dyslexia research paper

ORIGINAL RESEARCH article

The 100 top-cited studies on dyslexia research: a bibliometric analysis.

• 1 Department of Respiratory and Critical Care Medicine, West China Hospital/West China School of Medicine, Sichuan University, Chengdu, China
• 2 Department of Periodical Press and National Clinical Research Center for Geriatrics, West China Hospital, Sichuan University, Chengdu, China
• 3 Chinese Evidence-Based Medicine Center, West China Hospital, Sichuan University, Chengdu, China

Background: Citation analysis is a type of quantitative and bibliometric analytic method designed to rank papers based on their citation counts. Over the last few decades, the research on dyslexia has made some progress which helps us to assess this disease, but a citation analysis on dyslexia that reflects these advances is lacking.

Methods: A retrospective bibliometric analysis was performed using the Web of Science Core Collection database. The 100 top-cited studies on dyslexia were retrieved after reviewing abstracts or full-texts to May 20th, 2021. Data from the 100 top-cited studies were subsequently extracted and analyzed.

Results: The 100 top-cited studies on dyslexia were cited between 245 to 1,456 times, with a median citation count of 345. These studies were published in 50 different journals, with the “Proceedings of the National Academy of Sciences of the United States of America” having published the most ( n = 10). The studies were published between 1973 and 2012 and the most prolific year in terms of number of publications was 2000. Eleven countries contributed to the 100 top-cited studies, and nearly 75% articles were either from the USA ( n = 53) or United Kingdom ( n = 21). Eighteen researchers published at least two different studies of the 100 top-cited list as the first author. Furthermore, 71 studies were published as an original research article, 28 studies were review articles, and one study was published as an editorial material. Finally, “Psychology” was the most frequent study category.

Conclusions: This analysis provides a better understanding on dyslexia and may help doctors, researchers, and stakeholders to achieve a more comprehensive understanding of classic studies, new discoveries, and trends regarding this research field, thus promoting ideas for future investigation.

Introduction

Dyslexia is a common learning disorder that affects between 4 and 8% of children ( 1 – 3 ), and often persists into adulthood ( 4 , 5 ). This neurodevelopmental disorder is characterized by reading and spelling impairments that develop in a context of normal intelligence, educational opportunities, and perceptual abilities ( 4 , 6 ). Reading and spelling abilities can be affected together or separately. The learning abilities of children with dyslexia are significantly lower than those of their unaffected pairs of the same age. Generally, difficulties begin to show during the early school years. Dyslexia is a complex multifactorial disorder whose etiology has not been fully elucidated, and it has caused great social and economic burdens. Over the last few decades, the research on dyslexia has made some progress. For example, some studies have shown that dyslexia has a strong genetic background that can affect brain anatomy ( 7 , 8 ) and function ( 9 , 10 ). But a citation analysis on dyslexia that reflects these advances is lacking.

The publication of study results in scientific journals is the most effective strategy to disseminate new research findings. A high number of citations can indicate the potential of a paper to influence the research community and to generate meaningful changes in clinical practice ( 11 ). Citation analysis is a type of quantitative and bibliometric analytic method designed to rank papers based on their citation counts. The latest and up-to-date research findings on dyslexia are well-reflected in recent scientific papers ( 12 ), particularly in the most cited ones ( 13 , 14 ). By analyzing the most cited studies, especially the 100 top-cited studies, we can gain better insight into the most significant advances made in the field of dyslexia research over the course of the past several decades ( 15 ). This retrospective bibliometric approach has been used for many other diseases, such as diabetes ( 16 ), endodontics ( 17 ), cancer ( 18 ). However, to date, no bibliometric analyses have been conducted in the field of dyslexia. Therefore, the aim of the present study was to analyze the 100 top-cited studies in the field of dyslexia.

Materials and Methods

Search method and inclusion criteria.

This retrospective bibliometric analysis was conducted using the Web of Science Core Collection database. The Web of Science Core Collection is a multidisciplinary database with searchable authors and abstracts covering a vast science journal literature ( 19 ). It indexes the major journals of more than 170 subject categories, providing access to retrospective data between 1945 and the present ( 20 ). On May 20th, 2021, we conducted an exhaustive literature retrieval, regardless of the country of origin, publication year, and language. The only search term used was “dyslexia” and the search results were sorted by the number of citations.

Article Selection

Two authors independently screened the abstracts or full-texts to identify the 100 top-cited articles about dyslexia. Disagreements were resolved through discussion. Only studies that focused on dyslexia were included in subsequent analyses. Studies that only mentioned dyslexia in passing were excluded.

Data Extraction

The final list of the 100 top-cited studies on dyslexia was determined by total article citation counts. We extracted the following data for each article: title, authors, journal, language, total citation count, publication year, country, journal impact factor, type of article, and Web of Science subject category. If the reprint author had two or more affiliations from different countries, we used the first affiliation as the country of origin. If one article was listed in more than one subject category, the first category was selected. If one article had more than one author, we selected the first-ranked author as the first author and the last-ranked author as the last-author.

Data Analysis

SPSS 11.0 (Chicago, IL, USA) was used to count the frequency. We analyzed the following data: citation count, year of publication, country, the first author, journal, language, type of study, and Web of Science subject category.

Citation Analysis

The 100 top-cited studies on dyslexia based on total citations are listed in Table 1 . The total citation count for these 100 articles combined was 42,222. The total citation count of per study ranged from 245 to 1,456 times, with a median citation count of 345. Only 3 studies were cited more than 1,000 times, and the rest of the studies were cited between 100 and 1,000 times. The title of the top-cited study, which also had the largest mean citation per year count ( n = 91), was “Reading acquisition, developmental dyslexia, and skilled reading across languages: a psycholinguistic grain size theory,” which was published by Ziegler et al. in Psychological Bulletin in 2005 ( 21 ). The second top-cited study, which also had the second-highest mean citation per year count ( n = 80), was published by Vellutino et al. ( 22 ). In addition, we also identified the 100 top-cited studies on dyslexia based on mean citation per year, whose results were shown in Supplementary Table 1 .

Table 1 . The 100 top-cited studies on dyslexia based on total citations.

The different journals of the 100 top-cited studies on dyslexia and their associated impact factors are listed in Table 2 . The 100 top-cited studies on dyslexia were published in 50 different journals, with the top three in frequency being “Proceedings of the National Academy of Sciences of the United States of America” ( n = 10), “Brain” ( n = 6), and “Journal of Educational Psychology” ( n = 6).

Table 2 . Journals of the 100 top-cited studies on dyslexia.

The journal with the highest total citation count was “Proceedings of the National Academy of Sciences of the United States of America.” However, the highest average citation count per study belonged to the journal “Psychological Bulletin.” The journal impact factors of the 100 top-cited studies on dyslexia ranged from 1.315 to 74.699. Of the 100 top-cited studies, 29 were published in a journal with an impact factor greater than 10. The standard “CNS” journals, with the exception of “Cell,” “Nature,” and “Science” published 2 and 3 studies, respectively. Regarding the top four medical journals, while the “New England Journal of Medicine” and “Lancet” published 2 studies each, no top-cited study was published by the “Journal of the American Medical Association” or the “British Medical Journal.”

Language and Year of Publication

The 100 top-cited studies on dyslexia were all published in English and were published between 1973 [by Boder et al. ( 23 )] and 2012 [by Norton et al. ( 24 ) and Peterson et al. ( 25 )] ( Table 3 ). The most productive years were 2000, 2001 and 2003, with 9, 8 and 8 published articles, respectively. The year of 2003 had the most total citations with a total count of 3,788 and an average citation count per study of 474.

Table 3 . Publication year of the 100 top-cited studies on dyslexia.

Countries and Authors

Eleven countries contributed articles to the 100 top-cited studies on dyslexia ( Table 4 ). Most of the articles were from the USA ( n = 53), United Kingdom ( n = 21), Canada ( n = 7), and France ( n = 6). In addition, the USA had the highest total citation count (23,129), and Italy had the highest average citation count per study (665).

Table 4 . Countries of the 100 top-cited studies on dyslexia.

As shown in Table 5 , there were 18 first-authors and 13 last-authors who published more than one of the 100 top-cited studies on dyslexia. Among them, Shaywitz SE published the most top 100 articles ( n = 7) on dyslexia as the first author, followed by Galaburda AM ( n = 3) and Pugh KR ( n = 3). And for the last author, 8 studies of the 100 top-cited studies on dyslexia research were published by Shaywitz BA who was the most productive.

