Journals.lww.com



Reviewer CommentsRevisions Section Title, Paragraph #, Line #Reviewer #1: Although the findings are intriguing, the discussion is lacking in terms of explaining how the study fits into the literature. Thank-you for bringing this to our attention. We have added details in the Background and Discussion sections explaining how this study fits into and builds on the existing literature.BackgroundPrevious Neuroimaging Studies of Goal-directed BehaviorParagraph 4Lines 5-9 Discussion Section Paragraph 2Lines 15-18Paragraph 3Lines 4-9Paragraph 5Lines 8-10Furthermore, the authors fail to document the behavioral findings, thus it is not known whether there are performance differences due to BMI or if performance differences might be responsible for the functional differences reported. This is an excellent point. A discussion has been added of the behavioral findings. During the training phase of the task, the behavioral and neuroimaging results were consistent in that BMI positively correlated with accuracy and with dmPFC activation. Without a direct correlation between accuracy and dmPFC, we concluded that better performance during learning was associated with dmPFC activation. Results SectionBehavioral DataLines 1-7 1. Abstract: Although 30 women were recruited, only 19 were included in the final analyses. The abstract should therefore state that 19 women participated in the study. This has been added in the abstract, stating, “One participant was omitted from the analysis due to excessive motion (>4 mm), and six were omitted due to less than 50% correct responses on the exit survey. Four participants developed claustrophobia in the scanner and were disqualified from further participation. The remaining 19 participants were included in the final analysis.” AbstractParagraph 1Lines 2-62. Methods: Were psychiatric disorders ruled out solely by self-report, or where assessments given? Given the elevated risk for depression in women nearing menopause, it would seem prudent to assess this carefully. This is an important point for clarification. All diagnoses and medications, including antidepressants and other classes that could affect brain activations or hemodynamics were self-reported, and then verified in the EHR prior to scheduling the fMRI scan. MethodsParticipant CharacteristicsParagraph 2 Lines 11-123. Methods: The neuroimaging processing path description seems incomplete. Were the data spatially smoothed? To which template where the images aligned? Although the authors used AFNI (and hence the Talairach is assumed here), it is possible the authors chose a different template space (e.g., MNI).The description of the neuroimaging processing path has been revised and clarified in the methods section. Each of these points has been addressed. MethodsData AnalysisParagraph 1Lines 2-144. Methods: Please clarify whether the combined mask were bilateral or unilateral. A bilateral mask was created. This has been added in the data analysis section. In addition, Figure 3 was added displaying the mask regions of interest. Methods SectionData AnalysisParagraph 2Line 9AppendixFigure 35. Methods: What was the cutoff for determining excessive motion?The cut-off for excessive motion was >4 mm, and this has been included in the Abstract and the first paragraph of the Methods section. Abstract MethodsLine 3Methods SectionData AnalysisParagraph 1Line 4-86. Results: Within the final dataset, there are no details regarding their age, BMI range, education, etc. The demographic information for the sample has been compiled in a summary table (see Table 1) and described in the Results section. Results Section Demographic CharacteristicsParagraph 1Lines 1-6AppendixTable 17. Results: Other than stating in the methods that behavioral performance was used to exclude six participants, there is no further discussion on behavior. Did all participants show equivalent learning? Was there a difference due to BMI? How participants performed on the task has clear implications on interpreting the functional results. We agree, and have added the behavioral findings to support interpretation of the functional results. Overall, participants learned the task and were able to accurately implement goal-directed learning. However, accuracy on the training task was correlated with BMI. During training, the behavioral findings were consistent with the MRI results in that BMI positively correlated with accuracy and with dmPFC activation. Results SectionBehavioral DataParagraph 1Lines 1-88. Results: Was BMI associated with age? If so, is it possible that age, rather than (or in addition to) BMI might drive the effects reported? BMI was not associated with age, and this lack of relationship indicates that it is not likely that the observed effects are related to age, rather than BMI. Results SectionBehavioral DataParagraph 1Lines 8-109. Discussion: Although the introduction makes a point about using a task believed to support and dissociate goal-directed and habit-based systems, there is very little about how the findings relate to these systems in the discussion. Rather, the discussion seems to be focused upon the study's limitations rather than its findings and how they relate to the literature. Are there other studies showing associations of cognitive control with BMI in the literature? What information does this study add? What does this tell us clinically? This is an excellent point. The discussion section has been revised. The study limitations have been condensed, and the discussion now focuses on how the findings relate to the literature. Differences in cognitive control and self-regulation, specifically in terms of emotion processing in overweight and obese midlife women have implications for health behavior. This study builds on the descriptive findings in the literature. Future study is needed to assess the transfer of these findings to behavioral interventions in clinic populations. Discussion Section Paragraph 