Colour Vision Deficits in Children with Amblyopia: Impact of Angular Size of Stimuli on Detection

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This is a potentially useful paper. However, the significance and scientific soundness are spoilt without clarification of the following (see attachment) [1] Why were Ishihara results not compared between amblyopic and other eye? [2] If there was a difference, was the Ishihara test presented 1st to the amblyopic eye & then to the other eye? Or were RE and LE tested in that order, which would enable subjects with amblyopic left eye to learn from their previous response to the other eye? [3] 2 conflicting statements make it unknown whether computerized test was conducted monocularly or binocularly.

The conclusions are very vague and could be more specific and significant if the above issues are addressed.

Their are errors in the Introduction and in Materials & Methods which must be addressed.

Comments for author File: Comments.pdf

Author Response

Thank you very much to the reviewer for the important comments and suggestions. The corrections in the text are highlighted in yellow. I have answered the questions below.

Comments: [1] Why were Ishihara results not compared between amblyopic and other eye? [2] If there was a difference, was the Ishihara test presented 1st to the amblyopic eye & then to the other eye? Or were RE and LE tested in that order, which would enable subjects with amblyopic left eye to learn from their previous response to the other eye? [3] 2 conflicting statements make it unknown whether computerized test was conducted monocularly or binocularly.

Response: I'm sorry. It's my fault for not describing this in more detail. 

The Ishihara test was conducted monocularly, first for the amblyopic eye and then for the better-seeing eye. This order was maintained to eliminate any potential learning bias, as testing the better-seeing eye first could provide the subject with information that might influence responses during subsequent testing of the amblyopic eye.

Initial attempts were made to test children monocularly for both eyes in the computerised test. However, this extended the test duration significantly, averaging 50 minutes for both eyes combined (approximately 25 minutes per eye). This prolonged testing period led to fatigue in young participants, resulting in inconsistent and unreliable responses. Specifically, children tended to select answers randomly toward the end of the test, likely due to reduced concentration and a desire to complete the task quickly. Consequently, we opted for binocular testing to ensure more reliable engagement and completion of the computerized task.

Monocular testing of colour vision remains critical in evaluating amblyopic eyes, as subtle interocular differences may not be detectable under binocular conditions. Such differences, while absent in the Ishihara test results in our study, could provide insights into the visual processing alterations in amblyopia. This rationale underscores the initial use of monocular Ishihara testing despite the subsequent adaptation of the computerized test to accommodate the practical limitations of testing young children.

Figure 1 indicates that the Ishihara test was performed monocularly, whereas the computerized test was conducted binocularly.

Comments: Page 4, Line 174: This statement conflicts with sentence above: "administered BInocularly''.

Response: The statement about testing "administered binocularly" applies to the computerized test to compare results between children with amblyopia and those without amblyopia. While Ishihara testing was performed monocularly to evaluate interocular differences, the computerised test was administered binocularly to assess overall performance differences across these two groups. We acknowledge this distinction and hope this explanation resolves any confusion.

I hope this explanation clarifies the rationale behind our methodology and addresses the reviewer’s concerns.

Comments: Page 4, Line 159: Was this assessed just by questionning?

Response: This aspect was assessed through a detailed questioning process during anamnesis collection. Participants were asked whether they had chronic illnesses, used any medications, had allergies, or other relevant health-related inquiries. For children who visited a neurologist for preventive purposes, we relied on the neurologist’s conclusion that they were neurologically healthy.

Comment: How did chromatic distances compare between normal and colour deficient subjects for LARGE targets?

Response: In our study, chromatic distances between normal and colour-deficient subjects (children with amblyopia) significantly differed for small targets (1°), showing impairments along the red-green and blue-yellow axes in the amblyopic group. However, these chromatic differences were not observed for larger targets (2° and 3°). This suggests chromatic discrimination deficits in amblyopic children are more pronounced with more minor stimuli and diminish as the target size increases.

Comment: How did results on Ishihara compare for normal and colour deficient subjects?