Table 5 . Authors with at least two first-author or last-author publications in the 100 top-cited studies on dyslexia.

Publication Type and Web of Science Subject Categories

As shown in Table 6 , there were 71 studies in the form of an original research article, 28 studies in the form of a review article, and one study in the form of an editorial material publication. The total citation counts for each publication type were 27,812, 13,899, and 511, respectively. Although the type of original research article had the highest total citation count, it had the lowest average citation count per study. In addition, a total of 12 Web of Science subject categories were extracted. Among them, “Psychology” was the most frequent category associated with studies [35], followed by “Clinical Neurology” [15], and “Multidisciplinary Sciences” [15], “Neurosciences” [12], and “Education” [6]. Consistent with the number of studies, the subject categories of “Psychology” and “Clinical Neurology” also had the highest total citation counts (15,683 and 6,427, respectively). The “Behavioral Sciences” subject category had the highest average citation count.

Table 6 . Type of study and subject categories for the 100 top-cited studies on dyslexia.

Although retrospective bibliometric approach has been conducted in many other diseases, to our knowledge, no citation analyses have examined publications on dyslexia. Therefore, this study is the first comprehensive analysis summarizing several features of the most influential studies on dyslexia. It has been suggested that a highly cited study can be considered as a milestone study in a related field and has the potential to generate meaningful changes in clinical practice ( 26 ). We believe that the present analysis of the 100 top-cited studies on dyslexia may be beneficial to the research community for the following reasons. First, the present study not only provides a historical projection of the scientific progress with regards to dyslexia research, but it also shows associated research trends and gaps in the field ( 27 ). Second, our findings provide critical quantitative information about how both the classic studies and recent advancements in the field have improved our understanding of dyslexia ( 28 ). Third, the present analysis may help journal editors, funding agencies, and reviewers critically evaluate studies and funding applications ( 28 ).

Our analysis discovered that the 100 top-cited studies on dyslexia were published in 50 different journals. This may reflect the fact that the 100 top-cited studies on dyslexia were very multidisciplinary in nature, unlike the top studies of other fields (e.g., psoriatic arthritis) where there is a more inherent researcher bias for journal selection ( 29 ). Of the 100 top-cited studies, 29 were published in a journal with an impact factor >10, and 62 studies were published in journal with an impact factor >5. However, there were only five studies published in the standard “CNS” journals and only four published in the top four medical journals, which suggests that most dyslexia researchers are more inclined to choose the most influential journals in their respective professional fields when submitting articles ( 30 ). This is in marked contrast with some other fields (e.g. vaccines), where the majority of top-cited articles are published in either the standard “CNS” journals or in the top four medical journals ( 15 ). Several other factors, such as the review turnaround time, likelihood of manuscript acceptance, publication costs, journal publication frequency, will all invariably also affect a researcher's journal selection ( 13 , 20 ).

According to the results of our analysis, nearly 80% of the 100 top-cited studies on dyslexia were published between 1990 and 2005, and the years of 2000 was found to have the most publications. The increase of landmark publications between 1990 and 2005 might reflect an increase in the interest in dyslexia research or that researchers had made some important scientific breakthroughs during this time period. All the top-cited studies on dyslexia were published in English, likely because English is the most commonly used language for knowledge dissemination in the world.

The top countries with regards to total citation count and number of papers in the top 100 list were the USA ( n = 53) and United Kingdom ( n = 21), which accounted for ~75% of the 100 top-cited studies. The USA published the most studies from the list, and this is probably because some of the world's top research centers are located in the USA and likely also the USA receives more research funding ( 31 ). Furthermore, the most prolific first-author (Shaywitz SE) and last-author (Shaywitz BA) were also from the USA. It is also worth mentioning that China had two studies on the top 100 list, which attests to the improvement of our national scientific research community with regards to knowledge dissemination.

In the present study, there were more original research articles ( n = 71) than review articles ( n = 28), but the latter had higher average citation counts per study. These results indicate that even though researchers pay significant attention to new findings on dyslexia, they regularly use information from review articles to convey relevant points in their own papers. We found that “Psychology” was the most frequent subject category associated with the top 100 articles, which indicates that researchers have been working to find effective treatments for people with dyslexia and that research in this field will continue to progress.

Like with other bibliometric analyses, there are some study limitations that should be highlighted. First, the 100 top-cited studies were extracted from the Web of Science Core Collection, which might have excluded some top-cited studies from other databases, such as Scopus and Google Scholar. Second, there was no citation data for recently published studies. Third, self-citations might have substantially influenced the results of the citation analysis. Moreover, this was a cross-sectional study, which implies that the identified 100 top-cited studies could change in the future. Despite these limitations, this descriptive bibliometric study could contribute new information about the scientific interest in dyslexia.

In conclusion, the present analysis is the first analysis to recognize the 100 top-cited studies in the field of dyslexia. This analysis provides a better understanding on dyslexia and may help doctors, researchers, and stakeholders to achieve a more comprehensive understanding of classic studies, new discoveries, and trends regarding this research field. As new data continue to emerge, this bibliometric analysis will become an important quantitative instrument to ascertain the overall direction of a given field, thus promoting ideas for future investigation.

Data Availability Statement

The original contributions presented in the study are included in the article/ Supplementary Material , further inquiries can be directed to the corresponding author/s.

Author Contributions

YZ and HF designed the study. SZ and YZ acquired the data and performed statistical analyses. SZ, YZ, and HF drafted the manuscript. All authors critically revised the article and approved the final version of the manuscript.

This study was partly supported by National Clinical Research Center for Geriatrics, West China Hospital, Sichuan University (Z2018B016).

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Supplementary Material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpsyt.2021.714627/full#supplementary-material

1. Fortes IS Paula CS Oliveira MC Bordin IA de Jesus Mari J Rohde LA. A cross-sectional study to assess the prevalence of DSM-5 specific learning disorders in representative school samples from the second to sixth grade in Brazil. Eur Child Adolesc Psychiatry. (2016) 25:195–207. doi: 10.1007/s00787-015-0708-2

CrossRef Full Text | Google Scholar

2. Landerl K, Moll K. Comorbidity of learning disorders: prevalence and familial transmission. J Child Psychol Psychiatry Allied Discipl. (2010) 51:287–94. doi: 10.1111/j.1469-7610.2009.02164.x

PubMed Abstract | CrossRef Full Text | Google Scholar

3. Castillo A, Gilger JW. Adult perceptions of children with dyslexia in the USA. Ann Dyslexia. (2018) 68:203–17. doi: 10.1007/s11881-018-0163-0

4. Peterson RL, Pennington BF. Developmental dyslexia. Ann Rev Clin Psychol. (2015) 11:283–307. doi: 10.1146/annurev-clinpsy-032814-112842

5. Cavalli E, Colé P, Brèthes H, Lefevre E, Lascombe S, Velay JL. E-book reading hinders aspects of long-text comprehension for adults with dyslexia. Ann Dyslexia. (2019) 69:243–59. doi: 10.1007/s11881-019-00182-w

6. Wang LC, Liu D, Xu Z. Distinct effects of visual and auditory temporal processing training on reading and reading-related abilities in Chinese children with dyslexia. Ann Dyslexia. (2019) 69:166–85. doi: 10.1007/s11881-019-00176-8

7. Skeide MA, Kraft I, Müller B, Schaadt G, Neef NE, Brauer J, et al. NRSN1 associated grey matter volume of the visual word form area reveals dyslexia before school. Brain. (2016) 139:2792–803. doi: 10.1093/brain/aww153

8. Kraft I, Schreiber J, Cafiero R, Metere R, Schaadt G, Brauer J, et al. Predicting early signs of dyslexia at a preliterate age by combining behavioral assessment with structural MRI. NeuroImage. (2016) 143:378–86. doi: 10.1016/j.neuroimage.2016.09.004

9. Neef NE, Müller B, Liebig J, Schaadt G, Grigutsch M, Gunter TC, et al. Dyslexia risk gene relates to representation of sound in the auditory brainstem. Dev Cogn Neurosci. (2017) 24:63–71. doi: 10.1016/j.dcn.2017.01.008

10. Männel C, Meyer L, Wilcke A, Boltze J, Kirsten H, Friederici AD. Working-memory endophenotype and dyslexia-associated genetic variant predict dyslexia phenotype. Cortex. (2015) 71:291–305. doi: 10.1016/j.cortex.2015.06.029

