PONE-D-20-23193

Susceptibility of domain experts to color manipulation indicate a need for design principles in data visualization

PLOS ONE

Dear Dr. Christen,

Thank you for submitting your manuscript to PLOS ONE. Your manuscript has been reviewed by two external experts. Based on their recommendations, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

As you will see in the reviews below, both reviewers believe that this is an interesting study, but the also pose several concerns. In particular, both reviewers indicate that numerous important details are missing about the study methods. In addition, one of the reviewers is concerned that there may be alternative explanations of your results that should be addressed.

Please submit your revised manuscript by Nov 26 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

• A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
• A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
• An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Ronald van den Berg

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please include additional information regarding the survey or questionnaire used in the study and ensure that you have provided sufficient details that others could replicate the analyses. For instance, if you developed a questionnaire as part of this study and it is not under a copyright more restrictive than CC-BY, please include a copy, in both the original language and English, as Supporting Information.

3. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

Additional Editor Comments (if provided):

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This manuscript describes a study with three populations—one in neuroscience, the second in geovisualization, and a third of lay users---that measures subjective trust and interpretation of heatmap-style visualizations. The concept explored is intriguing and relevant: how does expertise influence colormap preference and use. Further, the paper explores an extensive range of relationships between visualization expertise and the collected measures. Overall, the study is interesting and has significant potential for challenging assertions around design knowledge and visualization use. However, there are several missing details and potential biases that may influence the results that should be addressed before the paper is ready for publication.

- Missing Details:

The description of the study data is reasonably clear. The use of real-world datasets and conventions from the tested fields is a significant strength of the work. However, there are several missing details that may help provide stronger context for interpreting the reported results.

While there are examples of how trust data collected, there is no description of how trust was defined to participants. Different conceptualizations of trust may influence how participants report trust. For example, how well do you trust yourself to use this visualization effectively seems like it may be more influenced by color choices than how well do you trust the data in this visualization to help you draw conclusions. While the latter framing is more ecologically valid (and seems closer to the likely framing based on evidence in the paper), it also leads to a bias whereby people may be willing to trust the visualization they’re using for a given task even though they know it is suboptimal (i.e., to report it as “good enough”). This is touched on in the discussion, but having more detail as to how trust was conceptualized (or interpreted) would significantly strengthen and contextualize the discussions of trust.

I appreciate the thoughtful considerations of CVDs. However, was any sort of CVD screening carried out? Many people do not know they have any sort of CVD until adulthood, making self-reporting challenging and at times inaccurate. This is a minor limitation as online CVD tests are imperfect due to a number of factors, like monitor calibration, but it is worth explicitly mentioning.

The manuscript notes that five common demographic questions were provided to all participants. However, no information about those five questions is provided. Reading the results section, it appears that these questions were used to draw conclusions about the data; however, having a summary of these questions would be helpful for understanding the study. For example, did any common questions assess lay participants’ comfort with data generally or experience with design?

For the statistical reporting, it would be helpful to have some sense of the precision of the provided means (e.g., standard deviations or confidence intervals) reported along side the means. It would also be helpful to explicitly define how variability is computed (e.g., standard deviation, variance, etc. and within subjects, over the same set of images, etc?). These details would significantly clarify the significance of the observed effects (or lack of effect in some cases).

- Potential Confounding Biases:

I appreciate the thorough discussion of potential confounding factors. The thoughtful analysis of factors around domain and design conventions is a nice addition to the paper. Further, the lack of evidence that expertise has any influence on trust and interpretation raises fascinating questions about the influence of design knowledge and thinking in visualization. However, two potential confounds that may explain at least some of the observations arise from the design of the study itself rather than from the colormaps or visualization experience, namely familiarity/risk and confirmation bias. These concerns seem problematic given the reliance on a lack of result used to drive many conclusions in the paper.

First, we expect that people are more familiar and comfortable with maps and space than with medical images. Medical images lead to both higher risk decisions and to less ability to personalize the data (see Peck et al, 2019), both of which are correlated with lower trust and confidence in an image and may provide an alternative explanation for the observed behaviors. This may lead to lower lay variability in the neuroscience images: they don’t feel they have the experience or authority to levy strong feelings or opinions and their response scales may be compressed as a consequence whereas geovis experts at least have familiarity and confidence in working with visualized data. It may also explain the higher criticality around brain death observed for neuroimaging experts over other populations.

