Separate regression models were computed for each ROI in each vocal contrast. We therefore examined a sub-group comprised of the 16 children with ASD who showed above chance identification accuracy and performed whole-brain TD vs. Results for all analyses were similar to those described previously for the entire ASD group. Specifically, whole-brain TD vs.
ASD group differences and social communication covariate results were evident in similar brain regions as those described for the larger ASD group. The overall effect size measured across all brain clusters identified in the TD vs. ASD Group Analyses is 0. The overall effect size measured across all brain clusters identified in the Social Communication Covariate Analysis is 0. In the interests of transparency, eLife includes the editorial decision letter, peer reviews, and accompanying author responses.
The Reviewing Editor's assessment is that minor issues remain unresolved. We appreciate that you have highlighted the sample size as a limitation, provided documentary support for the sample size, your deeper individual testing, and noted the need for future studies that are better able to capture the heterogeneity of ASD. Nonetheless, there is increased sensitivity to the out-of-sample reproducibility issues inherent in studies of this size, even given the challenges of clinical research of this note. While we support publication of the paper, this limitation has been raised in the peer review process.
Thank you for submitting your article "Impaired voice processing in reward and salience circuits predicts social communication in children with autism" for consideration by eLife. Your article has been reviewed by three peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by Michael Frank as the Senior Editor. The following individual involved in review of your submission has agreed to reveal her identity: Coralie Chevallier Reviewer 1.
If you have any questions, please do not hesitate to contact us. This is a very interesting and nicely presented study of cortical responses to mothers' versus strangers' voices on people with Autistic Spectrum Disorders. All reviewers felt the study was well designed in its stimuli and, in principle, impressed by the range of uni- and multivariate analyses undertaken. However, all three reviewers raise substantial concerns regarding the small sample size: these reflect the ability of the study to provide adequate cover of the heterogeneity of ASD, the possibility that some effects may be inflated particularly the regressions , the under-estimate of variance in the cross-validation tests and, most importantly, the likely challenges of reproducing the principle findings in the absence of an independent test data set.
Each analysis, on its own, may be valid, but the general feeling is that the authors are doing too much with this limited data set. In the absence of acquiring further data, there seems to be little additional work that the authors can do to address this fundamental concern. Given this is part of the eLife "trial", there does not seem to be a deep enough concern for the paper to be withdrawn from further consideration. However, if published, the important concerns of the reviewers will need to be published alongside the paper, pending the nature of the authors' responses.
Overall, I very much enjoyed reading the paper: clearly written, timely, and novel. The use of recordings that are specific to each participant is an important contribution. I am not a neuroscientist so I am not qualified to evaluate the quality of the neuroimaging work. I think the authors should tone down their claim about the limitations of behavioural studies to understand and tease apart different cognitive mechanisms.
Many behavioural studies use ingenious paradigms to tease apart various mechanistic possibilities. Recently, the use of computational modeling methods applied to behavioural data has also been very fruitful in this respect different model parameter reflect different mechanisms. Antonia Hamilton has published many papers demonstrating the validity of behavioural tools to measure social reward responsiveness. I have also published several papers on that same topic including one using signal detection theory, in PLOS One.
In these papers, no abstract judgment is required. Rather, participants' behaviours are thought to reflect underlying social motivation or social reward responsiveness. TD group? I would like to know a bit more the difference in accuracy detection between the groups: why did the ASD group perform below the TD group? The authors point out that 5 children performed below chance in identifying their own mother's voice. Is there evidence that this indicates a true deficit or is poor performance linked to other factors poor hearing? Where are these five children on the regression? Do they drive the effect?
Another naive question on the same question. The authors report that "fMRI activation profiles in children with ASD were not related to mother's voice identification accuracy". So I was left wondering what these activations reflect if they are not sensitive to the fact that 5 out of 21 children did not identify their mother, does it mean that these activations are picking up on something that is much more domain general than anything that might have to do with "mother's voice" specifically?
Is there a way to statistically correct all analyses for accuracy levels? ADOS scores are not meant to be used as a continuous severity scores unless they are transformed Gotham, K. Standardizing ADOS scores for a measure of severity in autism spectrum disorders.
Journal of autism and developmental disorders, 39 5 , Or published guidelines? Can the authors specify the exclusion decision criteria? Is there a group difference in average motion? In a behavioural study, a sample size of 21 is now considered too small. I realise that the change in standard is recent and I have published many papers using small sample sizes myself. However, I do think that we need to accelerate change, especially for conditions that are notoriously heterogenous, such as ASD. And especially when one is interested in explaining interindividual differences by using correlations.
Reviewers who specialise in neuroimaging methods may want to comment on this specific point but the paper would definitely be stronger is the sample size exceeded or at least matched the current average in cognitive neuroscience field ie 30, see Poldrack et al. Alternatively, the authors should report observed power.
This is I think most important for the regression part of the paper. If power is too low, I would recommend moving this part of the paper to the SM and rewriting the main text accordingly. Button, K. Power failure: why small sample size undermines the reliability of neuroscience. Nature Reviews Neuroscience, 14 5 , Poldrack, Russell A. To sum up, I think the paper raises an interesting question and would be even better if reproducibility concerns were addressed by increasing sample size.
This paper reports the uni- and multivariate analysis of fMRI acquired from children with ASD while listening to their mothers' or strangers' voice. The authors find quite striking correlations between social disability rating scales and both brain activity and functional connectivity. The experimental stimuli, writing, analysis and cohort characterization are all of a very high standard. There are some limitations in the actual design that limit the nature of the inferences drawn. In a revised manuscript, these should be addressed through a subtle reframing and possibly additional discussion points.
Therefore, in the absence of an explicit reward, the authors should be cautious of falling foul of reverse inference Poldrack, when framing the findings in terms of "reward circuits". I guess listening to a voice, particularly a mother's voice, is implicitly rewarding, but caution should be taken in drawing to direct an inference in regards to dysfunctional reward processes see Discussion section. Indeed, the whole positioning of functional neuroimaging studies as being more informative than behavioural tests should be mindful of the limitations of inferring disrupted functional circuits in disorders unless you are actually probing that function with an appropriate task.
But maternal voice is more than a simple cue but incorporates many other processes, including basic parental attachment and dyadic reciprocation. While this might be sufficient for reproducibility, it is inadequate here for two reasons: First, as eLife is a general not a clinical journal, more depth and context is required for the broader readership.
Second, if the diagnosis relies heavily on social deficits which I suspect it does , is it then surprising that the strength of between group differences covaries so markedly within the ASD with social deficit scores. If so, is there an implicit circularity between the identification of these regions Figure 2 and the very strong correlations against ADOS scores?
If not, what other effects may be driving these? Such strong correlations are bound to draw attention and I think the authors should therefore pay careful attention to this issue. N-fold cross-validation does always not provide strong control here Varoquaux, : Do the authors undertake a feature reduction step, such as a LASSO? Is there some logical circularity between identifying features with a group contrast, then using these same features in a between group classifier?
Also, I would avoid using the term "prediction" for contemporaneous variables, even when cross-validation is undertaken. Personally, I would prefer to have seen the author use a more dynamic, model-driven method of effective connectivity, using something like DCM to provide a deeper mechanistic insight into the changes in activation and information flow I also am not sure I buy into the choice of ROI's that do not show the group effects. However, given the classification success, the authors' approach seems entirely reasonable.
However, please do specify the nature of the gPPI model in the text and supplementary material what are the nodes, inputs and modulators; is it possible to represent this graphically? Poldrack, R. Can cognitive processes be inferred from neuroimaging data? Trends in cognitive sciences, 10 2 , Varoquaux, G. Cross-validation failure: small sample sizes lead to large error bars. Are these very impaired children also see point 4 above. In this study, Abrams and colleagues examined voice processing in children average age 10 years old with and without autism. The specifically were interested in reward and salience circuitry and relate the study back to predictions made by the social motivation theory of autism.
The authors applied group-level analyses, brain-behavior correlation analyses, as well as PPI connectivity analyses and applied multivariate classifier and regression analyses applied to the connectivity data. There are several issues which the authors may want to address. This sample size is likely not large enough to cover substantial heterogeneity that exists across the population of individuals with autism diagnosis. Thus, questions about generalizability arise. Can future studies replicate these findings? For multivariate classifiers and regressions, these too produce inflated and over-optimistic levels of predictions with smaller sample size e.
Statistical power would likely be increased for ROI analyses, and one can cite more unbiased estimates of effect size. Whole-brain analyses can show us is where the likely effects might be, and this is helpful when we don't have strong anatomical hypotheses. But here the authors do have strong anatomical hypotheses, yet they choose an analysis approach that is not congruent with that and penalizes them in terms of statistical power and doesn't allow for estimation of unbiased effect sizes.
Future studies that may try to replicate this study will need to know what the effect sizes are likely to be. Meta-analyses ideally need unbiased estimates of effect size. However, all we have to go off of here are the authors figures showing whole-brain maps, that likely just tell us where some of the largest effects may likely be. The reported correlations are negative, and yet the scatterplot shows what looks like a positive correlation. All this confusion is due to the inverted y-axes. The authors should correct the plots to avoid this confusion.
