Moving towards an Understanding of the Role of the Inferior Fronto-Occipital Fasciculus in Language Processing

Round 1

Reviewer 1 Report

Authors P. Eze and group investigated the role of inferior frontal occipital fasciculus (IFOF) in language processing. Here authors used behavioral tasks to test their three hypotheses 1) phonological vs. semantic processing hypothesis, 2) difficult vs. non-difficult task processing hypothesis and 3) automatic vs. non-automatic processing hypothesis to find out the functional differentiation between the dorsal and ventral components of IFOF. Through their findings authors reinforced the idea that the IFOF has semantic functionality and operates chiefly as a ventral language tract in adults.

Considering the entire antero-posterior span of IFOF, and its complex association with several other white matter tracts, such tractography-based studies have potential to provide novel insights into the IFOF’s structure and functional connectivity. Overall, the current manuscript is well-structured, elegantly designed and executed. Authors provided an elaborated introduction including the behavioral tasks and used methods, followed by adequate results and discussion. Authors have cited appropriate and recent references throughout the manuscript.

I have a few minor questions/concerns as follows.

1.       Authors should clarify if they examined both hemispheres or only the left hemisphere.

2.       Are there citations for the awareness tasks and letter probe task? A detailed exploration of their applications could benefit readers.

3.       Can the current study's tractography distinguish between the superficial and deep tracts of IFOF? Is there evidence of functional differences in language processing between these tracts?

4.       Did authors consider inclusion of motion correction parameters and stimulus duration as regressors?

Author Response

Authors P. Eze and group investigated the role of inferior frontal occipital fasciculus (IFOF) in language processing. Here authors used behavioral tasks to test their three hypotheses 1) phonological vs. semantic processing hypothesis, 2) difficult vs. non-difficult task processing hypothesis and 3) automatic vs. non-automatic processing hypothesis to find out the functional differentiation between the dorsal and ventral components of IFOF. Through their findings authors reinforced the idea that the IFOF has semantic functionality and operates chiefly as a ventral language tract in adults.

Considering the entire antero-posterior span of IFOF, and its complex association with several other white matter tracts, such tractography-based studies have potential to provide novel insights into the IFOF’s structure and functional connectivity. Overall, the current manuscript is well-structured, elegantly designed and executed. Authors provided an elaborated introduction including the behavioral tasks and used methods, followed by adequate results and discussion. Authors have cited appropriate and recent references throughout the manuscript.

I have a few minor questions/concerns as follows.

1.       Authors should clarify if they examined both hemispheres or only the left hemisphere.

Response: 

We have updated the manuscript the indicate that we only examined the left-based IFOF.

2.       Are there citations for the awareness tasks and letter probe task? A detailed exploration of their applications could benefit readers.

Response:

We have updated the manuscript to supplement the task descriptions and include references for the reader to further explore uses of the awareness tasks as well as the letter probe tasks. 

3.       Can the current study's tractography distinguish between the superficial and deep tracts of IFOF? Is there evidence of functional differences in language processing between these tracts?

Response:

The current study cannot distinguish between the superficial and deep tracts of the IFOF. Indeed, there are likely some functional differences in language processing between these tracts. Given that the superficial segments terminate primarily in the inferior frontal gyrus, and that the deep fibres are more prominently terminating in the middle frontal gyrus, it would be reasonable to make differential hypotheses about the potential role of these fibers in transmitting language-based information. We have updated out discussion to make mention of this potential avenue of future research.
“For example, an examination of the termination points of the IFOF, instead of the initiation points as conducted in this study, may yield different results. Sarubbo et al., [16] provided detailed information regarding the superficial and deep layers of the IFOF, with the former being comprised of upwardly directed fibres that terminated in the inferior frontal gyrus and the latter being comprised of posterior, middle and anterior fibres that primarily projected to the middle frontal gyrus. Given the extensive imaging work that describes differential function of the inferior and middle temporal gyri, it would be reasonable to hypothesize that the superficial and deep layers of the IFOF may correlate differentially with language function [18].”

4.       Did authors consider inclusion of motion correction parameters and stimulus duration as regressors?

Response:

Indeed, our pre-processing pipeline includes regressors for individual subject motion. However, stimulus duration was not a consideration in this study as the diffusion images were structural in nature (i.e., no stimuli were presented).

“Preprocessing steps included signal drift correction (quadratic), gibbs ringing correction (5 non-DWIs; Lamda = 100; iterations = 100; step size = 0.01), registration between diffusion images and structural images (i.e., normalization), masking to remove non-diffusion weighted signal (kernel = 9; 0.5 for non DWIs; 0.8 for DWIs), non-rigid EPI correction for distortions, and corrections for individual subject motion.”

Reviewer 2 Report

Authors presented an interesting and elegant study aiming to assess the different function of the dorsal and ventral IFOF. The work is very well presented, clear, with useful images and tables. I have a few minor suggestion that may help to improve the general quality of the manuscript. 