2Lines 15-18Paragraph 3Lines 4-9Paragraph 5Lines 8-1010. Figures: The authors should be sure to define any abbreviations within the figure caption. Also, Figure 4 had boxed question mark symbols on my copy; this might be due to the transformation from a word processing document to PDF.Abbreviations have been spelled out completely within the figure captions. All of the tables and figures, and specifically, Figure 4 (now Figure 5) were reviewed. We do not anticipate any problems with unintended characters in the revised version, and will request editorial assistance with the transformation to PDF files to avoid any issues with this revised manuscript. Appendix All Figures and TablesReviewer #3: I have two comments that will improve this manuscript.First, the description of the Dual Framework on page 4 does not point out the importance of the ability to use both emotional processing as well as task processing. Given your results, this aspect of the Dual Framework should be added to your description of it on page 4. We would like to thank the reviewer for this very insightful comment. We agree that given our results, emotion processing needed more attention. The discussion of the Dual System Framework on page 4 has been revised to include both cognition and emotion in the description. BackgroundFramework Guiding the Study AnalysisParagraph 1Lines 3-5Second, I suggest in your discussion section that you refer again to the Dual Framework, especially pointing out the findings related to the emotion processing and its possible importance in activation for health behavior. We agree. Our findings involving the cognitive control and self-regulation regions of the brain reveal the importance of emotion processing in goal-directed behavior, and adherence and maintenance of health behaviors over time. Discussion Section Paragraph 3Lines 4-9Perhaps I missed it, but a couple of reads of the manuscript and I could not find a statement that the protocol was IRB- approved.Prior to starting this study, IRB approval was received. Signed consent was obtained from all participants prior to enrollment. These statements have been added to the manuscript. Methods SectionParticipantsParagraph 1Lines 1-3Reviewer #4: The manuscript reports several such regions with activation levels that correlate with BMI, some of which fall within hypothesized brain regions. However, the link between the outcome measures (correlations of the differences in activation levels between task states of this particular goal oriented task to BMI) to predicting the success a particular subject would have in an interventional program is not as strongly made in this manuscript as would be preferred. These are important points that were used to improve the manuscript. The link between the outcome measures and predicting the success of participants in future interventions has been added. Our findings indicate that overweight women are sensitive to goals, but may have difficulty transforming goals into action, such as health behavior changes for fitness and weight loss. These findings indicate that overweight, compared to healthy weight midlife women will require assistance with different aspects of health behavior change. The literature indicates that activations in the cognitive control and self-regulation areas of the brain identified in this study predict those individuals who will be more successful with health behavior change. These findings provide opportunities to direct more intensive interventions to those individuals who will benefit most. Discussion SectionDiscussionParagraph 3Lines 5-8Paragraph 4Lines 4-9 Paragraph 6Lines 8-10Methods:1). The authors did not explicitly state that the study was performed under IRB approved methods. Standard IRB procedures were followed. A statement has been added that IRB approval was obtained prior to starting this study. All participants enrolled provided written consent to participate. MethodsParticipantsLines 1-32). There is no description of how the subjects were recruited and particularly given the small sample size it is difficult to extrapolate the findings from this study to the general population of middle aged women. For example, were the participants recruited from a cohort seeking to improve health? A description of how subjects were recruited has been added in the Methods section. Participants were recruited from an urban academic-affiliated Internal Medicine clinic. The first author has established an ongoing relationship with the physician providers in this group to recruit study participants. For the current study, a list of patients and their contact information was generated from the electronic health record (EHR) and provided to our research assistant for recruitment. Potential participants were contacted by phone and invited to participate. The women recruited from the Internal Medicine clinic were established patients. Health promotion and disease prevention are addressed at each clinic visit; however, this was not the primary focus of all patient visits, rather management of health conditions was the main reason these women were patients in this clinic. Methods SectionParticipantsParagraph 1Lines 5-63). Task performance and brain activity during task performance might be influenced by uncontrolled variables, such as participant IQ or educational achievement. The authors should report such information if this available. We agree with this point. Educational achievement was assessed, and a table with demographic characteristics has been added describing the education levels and employment status of the sample. A brief discussion of these findings has been added in the text. IQ was not measured, primarily due to subject burden (approximately 2 hours required to conduct functional scan, and pre- and post-surveys). However, we agree that this is important consideration in fMRI studies. ResultsDemographic CharacteristicsParagraph 1Lines 1-5AppendixTable 15). Overall task accuracy