11. Perazzo MF, Otoni ALC, Costa MS, Granville-Granville AF, Paiva SM, Martins-Júnior PA. The top 100 most-cited papers in paediatric dentistry journals: a bibliometric analysis. Int J Paediatr Dentist. (2019) 29:692–711. doi: 10.1111/ipd.12563

12. Daley EM, Vamos CA, Zimet GD, Rosberger Z, Thompson EL, Merrell L. The feminization of HPV: reversing gender biases in US human papillomavirus vaccine policy. Am J Public Health. (2016) 106:983–4. doi: 10.2105/AJPH.2016.303122

13. Kolkailah AA, Fugar S, Vondee N, Hirji SA, Okoh AK, Ayoub A, et al. Bibliometric analysis of the top 100 most cited articles in the first 50 years of heart transplantation. Am J Cardiol. (2019) 123:175–86. doi: 10.1016/j.amjcard.2018.09.010

14. Zhao X, Guo L, Lin Y, Wang H, Gu C, Zhao L, et al. The top 100 most cited scientific reports focused on diabetes research. Acta Diabetol. (2016) 53:13–26. doi: 10.1007/s00592-015-0813-1

15. Zhang Y, Quan L, Xiao B, Du L. The 100 top-cited studies on vaccine: a bibliometric analysis. Hum Vacc Immunother. (2019) 15:3024–31. doi: 10.1080/21645515.2019.1614398

16. Beshyah WS, Beshyah SA. Bibliometric analysis of the literature on Ramadan fasting and diabetes in the past three decades (1989-2018). Diabetes Res Clin Pract. (2019) 151:313–22. doi: 10.1016/j.diabres.2019.03.023

17. Adnan S, Ullah R. Top-cited articles in regenerative endodontics: a bibliometric analysis. J Endodontics. (2018) 44:1650–64. doi: 10.1016/j.joen.2018.07.015

18. Gao Y, Shi S, Ma W, Chen J, Cai Y, Ge L, et al. Bibliometric analysis of global research on PD-1 and PD-L1 in the field of cancer. Int Immunopharmacol. (2019) 72:374–84. doi: 10.1016/j.intimp.2019.03.045

19. Yu T, Jiang Y, Gamber M, Ali G, Xu T, Sun W. Socioeconomic status and self-rated health in China: findings from a cross-sectional study. Medicine. (2019) 98:e14904. doi: 10.1097/MD.0000000000014904

20. Yoon DY, Yun EJ, Ku YJ, Baek S, Lim KJ, Seo YL, et al. Citation classics in radiology journals: the 100 top-cited articles, 1945-2012. AJR Am J Roentgenol. (2013) 201:471–81. doi: 10.2214/AJR.12.10489

21. Ziegler JC, Goswami U. Reading acquisition, developmental dyslexia, and skilled reading across languages: a psycholinguistic grain size theory. Psychol Bull. (2005) 131:3–29. doi: 10.1037/0033-2909.131.1.3

22. Vellutino FR, Fletcher JM, Snowling MJ, Scanlon DM. Specific reading disability (dyslexia): what have we learned in the past four decades? J Child Psychol Psychiatry Allied Discipl. (2004) 45:2–40. doi: 10.1046/j.0021-9630.2003.00305.x

23. Boder E. Developmental dyslexia: a diagnostic approach based on three atypical reading-spelling patterns. Dev Med Child Neurol. (1973) 15:663–87. doi: 10.1111/j.1469-8749.1973.tb05180.x

24. Norton ES, Wolf M. Rapid automatized naming (RAN) and reading fluency: implications for understanding and treatment of reading disabilities. Ann Rev Psychol. (2012) 63:427–52. doi: 10.1146/annurev-psych-120710-100431

25. Peterson RL, Pennington BF. Developmental dyslexia. Lancet. (2012) 379:1997–2007. doi: 10.1016/S0140-6736(12)60198-6

26. Van Noorden R, Maher B, Nuzzo R. The top 100 papers. Nature. (2014) 514:550–3. doi: 10.1038/514550a

27. Fardi A, Kodonas K, Gogos C, Economides N. Top-cited articles in endodontic journals. J Endodontics. (2011) 37:1183–90. doi: 10.1016/j.joen.2011.05.037

28. Gondivkar SM, Sarode SC, Gadbail AR, Gondivkar RS, Choudhary N, Patil S. Citation classics in cone beam computed tomography: the 100 top-cited articles. Int J Dentistry. (2018) 2018:9423281. doi: 10.1155/2018/9423281

29. Berlinberg A, Bilal J, Riaz IB, Kurtzman DJB. The 100 top-cited publications in psoriatic arthritis: a bibliometric analysis. Int J Dermatol. (2019) 58:1023–34. doi: 10.1111/ijd.14261

30. Bullock N, Ellul T, Bennett A, Steggall M, Brown G. The 100 most influential manuscripts in andrology: a bibliometric analysis. Basic Clin Androl. (2018) 28:15. doi: 10.1186/s12610-018-0080-4

31. Shadgan B, Roig M, Hajghanbari B, Reid WD. Top-cited articles in rehabilitation. Arch Phys Med Rehabilit. (2010) 91:806–15. doi: 10.1016/j.apmr.2010.01.011

Keywords: dyslexia, bibliometric analysis, top-cited, citation analysis, citation

Citation: Zhang S, Fan H and Zhang Y (2021) The 100 Top-Cited Studies on Dyslexia Research: A Bibliometric Analysis. Front. Psychiatry 12:714627. doi: 10.3389/fpsyt.2021.714627

Received: 25 May 2021; Accepted: 28 June 2021; Published: 22 July 2021.

Reviewed by:

Copyright © 2021 Zhang, Fan and Zhang. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Hong Fan, fanhongfan@qq.com ; Yonggang Zhang, jebm_zhang@yahoo.com

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Int J Environ Res Public Health

Understanding Mental Health in Developmental Dyslexia: A Scoping Review

Adrienne wilmot.

1 Faculty of Health Sciences, Curtin School of Population Health, Curtin University, Perth, WA 6845, Australia

Penelope Hasking

2 Faculty of Health Sciences, Curtin enAble Institute, Curtin University, Perth, WA 6845, Australia

Suze Leitão

3 Faculty of Health Sciences, Curtin School of Allied Health, Curtin University, Perth, WA 6845, Australia

Elizabeth Hill

Associated data.

The article is a scoping review and no data were collected.

Children with dyslexia are at elevated risk of internalising and externalising mental health concerns. Our aim was to scope the extent and nature of the literature investigating factors which may influence this association. We systematically searched the peer-reviewed and grey literature with no restrictions on the date. We included both qualitative and quantitative studies. Inclusion criteria included: (1) a focus on childhood (≤18 years) reading/learning difficulties; (2) internalising and/or externalising symptoms; and (3) a potentially modifiable third factor (e.g., self-esteem). Ninety-eight studies met the inclusion criteria. We organised the studies according to individual, family, and community-level third factors. Whilst a range of third factors were identified, relatively few researchers tested associations between the third factor and mental health in the context of dyslexia. Furthermore, there was a focus on primary rather than secondary school experience and a reliance, in many cases, on teacher/parent perspectives on children’s mental health. Future researchers are encouraged to explore links between socio-emotional skills, coping strategies, school connectedness, and mental health in the context of dyslexia. Research of this nature is important to assist with the identification of children who are more (or less) at risk of mental health concerns and to inform tailored mental health programs for children with dyslexia.

1. Introduction

Dyslexia is characterised by difficulties with accurate and fluent word reading and poor spelling and decoding abilities that do not progress as expected with the provision of well-intentioned and targeted intervention [ 1 ]. Importantly, dyslexia is not related to more generalised cognitive difficulties or sensory deficits, rather, the difficulties are thought to stem from neuro-cognitive differences in the way speech sounds are processed [ 2 , 3 ]. Consistent with this account, oral language difficulties in early childhood are a frequent precursor to dyslexia [ 4 ] and difficulties in both the oral and written (reading, spelling, writing) domains of language often co-occur across the lifespan [ 5 , 6 , 7 ]. Furthermore, dyslexia is frequently associated with other learning and attentional difficulties that are believed to share genetic risk factors and/or underlying cognitive differences [ 8 , 9 ].

The Diagnostic and Statistical Manual of Mental Disorders-Fifth Edition (DSM-5); Ref. [ 10 ] has a category for dyslexia within the Specific Learning Disorders (SLD); a category which also includes dyspraxia (difficulties with writing) and dyscalculia (difficulties with mathematics). It is difficult to estimate prevalence rates for dyslexia as many children remain unidentified. Nevertheless, it is generally believed that 5–10% of children experience severe and persistent word reading difficulties consistent with dyslexia; a figure which equates to approximately 2–4 children in an average-sized classroom [ 3 , 11 ].