Response: The minimum acceptable visual acuity for the Ishihara test is 0.1 decimal units (equivalent to 20/200). For example, when visual acuity is 20/250 (0.08), it is expected that 1–2 errors might occur due to reduced acuity. The Ishihara test results showed no errors on plates 2–17 (transformation and vanishing plates) for participants in both the control and amblyopic groups. As visual acuity in the amblyopia group was consistently above 0.4 decimal units, performance was unaffected by the minimum acuity threshold required for the test. Therefore, no significant differences were observed between the groups in their Ishihara test results.

Comment: Line 230, Figure 3, Table 1 & figure 8 have a typo error:  "deitan"

Response: Thank you for pointing out the error; it has been corrected.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

In this manuscript, the authors report that chromatic thresholds are impaired in amblyopic children along protanopic, deuteranopic and tritanopic confusion lines in CIE (x,y) color space. This threshold impairment was found for targets subtending 1 deg of visual angle, but not for larger (2 or 3 deg) stimuli. As noted below, impaired color sensitivity has been reported previously for individuals with amblyopia. However, in the present study, observers’ chromatic thresholds were assessed during binocular viewing, which suggests that the reported impairments represent abnormal color processing using the observers’ non-amblyopic eyes. It has been observed previously that several aspects of visual functioning show abnormalities in the non-amblyopic eyes of amblyopic observers (e.g., Kandel et al., Am J Optom Physiol Opt, 1980; Meier & Giaschi, IOVS, 2017).

In their Introduction, the authors argue that amblyopia must be diagnosed at an early age, as treatment is presumed to be effective only when undertaken in early childhood. The notion that an absolute “critical period” exists for the treatment of amblyopia is a common misperception, as strong evidence to the contrary appears in a number of publications (e.g., Kupfer, Am J Ophthalmol, 1957; 1957; Levi & Li, Phil Trans Roy Soc Lond B, 2009; Kishimoto Jap J Opthalmol, 2014). The authors’ experiment appears to have been performed carefully; however, as detailed in the specific comments, below, several aspects of their Methods require elaboration and clarification. Finally, the authors note (e.g., lines 132-134) that the color perception of color deficient individuals exhibits less impairment for large compared to small chromatic stimuli. Central vision in amblyopic eyes has been suggested to be analogous to normal peripheral vision (e.g., Levi, Klein & Aitsebaomo, Vision Res, 1984; Wilson, Vision Res., 1991; Kalpadakis-Smith et al., J Vision, 2020). It may therefore be relevant that although small targets presented in the normal peripheral visual field appear desaturated, color perception for large peripheral stimuli is similar to that at the fovea (e.g., Gordon & Abramov, J Opt Soc Am, 1977).  

Below, I list a number of specific issues that the authors should address. Comments 22 and 23 identify some relevant earlier work that the authors probably should cite.

  

Specific comments.

1.     Line 16: The authors should provide a more specific description of their results in the Abstract. As a possibility, “Our findings indicate that, during binocular viewing, amblyopic children exhibit significantly impaired color discrimination along red-green and blue-yellow axes principally for 1 deg, but not for larger, chromatic stimuli.”

2.     Lines 31-32: This statement requires support from one or more references.

3.     Lines 37-38: As noted above, amblyopia has been treated successfully in adults as well as children.

4.     Lines 41-43: Amblyopia also generally produces impairments of binocular vision.

5.     Lines 65-65: Some instances of color deficiency are likely to reflect abnormal post-receptoral processing; e.g. Simunovic, Eye, 2010.

6.     Lines 77-79: Color deficient individuals also may be disqualified from some types of occupations, e.g., commercial vehicle driver, railroad engineer, class 1 airplane pilot.

7.     Line 84: To ensure clarity, replace “This condition” with “Amblyopia.”

8.     Line 92: Tritanopia or tritanomaly?  

9.     Lines 109-11: I would think the Farnsworth or Lanthony D-15 color test would be more appropriate for children, because of the smaller number of caps that need to be arranged.

10. Lines 132-135: The sentence is unclear. Do the authors mean that color-defective individuals require larger stimuli than persons with normal color vision to perceived highly saturated colors? In addition, one or more references should be provided for this statement.

11. Lines 153-160: The authors should explicitly state their criteria for amblyopia (e.g., between-eye acuity difference) and for anisometropia (e.g., between eye difference in refractive error). Also, they need to provide a definition for refractive amblyopia and specify whether they consider this a monocular or binocular condition.