However, if their manipulations were powerful enough to create a distinction between a stimulus that was heavily socially rewarding e. Figure 2B? In other words the main contrast of interest that might have been most relevant to the social motivation theory produces no group differences in activation in areas like the ventral striatum. Because this contrast doesn't really pan out the way the theory predicts, doesn't this cast doubt on the social motivation theory, or couldn't it be that some of the contrasts in this study can be better explained by some other kind of reverse inference than the social motivation theory?
This seems critical if the authors want to make strong reverse inferences back to the social motivation theory. Thank you for submitting your revised article "Impaired voice processing in reward and salience circuits predicts social communication in children with autism" for consideration by eLife.
Your article has been reviewed by two peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Michael Frank as the Senior Editor. The reviewers find that, on the whole, you have responded to the technical concerns raised on the prior round. Reviewer 3 remains concerned that the sample size is not sufficiently large to capture the heterogeneity of ASD.
The reviewer, BRE and senior editor have corresponded and agree that this is an important limitation regarding the broader generalizability of the study. In other words, even though you have adequately controlled for in-sample error, generalization beyond the study sample is limited by the size of the cohort.
We feel that a stronger statement that acknowledges this should be added to the manuscript. Given the advanced stage of the manuscript, we recognize that acquiring further subjects is highly unlikely to be feasible although we will note, in the published reviews, that this remained of concern at the conclusion of the reviewing process. The paper will be promptly be re-assessed by the Reviewing Editor upon resubmission. I would like to thank the authors for their extremely thorough response and for all the additional analyses they performed in response to my comments.
The authors have suitably addressed all of my comments and incorporated a limitations section to mention the modest sample size. Thank you again for this detailed and careful response. In this reviewer's opinion, the author's responses to previous comments largely dismissed many of the key issues in the comments, and the arguments brought to counter those comments were not at all convincing.
The authors need to more properly consider the issues and at the very least comment on them as potential drawbacks, limitations, caveats, etc. For example, none of the author's responses are adequate for addressing the main problem of small sample size. Discussion about symptom domains as 'constraining' heterogeneity are not relevant to this aspect of the issue.
If another study comes up with similar paradigm, but also small n, how likely will that study replicate the findings of the current study? It depends. If the new study samples a different strata of the autism population, then it may not replicate. Studies with larger sample size are more generalizable because they can cover a better range of the population and are not as susceptible to sampling biases that are more pronounced with small samples. In another example, the authors claim that within subject, the multiple runs somehow can counteract the issue of small sample size.
More data within subject can have an impact on statistical power. The estimates for each subject are more precise with more data within-subject. However, generalizing between-subjects or between studies is still an issue, for the reasons stated above. With regard to the comment about ROI analysis, an analysis with 16 regions is surely more highly powered than a whole-brain analysis with 20, voxels to correct for.
For anatomically constrained hypotheses which the authors have , the comment still stands that it is strange to take a whole-brain approach as if the authors had a less strong idea about the anatomical hypotheses. Whole-brain analyses are lower in statistical power and have the ability to over-inflate effect sizes, and the authors are merely capitalizing on those things with small sample size. The scatterplots must be plotted in the correct way, and not reversed, as this will mainly confuse readers.
However, all three reviewers raise substantial concerns regarding the small sample size: these reflect the ability of the study to provide adequate cover of the heterogeneity of ASD,. We thank the editors and reviewers for this comment. The ability of our sample to adequate cover heterogeneity in ASD is an important consideration for this study and all studies investigating the perceptual, cognitive, and biological bases of ASD. Given this widely-held view, sufficiently addressing heterogeneity in ASD represents a significant research challenge, particularly for pediatric brain imaging research in which gathering high-quality data in a modest sample such as that reported here can take many years.
Here we have addressed this challenge with a comprehensive approach which both: a constrains heterogeneity in ASD to a critical diagnostic symptom domain, social communication abilities, and b explores heterogeneity within this symptom domain by identifying neural features that covary as a function of social communication abilities. The rationale for constraining heterogeneity in ASD to a symptom domain is that all individuals with ASD necessarily have pronounced impairments in diagnostic domains, and these domains have considerably less heterogeneity in individuals with ASD relative to other behavioral and cognitive domains.
Language function and IQ in these individuals varies widely, from very high levels, which are commensurate with neurotypical individuals, through very low levels of language and cognitive function We argue that constraining analyses to an ASD diagnostic domain is an important approach for keeping a focus on critical areas that are central to the core deficits in the disorder, thereby providing important insight to clinicians, researchers, parents, and educators regarding the neurobiological bases of these core deficits.
For example, the Autism Diagnostic Observation Schedule ADOS-2 is the gold-standard diagnostic instrument for ASD 4 , and social communication subscores on the ADOS range from 0 to 22, with a 0 indicating no social communication deficit, a 7 indicating a more mild deficit, and a 22 indicating the most severe deficit. The high-functioning children included in our sample had a range of social communication abilities as measured with the ADOS between 7 and Children with ASD who have scored higher than 16 tend to have multiple comorbid symptoms, and low cognitive function, thus confounding neuroscientific interpretation of findings.
Importantly, understanding the link between social communication symptom severity and social brain processing is a critical question for understanding the neurobiological basis of autism, and is a question that has not been explored in previous studies of human voice processing in children with ASD. The reason this is an important question is that the severity of social communication deficits can play a crucial role in autism models. For example, the social motivation theory of ASD posits that impairments in representing the reward value of human vocal sounds impedes individuals with ASD from engaging with these stimuli and contributes to social interaction difficulties 5, 6.
Given the causal link proposed in this model, a key prediction of the social motivation theory is that children with more severe deficits associated with social reward will have greater social communication deficits compared to those children with less severe social reward deficits. Therefore, by examining heterogeneity within the social communication symptom domain, we are able to test an important prediction of an influential ASD model and understand the neural features that may contribute to this heterogeneity.
In our study, we have employed this approach and constrained heterogeneity in children with ASD by focusing on social communication abilities. We then examine heterogeneity within the social communication symptom domain and show that social communication abilities explain significant variance in activity and connectivity of social reward and salience brain regions during human voice processing. The possibility that some effects may be inflated particularly the regressions ,. We thank the reviewer for this comment. As we stated in the initial submission see Brain Activity Levels and Prediction of Social Function subsection of the Materials and methods , the goal for performing the regression analyses was to examine the robustness and reliability of brain activity levels for predicting SC scores.
The rationale for this analysis is that significant whole-brain covariate analysis can be the result of outliers in the data or other spurious effects 7. By extracting signal levels from ROIs, plotting the distributions, and performing additional regression analyses on these data, our goal was to show that the whole-brain covariate analysis was not driven by outlying data and was robust to confirmatory cross-validation regression analysis. We have made every effort to clarify this point in the revised manuscript. We acknowledge that proper estimation of variance is critical for robust and reliable cross-validation results.
An important consideration in the context of the current analyses is that the permutation test used to assess statistical significance of support vector classification SVC and regression SVR results undergoes the same cross-validation procedure as that used in the original cross-validation.
Therefore, the estimate of variance in the original cross-validation analysis is comparable to the null distribution from permutation. Importantly, this analytic approach accounts for any underestimate of variance in a modest sample size. The likely challenges of reproducing the principle findings in the absence of a independent test data set. Reproducibility in neuroimaging research represents a major challenge for the study of pediatric clinical populations, such as children with ASD, whose data is considerably more difficult to acquire compared to typically-developing children 8.
Importantly, resting-state fMRI and structural MRI studies are unable to address specific questions related to social information processing in ASD, such as biologically-salient voice processing, which is critical for understanding the brain bases of social dysfunction in affected children. Furthermore, our sample size is larger than or comparable to 13 other task-fMRI studies in children with ASD published since This is an important consideration given that sample size is not the only determinant for the replicability of fMRI task data. While previous studies have shown that increasing the sample size can improve the replicability of results 17 , an important consideration is that the replicability of task fMRI data is not solely contingent on a large sample size but also depends on the amount of individual-level sampling.
A recent report examining this question showed that modest sample sizes, comparable to those described in our submitted manuscript, yield highly replicable results with only four runs of task data with a similar number of trials per run as our study In the current study, we have used rigorous standards for inclusion that are, to the best of our knowledge, a first for autism and neurodevelopmental neuroimaging research.
Specifically, we required that each child participant had at least 7 functional imaging runs of our event-related fMRI task that met our strict head movement criteria. To our knowledge, these rigorous within-subject criteria, which have been shown to be a critical factor for producing replicable task fMRI findings 18 , are a first for autism and neurodevelopmental neuroimaging research. It seems that this confusion stems from the fact that reviewer 2 misunderstood the analysis and thought that the ADOS covariate analysis used ROIs that were identified from the TD vs.
ASD group analysis. This was not the case. Rather, the TD vs. By performing separate whole-brain analyses, we have avoided concerns related to collinearity in the ADOS covariate analysis. We have made every effort to clarify this point in the revised Materials and methods and Results sections.
Finally, the strong effect sizes we found are likely the result of our use of at least 7 functional imaging runs in each participant. We thank the reviewer for raising this point. We thank the reviewer for highlighting this important point, and we agree with the reviewer that we should highlight this important line of research. An important consideration is that the behavioral work that has been done in this area by Drs. Chevalier 19, 20 , Hamilton 21, 22 , and others 23 has been in the visual domain, in which researchers use eye-tracking or pupillary responses as a tool for studying implicit reward.