1.1. Academic and Psychosocial Correlates to Dyslexia

Dyslexia can negatively affect children’s motivation to read [ 12 ], vocabulary acquisition and reading comprehension. These skills are important for higher level learning and more complex academic content as the school years progress. Qualitative accounts suggest that children with dyslexia are often aware of the extra effort they need to put into their studies to achieve at comparable levels to their peers [ 13 , 14 ]. Similarly, longitudinal evidence suggests that many, but not all, experience relatively poor educational and occupational outcomes [ 15 , 16 ]. Furthermore, those who live and work with children with dyslexia frequently report concern for children’s mental health [ 11 , 14 , 17 ]. Indeed, there is considerable evidence to suggest that dyslexia is associated with a range of psychosocial difficulties in childhood including: reduced academic self-concept [ 18 ], poor reading self-efficacy [ 19 ], and elevated levels of internalising (e.g., anxiety) and externalising (e.g., aggression) symptoms indicative of poor mental health [ 20 , 21 , 22 ]. Taken together, these findings suggest that the mental health concerns of children with dyslexia can extend beyond the classroom into their everyday lives and may persist into adulthood. Understanding why children with reading difficulties, such as dyslexia, are at elevated risk of mental health concerns is now a stated research imperative [ 22 , 23 ].

1.2. The Current Review

The aim of the current review was to scope the extent and nature of the literature investigating factors which may influence the association between childhood dyslexia and internalising and externalising mental health concerns. In so doing, we aimed to identify gaps in the literature and use the findings of the review to inform decisions about directions for future research. Such research is needed to: (1) identify children who may be more (or less) at risk of mental health concerns, and (2) inform tailored mental health programs. Given this aim, a scoping rather than a systematic review was indicated. (For the purpose of this review, mental health concerns are defined according to an internalising/externalising dichotomy [ 24 , 25 ]. which is widely used in the child development literature and clinical settings. Internalising refers to expressions of emotional distress which are inwardly focused (e.g., withdrawal) whilst externalising refers to expressions of emotional distress which are outwardly focused (e.g., aggression).

A preliminary search for existing reviews and meta-analyses on this topic was conducted in April 2021 using the following databases: Prospero, Cochrane Database of Systematic Reviews, JBI Evidence Synthesis, Campbell Collaboration, and the Open Science Framework. We were unable to find any planned or existing reviews with the same focus as ours. Previous reviews have focused on investigating the strength and direction of the relationship between learning/reading difficulties and internalising symptoms [ 21 , 22 , 26 ] or learning difficulties and both internalising/externalising symptoms [ 27 ]. Another review by Haft and colleagues [ 28 ] offered an excellent preliminary discussion of protective factors for socio-emotional resilience in the context of developmental dyslexia but did not comprehensively search the literature of the time. Our review therefore offers an extension to the existing field.

We came to an a priori decision to present our findings using the three-tiered framework of (1) individual, (2) family, and (3) community-level factors developed by Haft and colleagues [ 28 ] and applied previously in research on dyslexia and child mental health [ 29 ]. This framework acknowledges that children’s mental health is influenced by a range of factors that relate to the child themselves (the individual), their family, and the communities in which they participate.

2. Materials and Methods

Our review was informed by the Joanna Briggs Institute (JBI) guidelines [ 30 ] and the PRISMA extension for scoping review checklist [ 31 ] which build on Arksey and O’Malley’s [ 32 ] foundational work on scoping review methodology. A protocol for the review was registered on the Open Science Framework on 04.05.2021 and is available to view at 10.17605/OSF.IO/GZ98X. Changes to the protocol that were applied during the review process are described in Appendix A .

2.1. Inclusion/Exclusion Criteria

Types of studies. Peer-reviewed primary research studies as well as theses/dissertations were considered for inclusion. When the same research was reported in a dissertation and journal article then the dissertation was excluded. No limits by date were set. Books, book chapters, study protocols, theoretical/opinion pieces, and previous reviews were excluded.

Population. Sources that included children (≤18 years) or adult perspectives (e.g., teachers, parents, or adult retrospective studies) on children’s mental health were considered so long as the child participants spoke an alphabetic language and were not being tested in a language other than their primary language.

Concept 1: Reading difficulties. Sources were included if they investigated internalising and/or externalising symptoms among children with recognised learning difficulties and/or word reading difficulties consistent with dyslexia. In keeping with DSM-5 diagnostic criteria, when the learning difficulties could be better explained by medical (e.g., hearing loss), neurodevelopmental (e.g., autism, intellectual disability), or socio-cultural factors (e.g., child poverty, second language learners, lack of educational opportunity) sources were excluded [ 10 ].

Concept 2: Mental health. Mental health concerns were defined as consisting of both/either internalising or externalising symptoms. Sources which used global measures of psychosocial functioning with sub-scales to measure internalising/externalising symptoms (e.g., Strengths and Difficulties Questionnaire (SDQ); Goodman, 1997 [ 33 ]), as well as those which measure a particular aspect of internalising/externalising symptoms (e.g., anxiety) were included. For qualitative research, self/parent/other description of emotional difficulties or mental health concerns were sufficient for inclusion.

Concept 3: Third factor. Sources were only included if they investigated a modifiable “third factor” in association with mental health concerns among children with learning/reading difficulties. By modifiable we are referring to a skill/attitude/behaviour (e.g., emotion regulation) which could become an intervention target for mental health programs designed for school-aged children with reading difficulties and/or their families and communities. For this reason, sources which solely investigated factors which are intractable (e.g., a child’s age); factors which cannot be modified once a child has reached primary school age (e.g., exposure to environmental toxins in-utero); or cognitive differences that may be difficult to modify (e.g., working memory, attention) were not included.

Context. Sources were not excluded based on geographical factors, the educational context of the children (e.g., mainstream, or specialised school), or whether children were receiving educational support or socio-emotional or reading intervention.

2.2. Search Strategy

An initial set of search terms were developed by the research team based on their current understanding of the literature. A 3-step search strategy was then followed as recommended by the JBI [ 30 ] (p. 418). At the first stage, the first author conducted a preliminary search of two databases (Scopus and ProQuest) on 10.06.2021. Based on the subject headings and keywords from relevant sources found in this preliminary search, an additional 4 keywords were added to the search terms ( Appendix B ).

The first author then conducted the second stage of the search on 11.07.2021 with an update on 17.05.2022. This included a search by subject heading and keyword of the OVID platform databases: Embase, PsychInfo, and OVID Medline, and a search by keyword on Scopus, ProQuest and CINHAL. Limits were set for English language and to peer-reviewed journals (where possible). The first author then conducted a search for dissertations/theses using the same keywords via the ProQuest Dissertation and Thesis database. See Table S2 for an example of a complete search of one database, PsychInfo. For the third and final stage of the selection process, the first author reviewed the reference lists of key sources and relevant reviews. An additional 8 sources were located at this stage.

2.3. Source Selection

A total of 10,810 sources (10,155 articles and 655 theses) were downloaded into Endnote [ 34 ] and duplicates removed by the Endnote de-duplication process and by hand. This resulted in a total of 7389 sources (6840 articles and 549 theses) being considered by title or title/abstract from the database search (a further 8 sources were added at a later stage from the third stage of the search). At this stage, a source selection tool (see Table S1 ) was developed and piloted by each member of the review team on a random selection of 25 sources. No one reviewer achieved less than 88% agreement with the first author and 4 members of the review team of 6 achieved 100% agreement with the first author. After this pilot, no changes to the subject headings or keywords of the search were considered necessary. However, the source selection tool was adjusted through team discussion to add further explanation and examples to aid the selection process. Throughout the source selection process, we further refined the inclusion/exclusion criteria as outlined in Appendix A .

Following this, a random sample of approximately 20% of sources was independently reviewed by two members of the review team (the first author and one other) by title and/or title/abstract. An inter-rater agreement of 88.6% was achieved. This was considered a satisfactory level of adherence to the selection criteria by the first author who then proceeded to select sources for consideration from the remaining 80% of total sources. Disagreement between reviewers was resolved by consensus by two members of the review team.