12. Lines 180-181: At the viewing distance of 40 cm, what is the range of dot sizes that comprise the Ishihara plates?

13. Lines 196-219: More information is required about the computerized color test. (A) From what distance did the participants view the computer screen? (B) At this viewing distance, what was the angular subtense of the screen? (C) The linear equations that were used [to define confusion lines??] should be specified. (D) How many trials were presented for each stimulus size along each color axis? (E) What was the order of target presentations? For example, were trials along different confusion lines and for different target sizes interleaved or presented in separate blocks of trials? (F) What was the total number of trials for each participant and how long did testing take? (G) For what duration was each stimulus presented? (H) At what eccentricity from the screen center were the color stimuli presented? (I) How did participants report the direction of the colored target (up, down, right or left) from the screen center following each presentation? 

14. Figure 2: My understanding is that the targets were hexagonal patches of color (line 216) and the background consisted of an array of tessellated hexagons that were set randomly to 12, 16, 18 or 20 cd/sq.m., presumably to mask possible luminance differences between the chromatic target and background (lines 209-212). Presumably, the size of the background elements matched that of the hexagonal chromatic stimulus.  Unfortunately, neither the hexagonal shape of the targets nor the variation in the luminance of the background-elements is clear in the panels of Figure. I recommend that the authors describe the above aspects of the stimuli more clearly in the text. In addition, I suggest that they add another panel to Figure 2, in which the tessellated pattern of the target and background elements is depicted clearly, possibly using thin lines to mark the boundary of each hexagon and by enhancing the contrast differences between the adjacent achromatic background elements.

15. Lines 204-208: Stimuli of smaller angular subtense could be presented by changing the participants’ distance from the monitor screen. Based on the authors’ results, there is not much reason to test using larger targets.

16. Lines 223-224: Amblyopes have been reported previously to give normal responses on the Ishihara test, e.g., Bradley et al., IOVS, 1986.

17. Figures 4, 8 and Tables 1 and 2: “Deitan” should be “Deutan.”

18. Lines 256-258: How, specifically, do the results of this study impact the understanding of color deficiencies and their diagnosis and treatment?

19. Lines 268-271: What were the multiple dependent variables that were included in the MANOVA?

20. Line 276: I believe the authors mean “as the p value for each comparison was greater than 0.05…”.

21. Tables 1 and 2: The Tables unnecessarily present each comparison twice. There need to be only 3 pairwise comparisons presented in each cell (e.g., Protan-red) of the Table.  In addition, Table 1 can be eliminated entirely, as all p values are greater than 0.20. The authors can simply state in the text that p values for pairwise comparisons between target sizes for the control participants ranged from 0.209 to 0.999.

22. Lines 285-286: Bradley et al. (IOVS, 1986) and Mullen et al. (Vision Res, 1995) reported that chromatic contrast sensitivity for red-green gratings is reduced in amblyopic eyes. Contrast-sensitivity deficits occurred primarily for middle and not low spatial-frequency targets corresponding, respectively, to stimuli of smaller and larger angular subtense. However, Bradley et al did not find a reduction of chromatic contrast sensitivity in their observers' non-amblyopic eyes. On the other hand, Davis et al. (IOVS, 2006) reported reduced color contrast sensitivity, compared to normal observers, for 3.3 cpd red-green gratings in both eyes of amblyopic observers.

23. Lines 285-286: Mangelschots et al. (Eye, 1996) reported that tritan discrimination thresholds in both the amblyopic and non-amblyopic eyes of amblyopic observers are similar to those of normal observers. However, their chromatic stimulus was an approximately 3.5 x 3.5 deg capital letter.

24. Line 306: Replace “this difference is” with “these differences are …”.

25. Line 324: In the Results section, the authors should include comparisons between normal and amblyopic observers for the three target sizes. Also, the authors should address whether the results differ for the three different etiologies of amblyopia that were included in their participant sample.

26. Lines 329-331: What is the difference between “reduced visual acuity” and “the degree of amblyopia”? Perhaps the authors do not mean “whereas…”.