Studying implicit reward in the context of biologically-salient voice processing presents additional challenges given that, to our knowledge, validated behavioral methods for ascertaining whether children are directing their neural resources to a specific vocal source, thereby measuring auditory social reward responsiveness, have not yet been developed. Therefore, we argue that using brain imaging methods represents a critical approach in the context of implicitly rewarding voice processing. We have clarified these important points in the revised Introduction. To test this hypothesis, we have performed additional analyses to examine whether there are any identifying clinical, cognitive, or neural or characteristics regarding these 5 children with low identification accuracy.
We next examined whether there were any differences for children with low vs. We next examined neural response profiles for the 5 children with low vs. We next examined low vs. One possible explanation is that all children performed the mother's voice identification task immediately after the fMRI scan, which took approximately 2 to 2. The reason children performed this task after the fMRI scan rather than before it i. We have included these additional analyses and information in the revised Appendix.
This is an important question, and one that we have spent much time considering. One additional domain-general possibility, which was discussed in section 1. However, we do not have additional data to further probe this possibility. Separate regression models were computed for each ROI in each contrast. We have included this result in the revised Appendix. We thank the reviewer for this astute point. Given that all of our participants were administered module 3 of the ADOS, which is stated in the Participants subsection of the Materials and methods, standardization of ADOS scores here is unnecessary.
Our study incorporated stringent a priori criteria for exclusion based on head motion during all fMRI runs, which is consistent with our previous work 30 and is described in the Materials and methods subsection entitled Movement criteria for inclusion in fMRI analysis. This sections states:. We have further clarified this point in the revisedParticipants subsection of the revised Materials and methods. We thank the reviewer for this comment and their concern for the ability of our study to address heterogeneity in ASD.
Here we have addressed this challenge with a comprehensive approach which both: a constrains heterogeneity in ASD to a critical diagnostic symptom domain, social communication abilities, and b explores heterogeneity within this symptom domain i. Finally, it is important to note that our inclusion criteria required each participant to have at least 7 functional imaging runs of our event-related fMRI task that met our strict head movement criteria.
Requiring a large number of functional runs for each participant is an important approach for increasing statistical power, an issue further elaborated below. This approach has been used primarily in basic human vision experiments, which often use small samples of participants with intense scanning in each participant , and, to our knowledge, is a first for neuroimaging studies in autism, and in children.
Specifically, we required that each child participant had at least 7 functional imaging runs of our event-related fMRI task 4 min each that met our strict head movement criteria. To provide a better estimate of effect size, we used the originally computed t -scores from the whole-brain GLM analysis. Instead of examining the peak, we averaged the t -scores in each cluster to compute effect sizes.
To estimate effect sizes for the TD vs. ASD group comparisons i. To provide estimates of effect sizes within regions identified in the ASD Social Communication covariate analysis i. We report an overall effect size of 0. ASD group analysis Figure 2 and an overall effect size of 0. In a revised ms, these should be addressed through a subtle reframing and possibly additional discussion points. We thank the reviewer for this comment and acknowledge that, as with all naturalistic and biologically salient stimuli, we cannot know for certain whether aberrant activation and connectivity patterns measured in nucleus accumbens NAc and ventromedial prefrontal cortex vmPFC in children with ASD reflect reward processing in these regions.
However, previous empirical evidence and theory, which we have highlighted in our manuscript, provide a strong theoretical foundation for considering vocal stimuli in the context of reward, even in the absence of an explicit reward task. Second, behavioral evidence shows that children with ASD often fail to be attracted to human vocal sounds, even when they are able to engage with other sounds in their environment, which suggests that they may not find these sounds rewarding 39, Third, an influential theory posits that social reward processing, such as weak reward attribution to vocal communication, may substantially contribute to pronounced social deficits in children with ASD 5, 6.
Given these converging results and theory, we believe that considering reward in the context of diminished activity and connectivity in response to vocal sounds in brain regions that are closely associated with reward processing i. Importantly, we have made every effort to temper statements to avoid issues with reverse inference in the revised manuscript. We apologize for any confusion here.
Our goal for these statements was to highlight neuroimaging research as an additional tool to test important hypotheses regarding voice and reward processing in children with ASD. We have clarified these statements in the revised manuscript. While parametrically manipulating attention to human voices in children with ASD is an important question, unfortunately this was beyond the scope of the current work. Importantly, we hypothesize that a lack of interest in human voice processing is a fundamental prediction of the social motivation theory 5 : humans engage and pay attention to rewarding stimuli in their environment, and a consistent lack of attention to a category of stimuli strongly suggests that these stimuli may not be rewarding to an individual 5.
It is hoped that future studies will test this prediction and examine the relative contributions of attention and reward for human voices in children with ASD. We thank the reviewer for requesting clarification here. The relationship we had intended to highlight here was that of the link between social communication abilities in children with ASD and brain activation in social and reward brain areas during voice processing.
We have clarified this important point in the revised manuscript. Autism spectrum disorder is characterized by pronounced social communication deficits, particularly in the areas of social-emotional reciprocity and verbal and non-verbal communication, and repetitive and restricted behaviors RRB and interests 1. Nevertheless, these children are generally characterized as having moderate-to-severe communication impairments, especially in the area of reciprocal conversation 1.
We have included additional information regarding the defining characteristics of ASD in the revised Participants section. It seems that this confusion stems from the fact that the reviewer misunderstood the analyses and thought that the ADOS covariate analysis used ROIs that were identified from the TD vs.
The use of separate whole-brain analyses in this context avoids circularity mentioned by the reviewer. We thank the reviewer for highlighting this important point. We thank the reviewer for inquiring about this point. As we stated in Materials and methods section of the initial submission:. We believe that this is not a circular approach for two reasons: 1 the group contrast used to identify ROIs for the functional connectivity analysis was from a previous study 41 in an independent sample of children with ASD relative to the participants in the current study, and 2 the brain imaging approach and analysis employed in that previous study was intrinsic functional connectivity using resting-state data and seed-based analyses, which provides complementary information regarding brain network organization relative to the task-based data and gPPI analysis used in the current study.
The importance of this approach was highlighted in the subsection entitled A voice-related brain network approach for understanding social information processing in autism in the Discussion section of the initial submission. We would like to bring special attention to the final sentence in this paragraph quoted below which highlights the fact that the networks approach used in the current study bridges a critical gap between findings from intrinsic connectivity analyses 41 and task-based social information processing that is fundamental to social communication deficits in children with ASD.
Findings bridge a critical gap between the integrity of the intrinsic architecture of the voice-processing network in children with ASD and network signatures of aberrant social information processing in these individuals. Consistent with many papers in the fMRI literature , the use of prediction to describe cross-validated results is a widely used convention and therefore we would prefer to use this nomenclature in our study.
Personally, I would prefer to have seen the author use a more dynamic, model-driven method of effective connectivity, using something like DCM to provide a deeper mechanistic insight into the changes in activation and information flow. The reason for this long TR is that it allowed the auditory stimuli to be presented in silent periods between volume acquisitions. Moreover, serious concerns have been raised in the literature regarding DCM 46 especially when estimating causal influences with a large set of nodes as we did in the present study.
The gPPI models used in our study are not faced with estimability issues. I also am not sure I buy into the choice of ROI's that do not show the group effects. The reason that ROIs were not selected based on a group effect is that this approach could be considered circular, and our goal was to provide a more generalizable set of results compared to a network defined based on nodes identified using the current sample of children and task conditions.
The use of an a priori network is it is an established method of network identification that preempts task and sample-related biases in region-of-interest ROI selection The gPPI model is summarized in Equation 1 below:. Briefly, in each participant, the regional timeseries from a seed ROI is deconvolved to uncover quasi-neuronal activity and then multiplied with the task design waveform for each task condition to form condition-specific gPPI interaction terms.
In the equation above, ROItarget and ROIseed are the time series in the two brain regions, and taskwaveformcontains three columns corresponding to each task condition. We have included this description in the revised Functional Connectivity Analysis subsection of the Materials and methods. We have added this information to the Participants subsection of the revised Materials and methods. No, there were no between-group differences in head motion parameters. Mean and standard deviation for maximum head motion for TD and ASD groups are included in Table 1, and we have further clarified this point in the revised Participants subsection of the revised Materials and methods.
Network identification in brain imaging studies presents several challenges. One important consideration is that selecting ROIs based on a GLM contrast from a sample of participants may be considered circular when those ROIs will then be used in a subsequent functional connectivity analysis on that same contrast and sample of participants. Therefore, by using ROIs from a previous study of the intrinsic architecture of voice-selective cortex in a separate sample of children with ASD 41 , we have preempted biases associated with both task contrast and participant sample that would have emerged had we used task-based GLM results from the current sample to generate ROIs.
Finally, we have provided an explanation for why we have not performed additional causal analyses in section 2. As discussed in response to similar reviewer comments, here we have addressed this challenge with a comprehensive approach which both: a constrains heterogeneity in ASD to a critical diagnostic symptom domain, social communication abilities, and b explores heterogeneity within this symptom domain i. An important consideration is that sample size is not the only determinant for the replicability of fMRI task data. A recent report examining this question showed that modest sample sizes, comparable to those described in our submitted manuscript, yield highly replicable results with only four runs of task data We thank the reviewer for drawing our attention to these important concepts.