A similar process was followed for the next stage of the selection process; selection for inclusion/exclusion based on full text review. Approximately 20% of sources were independently screened by two members of the review team (the first author and one other) (inter-rater agreement of 82%) and the remainder was reviewed by the first author alone. Disagreements between authors, and/or any uncertainty by the first author, were resolved by group consensus. A record of the number of sources included/excluded at each stage of the source selection process and the reasons for exclusion are included below in Figure 1 .

PRISMA flow diagram http://prisma-statement.org/ (accessed on 28 January 2021) and adapted for this scoping review.

2.4. Data Extraction

A data extraction form for this review was adapted from models presented by the JBI manual and other authors [ 30 , 32 ] and developed with input from 3 members of the research team. The first author extracted the relevant data from each source and the fourth author (EH) checked the data extraction of 20% of sources. Based on EH’s review, one source was excluded and the participant details of another were corrected.

A total of 98 studies (12 theses and 86 peer-reviewed journal articles) met the inclusion criteria. Geographically, the scope of the studies was widespread. A total of 20 countries were represented; the largest numbers of studies originated from the USA (32), Italy (11), United Kingdom (10), and Israel (7). One study was a cross-cultural analysis of children’s mental health and coping strategies and included participants from Germany and Indonesia [ 35 ]. In terms of design, nine studies were qualitative, 10 incorporated a mixed methods approach, and the remainder were quantitative by design, incorporating either a longitudinal (21 studies) or cross-sectional (58 studies) approach. One of the longitudinal studies was a secondary analysis of four longitudinal studies from English-speaking countries [ 36 ].

3.1. Sampling Characteristics

Approximately one quarter of the included studies (27 studies; 27.5%) were not dyslexia specific, that is the researchers conducted their investigations using samples of children/adolescents with learning difficulties without specifying sub-type (reading, writing, mathematics difficulties) or specified that the children all had a diagnosis of SLD mixed-type (difficulties in at least two domains of reading, writing, and/or mathematics), see for example [ 37 ]. Researchers of the remaining 71 studies focused on developmental dyslexia or word reading difficulties consistent with a dyslexia diagnosis. A wide range of methods were used to determine group status, ranging from self-identification of reading difficulties, see for example [ 38 ], to extensive diagnostic and validation measures, see for example [ 39 ]. Furthermore, the severity of dyslexia (when reported) ranged from mild [ 40 ] to severe [ 41 ].

Sample sizes ranged from one (a case study) [ 42 ] to a study incorporating findings from four databases totaling 23,799 children [ 36 ]. In terms of the age and the stage of development of participants, the researchers of 46 studies focused exclusively on primary school-aged children (≤12 years; 46 studies); 17 on adolescents (≥12 years); and 27 included samples of young people whose ages ranged across these developmental periods (childhood–adolescence). Of the remainder, there were four longitudinal studies in which children were tracked from their primary to secondary school years [ 43 , 44 , 45 , 46 ]; three studies in which adults were interviewed about their childhood experiences with dyslexia, see for example [ 47 ]; and one study in which educators of children with dyslexia aged 5–18 years were interviewed [ 11 ]. In many of the longitudinal studies the researchers stopped assessing children in the primary school years, see for example [ 36 , 48 , 49 , 50 , 51 , 52 , 53 , 54 , 55 , 56 , 57 , 58 ].

3.2. Study Aims and Mental Health

There was great variety with regard to the aims of the included studies. For example, some researchers primarily aimed to investigate the relative contribution of SLD or attentional difficulties (e.g., co-morbid ADHD) in explaining internalising/externalising symptoms, see for example [ 59 ]. Many researchers had the primary aim of investigating the direction of effects between learning/reading difficulties and a range of psychosocial (e.g., self-esteem, social skills) variables, see for example [ 36 , 50 , 54 , 55 , 56 , 60 , 61 ]. In other studies, mental health was not included as an outcome variable. Rather, the relative influence of learning difficulties and/or behaviour/mental health on children’s social skills [ 62 ] or academic outcomes was investigated (e.g., grades, reading comprehension, secondary school completion), see for example [ 43 , 57 , 63 , 64 ]. However, in our estimation, the aim of 39 (40%) of the included studies aligned to that of this review; that is, the researchers aimed to better understand mental health in the context of childhood learning/reading difficulties. Not all researchers found dyslexia-related group differences in mental health concerns, see for example [ 65 ]. However, when differences were found, the balance of evidence suggests that reading difficulties precede internalising symptoms [ 36 , 46 , 48 , 50 , 52 , 54 ] whereas the direction of effects between dyslexia and externalising symptoms was less certain. Specifically, there is evidence that externalising symptoms are associated with attentional difficulties which may be present before school entrance but can worsen in response to school demands [ 46 , 66 , 67 ].

In the qualitative studies (and one mixed method study), mental health was investigated among children with reading difficulties broadly. Among the studies incorporating a quantitative approach, the researchers of 25 studies focused on internalising symptoms only and those of one study focused on externalising symptoms only [ 59 ]. Most researchers in this review (62; 63.3%) investigated both internalising and externalising symptoms. Many conducted their investigations of internalising/externalising symptoms with well-validated measures of global psychosocial functioning such as the Strengths and Difficulties Questionnaire (SDQ) (10 studies) and the Child Behaviour Checklist (CBCL); Achenbach (various versions, for example, [ 24 ]) (27 studies). Of the total sample of the included studies (qualitative and quantitative), the researchers of 39 studies combined perspectives (e.g., child and/or teacher and/or parent and/or both parents) on children’s mental health. However, in nearly half of the included studies (45; 45.9%, excluding the adult retrospective studies where adults reflected on their own childhood experiences) children’s perspectives on their mental health were not considered. Furthermore, a wide variety of child-reported measures of anxiety (11 measures) and depression (10 measures) were included in this review.

3.3. Third Factors

The research team formed 26 third factors categories (see Table 1 below) on the basis of common tools or measures and definitions provided in the publication. Each category was allocated to individual, family, and community-level factors. Many studies investigated factors on more than one of these levels. Where constructs are included in parathesis, they were deemed to be related to the primary third factor. Studies have been placed in a column depending on whether they primarily dealt with the third factor using qualitative, quantitative or both approaches. When a quantitative approach was taken, we have provided an indication of whether a statistical association between the third factor and the mental health concern in the context of learning difficulties was found, not found, or was not tested. Table S3 presents the summary of findings from all 98 studies included in this review including information regarding the direction of association between the variables.

“Third factors” and associations with internalising/externalising mental health (MH) concerns among children with reading/learning difficulties.

* Check Supplementary Materials for direction of association. † This category includes many domains of self-esteem (or related constructs), such as social self-esteem, scholastic self-esteem etc. Check Supplementary Materials for specific domain of self-esteem explored in the study.

4. Discussion

The aim of the current review was to scope the extent and the nature of the literature investigating factors which may influence the association between childhood dyslexia and internalising and externalising mental health concerns. In so doing, we aimed to identify gaps in the literature and use the findings of the review to inform decisions about directions for future research. Such research is needed to (1) identify children who may be more (or less) at risk of mental health concerns and (2) inform tailored mental health programs. To this end, our review identified a total of 98 sources (86 peer-reviewed journal articles and 12 theses) for inclusion dating from 1968. Twenty different countries were represented highlighting a global concern for the mental health of children who struggle to read. Our review includes: 70 studies in which individual-level factors (e.g., self-esteem) were explored; 39 in which family-level factors (e.g., the parent-child relationship) were explored; and 54 in which community-level factors (e.g., the teacher-child relationship) were explored. Most researchers investigated both internalising and externalising symptoms but those of 25 studies focused exclusively on internalising symptoms. One explanation is that internalising symptoms, specifically anxiety, have been highlighted as a particular mental health concern among children with dyslexia, see for example [ 21 , 22 ].

4.1. Social Experiences

The social skills (20 studies) and the social problems (29 studies) of children with word reading/learning difficulties were amongst the most studied “third factors” in this review. However, social difficulties were most often studied as correlates of learning difficulties, see for example [ 48 , 50 , 110 ], or poor academic performance [ 62 ] or attentional difficulties [ 59 , 129 ] that may co-occur, rather than as possible risk/protective factors for mental health. Furthermore, there was a focus on investigating children’s challenges (e.g., peer difficulties) from the parent and/or the teacher perspectives rather than their strengths. Children’s subjective feelings of loneliness were also under-studied, a finding consistent with Kwan and colleagues’ (2020) recent review [ 130 ]. Nevertheless, our review highlighted consistent links between bullying involvement and mental health concerns [ 29 , 58 , 124 , 125 ] and the protective function of friendship, see for example [ 77 , 117 ], in the context of childhood dyslexia. Differences between child, parent, and teacher reports of social difficulties were reported by some researchers, see for example [ 41 , 61 ], highlighting the importance of examining the context of social difficulties (e.g., school or home) in addition to gauging children’s own perspectives on their social strengths and challenges in future research.