27. Lines 332-333: As indicated in comment #25, above, the analyses reported in the Results address neither the magnitude of acuity deficits nor the type of amblyopia.

28. Lines 334-335: As viewing of the chromatic stimuli was binocular, one must assume that impaired chromatic discrimination for small stimuli is a characteristic of the fellow, non-amblyopic eyes. Another, much less likely possibility, is that the impaired performance of the amblyopic observers during binocular viewing reflects reduced or absent binocular summation.

29. Lines 342-343: The authors should report the average (±SD) testing time in Methods section. I agree that 25 minutes is too long for 6 – 9 year-old children to remained engaged in a threshold task. In future work, they may be able to reduce testing time using an adaptive psychophysical procedure (e.g., ZEST, PEST, QUEST, MOBS, etc.; see, e.g., Treutwein, Vision Res, 1996; Anderson & Johnson, Vision Res, 2006.

Author Response

Thank you very much to the reviewer for the important comments and suggestions. The corrections in the text are highlighted in yellow. I have answered the questions below.

Comment: Line 16: The authors should provide a more specific description of their results in the Abstract. As a possibility, “Our findings indicate that, during binocular viewing, amblyopic children exhibit significantly impaired color discrimination along red-green and blue-yellow axes principally for 1 deg, but not for larger, chromatic stimuli.”

Response: Done! We thank the reviewer for the helpful suggestion to provide a more specific description of the results in the abstract.

Comment: Lines 31-32: This statement requires support from one or more references

Response: Done!

Comment: Lines 37-38: As noted above, amblyopia has been treated successfully in adults as well as children. 

Response: We add information.

Comment: Lines 41-43: Amblyopia also generally produces impairments of binocular vision.

Response: Thank you for your comment. Indeed, amblyopia usually results in impaired binocular vision. Although reference [13] provides relevant information, I have added an additional reference to further support this assertion. This reference is highlighted in the revised text.

Comment: Lines 65-65: Some instances of color deficiency are likely to reflect abnormal post-receptoral processing; e.g. Simunovic, Eye, 2010.

Response: Thank you for your insightful comment. We agree that some instances of colour deficiency can reflect abnormalities in post-receptoral processing, as noted in the reference by Simunovic and Sharpe et al. To address this, we have expanded the sentence to include post-receptoral contributions to colour perception defects, highlighting that while cone malfunction is a primary cause, abnormalities in neural processing pathways can also play a significant role.

This addition better reflects the multifaceted nature of colour perception defects and appropriately acknowledges the complexity of the underlying mechanisms.

Comment: Lines 77-79: Color deficient individuals also may be disqualified from some types of occupations, e.g., commercial vehicle driver, railroad engineer, class 1 airplane pilot.

Response: Thank you for the advice; we have also included this information.

Comment: Line 84: To ensure clarity, replace “This condition” with “Amblyopia.”

Response: Done

Comment: Line 92: Tritanopia or tritanomaly?  

Response: Yes, you are correct; it was my mistake. It is about tritanomaly; I have corrected it in the text.

Comment: Lines 109-11: I would think the Farnsworth or Lanthony D-15 color test would be more appropriate for children, because of the smaller number of caps that need to be arranged.

Response: Thank you for your comment. We appreciate your observation regarding the appropriateness of the Farnsworth or Lanthony D-15 test for children due to the smaller caps. However, our discussion refers to the colour vision assessment tests most commonly used in general populations rather than specifically for children. To clarify, we have adjusted the text to emphasize this broader context.

Comment: Lines 132-135: The sentence is unclear. Do the authors mean that color-defective individuals require larger stimuli than persons with normal color vision to perceived highly saturated colors? In addition, one or more references should be provided for this statement.

Response: Thank you. References to the publications have been added.

Comment: Lines 153-160: The authors should explicitly state their criteria for amblyopia (e.g., between-eye acuity difference) and for anisometropia (e.g., between eye difference in refractive error). Also, they need to provide a definition for refractive amblyopia and specify whether they consider this a monocular or binocular condition.