Second, the question used in the geovis context appears like one that would be highly subject to confirmation bias. Most lay people have strong opinions on whether or not climate change is caused by human activities. As a result, confirmation bias may have a significant impact on responses to geovis questions compared to the neuroimaging questions, which focus on specific examples. This appears to play out in the high overall scores (mean of above 7 out of 8 for experts populations may indicate ceiling effects).

The discussion appears to point to that would invalidate this bias, “Our results also suggest that (c) potentially ideologically charged opinions related to the data context (in our case related to brain death as a death criterion or the degree to which human activities cause climate change) do not influence response variability, which is in line with our hypothesis.” However, it isn’t clear from the discussion of results (which primarily focus on expertise as relates to visualization) where the direct evidence of this division arises from. It would be helpful to the reader to more explicitly bridge this claim with the data itself.

While these alternative explanations do not invalidate the results, they do suggest limitations in considering the magnitude and outcomes of the provided results and for guiding future study. A revised manuscript should include an evidence-based discussion of how the general problem framing may invite specific kinds of biases and/or how those biases are mitigated by the infrastructure itself.

In summary, the research presents an interesting line of inquiry into understanding the role of expertise in assessing visualizations with various color encodings. While the study offers a number of interesting observations, there are several missing details and potential alternative explanations that should be addressed in the manuscript before it is ready for publication.

- Grammatical Issues:

“each of the tree ranking tasks”—three

“that that (b) response”—extra “that”

Reviewer #2: This paper presents a study whose focus about color usage in visualizations has the potential to offer value to research and practitioner communities. It posits interesting hypotheses that certainly are well-posed and of interest. Some aspects of the methods, results, and presentation need some attention, though - these few, probably correctable, problems with the manuscript at present make it not yet suitable for publication.

In my view, the conclusion here that certain experts exhibited a higher-than-predicted trust in the rainbow scale is in agreement with some other reported studies and with some small-sample informal tests by colleagues. Those other reports have been to the surprise of some who have had discussions about them. This work here I think adds to the evidence that we are going to have to have further, deeper analysis and care in our views of the rainbow scale. Thus, the work could both increase our current knowledge status and lead to future increases, too. That is a nice plus, counseling for a revised, improved manuscript that PLOS ONE might later accept.

Some more detailed points about the areas needing attention follow.

1. Of the 134 neuro email addresses, is there a possibility that some of these belong to a common person? That is, could some individual(s) have been represented

multiple times in the responses, or were steps taken to somehow remove responses from an individual with multiple email addresses?

2. Of the 574 geographer email addresses, is there a possibility that some of these belong to a common person? That is, could some individuals have been represented multiple times in the responses, or were steps taken to somehow remove responses from an individual with multiple email addresses? The problem seems potentially more severe here (than with the neuro people) as there were also some addresses from ICA here.

3. I wonder if some "master experts" who were contacted handed off their response to someone else. One way to test that would be to determine if any of the inputs came from email addresses different than those who were emailed (at least for the neuro people).

4. How much were the Mechanical Turk people paid? How much were the domain experts paid?

5. The term "locked-in" in "Image Creation" is not well-known; a different term is needed.

6. There is a typo in "Image Creation", Sentence 3: "graphic" -> "geographic".

7. There are two grammar errors in "Image Creation": Sentence 6, "such as, residential" should be ", such as residential"; Sentence 7, "set is" should be "set was"

8. In "Structure of the online...", it should be mentioned if any of the Mechanical Turk participants were classified as experts? In addition, if any from the neuro or geography groups were classified as lay (or if none were), that should be mentioned.

9. In "Structure of the online...", Paragraph 3, "likert" should be "Likert".

10. In "Structure of the online....", Paragraph 3, it should be made clear if the lay people answered all these questions, also.

11. In "Structure of the online....", Paragraph 4, "more a consumer or a producer" should be "more consumers or producers".

12. In "Structure of the online....", Paragraph 4, what is "of main text" referring to? Is there another Figure 1 in another component that was not submitted?

13. In "Structure of the online....", Paragraph 6, Sentence 2, this point should also be made earlier--just before Paragraph 2 of "Structure of the online..." (or perhaps only made there).