4.2. Self-Esteem

Other individual-level third factors which have been widely studied in this field are self-esteem (and related constructs, e.g., self-concept/self-efficacy) (30 studies) and stress, coping and resilience (18 studies). A secondary analysis of four different longitudinal studies by McArthur and colleagues (2022) found a link between early reading difficulties and later anxiety, depression, and poor reading self-concept (beliefs about oneself as a reader) [ 36 ], suggesting that poor reading self-concept may be a risk factor for anxiety and depression. Consistent with this, Terras and colleagues (2009) found an association between low scholastic self-esteem (beliefs about oneself as a learner) and parent/child reported internalising symptoms [ 73 ] and Giovagnoli and colleagues (2020) found that adolescents with dyslexia who reported low levels of self-efficacy (belief in one’s ability) with regard to school tasks experienced more somatic (headaches, stomach aches) symptoms [ 39 ]. However, there were some mixed results across the field regarding both the domain of mental health (anxiety, depression, externalising) and the domain of self-esteem/self-efficacy (e.g., scholastic self-esteem, social self-esteem) involved. Furthermore, most studies which examined the associations between self-esteem and internalising/externalising symptoms were cross-sectional by design meaning that the direction of effects could not be determined. Disentangling the strength and direction of these associations, and factors that support children’s self-esteem in the context of dyslexia, would allow for the provision of timely and targeted support for children at risk of mental health concerns.

4.3. Coping and Resilience

Although over a decade old, Singer’s (2005, 2007) foundational work suggests that the strategies that children with dyslexia use to cope with school-related difficulties (such as teasing and poor grades) are instrumental in supporting their self-esteem (or not) [ 68 , 69 ]. For example, the children she interviewed used self-talk to either support or hinder their self-esteem. Consistent with this, Hossein and colleagues (2022) found that children with reading disorder who had more “grit” (i.e., perseverance) and “resilience” were less likely to experience anxiety (reported by teachers) and depression (reported by parents) [ 91 ]. Similarly, Giovagnoli and colleagues (2020) found that internalising symptoms among adolescents with dyslexia were associated with a tendency to react to school-related problems with hypervigilance, defined as an “excessive sense of alertness” [ 39 ] (p. 461). However, two other groups of researchers in this review did not find any group differences in coping related to learning/reading difficulties although higher levels of internalising symptoms were observed [ 35 , 94 ]. As such, we believe that an investigation of coping strategies, especially with regard to the school context, in relation to children’s mental health warrants further exploration.

4.4. Emotion Regulation and Academic Factors

Emotion regulation (the ability to understand and manage one’s emotions) is one aspect of coping which appears to be relatively under-studied in this field. Several studies in this review included an examination of factors related to emotion regulation, such as emotional intelligence, see for example [ 99 , 106 ]; focused on one aspect of emotion regulation, namely rumination [ 97 ]; or measured emotion regulation with a single item on a parent survey [ 29 ]. As a result, we believe that emotion regulation in the context of dyslexia has not been comprehensively explored. This represents a gap in the literature given that emotion regulation is strongly associated with mental health across the lifespan [ 131 , 132 , 133 ] and may be hindered in children who experience language/literacy difficulties in early childhood and beyond [ 134 , 135 , 136 ]. Certainly, there is evidence from experimental research that children with dyslexia, relative to controls, have more difficulty with recognising emotions in others (from facial and vocal cues) [ 137 , 138 ]. This may suggest difficulty with understanding their own emotions, a basic building block to emotion regulation [ 139 ]. Furthermore, many children with reading difficulties report experiencing heightened negative emotions (e.g., frustration, anger, sadness) in the school context [ 14 , 55 ] suggesting that effective emotion regulation may be a particularly salient intervention target for their mental health. Longitudinal studies using well-validated (child and parent/teacher versions) measures of specific socio-emotional competencies (e.g., emotion regulation, spoken language skills) are needed to test these associations over time and to build upon the existing work described in this review. Relatedly, children’s attitude to their learning and whether they succeed in their studies despite their learning challenges has been linked to their mental health, see for example [ 57 , 65 , 108 , 109 ], and warrants replication.

4.5. Family Factors

In terms of family factors, the association between children’s mental health and parental psychological variables (e.g., parental stress, anxiety, self-esteem, coping, emotion regulation); the quality of the child-parent relationship; family functioning/support; and parenting values/practices is well researched. A strength of this field is the widespread use of well-validated measures such as the Parenting Stress Index (PSI) and its short form [ 140 ] enabling a comparison of the findings across the field. Importantly, the PSI-SF has been validated for use with parents of children with a wide variety of mental, emotional, and behavioural difficulties [ 141 ]. However, much of the research in this section of the review has employed samples of children with a wide range of learning and attentional difficulties without specifying the sub-type, see for example [ 98 ]. This is a limitation given that parents of children with dyslexia may have unique strengths and challenges, see for example [ 83 , 113 ]. Furthermore, the perspectives of fathers and siblings were often absent and there are mixed results which require further exploration. Nevertheless, future researchers are encouraged to investigate the value of whole-family support within mental health promotion programs for children with dyslexia and other learning difficulties.

4.6. School and Community Factors

Certainly, the importance of having parents who understand dyslexia and provide both emotional and academic support was highlighted by many studies in our review, see for example [ 14 , 73 , 117 ]. However, when interviewed, people with lived experience of dyslexia (especially parents/teachers and older adolescents/young adults) also raised school/community-level concerns such as: teachers’ misunderstanding of dyslexia; a lack of accommodation and support for reading-related challenges; and experiences of stigma, shame, and discrimination in learning environments, see for example [ 11 , 14 , 42 , 47 , 121 ]. Findings such as these suggest that children with dyslexia and other learning difficulties may be vulnerable to low levels of “school connectedness”, a concept that describes perceptions of being understood, supported, and treated fairly at school (by peers and by teachers alike) [ 142 , 143 ]. Consistent with this, two sets of researchers in this review found evidence to suggest that school connectedness may be a particularly salient protective factor for the socio-emotional wellbeing of children with learning difficulties [ 107 , 126 ]. Similarly, Chiappedi and Baschenis [ 122 ] found that children with SLD who believed that their teacher understood and supported their learning disability reported significantly lower levels of anxiety. Future researchers are encouraged to replicate the aims of these studies using well-validated measures of school connectedness and dyslexia-specific samples. There is evidence from our review that children with dyslexia experience a different trajectory of socio-emotional difficulties during the school years than those with other special educational needs, see for example [ 50 ], and therefore need to be differentiated in future research.

4.7. Methodological Considerations

A strength of the field is the widespread use of well-validated measures of children’s psychosocial functioning such as the Strengths and Difficulties Questionnaire (SDQ)and the Child Behaviour Checklist (CBCL).This aids cross-cultural comparisons and the generalisability of each study’s findings. However, many studies included in this review examined group differences (e.g., children with and without reading difficulties) with regard to the third factor rather than testing associations between the third factor and mental health concerns. This limits our understanding of risk and protective factors for mental health. Furthermore, in many cases, researchers relied on parent/teacher, rather than child, perspectives on internalising symptoms, see for example [ 49 , 126 ]. Internalising symptoms may be difficult for an outsider to accurately assess; therefore, future research which gauges children’s own perspectives is encouraged. Furthermore, to advance understanding of risk/protective factors for specific (e.g., depression) mental health concerns then the use of well-validated measures which align to DSM criteria (such as the Revised Children’s Anxiety and Depression Scale (RCADS; Chorpita et al., 2000) [ 144 ] and include parent/child versions, rather than (or in addition to) broad-band measures of psychosocial functioning such as the SDQ, are needed.

Additionally, our review highlighted a bias in the field towards investigating the psychosocial wellbeing of children rather than adolescents. Indeed, several researchers of longitudinal studies included in this review stopped tracking children’s socio-emotional functioning by the time they reached the upper primary school years, see for example [ 48 , 50 , 55 ]. This presents a significant gap in the literature. Adolescence is a risk period for the onset of several mental health concerns [ 145 ] which may exacerbate for children with dyslexia due to concerns about the changes in the secondary school environment such as the increased difficulty of reading and workload, see for example [ 39 ]. To complement the field, future researchers are encouraged to track children with and without dyslexia as they transition from primary to secondary school and beyond. Current research suggests that this school transition may be a risk period for both school connectedness and mental health concerns among children broadly [ 146 , 147 ] and may have specific relevance for children with reading difficulties [ 46 ].