Response: Thank you for the valuable suggestion to include detailed participant information. This feedback has helped improve the clarity and depth of the research, and we truly appreciate the insight provided. We defined amblyopia as a condition where the difference in best-corrected visual acuity between the two eyes was greater than or equal to 2 optotype lines. Anisometropia was considered when the difference in refractive error between the two eyes was greater than or equal to 1 D.  Refractive amblyopia was considered when an uncorrected refractive error in one eye caused a significant visual acuity difference.

Lines 180-181: At the viewing distance of 40 cm, what is the range of dot sizes that comprise the Ishihara plates?

Response: The dot sizes on Ishihara plates, when viewed at 40 cm, generally range from 1 mm to 3 mm in diameter.

Comment: Figures 4, 8 and Tables 1 and 2: “Deitan” should be “Deutan.”

Response: Thank you! Done

Lines 196-219: More information is required about the computerized color test.

(A) From what distance did the participants view the computer screen?

Response: All measurements were taken with both eyes at 90 cm from the monitor.

(B) At this viewing distance, what was the angular subtense of the screen? 

Screen angular size was calculated with formula , where h – screen size (cm), d – distance till monitor (cm)  

The vertical angular size = ~13.46°

The horizontal angular size = ~19.18°

(C) The linear equations that were used [to define confusion lines??] should be specified.

Coefficients (k -slope of confusion line, b – confusion line intercept with y axis) of confusion lines were calculated suggesting that all confusion lines include monitor white point (x_w, y_w) and confusion point which is characteristic to particular confusion line (protan (x_p, y_p), deutan (x_d, y_d), tritan (x_t, y_t))

Equations for protan, deutan and tritan confusion lines

 

(D) How many trials were presented for each stimulus size along each color axis?

It depends on correctness of individuals response. However, in general particular size stimulus in particular direction had on average 20 trials.

(E) What was the order of target presentations? For example, were trials along different confusion lines and for different target sizes interleaved or presented in separate blocks of trials?

We used multiple random staircase method. Before stimuli presentation program automatically prepared 6 stimuli – protan red, protan green, deutan red, deutan green, tritan yellow, tritan blue in random order. After stimuli presentation answers are analysed and next block of stimuli is generated. Process continues until predefined number of thresholds are measured (2 thresholds). In case in particular direction predefined number of thresholds is measured then in following block stimuli in that direction are not included.

(F) What was the total number of trials for each participant and how long did testing take?

Colour vision examination in 6 directions in colour space takes approximately 8 minutes. Whole colour vision examination with 3 test stimulus sizes takes 25 minutes. Each examination includes approximately 120 trials.

(G) For what duration was each stimulus presented?

Each test stimulus was presented for 3 seconds. Participant was instructed to submit answer by pressing button on keyboard within 3 seconds after stimuli presentation.

(H) At what eccentricity from the screen center were the color stimuli presented?

Distance between chromatic stimulus and background centre was 2.22^o.

(I) How did participants report the direction of the colored target (up, down, right or left) from the screen center following each presentation? 

Participants reported coloured stimulus location by pressing corresponding arrow button on computer keyboard. Test stimulus was presented for 3 seconds, after stimulus presentation participants were given 3 second response window to submit answer.

14. Figure 2: My understanding is that the targets were hexagonal patches of color (line 216) and the background consisted of an array of tessellated hexagons that were set randomly to 12, 16, 18 or 20 cd/sq.m., presumably to mask possible luminance differences between the chromatic target and background (lines 209-212). Presumably, the size of the background elements matched that of the hexagonal chromatic stimulus.  Unfortunately, neither the hexagonal shape of the targets nor the variation in the luminance of the background-elements is clear in the panels of Figure. I recommend that the authors describe the above aspects of the stimuli more clearly in the text. In addition, I suggest that they add another panel to Figure 2, in which the tessellated pattern of the target and background elements is depicted clearly, possibly using thin lines to mark the boundary of each hexagon and by enhancing the contrast differences between the adjacent achromatic background elements.

Test stimulus background and chromatic stimulus consist of equal sided triangles. Each triangle (background of coloured stimulus triangle) has one of 5 luminance levels (equal numbers of triangles has one of 5 luminance level). Different luminance levels are necessary to mask chromatic signal (coloured triangles have same luminance levels). Images provided there are same as used in actual colour vision examinations which are carried out in mesopic conditions.