From one perspective, identifying more subtle effects comes with its own challenges: weak effects are often viewed with caution irrespective of statistical power. While we agree that a larger sample size would have been preferred, we argue that the rigorous within-subject criteria implemented for the current study, which is described in detail in response to the points raised in 3.
It should also be noted that we also identified significant between-group Figure 4 and brain-behavior relationships Figure 5 using an a priori brain network identified from an independent sample of children with ASD We again thank the reviewer for this comment regarding sample size, and have considered the important report by Woo et al. While we agree that a larger sample size would have been preferable, we note that the current study avoided analysis procedures identified by Woo et al.
Crucially, as discussed previously, we required that each child participant had at least 7 functional imaging runs of our event-related fMRI task 4 min each that met our strict head movement criteria. The acquisition of high quality brain imaging data, including at least 7 functional runs from each child, is extremely difficult in the context of pediatric clinical populations 8 and is unprecedented in studies of autism. We thank the reviewer for this remark. First, intrinsic connectivity results from this previous study identified 16 ROIs, and including all of these ROIs would have required FDR correction, which would have reduced the ability to detect subtle effects.
Furthermore, had we limited the number of ROIs included in the analysis to reduce the multiple comparisons issue, it may have appeared that we were selecting and reducing the ROIs after results are known i. While the use of ROIs based on anatomical hypotheses would have been justified for GLM analyses, we do not believe that the use of whole-brain analysis in the context of the current study presents a methodological weakness.
The plots in Figure 3 were meant to aid visualization of regional brain responses, and were not intended to reflect effect size. In general, there is no good solution to this problem. A contributing factor is the more stringent GLM activation thresholds that are published in more recent fMRI papers: effect sizes increase when higher voxel-wise t -scores are used to compute them — there is no good solution to this problem. In the revised manuscript, these effect sizes have replaced the R and P values previously listed in Figure 3 scatterplots, and are also provided in Appendix 1—tables 1 and 2.
To provide additional guidance for future studies that seek to replicate our findings, we report an overall effect size of 0. We inverted the y-axes in the initial submission since greater values of ADOS Social Communication scores are associated with more severe deficits, and often readers prefer to see reduced abilities at the bottom of the y-axis rather than at the top. If the reviewer still feels that we should make the change, we would be happy to. Our interpretation of the results is that there was a high degree of variance within the ASD group with regards to neural responses to vocal stimuli, and that variance within the ASD group was comparable to or exceeded the between-group variance, prohibiting significant between-group differences.
Specifically, these latter results showed that the greater the social function in the children with ASD, the greater the activity in regions associated with reward processing, including NAc and vmPFC. Results provide new evidence for the social motivation theory 5 by suggesting that the degree of social reward impairment varies as a function of social abilities, supporting a link between being tuned into the social world and being rewarded by the social world. Furthermore, to our knowledge, there are no validated behavioral measures for auditory processing analogous to eye-tracking 19, 21, 22 that might have been used for children in this age range to infer reward processing for these vocal sounds.
We have added a paragraph that identifies limitations of the current study to the revised Discussion section. Association AP. Diagnostic and statistical manual of mental disorders: DSM Washington, D. Lang Cogn Process. Tager-Flusberg H, R. Paul, and C. Language and Communication in Autism.
In: F. Volkmar RP, and A. Klin, editor. Autism diagnostic observation schedule, second edition. The social motivation theory of autism. Trends Cogn Sci. Neural correlates of face and object recognition in young children with autism spectrum disorder, developmental delay, and typical development. Child Dev. Decoding continuous variables from neuroimaging data: basic and clinical applications. Front Neurosci. Hum Brain Mapp. Impaired downregulation of visual cortex during auditory processing is associated with autism symptomatology in children and adolescents with autism spectrum disorder.
Autism Res. Action simulation and mirroring in children with autism spectrum disorders. Behav Brain Res. Young adolescents with autism show abnormal joint attention network: A gaze contingent fMRI study. Neuroimage Clin. The Action Imitation network and motor imitation in children and adolescents with autism. Three stages of carcinoma in situ have been previously distinguished.
Invasion of epithelial cells into deeper tissue layers, anywhere within the larynx but not involving the true vocal folds. The most common sites are the ventricular folds, epiglottis, and pyriform sinuses. Lesions are usually identified later in the course of development as their increased size begins to interfere with voice, swallowing, or breathing.
Hoarseness may result when the lesion interferes with airflow or when the mass interferes with the movement of one or both vocal folds. The malignancy may extend deep into the larynx, limiting vocal fold mobility. Swallowing difficulties, breathing difficulties, weight loss, fetid breath, pain in the throat, and ear pain may occur. The degree of vocal impairment is related to the size of the lesion and its effects on respiration and movement of the vocal folds.
Voice quality, therefore, may range from unimpaired to severely impaired depending on the size and location of the lesion. Voice lab test results vary with the same factors. Minimum and average phonatory airflows may be increased if lesions limit vocal fold closure or decreased if lesion limits airway size. Typically a disease of adulthood. The lesion begins as a small abnormality of the epithelium and if left untreated will invade the deeper structures of the larynx. Without medical or surgical treatment, malignant laryngeal lesions are life threatening.
Behavioral therapy is often required following medical or surgical treatment, to assist the rehabilitation of vocal and swallowing functions. As a laryngeal malignancy progresses, dysphagia and respiratory interference may result. When left untreated or advanced, lesions will most likely spread to regional lymph nodes and may further metastasize to other organs of the body. Untreated malignant laryngeal lesions lead to death. Some data indicate a role of previous radiotherapy to the neck, dietary deficiencies, and exposure to environmental carcinogenic agents in some cases.
Laryngeal lesions glottic and extraglottic are reported to be between 2 and 5 percent of all malignancies. The ratio of men to women who develop malignant laryngeal lesions glottic and extraglottic is 4 to 5 : 1, with the ratio narrowing steadily for the past 20 years. Chondrosarcoma, osteosarcoma, salivary gland carcinoma, mucoepidermoid carcinoma, adenocystic carcinoma, and lymphoma. Laryngeal anatomic subsites comprised within the supraglottic staging system are false vocal folds, arytenoids, suprahyoid epiglottis, infrahyoid epiglottis, and aryepiglottic folds. The subglottic region is not subdivided.
Supraglottic T1: Tumor limited to one site of the supraglottis with normal vocal fold mobility. T2: Tumor invades mucosa of more than one adjacent subsite of the supraglottis or glottis or region outside the supraglottis, without fixation of the larynx.
T3: Tumor limited to larynx with vocal fold fixation or invades any of the following: post-cricoid area, pre-epiglottic tissues, or base of tongue. Subglottis: T1: Tumor limited to the subglottis. T2: Tumor extends to the vocal folds with normal or impaired vocal fold mobility. T4: Tumor invades through the cricoid or thyroid cartilage or extends to other tissues beyond the larynx, for example the trachea, soft tissues of the neck, thyroid, or esophagus.
Both supraglottis and subglottis: NX: Regional lymph nodes cannot be assessed. Benign Lesions of the Lamina Propria This category represents one of the most common, yet most controversial areas in voice disorders. Debate arises about definitions of focal, benign lesions of the vocal folds. In the past, many benign lesions were grouped together and collectively referred to as nodules or polyps.
Careful examination of vocal fold lesions, especially with stroboscopy, has led to the improved ability to differentiate characteristics of lesions. Also, recent histologic studies provide further biological insights. Thus, there is increased ability to differentiate subtypes of benign lesions including vocal fold nodules, fibrous masses, polyps, cysts, and reactive lesions of the superficial lamina propria.
However, understanding of the lesions is complicated by inconsistent terminologies across clinical centers, and lack of consensus regarding a priori definitions of lesions. A further complicating factor in interpreting results from histological studies has been that investigations have targeted different substrates across lesions, using different experimental approaches.
Therefore, comparisons across lesions must be approached with caution. What is known with certainty is that all of these lesions involve an imbalance of the constituent proteins of the extracellular matrix. Current expert opinion indicates that benign lesions of the vocal folds are generally thought to have the following characteristics: a Vocal fold nodules are bilateral, symmetrical lesions. Underlying tissue properties involve a continuum of histologic changes—not to be understood as a clinical progression. Nodules involve a minimal disruption of the mucosal wave on stroboscopy.
In nodules, the basement membrane zone is thickened, having increased fibronectin. Anchoring fibers in the basement membrane zone are also decreased and disorganized. Gaps are present between cells of the basement membrane zone and the epithelium. Compared to nodules, a fibrous mass has the predominant feature of increased deposition of unorganized collagen in the superficial lamina propria or near the vocal ligament.
Therefore, the mucosal wave is typically reduced. There is also thickening of the basement membrane zone, but this feature is less prominent than in nodules. Mucous retention cysts are believed to be related to obstruction of a mucous-producing gland. The stroboscopic features of a vocal fold cyst include reduction of mucosal wave amplitude, especially for a vocal fold cyst near the vocal ligament. The lesions are often exophytic in nature and may be associated with an enlarged blood vessel or hemorrhage. In contrast to vocal nodules, polyps are not typically associated with a thickening of the basement membrane.