4.8. Limitations and Future Research

The current review is limited by our exclusion of studies in which reading ability was measured as a continuous variable. We are aware of studies of this kind which would have addressed our research aim, see for example [ 148 ]. The decision to exclude these studies was influenced by the feasibility of the review process and the readability of a review with 100+ sources. Additionally, a quality assessment of studies in this field is warranted given the range of methodological limitations which has been briefly discussed. Nevertheless, we identified a variety of individual, family, and community-level factors which may influence mental health in the context of developmental dyslexia, located gaps in the literature, and offered suggestions for future research. Future research which examines associations between aspects of children’s socio-emotional competencies (e.g., emotion regulation), domains of self-esteem, coping strategies, school connectedness, and sub-types of mental health (e.g., depression) in the context of childhood dyslexia is encouraged. This will improve understanding about risk and protective factors for the mental health of children with dyslexia. In terms of methodological factors, our review highlighted a need for more longitudinal work (especially over the transition to secondary school and beyond), which includes dyslexia-specific samples alongside typically developing comparison groups and includes child as well as parent/teacher perspectives wherever possible.

5. Conclusions

The current review highlighted a broad range of individual, family, and community-level factors which may influence mental health in the context of developmental dyslexia but relatively few studies which tested associations between third factors and mental health. We identified several gaps in the literature regarding both the content (e.g., school connectedness) and methods (e.g., child perspectives on mental health) of current research and proposed recommendations for future research. Such research is needed to help to identify children who are more (or less) at risk of mental health concerns and to inform tailored mental health promotion programs for children with dyslexia.

Acknowledgments

With thanks to Sophie Haywood and Lachlan Toms for their assistance with the source selection process and the Health Sciences Librarians atCurtin University, for ongoing support throughout the review process.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/ijerph20021653/s1 , Table S1: Source selection tool, Table S2: An example of a database search, Table S3: Summary of findings from studies included in the review.

Appendix A. Changes to the Protocol and Inclusion/Exclusion Criteria

• Changes to title and objective/aim of the review : We removed the reference to “risk and protective factors” in the wording of the title and the objective/aim of the review to better reflect the broad range of studies included.
• Changes to the inclusion/exclusion criteria : We decided to exclude studies which measured reading as a continuous variable for feasibility reasons. The focus of our review is children with developmental dyslexia rather than other types of reading difficulty (e.g., difficulty with reading comprehension). For this reason, we specified dyslexia or word reading difficulties in our inclusion criteria. Further refinement to the inclusion/exclusion process occurred by group consensus during the source selection process. Specifically, we decided to exclude studies which measured mental health specifically within the context of the COVID-19 pandemic, as well as past reviews, intervention studies, and all “cognitive” third factors such as a child’s Theory of Mind ability and their executive functioning.
• Changes to the source selection process: Due to feasibility reasons, approximately 20% of sources selected for full text review and 20% of data extraction was independently screened by a second author. We used Excel spreadsheets in addition to Endnote to manage the source selection process, as this method facilitated better sharing of sources between the review team.
• Changes to the data extraction form: We applied some changes to the data extraction form by group consensus during the source selection/data extraction process. Specifically, we removed reference to the strengths and the limitations of each study and suggestions for future research and added a column to describe the context of the study and the “third factor” included in the study.

Appendix B. Keyword Search Terms

• The keywords in bold were added after the first stage of the search.
• Child* or teen* or adolescen* or youth or young AND dyslexi* or “poor reader*” or “reading abilit*” or “reading skill*” or “reading achievement” or “reading difficult*” or “reading disorder*” or “reading impair*” or “reading delay” or “reading defici* ” or “reading disabilit*” or “literacy difficult*” or “learning disorder” or “struggl* adj3 read*” AND internali?ing or “somatic complaint*” or anxious or anxiety or depression or depressed or “emotion* difficult*” or “emotion* problem*” or psycho?social or socio?emotion* or withdrawn or “mental health” or hyperactiv* or impulsiv* or externali?ing or “conduct problem*” or “behaviour* problem*” or “behavior* problem*” or aggress*

Funding Statement

Adrienne Wilmot is supported by an Australian Government Research Training Program scholarship. Mark Boyes is supported by the National Health and Medical Research Council, Australia (Investigator Grant 1173043).

Author Contributions

Conceptualisation and methodology, A.W., P.H., S.L. and M.B. Data curation and writing—original draft preparation, A.W. Writing—review and editing, A.W., P.H., S.L., E.H. and M.B. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Data availability statement, conflicts of interest.

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

A Systematic Review of Research Dimensions Towards Dyslexia Screening Using Machine Learning

• REVIEW PAPER
• Published: 23 January 2023
• Volume 104 , pages 511–522, ( 2023 )

Cite this article

• Tabassum Gull Jan   ORCID: orcid.org/0000-0002-4882-3247 1 &
• Sajad Mohammad Khan 1  

747 Accesses

5 Citations

Explore all metrics

Dyslexia is the hidden learning disability, neurobiological in origin wherein students face hard time in accurate or fluent word recognition, connecting letters to the sounds. In India, index of dyslexia is increasing exponentially. The level of difficulty of dyslexic children varies from person to person. Their brain is normal; often very “intelligent,” but with strengths and capabilities in areas other than the language area. Henceforth, such students are suffering from low self-esteem, are bipolar in nature, have negative feelings and depression. Therefore, early detection and evaluation of dyslexic students is very important and need of the hour. In this review paper, the authors have summed up various research dimensions toward dyslexia detection. This paper principally focuses on the machine learning techniques for dyslexia screening which includes applications covering different machine learning-based approaches, game-based techniques and image processing techniques for designing various assessments and assistive tools to support and ease the problems encountered by dyslexic people. This review paper identifies various knowledge gaps, current issues and future challenges in this research domain. It mainly focuses on various machine learning applications toward detection of dyslexia.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price includes VAT (Russian Federation)

Instant access to the full article PDF.

Rent this article via DeepDyve

Institutional subscriptions

Similar content being viewed by others

Applications of artificial intelligence in forensic sciences: Current potential benefits, limitations and perspectives

Nicola Galante, Rosy Cotroneo, … Michelangelo Bruno Casali

Advances and challenges in the detection of academic stress and anxiety in the classroom: A literature review and recommendations

Laura P. Jiménez-Mijangos, Jorge Rodríguez-Arce, … José Javier Reyes-Lagos

Detection of Autism Spectrum Disorder in Children Using Machine Learning Techniques

Kaushik Vakadkar, Diya Purkayastha & Deepa Krishnan

Abbreviations

Support Vector Machine

Decision Tree

Random Forest

Naive Bayes

Magnetic Resonance Images

Electroencephalogram

Support Vector Classifier

Standard Deviation

Particle Swarm Optimization

Convolutional Neural Network

Artificial Neural Network

Logistic Regression

A. Frid, Z. Breznitz, An SVM based algorithm for analysis and discrimination of dyslexic readers from regular readers using ERPs. In: IEEE 27th convention of electrical & electronics engineers in Israel (IEEEI) , Israel. pp 1–4, (2012)

A. Palacios, L. Sánchez, I. Couso, S. Destercke, An extension of the FURIA classification algorithm to low quality data through fuzzy rankings and its application to the early diagnosis of dyslexia. Neurocomputing 176 , 60–71 (2016). https://doi.org/10.1016/j.neucom.2014.11.088

Article   Google Scholar  

Y. Lakretz, G. Chechik, N. Friedmann, M. Rosen-Zvi, Probabilistic graphical models of dyslexia in proceedings of the 21th ACM SIGKDD international conference on knowledge discovery and data mining. ACM , pp. 1919–1928. (2015). doi: https://doi.org/10.1145/2783258.2788604 .

L. Rello, M. Ballesteros, Detecting readers with dyslexia using machine learning with eye tracking measures in proceedings of the 12th international web for all conference. ACM , pp.1–8, (2015). doi: https://doi.org/10.1145/2745555.2746644

A.Z.A. Zainuddin, K.Y. Lee, W. Mansor, Z. Mahmoodin, Optimized KNN classify rule for EEG based differentiation between capable dyslexic and normal children. In: 2016 IEEE EMBS Conference on Biomedical Engineering and Sciences (IECBES) . pp. 685–688, (2016). doi: https://doi.org/10.1109/IECBES.2016.7843537 .