By setting the luminance values of the test stimulus and the background elements at 40, 43, 46, 49, and 52 cd/m², we ensure no contrast noise.

15. Lines 204-208: Stimuli of smaller angular subtense could be presented by changing the participants’ distance from the monitor screen. Based on the authors’ results, there is not much reason to test using larger targets.

Yes, it seems so. Still for proposes of this research we used 3 different sizes of stimuli to avoid necessity to change individuals’ distance from screen to avoid distance errors and save time to setup examination procedure.

16. Lines 223-224: Amblyopes have been reported previously to give normal responses on the Ishihara test, e.g., Bradley et al., IOVS, 1986.

Koçak-Altintas, A. G., Satana, B., Koçak, I., & Duman, S. (2000). Visual acuity and color vision deficiency in amblyopia. European journal of ophthalmology, 10(1), 77–81. 

In this publication, it is confirmed that individuals with amblyopia exhibit lower performance on the colour vision test. However, this study has a limitation: the participants with amblyopia were younger. The examination was conducted using the FM100 test, which is a reliable tool; however, younger children may struggle to comprehend the instructions for this test (it has been demonstrated that performance on this test correlates with nonverbal IQ in children). Additionally, it is well-established that colour vision improves in all children up to the age of 18. Therefore, the observed differences are not necessarily attributable to the pathology of amblyopia. Nevertheless, there is one noteworthy finding: regardless of the type of amblyopia, performance on the colour vision test did not differ. This suggests that individuals with amblyopia can be grouped together in this context.

18. Lines 256-258: How, specifically, do the results of this study impact the understanding of color deficiencies and their diagnosis and treatment?

The aim of this study was to determine whether the development of colour vision in children diagnosed with amblyopia is comparable to that of children with normal vision. The findings of this study indicate that children's performance is influenced by reduced visual acuity rather than deficiencies in colour vision.

19. Lines 268-271: What were the multiple dependent variables that were included in the MANOVA?

In this study, the multiple dependent variables included in the MANOVA were the confusion line errors measured for different colour vision channels: protan-red, protan-green, deutan-red, deutan-green, tritan-yellow, tritan-blue. These variables represent the specific errors or performance metrics assessed for each confusion line under varying angular sizes of the target stimulus in the computerized colour assessment test.

20. Line 276: I believe the authors mean “as the p value for each comparison was greater than 0.05…”.

Yes, p-value was greater than 0.05.

21. Tables 1 and 2: The Tables unnecessarily present each comparison twice. There need to be only 3 pairwise comparisons presented in each cell (e.g., Protan-red) of the Table.  In addition, Table 1 can be eliminated entirely, as all p values are greater than 0.20. The authors can simply state in the text that p values for pairwise comparisons between target sizes for the control participants ranged from 0.209 to 0.999.

Thank you for your thoughtful feedback! We have removed Table 1 as suggested, incorporating the relevant information into the text, where we noted that p-values for pairwise comparisons between target sizes for the control participants ranged from 0.209 to 0.999. However, we would like to retain Table 2, as we believe that presenting the information in this format offers greater clarity and detail for readers. Of course, we are open to further suggestions if you feel this approach could be improved. Thank you again for your valuable input!

22. Lines 285-286: Bradley et al. (IOVS, 1986) and Mullen et al. (Vision Res, 1995) reported that chromatic contrast sensitivity for red-green gratings is reduced in amblyopic eyes. Contrast-sensitivity deficits occurred primarily for middle and not low spatial-frequency targets corresponding, respectively, to stimuli of smaller and larger angular subtense. However, Bradley et al did not find a reduction of chromatic contrast sensitivity in their observers' non-amblyopic eyes. On the other hand, Davis et al. (IOVS, 2006) reported reduced color contrast sensitivity, compared to normal observers, for 3.3 cpd red-green gratings in both eyes of amblyopic observers.

Thank you for your comment. We appreciate your perspective on the perceptual mechanisms involved. Other authors' observations are undoubtedly valid. However, colour vision assessments typically determine the minimum stimulus saturation required for detection. This study's results suggest that children with amblyopia have underdeveloped perceptual mechanisms necessary for detecting edges and boundaries.