The stroboscopic features of a vocal fold polyp are variable, dependent on the size and morphology of the lesion. Typically, there is only minor if any disruption of the mucosal wave vibratory activity. Differentiation of lesions of the vocal fold is important for communication among patient and voice care providers and often has treatment decision and prognostic implications. Presence of localized benign masses, bilateral, although not always symmetrical, located within the lamina propria typically at the midpoint of the membranous vocal folds or, stated another way, at the junction of the anterior third and posterior two thirds of the full length of the vocal fold.
These lesions often or typically interfere with the vibratory behavior of the vocal folds, creating increased aperiodicity that contributes to abnormal voice production. Incomplete vocal fold 38 O However, small nodules that compress on closure often result in complete closure of the vocal folds without the more typical hourglass configuration. Vocal fatigue is common. Problems with the upper vocal pitch range are often the first symptom noticed by the patient. The quality of the voice worsens with use, particularly if there is extensive, loud, pressed voice use.
Subglottic pressures are typically increased. Fundamental frequency and intensity are often normal in adults, however frequency and intensity ranges may be reduced. In children, the onset of benign mass lesions is often in the pre-school years, primarily among boys, who have a tendency to use their voices aggressively and strenuously, e. In young adults, the most frequent age at onset is the teen years for girls who are heavy voice users e.
In adults, lesions occur most frequently in women between the ages of 20 to 40 years. Benign lesions of the lamina propria and their symptoms may be a fairly acute reaction to an intense period of phonotraumatic behavior. Conversely, lesions and symptoms may have a gradual course of development.
Lesions and symptoms from acute phonotrauma may resolve with reduced voice use. However, in other cases, lesions and symptoms persist and may require treatment. Vocal fold lesions do not ordinarily disturb any physical functions other than phonation. Patients may complain of pain lateral to the larynx that may result from increased effort used in producing voice, or sometimes patients report a feeling of a lump in the Vocal fatigue may be severe enough to impact occupational and social functioning and general quality of life.
Professional voice users can be significantly affected even with minor vocal fold lesions. Children who are teased about the abnormal sound of their voices or who have difficulty being heard or understood in group settings or in noisy environments may experience a negative impact on school performance. Perpendicular impact stress to the vocal folds is thought to be the primary causal factor for the development of benign vocal fold lesions of the lamina propria. Specific factors leading to high impact stress are high subglottic pressure, vocal fold hyperadduction and vocal fold elongation high pitch within a given register.
Some literature indicates that an extroverted personality with reduced abilities to inhibit may contribute to phonotraumatic behaviors in some individuals. Behaviorally, activities involving extremes of voice use, such as cheerleading, untrained singing, speaking above noise, and unamplified teaching and aerobics instructing tend to involve high inter-vocal fold impact stress and thus are predisposing behaviors for nodules. The presence of dehydration, respiratory infections, or inflammatory factors e. No reliable data are available concerning the frequency of benign lesions of the lamina propria.
Similar percentages appear to apply to adults. In pre-pubertal children, benign vocal fold lesions affecting the lamina propria appear to occur more frequently in boys than in girls. In adult age, women are substantially more often affected. Vocal fold polyps, cysts, fibrous mass lesions, reactive vocal fold lesions, and epithelial lesions of the vocal folds recurrent respiratory papillomatosis and leukoplakia. Severity criteria. Fibrous Mass—Sub-Epithelial Essential features. This lesion is composed of fibrous material within the sub-epithelial space, typically occurring at the midpoint of the membranous folds.
Intra-operatively, the lesion is usually found to be amorphous in nature and often has thin extensions both anteriorly and posteriorly. A fibrous mass has mild to moderate negative impact on the mucosal wave vibratory activity as seen during stroboscopy. Fibrous mass sub-epithelial can be unilateral or bilateral and is often associated with a contra-lateral reactive lesion. This lesion interferes with the vibratory behavior of the vocal folds, creating increased aperiodicity that leads to a rough, hoarse voice. Incomplete vocal fold closure pattern on stroboscopy is typically present hourglass configuration and may result in breathiness.
Changes in quality or control of the upper pitch range may be the initial symptom. Such effect can be related to the size of the lesion s , the degree of lamina propria Signs and symptoms may include varying degrees of hoarseness roughness and breathiness , loss of high notes, vocal fatigue, and restricted dynamic range. For a professional voice user, any amount of phonatory impairment can be disabling.
Aerodynamic characteristics: Lack of vocal fold closure will result in increased average phonatory flows. Additional mass may increase phonatory subglottic pressures. Acoustic characteristics: Effects on fundamental frequency and intensity vary, depending on lesion characteristics. Increased stiffness may increase fundamental frequency. Possible at any age. Fibrous mass development is commonly associated with repeated, chronic phonotrauma, possibly due to associated hemorrhage.
Vocal fold lesions do not ordinarily disturb any function other than phonation. However, the presence of a fibrous mass may increase the risk of vocal fold hemorrhage. In addition, patients may complain of pain lateral to the larynx that may result from increased effort used in producing voice. In other cases, patients report a feeling of a lump in the throat.
Vocal fatigue may be severe enough to impact occupational and social functioning and quality of life. Repeated incidence of perpendicular impact stress to the vocal folds is thought to be the primary causal factor for the development of benign vocal fold lesions of the lamina propria. Specific factors leading to high impact stress are high subglottic pressure, 42 O Screaming has a high likelihood of being phonotraumatic.
Behaviorally, activities involving extremes of voice use, such as cheerleading, untrained singing, speaking above noise, and unamplified teaching and aerobics instructing tend to involve high inter-vocal fold impact stress and thus are predisposing behaviors for developing a fibrous mass. The presence of dehydration or inflammatory agents allergies, tobacco and alcohol use, drugs, dry environments, laryngopharyngeal reflux may be predisposing or aggravating factors.
Vocal fold polyp, cyst, nodules, reactive vocal fold lesion, vocal fold scar, sulcus vocalis, epithelial lesions of the vocal folds recurrent respiratory papillomatosis and keratosis , and laryngitis. Fibrous Mass—Ligament Essential features. This lesion is composed of fibrous material near the vocal ligament, typically occurring at the midpoint of the membranous fold.
Intra-operatively, the lesion is usually found to be amorphous in nature and often with thin extensions both anteriorly and posteriorly. A fibrous mass has moderate to severe negative impact on the mucosal wave vibratory activity as seen during stroboscopy. Fibrous mass—ligament can be unilateral or bilateral and is often associated with a contra-lateral reactive lesion. Due to the deep location of a Fibrous mass—ligament within the vocal fold, the morphological impact on the vocal fold free edge is minimal and visualization of the lesion is often best during vocal fold abduction.
This lesion interferes with the vibratory behavior of the vocal folds, creating increased aperiodicity that leads to rough, hoarse voice. In contrast to sub-epithelial masses, a ligamentous fibrous mass does not tend to impede vocal fold closure. Vocal fold asymmetry due to differential vocal fold masses is presumed source of dysphonia. For a professional voice user, even the least amount of phonatory impairment can be disabling. Children who are teased about the abnormal sound of their voices or who have difficulty being heard or understood in group settings or in noisy environments, may experience a negative impact on school performance.
Aerodynamic characteristics: Additional mass may increase phonatory subglottic pressures. The prognosis for post-operative voice recovery is less favorable for ligamentous fibrous masses as compared to sub-epithelial masses. Similar to a subepithelial mass, a ligamentous fibrous mass may place patients at increased risk for vocal fold hemorrhage. Patients may complain of pain lateral to the larynx that may result from increased effort used in producing voice.
Behaviorally, activities involving extremes of voice use, such as cheerleading, untrained singing, speaking above noise, and unamplified teaching and aerobics instructing tend to involve high inter-vocal fold impact stress and thus are predisposing behaviors for the formation of a fibrous mass. Vocal Fold Polyp s Essential features. Presence of benign mass es , located within the lamina propria typically at the midpoint of the membranous vocal folds.
Lesions may be sessile broad attachment to the vocal fold surface or peduncular hanging. Polyps are most often unilateral. These lesions interfere with the vibratory behavior of the vocal folds, creating increased aperiodicity that leads to a rough, hoarse voice quality. Vocal fold polyp s are generally understood to have minimal to moderate impact on the mucosal wave. Incomplete vocal fold closure pattern on stroboscopy may be present. Peduncular polyps can vibrate independently from the fundamental frequency of the vocal folds, resulting in diplophonia.
The patient typically reports difficulty with the upper pitch range. The quality of the voice worsens with use, particularly loud, pressed, and extensive voice use. For a professional voice user, even 46 O Children who are teased about the abnormal sound of their voices or who have difficulty making themselves heard or understood in group settings or in noisy environments, may experience a negative impact on school performance.
Aerodynamic characteristics: Lack of vocal fold closure may result in increased average phonatory flows.
Pathological Voice Analysis and Classification Based on Empirical Mode Decomposition
Additional mass added by the polyp may also increase phonatory subglottic pressures. Largely unknown. In young adults, the most frequent age at onset for benign mass lesions is the teen years for girls who are heavy voice users, e. Benign lesions of the lamina propria and their symptoms may be a fairly acute reaction to an intense period of phonotraumatic behavior, or lesions and symptoms may have a gradual course of development.
However, in other cases, lesions and symptoms persist and may require more aggressive treatment.
- Kirlian Quest (Cluster, Book 3).
- The Indian Frontier of the American West, 1846-1890 (Histories of the American Frontier).