was not reported nor was accuracy reported on a per condition basis. Brain activation level will change if a subject stops performing the task, even when the subject knew how to perform the task. A discussion of overall task accuracy or behavioral performance was added. We agree that it is important to consider the possibility that a subject may stop performing the task, resulting in brain activation changes. However, it is our belief that participants who are fully informed of the study protocol, have all of their questions answered, agree to participate and sign the consent form will be motivated to complete the protocol to the best of their ability. As part of standard protocol, we remain in close verbal contact with participants during the scan, and are alerted to any problems, and attempt to resolve any issues that may arise during the scanning process that could affect the results. ResultsBehavioral DataParagraph 1 6). It is not clear why "earning points" was actually motivating to some or all of the participants in the study. Was there any tangible result provided based on points scored? Differences in participant motivational level could drive the differences in brain activation. Points served as an intrinsic reward to motivate participants to perform their best and did not impact participant payment for participation in the study. Methods SectionParagraph 4Lines 13-15The data analysis methods used in this manuscript are adequate and for the most part adequately described. Restricting the parts of the brain in which the analysis was performed to specific regions of hypothesis is a reasonable way to help control for multiple comparisons. Including a figure which shows these regions, a figure similar to supplement figure 1 from de Wit et al 2009, would help the reader better visualize which brain regions were under investigation. We agree that a figure will help the readers better visualize the brain regions under investigation. Figure 3. Mask Regions (Regions of Interest) was added. Appendix Figure 3The exclusion of participants that failed to learn at least 50% of the correct responses is reasonable, although this should be (and is) noted as a limitation in the discussion. The authors are justified in excluding the subject with "excessive" motion however the authors should state what motion threshold was used as the criteria for rejection.The motion threshold and criteria for rejection has been added. Participants were excluded from analysis if they had a maximum displacement of greater than 4 mm (one participant was excluded for excessive motion). Methods SectionData AnalysisParagraph 1Lines 5-8Also, because motion during fMRI can impact measured activation levels in fMRI analysis the authors should consider computing a motion metric, for example mean RMS motion during task, for each subject. It would be helpful for the authors to confirm motion RMS does not correlate to BMI, as motion effects are known to lead to artifactual correlations in fMRI.We agree that it is important to know if motion correlates with BMI and this was not the case. The following statement has been added: Timepoints during which participants moved more than 0.3 mm within a TR (3000 ms) were censored. Motion parameters did not correlate with BMI (r = .10, p=.69). Methods SectionData AnalysisParagraph 1Lines 5-8Only 2 participants with BMI above 30 were included in the analysis. This limits the interpretation of the results from this study for seriously overweight women and should be mentioned as a limitation. This has been added as a limitation in the Discussion section. BMI of 40 kg/m2 was established as the cut-off for recruitment. This omitted severely (morbidly) obese individuals, who may differ from the healthy weight and overweight participants in some ways. Future study is needed to identify these differences. Discussion SectionParagraph 5Lines 11-14Of the 30 participants originally enrolled in the study, 11 participants were not included in the analysis for various reasons. If these 11 participants had higher BMI on average than the 19 included the authors should state this in the manuscript. This is important to consider. There were no significant differences, on average in the characteristics (including, BMI) of the 11 participants omitted and those who remained in the final analysis. Results Section Demographic CharacteristicsParagraph 1Lines 3-6Increased exclusion rates for higher BMI participants might bias the same. Also, future studies may use this finding to recruit higher rates of overweight participants to achieve a more balanced weight distribution in their final analysis. This is an excellent point. Future study is needed to assess those individuals in the higher BMI categories. This has been stated in the Discussion section, and we acknowledged a narrow BMI range is a limitation of the current study. Discussion SectionParagraph 5Lines 11-14The results from the analysis are described adequately. Tables 1 and two are helpful, although they would be more helpful if the reader knew whether the coordinates were for the centroid of the cluster or were for the voxel within the cluster with maximum activation. Figures 3 and 4 are helpful in visualizing the brain locations of the various finds. The coordinates in Tables 1 and 2 were for voxel within the cluster with maximum activation. “Data analyses focused on voxelwise correlation analyses to determine the association between BMI and brain activation (i.e., percent signal change from baseline) during goal-directed learning and implementation of goal-directed behavior.” Data Analysis SectionParagraph 2Lines 1-6Several minor issues with the organization and style of manuscript are noted:1). The abstract can be improved. The last sentence of the results of the abstract belongs in the discussion while the discussion part of the abstract describes brain regions associated with emotional processing, et al without references which regions the authors