23. Lines 285-286: Mangelschots et al. (Eye, 1996) reported that tritan discrimination thresholds in both the amblyopic and non-amblyopic eyes of amblyopic observers are similar to those of normal observers. However, their chromatic stimulus was an approximately 3.5 x 3.5 deg capital letter.

In amblyopia cases, colour vision deficiencies can occur, but they are not uniformly distributed across all colour discrimination axes. Research indicates that red-green deficiencies (protan and deutan) are more commonly observed than blue-yellow (tritan) deficiencies. Therefore, while both types of colour vision deficiencies can be present in individuals with amblyopia, red-green deficiencies are more frequently observed compared to blue-yellow deficiencies. In general.

Kocak-Altintas et al. (Eur J Ophthalmol, 2000) reported that the error scores of all axes were lower in the control group than the amblyopic groups, but the differences within amblyopic groups were not significant. For example, in Rajavi's et al. (J Ophth Vis Res, 2015) paper on the correlation between amblyopia and colour disturbance, participants were examined only for red-green deficits, but not for blue-yellow deficits.

24. Line 306: Replace “this difference is” with “these differences are …”.

Thank you, done!

26. Lines 329-331: What is the difference between “reduced visual acuity” and “the degree of amblyopia”? Perhaps the authors do not mean “whereas…”.

Thank you for your insightful comment. To clarify, “reduced visual acuity” refers to the measurable decrease in the ability to resolve fine details, which can be quantified using standard visual acuity tests. On the other hand, “the degree of amblyopia” reflects the severity of amblyopia as a clinical condition, encompassing factors beyond visual acuity alone, such as contrast sensitivity deficits or binocular function impairment.

28. Lines 334-335: As viewing of the chromatic stimuli was binocular, one must assume that impaired chromatic discrimination for small stimuli is a characteristic of the fellow, non-amblyopic eyes. Another, much less likely possibility, is that the impaired performance of the amblyopic observers during binocular viewing reflects reduced or absent binocular summation.Top of Form

Thank you for your thoughtful comment. When the examination is conducted binocularly, involving both the amblyopic and the non-amblyopic eye, it is indeed possible that the results may not differ significantly in terms of either colour vision or visual acuity, as the non-amblyopic eye could compensate for the deficits of the amblyopic eye. As we mentioned earlier, the assessment was performed binocularly due to the prolonged testing time, which would have been challenging for young children to tolerate. We acknowledge that future studies should address this limitation by optimizing the speed of examination and stimulus presentation to enable the computerized colour vision test to be performed monocularly. Thank you again for highlighting this important consideration.

29. Lines 342-343: The authors should report the average (±SD) testing time in Methods section. I agree that 25 minutes is too long for 6 – 9 year-old children to remained engaged in a threshold task. In future work, they may be able to reduce testing time using an adaptive psychophysical procedure (e.g., ZEST, PEST, QUEST, MOBS, etc.; see, e.g., Treutwein, Vision Res, 1996; Anderson & Johnson, Vision Res, 2006.

Thank you for your question. We hope this explanation clarifies our methodology. We utilized the random staircase method, where each individual staircase was governed by a 1-up/1-down rule. In total, two threshold values were determined for each direction. The assessment for each threshold began with the maximum saturation level.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The revised MS is an improvement on the original, particularly regarding description of the method.

My comment [1] Why were Ishihara results not compared between amblyopic and other eye? has not been addressed and perhaps would be best as part of a new scientific paper. Also, your study could be extended to include adults.

Please revise abstract: particularly for red-green and blue-yellow axes.

This should read: both for red-green and blue-yellow axes

 

Author Response

Comments 1: Why were Ishihara results not compared between amblyopic and other eye? has not been addressed and perhaps would be best as part of a new scientific paper. Also, your study could be extended to include adults.

Response: Thank you for your question. The revised version addresses this issue in lines 196-201.

The Ishihara test was conducted monocularly, first for the amblyopic eye and then for the better-seeing eye. This order was maintained to eliminate any potential learning bias, as testing the better-seeing eye first could provide the subject with information that might influence responses during subsequent testing of the amblyopic eye. All participants in the experiment responded correctly without making a single error. 