14. Who were the 27 participants that tested questionnaire usability--were any of them the same ones that took part in the other parts of the study? Did any of the 27 that tested usability drop out due to the length of the survey completion?

15. There is a sentence frag. just above Table 1 ("Specifically....").

16. Figure 3 seems misleading as it aggregates the Table 1 information without regard to context. I'm not convinced of the conclusions related to such aggregated material...domain experts, especially if they have a long history in their domains, may tend to become very domain-centric and thus perhaps more likely to use their own training as they approach another domain, which may be a serious error for the other domain. I think the Figure 3 material should be de-aggregated, with the study then considering each context for each expert set, without considering "geo" as a whole, for example. Alternately, the focus could be just neuro experts on neuro data and geo experts just on geo data.

17. In "Factors potentially," the focus suddenly shifts to "preference." I don't see where that comes from. Moreover, the Figure 3 material is suitability versus trust, NOT preference versus trust. There is a difference between something being suitable and something being preferred. This part of the paper needs some re-thought and re-work.

18. The current Figure 3 is not convincing to me that "geovis experts seem to trust the heated body color scale most" (statement just before the "Other variables" section) - the trust levels are pretty close for all. Either a statistical showing there is needed or the statement needs to be removed.

19. Table 2 has a low information-to-space ratio. The table should be reformatted in some way to be made much more compact - nearly every entry there is "ns" but the table takes up about 5% of the entire page length of the paper.

20. p. 19, "Our hypotheses" paragraph, line 5, "that that" should be "that".

21. Ref. 2 is incomplete.

Lastly, a final positive remark:

* Next-to-last paragraph last sentence makes a very good point!

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PONE-D-20-23193R1

Susceptibility of domain experts to color manipulation indicate a need for design principles in data visualization

PLOS ONE

Dear Dr. Christen,

Your revised manuscript has now been reviewed by the two original reviewers. As you will see, they were generally happy with the revision, but reviewer #1 still has a few minor points that they would like to see addressed. Therefore, we are inviting you to submit another revision. Since the points are all very small, I will most likely not send the manuscript back to the reviewers after your revision, but instead render a final decision based on my own evaluation.

Please submit your revised manuscript by Feb 19 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

• A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
• A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
• An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Ronald van den Berg

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: I thank the authors for their thoughtful and thorough revision. The clarifications significantly improve the legibility of the paper and the connection between the conclusions and data. I have a few small points that may benefit from consideration and possible revision, but these are not points that would require rereview. The work will be an interesting addition to our understanding of color use in visualization.

To clarify the concerns about ceiling effects, my assumption based on the methods description of the prior paper had been that the scale ranged from 0-8 rather than 1-9. With a cap at 9, I agree that there are not likely ceiling effects. Thanks for the clarification.

The definition of trust makes the rest of the conclusions a lot more logical and easier to follow. Given the definition, the results seem better supported compared with more conventional definitions. I wonder whether “trust” is really what this metric captures (it seems more like a measure of the validity of the conclusions), but so long as a clear definition is maintained, this is a minor point that I simply want to raise for the authors to consider as a potential opportunity for further clarity. It would similarly be useful to reiterate this definition when introducing these measures in the Structure section.

As another minor point for clarification, simply being blue-white-red does not necessarily mean the scale will be color-blind friendly (it depends on the internal variations in each of the blue and red sections). It’s worth noting if the scale has been externally confirmed to be colorblind safe (e.g., by the source document or by an external simulator). Most such encodings are, but explicit validity would be useful here.

- “In the methods part”—Methods section

- “Nine, respectively ten persons”—It seems like there’s an issue here: is it nine or ten? Also, were these participants removed or reclassified as they don’t fit the definition of a “lay” participant?

- “color vision deficiencies (no deficiency… (i.e., color blindness)”—Missing closing parentheses

- Some of the paragraphs in the discussion are quite long. For legibility, it may be worth breaking these into smaller pieces.

Reviewer #2: This manuscript is a nice improvement on the initial submission. I recommend its acceptance. I did notice one small typo to fix - p. 22, line 3, "assessing" -> "assess"

I appreciate the nice work of the authors!

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Susceptibility of domain experts to color manipulation indicate a need for design principles in data visualization

PONE-D-20-23193R2

Dear Dr. Christen,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Ronald van den Berg

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments: