if possible. The words provided do not seem to align with standard word anchors for the 5th position on a 5 point Likert scale. Were these for the 4th position?
Author Response
Thank you for your comments. Following your comments, we've reviewed and revised our manuscript carefully.
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This manuscript aims to investigate the ability of subjects to shift weight and balance in ways that may be required when driving a SPMD. Overall I believe the paper is well written and appropriate for the journal. I only have small editorial issues I'd like to see prior to publication.
- For all subjects the weight shifting task was conducted first and the balancing task second. Can the authors comment on how the extra 'practice time' of the weight shifting task could influence the results of the balancing task? Presumably the subjects had time to get comfortable with the test methodology and setup during the first task, which may affect their performance on the second task.
-
Introduction: The first paragraph could be split into 2 (first, general PMD info, and second, SPMDs) to more easily distinguish the background information.
-
Methods: Was any IRB approval needed?
-
Methods: Fig 1 and 2 define the tasks as ‘weight-shifting task’ and ‘balance task’, but those tasks are not defined as such in the text until later. It would be helpful to use that terminology in the methods as well, and consistently throughout the paper, for clarity.
-
Methods: In Figure 3 it is difficult to distinguish between the blue and purple portions of the curve
-
Results: Two tables are listed as "Table 1", please fix the captioning and make sure they're properly referenced in the text.
- Discussion: This study has a relatively small sample size (n=21) that skews very young (~30yo). Please comment on the limitation of using such a young, small sample size. How does it compare to the average person who uses SPMDs?
Author Response
Thank you for your comments. Following your comments, we've reviewed and revised our manuscript carefully.
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
The purpose of this study was: (1) to investigate selected parameters of COP trajectories during body weight shifting from initial position to several targets located close to the stability border as a function of distance and direction of movement, (2) to assess postural steadiness while holding the COP for 3 s in the previous locations.
General comments:
The paper reads very well until the results appear. As I continued reading my interest diminished. The amount of entries in the tables is discouraging to the reader. It is difficult to understand the data in the tables and what they are compared to (e.g. the first entry in Tab. 2: 1.10 Abc (0.07)). As far as I am concerned, the notes under the tables are not helpful. I believe that explanation of the meaning of the superscripts should be clearer and more precise. Perhaps it is worth showing on one brief example how to decrypt them. If any of the results bring something more original to those commonly known, it would be worth presenting them in a graphic form as Figures.
I have a general impression that substantial contribution of the author's work to this research significantly exceeded the value of the results obtained. There seems to be enough evidence in earlier works to predict the outcome of this study. Therefore, it would be important to highlight even the subtle differences that indicate the originality of the approach presented in this Ms. For this purpose, it would be necessary to point out the differences between this experimental setup and the like of other authors. Perhaps it was mentioned somewhere and I didn't notice it. If so, it should be much more emphasized. I expressed my doubts about the novelty of this work in the Review Report Form indicating “No answer” to questions regarding Novelty and Significance of content.
I am almost sure that you are the first researchers to measure postural steadiness during 3 s trials. This undoubtedly deserves a few words in the discussion.
I would also like to learn more about how the results of these studies can help in designing SPDMs and learning to ride on them. Without this information, I have little confidence in the practical utility of these results.
It seems to me that the conclusions of this article take up too much space. For the most part, this is repeating the results. Moreover, this is not the place for the limitations that should be included in the Discussion section.
“Controllability” comes from control engineering and is certainly well defined in this field. One may have mixed feelings here whether to apply it or not to posturographic studies. On reflection, I don't mind the term.
One specific comment:
There are two Tables 1. The table numbering must be corrected.
Author Response
Thank you for your comments. Following your comments, we've reviewed and revised our manuscript carefully.
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
I would like to thank the authors for the extensive revisions made to their manuscript. They have addressed most of my concerns related to applicability of collected data to SPMD CoP movements. I found the limitations and future work section to be well written and a reasonable accounting of these concerns.
There is one remaining area of concern and one new concern.
Remaining area of concern:
From the update to the Introduction section (page 1, lines 41 to 44), it appears that SPMD control is either anterior/posterior (most) OR lateral (electric one-wheel skateboard) control. You do not provide an example where control is BOTH anterior/posterior AND lateral control. Therefore, W1 to W3, E1 to E3, and N1 to N3 seem to match with SPMD control mechanisms. However, NE1 to NE3 and NW1 to NW3 do not match with the SPMD control mechanisms described in the introduction.
If there is a SPMD device that requires both anterior/posterior AND lateral control, it needs to be added to the Introduction (or more clearly described if already listed).
If there is not a SPMD device that requires BOTH, then the diagonal (NE & NW) data points should be removed from the paper as they are not relevant to this application.
New concern:
Thank you for providing the full questions in the Appendix. My issue with the words "better", "easier", etc. in the manuscript is that they sound like word anchors from a combined Likert and Semantic Differential scale. However, you used a strongly agree to strongly disagree scale. I would suggest taken them out of quotations and adding 'respectively' to that sentence.
Additionally, some of the questions are double-barreled (asking two things in one question): Weight-shifting Q1 & Q3; Balance task Q3. Of these, Q1 is the most concerning as there is often a trade-off between speed and accuracy (see Fitt's Law) making it difficult to interpret the results.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 3 Report
In principle, the authors' responses are only partially satisfactory, but I will not ask for any further clarification. I believe that the discussed topic is original and the work does not contain methodological flaws. However, the excess of experimental data and the not entirely successful interpretation of some of the results make it difficult to understand the whole work. The same applies to the practical value of the obtained results and their suitability for designing the SPMD. I have the impression that the very effect of writing this work and its following reviews revealed some difficulties in the direct application of posturography to solve new problems undertaken by the authors. Two difficult questions remain: (1) Should more basic work be done before this article is published? (2) Is the work inspirational enough to deserve publication and stimulate the interest of other researchers who fill these gaps? I leave the answers to the editor’s discretion.
Author Response
As the comment by the reviewer, although the results may not be directly applied to a specific design problem or model of the device, we believe that it can be utilized to provide a better understanding of balance control of the rider when designing the SPMD.