We appreciate your thoughtful comment about extending the study to include adults. While the current study focuses on pediatric populations, future research could explore how colour vision deficits manifest in adult amblyopic patients, contributing further to understanding this condition across different age groups.

Comments 2: Please revise abstract: particularly for red-green and blue-yellow axes. This should read: both for red-green and blue-yellow axes.

Response: Thank you, that's been fixed.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

      The authors provided much of the requested experimental detail in their reply to my initial review. Unfortunately, they failed to include all of this information in their revised manuscript leaving it unavailable to potential readers of their study. To allow their study to be fully understood and potentially reproduced, the authors need to include details about the participants’ viewing distance, the angular screen dimensions, the average numbers of trials, the staircase strategy and trial-presentation order, the testing time, the target duration, and the target eccentricity in the vicinity of lines 231 – 242 of the revised manuscript text.

      In addition, in lines 286 – 291, the authors should state explicitly that the MANOVA assessed chromatic-difference thresholds along the six color-confusion axes as a function of the angular stimulus size

      The authors’ conclusion in lines 347-349 that color deficits are unrelated to the severity or type of amblyopia currently is not supported by the reported results. (Actually, this outcome makes sense if the size-specific color deficits reflect abnormalities in the participants non-amblyopic eyes – see below.) Should the authors wish to retain this statement, they would need to include analyses in their Results section that specifically consider the etiology and magnitude of acuity loss of their amblyopic participants.

      As noted in my initial review, because amblyopes are characterized by abnormal binocularity, suppression, and non-foveal viewing (in the participants with strabismus), the size-specific color deficits reported by the authors almost certainly reflect processing in the non-amblyopic, rather than the amblyopic, eyes of the participants in the authors study. This should be stated explicitly in the Conclusions, e.g., in the vicinity of lines 353 – 354. Indeed, it is possible that more profound color deficits would be found if testing were performed monocularly using the participants’ amblyopic eyes.

      Finally, after the changes in the text made by the authors, the sentence in lines 72 – 73 no longer makes sense. In addition, I think the references cited at the end of the sentence should be [33-34].

Author Response

Comments 1: ....the authors need to include details about the participants’ viewing distance, the angular screen dimensions, the average numbers of trials, the staircase strategy and trial-presentation order, the testing time, the target duration, and the target eccentricity in the vicinity of lines 231 – 242 of the revised manuscript text.

Response: Thank you for this valuable feedback. We understand the importance of providing comprehensive experimental details to ensure our study is fully understood and reproducible. In response to your comment, we have revised the manuscript to include all requested information.

Comments 2: In addition, in lines 286 – 291, the authors should state explicitly that the MANOVA assessed chromatic-difference thresholds along the six color-confusion axes as a function of the angular stimulus size.

Response: Thank you for pointing this out. We agree that the role of the MANOVA in assessing chromatic-difference thresholds needs to be explicitly stated in the manuscript for clarity. We have revised the text to include this information.

Comments 3: ...Should the authors wish to retain this statement, they would need to include analyses in their Results section that specifically consider the etiology and magnitude of acuity loss of their amblyopic participants.

Response: Thank you for your valuable feedback. We appreciate your suggestion and agree that further clarification is necessary. To address your concern, we have expanded the Results section to include a more detailed analysis of the aetiology and magnitude of acuity loss in our amblyopic participants. We have also examined whether visual acuity correlates with colour deficits. This additional analysis will help clarify the relationship between acuity and colour vision deficits in the amblyopic group.

Comments 4: ...This should be stated explicitly in the Conclusions, e.g., in the vicinity of lines 353 – 354. 

Response: Thank you for your insightful comment. We agree that the observed size-specific colour deficits are most likely due to processing in the non-amblyopic eyes, given the characteristics of amblyopia, such as abnormal binocularity and suppression in the amblyopic eye. We have revised the Discussion section to mention this point.

Comments 5: Finally, after the changes in the text made by the authors, the sentence in lines 72 – 73 no longer makes sense. In addition, I think the references cited at the end of the sentence should be [33-34].

Response: Thank you, done!

Author Response File: Author Response.pdf