are referring to. In the abstract, the last sentence of the Results section has been moved to the Discussion section. Also in the Discussion section, the regions associated with emotion processing (anterior insula, lateral orbital frontal cortex) have been added. AbstractDiscussion Section Lines 13-15Line 62). In the fMRI data acquisition section (page 8 line 12-13) the authors report that the data were "?warped to standard space?" This statement belongs in the data analysis discussion.The fMRI data analysis information has been moved from the data acquisition section to the pre-processing information in the data analysis section. The statement “warped to standard space” has been revised. MethodsData AnalysisLines 2-53). In the fMRI data acquisition section (page 8 line 17-20) the authors report that the subjects were "...positioned in Talairach space?" It is unclear what this statement means. Did the authors position the participants in the head coil such that they had a specific range in the tilt of the AC-PC plane relative to the scanner coordinates?This statement has been clarified in the pre-processing information in the data analysis section. MethodsData AnalysisParagraph 1Lines 10-124). It would be helpful to the reader if the authors would use both the abbreviations and full names when describing the regions of correlations in tables 1 and 2, i.e. superior frontal gyrus (dlPFC) or just refer to the regions as superior frontal gyrus or dlPFC. Full names were added next to the abbreviations of the cognitive control areas in the tables. Please note that the table numbers were adjusted based on new tables added. Tables 1 and 2 are now labeled Tables 2 and 3. AppendixTables 2 and 35). The first sentence of the discussions section states: "?healthy weight to overweight women with no routine exercise program." This is helpful information regarding the study population however this information should be included in the methods section rather than introduced for the first time in the discussion section. The statement “no routine exercise program” was introduced in the Method section to describe the study population. MethodsParticipant CharacteristicsLines 4-56). Hare, et al. is referenced in the discussion section only. Literature citations should not be introduced in the discussions section. This is a good point, and Hare, et al was introduced in the Background section, and then referenced in the Discussion. Background Section Previous Neuroimaging StudiesParagraph 2Line 9The fMRI methods and analysis methods are reasonable and the task has been successfully used in previous studies. The weakness includes a relatively small sample size with few participants with BMI >30. The authors acknowledge this as a limitation. Omitting severely obese individuals due to potential differences limits inference of the study findings. A statement has been added addressing the need for future study to assess differences in healthy weight, overweight and obese BMI categories. Discussion SectionParagraph 5Lines 11-14A second weakness is that it is unclear how relevant this task is to the global questions of better understanding how to better design motivational interventions for this population. We would like to thank the reviewers who identified this issue, as we were able to strengthen the case for the relevance of this study to the design of better behavior change interventions. The goal-directed decision task elicited differences in brain activations between overweight and healthy weight women. To date, there is minimal clinical trial evidence addressing the underlying cognitive control and self-regulation areas of the brain related to health behavior. This study addresses this gap. The findings have implications for the development of cognitive interventions to improve cognitive control and emotion self-regulation. Ultimately, the goal of this research is translation of an effective intervention to clinical practice to address one of the greatest health challenges today. Background Previous Neuroimaging StudiesParagraph 4Lines 5-10DiscussionParagraph 3Lines 5-8Paragraph 4Lines 4-9EDITOR COMMENTS: In the abstract, include both the number recruited, the number disqualified from further participation, and number analyzed. The details about the sample have been clarified in the abstract, as follows: “Thirty non-diabetic, midlife (age 47-55 years) women with BMIs ranging from 18.5 to 40 kg/m2 were recruited. One participant was omitted from the analysis due to excessive motion, and six were omitted due to less than 50% correct responses on the exit survey. Four participants developed claustrophobia in the scanner resulting in disqualification from further participation. The final sample of 19 participants was included in the final analysis.”AbstractMethodsParagraph 1Lines 2-6IRBIn the paper, include information about approval of the protocol by the IRB. The statement of IRB approval has been added.MethodsParticipant CharacteristicsLines 1-3COPYRIGHT TRANSFER AGREEMENTSA completely filled out, signed copyright transfer agreement is needed from Author 2. The second author has completed and signed the copyright transfer agreement. Copyright Transfer Agreement FormsTITLE PAGE ----Supply professional working title for each author. Professional titles have been included for each author. Title Page --Supply running head of less than 50 characters (no abbreviations). This has been met with the current title. --Include a Conflict of Interest statement here.The authors report no financial or non-financial competing interests, including, political, personal, religious, ideological, academic, intellectual, commercial or any other to declare in relation to this manuscript.FIGURES ----In Figure 3, please correct the label for Part a to "% Signal Change"Thank-you for bringing this to our attention. Corrections have been made in the figure. AppendixFigure 3 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download