INTRODUCTION

- authors may consider adding a numeration to the subsections (e.g., 1.1, 1.2, etc)

- authors may consider changing the name of the last paragraph from "summary" to "objectives" or "aims of the study"

METHODS

- The specific characteristics of the population (demographics, gender, years of schooling, TOWRE) should be listed first thing in the results, not among the material and methods. Here, authors should instead list the characteristics of the desired population (a priori), with specific inclusion and exclusion criteria. 

- has the DTI (or in general the functions of the IFOF) been considered unilaterally or bilaterally? A few statements could be added to support the answer. 

- same consideration about adding a numeration to the subsections (e.g., 2.1, 2.2, etc). Consider putting the behavioral task in a separate subsection as done with DTI, etc. 

- statistical analysis paragraph: please write how data will be presented, eg, mean values (standard deviation)

RESULTS

- always the suggestion to  suggestible to add a numeration to the subsections, same thing even for the DISCUSSION

DISCUSSION

- consider among the limitations a lacking sub analysis between male and females? Same consideration for left- vs right-handed subjects?

Author Response

Authors presented an interesting and elegant study aiming to assess the different function of the dorsal and ventral IFOF. The work is very well presented, clear, with useful images and tables. I have a few minor suggestion that may help to improve the general quality of the manuscript. 

INTRODUCTION

- authors may consider adding a numeration to the subsections (e.g., 1.1, 1.2, etc)

Response:

The subsections are now numbered.

- authors may consider changing the name of the last paragraph from "summary" to "objectives" or "aims of the study"

Response:

We now label ‘summary’ as ‘Aims of the Study’

METHODS

- The specific characteristics of the population (demographics, gender, years of schooling, TOWRE) should be listed first thing in the results, not among the material and methods. Here, authors should instead list the characteristics of the desired population (a priori), with specific inclusion and exclusion criteria. 

Response:

We have made the recommended changes to the methods and results. 

- has the DTI (or in general the functions of the IFOF) been considered unilaterally or bilaterally? A few statements could be added to support the answer. 

Response:

We have updated the manuscript to be clear that we are looking specifically at the left IFOF.

- same consideration about adding a numeration to the subsections (e.g., 2.1, 2.2, etc). Consider putting the behavioral task in a separate subsection as done with DTI, etc. 

Response:

Adjusted according to the reviewer's suggestion.

- statistical analysis paragraph: please write how data will be presented, eg, mean values (standard deviation)

Response:

We have adjusted the analysis paragraph in line with the reviewers suggestions. 

“Means and standard deviations associated with the letter probe tasks (i.e., phonological, orthographic-phonological, orthographic) and accuracy rates from the awareness tasks (i.e., phonological, orthographic, morphological) were extracted (Table 2). In addition, means and standard deviations associated with FA and MD values from the dorsal IFOF and ventral IFOF were obtained (Tables 3). To address our hypotheses, a series of regression analyses were run with the FA (and MD) values of the ventral and dorsal components of the IFOF as the criterion, and the awareness tasks or letter probe tasks as the predictors. This served to control for shared variance among the behavioural tasks, which allowed us to assess the unique contributions of each process to the IFOF segments. Standardized regression coefficients, and associated p-values, are reported.”

RESULTS

- always the suggestion to  suggestible to add a numeration to the subsections, same thing even for the Response:

Adjusted.

DISCUSSION

- consider among the limitations a lacking sub analysis between male and females? Same consideration for left- vs right-handed subjects?

Response:

We have updated the limitations sections to include these additional limits to our generalizability.

Reviewer 3 Report

In this manuscript, Eze et al. explored 3 different hypotheses on the inferior frontal occipital fasciculus (IFOF) processing. Some important points regarding the methodology should be considered as follows.

The IFOF connectivity has been related to several other networks and precise anatomical subdivision could be beyond its tractography (Conner et al., 2018). In lines #271, 272and 273 the authors mentioned: “The ROIs differed between each participant, but ultimately the tracts were separated at their posterior coronal cross-sections where the dorsal tract began to trend upwards”. The authors should show the segmentation of the IFOF for each participant (N=32), to demonstrate if the segmentation is consistent. Regarding images acquisition, a slice thickness acquisition of 1.5mm can generate errors in the tractography analysis. That is another reasoning to show the IFOF segmentation per participant.

Quality of English is satisfactory. 

Author Response

In this manuscript, Eze et al. explored 3 different hypotheses on the inferior frontal occipital fasciculus (IFOF) processing. Some important points regarding the methodology should be considered as follows.

The IFOF connectivity has been related to several other networks and precise anatomical subdivision could be beyond its tractography (Conner et al., 2018). In lines #271, 272 and 273 the authors mentioned: “The ROIs differed between each participant, but ultimately the tracts were separated at their posterior coronal cross-sections where the dorsal tract began to trend upwards”. The authors should show the segmentation of the IFOF for each participant (N=32), to demonstrate if the segmentation is consistent. Regarding images acquisition, a slice thickness acquisition of 1.5mm can generate errors in the tractography analysis. That is another reasoning to show the IFOF segmentation per participant.

Response:

Thank you for the recommendation. We have updated the manuscript to include the work of Conner et al., 2018 and now include an Appendix with the segmentation images for each participant.

Round 2

Reviewer 3 Report

The authors have included segmentation images for each individual participant as previously requested.