- Young Jesus: Restoring the Lost Years of a Social Activist and Religious Dissident!
- Talking Back to Ritalin: What Doctors Arent Telling You About Stimulants and ADHD (Revised Edition);
- Structured Electronic Design: Negative-Feedback Amplifiers.
- Search results for A C Clark (editor) Blackwell's.
- The Buddhist handbook: a complete guide to Buddist teaching, practice, history and schools;
Patients may complain of pain lateral to the larynx that may result from increased effort used in producing voice, or patients may report a feeling of a lump in the throat. Professional voice users can be significantly affected, even with minor vocal fold lesions. The primary causal factor of benign mass lesions of the lamina propria is thought to be perpendicular impact stress to the vocal folds. Specific factors leading to high impact stress are high subglottic pressure, and vocal fold elongation high pitch within a given register.
Behaviorally, activities involving extremes of voice use, such as cheerleading, untrained singing, speaking above noise, teaching without amplification to a large group of students and aerobics in- Vocal fold hemorrhage or capillary leaks may further predispose to the formation of a vocal fold polyp. Hemorrhagic polyps appear as a localized reaction similar to a blood blister, possibly associated with an intense period of extreme voice use such as yelling at an athletic event or oversinging.
Certain medications, such as hormonal treatments and anti-inflammatory agents, may predispose individuals to these lesions. In adults, women seem to be more affected by benign mass lesions of the lamina propria in general. Presence of localized benign mass es , typically unilateral, located within the superficial lamina propria usually on the medial edge at the midpoint of the membranous vocal folds. However, cysts may also be observed on the cephalic surface of the fold.
Vocal fold cysts are an encapsulated structure sometimes attributed to glandular duct blockage. Phonotrauma has also been implicated as the cause of vocal fold cysts, and there is also evidence that some cysts of the vocal fold may be congenital. The primary stroboscopic feature of a vocal fold cyst is reduction of mucosal wave with an hourglass closure pattern.
This interference may create increased aperiodicity that leads to rough, hoarse voice. Incomplete vocal fold closure pattern on stroboscopy may be present and may result in breathiness. Feeding vessels in the form of a dense vascular network may be seen in association with vocal fold cysts. The upper pitch range is often the first vocal feature affected. The quality of the voice worsens with use, particularly if there is loud, pressed, or extensive voice use. Additional mass may also increase phonatory subglottic pressure. Acoustic characteristics: Effects on aperiodicity, fundamental frequency, and intensity vary, depending on lesion characteristics and patterns of compensation.
In some cases, acoustic deviations are only present during singing in frequencies produced at the upper end of the register. Symptoms and signs of a vocal fold cyst may gradually increase over time if left untreated, regardless of its origin. Alternatively, effects may remain stable or occasionally improve over time, depending on adaptive strategies and other factors. Patients may complain of pain lateral to the larynx that may result from increased effort used in producing voice, or may report a feeling of a lump in the throat.
Professional voice users can be significantly affected, even with minor vocal fold lesions Predisposing factors. Various reports exist regarding potential causes of vocal fold cysts. Some literature reports cases of congenital cysts. Still other reports indicate a possible role of phonotrauma. In adults, women seem to be more affected. Vocal fold polyp, nodules, fibrous mass, reactive vocal fold lesion, rheumatoid lesion, vocal fold cyst—ligament Classification criteria.
Vocal Fold Cyst—Ligament Essential features. Presence of localized benign mass es , typically unilateral, located within the deeper layers of the lamina propria or vocal ligament. Vocal fold cyst—ligament may also be visible on the cephalic surface of the fold. Vocal fold cyst is an encapsulated structure sometimes attributed to glandular duct blockage.
Phonotrauma has been implicated as the cause of vocal fold cysts. There is also evidence that some cysts of the vocal fold may be congenital. The primary stroboscopic feature of a vocal fold cyst is severe reduction of mucosal wave vibratory activity as seen during stroboscopy. Symptoms and signs of a vocal fold cyst may gradually increase over time, if left untreated, regardless of its origin.
Various reports exist regarding potential causes of a vocal fold cyst. Some literature reports cases of congenital cysts, while other literature poses glandular blockage or vascular feeding vessels as possible predisposing factors. Other reports indicate a possible role of phonotrauma. Vocal fold polyp, nodules, fibrous mass lesion, reactive vocal fold lesion, rheumatoid mass lesion, vocal fold cyst sub-epithelial. Reactive Vocal Fold Lesion Essential features. Reactive lesion, also known as contrecoup, contralateral, or secondary lesion, occur opposite lesions of the vocal fold cysts, polyps, fibrous masses and are thought to be a consequence of altered contact forces caused by the lesion.
These benign lesions are located within the superficial lamina propria and have minimal negative impact on the mucosal wave vibratory properties. Reactive vocal fold lesions will reduce or subside with voice rest, reduced voice use, and voice therapy. Reactive lesions may interfere with the vibratory behavior of the vocal folds creating increased aperiodicity that leads to rough, hoarse voice.
Incomplete vocal fold closure pattern on stroboscopy is typically present and may result in breathiness. Reduction of upper pitch range is often the first difficulty noticed by the patient. The ef- Signs and symptoms vary with the severity of the reactive lesion and the contralateral lesion. The impairment may include varying degrees of hoarseness roughness and breathiness , loss of high notes, vocal fatigue, and restricted dynamic range.
Can occur at any age. Lesions and symptoms from acute phonotrauma may resolve with reduced or modified voice use. Some patients report a feeling of a lump in the throat. These lesions are a complication of the primary contralateral benign vocal fold lesion, which is the primary source of the dysphonia. Contralateral benign vocal fold lesion see , , , , Vocal fold nodules, polyps, cysts, and all fibrous mass lesions.
Essential features. Presence of increased gelatinous material in the superficial layer of the lamina propria on a chronic basis. The membranous vocal fold s is edematous along much or even the entire length of the membranous folds. The lesion is usually bilateral although it may be asymmetrical. The mass of the cover is increased by the presence of the gelatinous material within, which varies widely in viscosity from fairly watery to thickened. Stiffness is presumed to increase due to the extra mass, although mucosal wave may appear increased due to increased phonatory effort.
The primary associated feature is lowered vocal pitch due to increased mass of the vocal fold. Hoarseness or roughness, The vocal folds close completely during phonation with an overclosure when both folds are involved. Stridor and shortness of breath may be present if the lesions are large enough to narrow the glottal airway. In addition to increased mass of the superficial lamina propria, erythema may be present.
Fundamental frequency is typically lower than average and diminished in the upper range. This impairment is more striking in females. Hoarseness and roughness are present. Aerodynamic characteristics: Average phonatory airflow may be decreased due to long closed phase of the vibratory cycle. If vocal fold closure is complete, subglottic pressure may be normal. However, substantial added mass may increase subglottic pressure.
Acoustic characteristics: The predominant acoustic feature is low fundamental frequency in speech. Typically seen in middle-aged individuals with a longstanding history of smoking. Onset and progression are often gradual, following consistent and prolonged exposure to some combination of smoke, environmental toxins, reflux, or phonotrauma. Allowed to advance, airway compromise may occur. Obstruction of the airway is the most serious complication that can accompany this condition.
Heavy smoking is a clear predisposing factor. Additionally, laryngopharyngeal reflux may be strongly implicated. Phonotraumatic behavior may be present, possibly as a consequence of the presence of the pathology. Exposure to environmental pollutants may also be a contributing factor. However, the fact that the lowered pitch of the voice is one of the characteristic features may skew the sample.
Lowering of the pitch of the male voice would not be as likely recognized as a problem. Discrete polyps, myxedema associated with hypothyroidism, virilization of the voice due to the use of male hormones by a woman, or other hormonal changes female adolescence and menopause. Edema related to acute conditions e. Moderate: Moderate diffuse edema.
Severe: Severe diffuse edema, possibly lobulated. Profound: Airway compromise due to diffuse edema. Vocal Fold Scar Scars involve a permanent change to the microarchitecture of the lamina propria, consisting of a loss of the viscoelastic properties of the tissue. Scars may involve a reduction sulcus or increase scar proper in the amount of extracellular matrix proteins.
Both types of scar are described in the following sections. Vocal Fold Scar Proper Essential features. A permanent change to the microarchitecture of the lamina propria consisting of a loss of the viscoelastic properties Scar may be accompanied by increased vascularization. Hoarseness and vocal weakness are common with these vocal fold abnormalities.
Vocal fold scar proper can often be seen in combination with a mass lesion of the vocal fold, such as a cyst, fibrous mass, etc. Reduction of mucosal wave at the site of lesion is characteristic. In severe cases, mucosal wave at the site of the scar may be absent. In addition, vocal fold closure may be impacted. Vocal impairment can range from minimal to profound including loss of voice. Aerodynamic characteristics: Variable; subglottic pressure may be increased.
Acoustic characteristics: Due to stiffness of the tissue, the voice is characterized by increased fundamental frequency in speech, reduced signal-to-noise ratio, presence of higher harmonics, voice breaks, and reduced fundamental frequency and intensity range. Intensity may be decreased due to reduced mucosal wave amplitude or decreased vocal fold closure time. The scar is typically stable unless aggravating factors are present. However, severity of voice symptoms and signs can progress if phonotraumatic behaviors induce further injury by aggravating underlying vocal pathology.
Behavioral management may improve voice symptoms. The primary complications are aggravation or addition of compensatory lesions e. Scar proper may be the result of extended phonotrauma, surgery, radiation, trauma, hemorrhage, or prolonged laryngopharyngeal reflux. The presence of scarring is associated with decreased lamina propria and reduced mucosal wave, resulting in incomplete or very short vocal fold closure; these may be confused with sulcus vocalis, vocal fold paresis, or vocal fold atrophy.
Vocal fold margins surrounding the scar proper may appear as small divots, suggestive of invasion of the epithelial layer. Moderate: Focal regions of abnormal lamina propria tissue. Severe: Majority of one or both vocal folds with abnormal lamina propria. Vocal Fold Sulcus Essential features. A permanent change to the microarchitecture of the lamina propria consisting of a loss of the viscoelastic properties of the tissue caused by a reduction in the amount of the lamina propria.
The characteristic appearance is a furrow or groove that extends partially or completely along the medial membranous edge of one or both vocal folds. Sulcus may be seen in conjunction with a mucosal bridge band of isolated epithelium , or vocal fold pit an extension of the sulcus into the vocalis ligament or muscle. Vocal fatigue is also common. Vocal impairment can range from minimal to profound.
Stroboscopy shows a linear delineation of a furrow or groove along the medial edge of the vocal fold due to a loss of lamina propria. Reduction of mucosal wave at the site of lesion is characteristic, due to tethering of epithelium to deeper tissues. In severe cases, mucosal wave may be absent.
In addition, vocal fold closure may be reduced. Minimum and average phonatory airflows and subglottic pressure may be increased due to incomplete vocal fold closure. A common finding is decreased signalto-noise ratio. Sulcus may be congenital or acquired. Severity of voice symptoms and signs can progress if the lesion s worsen s over time, or if compensatory behaviors are maladaptive. The primary complication is progression of the lesions and worsening of voice. There is speculation that some sulci may be congenital. Other speculation suggests that sulci can be the result of ruptured cysts.
Incomplete or very short vocal fold closure as a result of sulcus may be confused with vocal fold paresis or vocal fold 60 O Intraoperative palpation of the vocal fold may be required for definitive diagnosis. Physiologic Sulcus: Preserved vocal fold vibration on laryngeal examination, implying intact lamina propria. No dysphonia. Type II. Sulcus Vergerture: Disturbance of lamina propria with lack of separation between the deep and superficial layers of vocal fold tissue.
Stiffness is observed on videostroboscopy. Dysphonia may be severe. Type III. Sulcus Vocalis proper: Focal lesion, extending into the ligament, with inflammation. Vocal Fold Granuloma Vocal fold granuloma may be distinguished between non-intubation related and intubation related lesions, because these two entities arise from different etiologies and exhibit dissimilar clinical courses.
Although the pathology is consistent, clinical experience has shown that intubation related granulomas are more responsive to treatment and have a high probability of resolution without recurrence. In contrast, non-intubation related vocal fold granulomas tend to be difficult to treat and have a high probability of recurrence. The essential feature of a non-intubation related contact ulcer is a superficial ulcerated area on the medial surface of the arytenoid that is typically unilateral, but can be bilateral, not associated with recent oro-tracheal intubation.
Granuloma is an exophitic mass arising from the arytenoid cartilage. The most common area of occurrence is at the region of the vocal process, but lesions may be seen as small projections on the medial surface of the arytenoid below the apex. Vocal fold granulomas are inflammatory lesions arising from a perichondritis of the arytenoid cartilage. Throat pain is common and often quite localized to the side of the neck at the level of the arytenoids.
Ear pain may also be present due to shared innervation of Cranial Nerve X. Persistent throat tickle, sensation of dryness, lump in the throat, and feeling the need to clear the throat, and non-productive cough are typical symptoms. Dysphonic symptoms are sometimes present including mild hoarseness and aphonic breaks. In addition to the presence of lesions, laryngeal examination findings often include edema and erythema of the arytenoid cartilages and of the posterior larynx.
Laryngopharyngeal reflux is often an associated condition. Vocal impairment: Vocal impairment is usually mild since the lesions do not directly affect the vibratory portions of the vocal folds. The voice may be hoarse if the membranous vocal folds are affected. If the lesion s are sufficiently large, limit adduction, or result in asymmetric movement, vocal impairment may be significant. The degree of pain accompanying phonation varies from person to person.
Aerodynamic characteristics: Usually, airflow and subglottic pressures are normal if the lesions do not affect the membranous vocal folds.
Acoustic characteristics: Literature indicates a possible finding of decreased fundamental frequency, which may be a contributing etiologic factor. Typically a lesion of adulthood usually presenting between 20 to 40 years. The appearance of the lesion at presentation may depend on the stage of its development.
Initially, ulcers may appear as small, pale, and concave lesions with the cartilage exposed in the center. Lesions may become covered with necrotic tissue. In more active stages, the margins become hyperemic. A raised, sessile, pale-appearing granuloma may then develop. These lesions do not usually regress spontaneously.
Recurrence following surgical excision is common. With recalcitrant disease, hoarseness and pain frequently persist. In extreme cases, granuloma formation may obstruct the airway. Primary etiologic factors, alone or in combination, are chronic cough, hyperadduction of the vocal processes, persistent low pitch in speech or singing causing large-amplitude arytenoid oscillations , and laryngopharyngeal reflux. Sinonasal allergic or infectious disease has also been implicated due to chronic throat clearing. Lack of membranous vocal fold closure has been implicated in granuloma formation due to compensatory hyperadduction of the vocal processes during phonation.
Due to the underlying vocal fold incompetence, recurrent granuloma formation following surgical removal is common. This condition appears to occur more often in males than females, perhaps because of a persistent glottal chink seen in females. There is no known familial pattern.
Carcinoma, infection tuberculosis, Klebsiella. Moderate: Unilateral or bilateral granulomas of the vocal process es. Severe: Unilateral or bilateral granulomas of the vocal process es or vocal folds compromising the airway. The essential feature of an intubation contact ulcer is a superficial ulcerated area on the medial surface of the arytenoid that is typically unilateral, but can be bilateral, associated with recent oro-tracheal intubation. Intubation granuloma is an exophitic mass arising from the arytenoid cartilage s , following intubation.
The most common area of occurrence is at the region of the vocal process. Regardless of their origin, vocal fold granulomas are inflammatory lesions arising from a perichondritis of the arytenoid cartilage s. Throat pain is common and often localized. Ear pain may also be present due to radiated neuralgia from Cranial Nerve X. Persistent throat tickle, sensation of dryness, lump in the throat, feeling the need to clear the throat, and non-productive cough are typical symptoms. Dysphonic symptoms are sometimes present including 64 O In addition to the presence of the lesions, laryngeal examination findings often include edema and erythema of the arytenoid cartilages and of the posterior larynx.
Laryngopharyngeal reflux is commonly seen as a comorbid condition. In some cases, symptoms may first appear a few weeks to a few months following intubation. Vocal impairment: The voice impairment is usually mild since the lesions do not directly affect the vibratory portions of the vocal folds. If the lesion limits adduction of the vocal folds or results in asymmetric movement of the folds, voice impairment may be present. Aerodynamic characteristics: Usually, average airflow and subglottic pressure are normal if the lesions do not affect the membranous vocal folds.
Intubation lesions may occur at any age. These may become covered with necrotic tissue or in the more active stage, the margins become hyperemic. Typically, complications from intubation lesions are limited, because these lesions usually regress with little or no treatment unless reflux or severe phonotrauma persists following intubation, thus not allowing the lesion to heal.
The predisposing factor is intubation. Long duration of intubation, emergent intubation, large intubation tube, and excessive motion of the endotracheal tube during the intubation may increase the risk of these lesions. Other factors that may increase the risk include post-extubation chronic cough and laryngopharyngeal reflux, hyperadduction of the vocal processes, persistent low pitch in speech or singing, and sinonasal allergic or infectious disease.
Differential diagnosis is usually limited because of the temporal association of symptoms and lesion appearance, and recent history of intubation. If relevant, differential diagnosis involves the same factors as for non-intubation related lesions: ruling out carcinoma or infection tuberculosis. Malignancy must be ruled out if associated risk factors are present and chest X-ray is negative for tuberculosis.
Moderate: Unilateral or bilateral granulomas of the vocal process es without impact on vocalization. Keratosis Leukoplakia, Erythroplasia Essential features. Keratosis is a change in the microarchitecture of the epithelial layer of the vocal fold. The epithelial cells are normal, but redundancy is present. If keratin is present and the surrounding epithelium is normal, the condition is called keratosis without atypia. If keratin is present and the surrounding epithelium is abnormal, the condition is called keratosis with atypia.
Clinically, the lesions are epithelial abnormalities that appear as red erythroplasia or white leukoplakia thickened patches, covering the vocal fold. The lesions can be unilateral or bilateral. They can occur anywhere on the vocal folds and often have an irregular pattern and a rough surface. A connection between keratosis with atypia and cancer is present, but inconsistent across patients. Voice can be abnormal, but is not necessarily disrupted.
Voice may be normal or near normal, or may be more severely affected, depending on lesion characteristics. Aerodynamic characteristics: Variable, depending on lesion characteristics and patient adaptation. Acoustic characteristics: Variable, depending on lesion characteristics and patient adaptation. Keratosis is a disease of adulthood, most frequently occurring between the ages of 40 and 80 years. The course can be variable, in that lesions may be stable and unchanging over time, or they can progress in the dimensions of size or regional spread. Keratosis with atypia may progress to invasive carcinoma.
The primary complication is the progression of keratosis with atypia to carcinoma. Predisposing factors are chronic irritations from chemicals and tobacco smoke, alcohol, or laryngopharyngeal reflux. Phonotraumatic behaviors, which may be compensatory, could aggravate the clinical condition. Sex Ratio. Appears to be higher in males. Familial Pattern. Carcinoma, dysplasia, infectious processes candida , papilloma. Lack of abnormal cells in the surrounding epithelium leads to a diagnosis of keratosis without atypia, whereas presence of abnormal cells in the surrounding epithelium points to keratosis with atypia.
Severe: Multiple lesion sites. RRP is characterized by epithelial growths anywhere in the upper aerodigestive tract, most commonly on the vocal folds. The lesions are epithelial hyperplasia induced by human papilloma virus HPV infection. Chronic cough may be present, and voice disruption is typical. Laryngeal endoscopy usually reveals mass lesions. Reduced mucosal wave is often seen during stroboscopic examination, due to increased stiffness of the lesions. Closure may be compromised. In severe cases, the airway can be compromised.
Voice is typically hoarse. Extreme cases may involve voice loss. Aerodynamic characteristics: Findings are variable, depending on lesion characteristics. High subglottic pressures are common. Acoustic characteristics: Fundamental frequency may be increased if stiffness predominates. If the vocal fold surface is irregular or closure is incomplete, aerodynamic turbulence results in reduced signalto-noise ratio.
This condition is a disease of both child and adulthood. Onset is usually in the first decade of life.
Classification Manual For Voice Disorders I Book (ePUB/PDF) Free
Adult onset of RRP may also occur, but is less common. Course is highly variable. The condition can be stable with no new growths over a lengthy period of time or may exhibit rapid regrowth and extension to new areas after surgical excision. The primary complication is the potential for airway compromise. The disease can become malignant, and can also invade pulmonary parenchyma. In extreme cases, death may ensue. Vocal fold scar commonly occurs following repeated surgical removal of the disease. For juvenile onset, risk factors appear to include young mothers, first-born children, and low socio-economic status.
However, the link is not fully understood. Estimated incidence of laryngeal RRP in children is 4. In children, the sex ratio is approximately After age 20, the male to female ratio is closer to for adult onset RRP. The primary familial link may be through the presence of HPV infection in the mother during pregnancy.
The route of transmission is not well understood. Cancer, keratosis, and non-HPV infection. Light microscopy findings show a thickened spinous layer within the epithelium, projections of keratinized, stratified, squamous epithelium overlying a fibrovascular core. Hyperplastic basal zone of the epithelium and koilocytes, which appear as vacuolated cells with clean cytoplasm, are commonly seen. O 70 Indicate total score all sites. A staging system for assessing severity of disease and response to therapy in recurrent respiratory papillomatosis.
Jun 6 : — Subglottic Stenosis Essential features. Narrowing of the subglottic area. Causes may be congenital, or due to scar or cartilage deformation from prolonged endotracheal intubation, subglottic lesions, trauma, or other factors. Biphasic stridor, hoarseness, or dyspnea may be associated features. Physical activities may be restricted as a result of the compromised airway.
Voicing and oral communication may be impaired. Voice may be impaired due to insufficient respiratory control for phonation, or sequelae of tracheotomy. Descriptive reports of voice quality range from mild impairment in vocal function to aphonia. Acoustic characteristics: Variable, depending on lesion properties and patient adaptation. Congenital or post-natal at any age. In infancy, airway compromise with upper respiratory infection may recur. Some individuals tolerate mild airway restrictions during daily living without intervention.
Other individuals require surgical intervention to bypass the restriction tracheotomy or subglottic reconstructive surgery. Respiratory distress requiring the need for tracheotomy; feeding dysfunction, including aspiration; phonation problems, including abnormal cry; aphonia or hoarseness.
Vocal fold paralysis may also be sequelae of surgical management or prior intubation. For acquired conditions, the rate of post-intubation stenosis ranges from 0. An anterior glottic web involves the presence of tissue between the vocal folds forming a membrane that spans the glottis. This most commonly involves scar tissue between the vocal folds and can range in severity from a mild, anterior commissure micro-web to complete closure of the glottis due to vocal fold fusion. An anterior glottic web typically causes airway and voice problems.
If the web is of significant size, it will diminish the airway resulting in symptoms that may include shortness of breath, dyspnea, and possibly stridor. Given that most glottic webs occur from trauma, dysphonia is also a common finding. The length of vocal fold involved in vibration is shortened, and thus pitch may be increased. Mucosal wave is often reduced. The voice is typically hoarse. Diplophonia may also be present. Aerodynamic characteristics: Airflow may be reduced consistent with the extent of the webbing.
Acoustic characteristics: Fundamental frequency may be increased. Age at Onset. Post-natally, acquired glottic web can occur at any age. Pediatric glottic webs often form subsequent to repeated laryngeal surgery for recurrent respiratory papillomatosis. Adult glottic webs often form following laryngeal trauma or following bilateral mucosal surgeries, particularly in or near the anterior commissure. The course for a patient with a glottic web is highly variable across patients, but typically stable within individuals.
If the web is sufficiently extensive as to interfere with respiration or phonation, surgical lysing is usually required. If this procedure is successful, there is typically no need for further follow-up. The most common complication is scarring postoperatively at the site of the web. Reduced vocal fold vibration and poor voice quality may result. Surgical attempts to correct the scar may result in further scarring and web formation. A rare complication is sub- 74 O Common predisposing factors for non-congenital glottic webs include previous intubation, especially for microwebs, and laryngeal surgery, particularly if repeated, for papilloma, leukoplakia, cancer, trauma reconstruction, or other conditions.
Ingestion of caustic substances or thermal damage may lead to the formation of a glottic web. Anterior commissure micro-web. Complete connection of the vocal folds from the anterior commissure to the vocal processes. Vocal Fold Hemorrhage Essential features. Extravasation of blood into the lamina propria of the vocal fold due to disruption of the intra-vocal fold vasculature. The condition can disrupt vocal fold vibratory function both acutely due to accumulation of blood, and chronically due to deposition of fibrous material following blood resorption.
Voice changes can vary from minor to complete voice loss. Some may experience no voice difficulties. Vocal impairment can occur suddenly e. Hemorrhage is commonly unilateral, but may be bilateral. Hoarseness may or may not be present. Laryngeal endoscopy reveals extravasation into the vocal fold tissue. Typically, mucosal wave is diminished on stroboscopy. Aerodynamic characteristics: Variable across individuals. Acoustic characteristics: Variable across individuals. Typically occurs in adults, but may occur in children following trauma. Hemorrhage onset is characteristically abrupt following trauma due to voice use, coughing, crying, intubation, or external laryngeal injury.
Resorption is usually gradual following a period of voice rest 1 to 2 weeks. Hematoma, which is a distinct collection of blood with a raised surface, may also occur from a vocal fold hemorrhage. Phonotraumatic behaviors, laryngeal trauma, the ingestion of anti-coagulants [e. During the premenstrual period, women may experience increased fragility of the vocal fold vascular network. Case studies indicate a higher number of females than males. Moderate: Intermediate unilateral or bilateral bleed. Severe: Large bleed either unilateral or bilateral. Varix and Ectasia of the Vocal Fold Essential features.
In the healthy state, blood vessels are organized parallel to the longitudinal axis of the vocal fold, and are less than 1 mm in diameter. A change in either orientation or size of the vessels constitutes a vascular abnormality that may be a varix or an ectasia. Varices are considered to be enlarged or acutely tortuous blood vessels. Ectatic vessels are dilated tubular structures involving greater mass, appearing as a relatively minor pooling of blood clearly contained within a blood vessel lining.
- Search results for A C Clark (editor) Blackwell's.
- Account Options.
- Acoustic voice analysis: comparison between two types of microphones.
- Optical Waveguide Theory.
Clinically, the terms varix and ectasia are frequently used interchangeably. Varix and ectasia may occur in isolation or in association with hemorrhage, mass lesion abnormalities of the mucosa, or with neoplastic lesions. In isolation, voice may be normal, or there may be voice changes. Clinically, a common complaint is vocal fatigue. Stroboscopic findings may be minimal, apart from the observation of the characteristic vessel changes. Can occur at any age, although the condition is probably more common in adulthood. The onset is presumed to be sudden, in conjunction with phonotrauma.
The lesions are often stable and usually do not resolve spontaneously. Vessel changes may proceed to hemorrhage. Evidence of repeated hemorrhage due to these blood vessel abnormalities is an indication for surgical removal. Vessel changes may lead to hemorrhage. Assumed to be similar to hemorrhage: phonotraumatic behaviors, laryngeal trauma, the ingestion of anti-coagulants [e. Vocal fold hemorrhage. Moderate: Moderate appearance of vessel change, and possibly minor associated mass lesions or minor hemorrhage.
Severe: Significant appearance of vessel change, and possibly associated mass lesions or hemorrhage.