P. Plonski et al., Multi-parameter machine learning approach to the neuroanatomical basis of developmental dyslexia. Hum. Brain Mapp. 38 (2), 900–908 (2016). https://doi.org/10.1002/hbm.23426

M.N. Benfatto, G. Seimyr, J. Ygge, T. Pansell, A. Rydberg, C. Jacobson, Screening for dyslexia using eye tracking during reading. PLoS ONE 11 (12), e0165508 (2016). https://doi.org/10.1371/journal.pone.0165508

H. M. Al-Barhamtoshy, D.M. Motaweh, Diagnosis of dyslexia using computation analysis. In: 2017 International Conference on Informatics, Health Technology (ICIHT) , pp. 1–7, (2017). doi: https://doi.org/10.1109/ICIHT.2017.7899141

I. Smyrnakis et al., RADAR: a novel fast-screening method for reading difficulties with special focus on dyslexia. PLOS ONE. 12 (8), e0182597 (2017). https://doi.org/10.1371/journal.pone.0182597

L. Rello, E. Romero, M. Rauschenberger, A. Ali, K. Williams, J.P. Bigham, N.C. White (2018) Screening dyslexia for English using HCI measures and machine learning. In: Proceding International Conference Digital Health . pp. 80–84. doi: https://doi.org/10.1145/3194658.3194675

S.S.A. Hamid, N. Admodisastro, N. Manshor, A. Kamaruddin, A.A.A. Ghani, Dyslexia adaptive learning model: student engagement prediction using machine learning approach. In: Recent Advances on Soft Computing and Data Mining: Advances in Intelligent Systems and Computing . R. Ghazali, M. Deris, N. Nawi, J. Abawajy, (Eds.) Cham, Switzerland: Springer, pp. 372–384, (2018). doi: https://doi.org/10.1007/978-3-319-72550-5_36

R.U. Khan, J.L.A. Cheng, O.Y. Bee, Machine learning and dyslexia: diagnostic and classification system (DCS) for kids with learning disabilities. Int. J. Eng. Technol. 7(3): 97–100, (2018)

H. Perera, M. Shiratuddin, K. Wong, K. Fullarton, EEG signal analysis of writing and typing between adults with dyslexia and normal controls. Int. J. Interact. Multim. Artific. Intell. 5 (1), 62 (2018)

Google Scholar  

F.J. Martinez-Murcia, A. Ortiz, R., Morales-Ortega, P.J. Lopez, J.L. Luque, Castillo-Barnes, J.M. Górriz, Periodogram connectivity of EEG signals for the detection of dyslexia. In: International Work-Conference on the Interplay between Natural and Artificial Computation . Springer: Cham. pp. 350–359. (2019)

A Jothi Prabha, R. Bhargavi, R. Ragala Predictive Model for Dyslexia from Eye Fixation Events.  Int. J. Eng. Adv. Technol. 9(13): 235–240, (2019)

K. Spoon, D. Crandall, K. Siek, Towards detecting dyslexia in children’s handwriting using neural networks. In: Proc. Int. Conf. Mach. Learn. AI Social Good Workshop , (2019). pp. 1–5.

K. Spoon, K. Siek, D. Crandall, M. Fillmore, Can we (and should we) use AI to detect dyslexia in children’s handwriting? In: Proc. Artif. Intell. Social Good (NeurIPS) , (2019) pp. 1–6.

A. Jothi Prabha, R. Bhargavi, Predictive model for dyslexia from fixations and saccadic eye movement events. Comput. Methods Prog. Biomed. 195 , 105538 (2020). https://doi.org/10.1016/j.cmpb.2020.105538

P. Tamboer, H.C.M. Vorst, S. Ghebreab, H.S. Scholte, Machine learning and dyslexia: classification of individual structural neuroimaging scans of students with and without dyslexia. NeuroImage Clin. 11 , 508–514 (2016). https://doi.org/10.1016/j.nicl.2016.03.014

X. Feng et al., Dyslexic children show atypical cerebellar activation and Cerebro-cerebellar functional connectivity in orthographic and phonological processing. The Cerebellum. 16 (2), 496–507 (2016). https://doi.org/10.1007/s12311-016-0829-2

Article   MathSciNet   Google Scholar  

Z. Cui, Z. Xia, M. Su, H. Shu, G. Gong, Disrupted white matter connectivity underlying developmental dyslexia: A machine learning approach. Human Brain Mapp. 37 (4), 1443–1458 (2016). https://doi.org/10.1002/hbm.23112

F. Morken, T. Helland, K. Hugdahl, K. Specht, Reading in dyslexia across literacy development: a longitudinal study of effective connectivity. Neuroimage. 144 , 92–100 (2017). https://doi.org/10.1016/j.neuroimage.2016.09.060

A. Frid, L.M. Manevitz, Features and machine learning for correlating and classifying between brain areas and dyslexia. (2018) arXiv preprint arXiv:1812.10622

Z. Rezvani, M. Zare, G. Zaric, M. Bonte, J. Tijms, M. Van der Molen, G.F. Gonzalez, Machine learning classification of dyslexic children based on EEG Local network features. Bio Rxiv. 57 , 1–23 (2019)

G. Richard, M. Serrurier, Dyslexia and Dysgraphia prediction: a new machine learning approach. (2020) arXiv preprint arXiv:2005.06401

L. Rello, K. Williams, A. Ali, N. Cushen White, J.P. Bigham, Dytective: towards detecting dyslexia across languages using an online game. In: Proc. W4A’16, Montreal, Canada. ACM Press (2016)

L. Rello, M. Ballesteros, A. Ali, M. Serra, D. Alarc´on, J.P. Bigham, Dytective: Diagnosing risk of dyslexia with a game. In: Proc Pervasive Health’16. Cancun: Mexico, (2016)

N. Zygouris, F. Vlachos, A. Dadaliaris, P. Oikonomou, G.I. Stamoulis, D. Vavougios, E. Nerantzaki, A. Striftou, The implementation of a web application for screening children with dyslexia. In: 19th international conference on interactive collaborative learning. Springer: Cham. pp. 415–423, (2016)

N.C. Zygouris, F. Vlachos, A.N. Dadaliaris, E. Karagos, P. Oikonomou, A. Striftou, G.I. Stamoulis, New tasks for a dyslexia screening web application. In: international conference on interactive collaborative learning. Springer: Cham. pp. 263–271, (2018)

A. Jothi Prabha, R. Bhargavi, Prediction of dyslexia from eye movements using machine learning. IETE J. Res. (2019). https://doi.org/10.1080/03772063.2019.1622461

V.F. Martins, T. Lima, P.N.M. Sampaio and M. de Paiva, Mobile application to support dyslexia diagnostic and reading practice.  2016 IEEE/ACS 13th International Conference of Computer Systems and Applications (AICCSA) , (2016), pp. 1–6, doi: https://doi.org/10.1109/AICCSA.2016.7945710 .

S. Zahia, B. Garcia-Zapirain, I. Saralegui, B. Fernandez-Ruanova, Dyslexia detection using 3D convolutional neural networks and functional magnetic resonance imaging. Comput. Methods Prog. Biomed. 197 , 105726 (2020). https://doi.org/10.1016/j.cmpb.2020.105726

S. Karande, R. Sholapurwala, M. Kulkarni, Managing specific learning disability in schools in India. Indian Pediatr. 48 (7), 515–520 (2011). https://doi.org/10.1007/s13312-011-0090-1

S. Singh et al., Specific learning disability: a 5 year study from India. Int. J. Contemp. Pediatr. 4 (3), 863 (2017). https://doi.org/10.18203/2349-3291.ijcp20171687

S. Bandla, G.D. Mandadi, A. Bhogaraju, Specific learning disabilities and psychiatric comorbidities in school children in South India. Indian J. Psychol. Med. 39 (1), 76–82 (2017). https://doi.org/10.4103/0253-7176.198950

A. Ortiz, Periodogram connectivity of EEG signals. Int. Work. Conf. Interp. Nat. Artif. Comput. 1 , 350–359 (2019). https://doi.org/10.1007/978-3-030-19591-5

A. Ortiz, P.J. López, J.L. Luque, F.J. Martínez-Murcia, D.A. Aquino-Britez, J. Ortega, An anomaly detection approach for dyslexia diagnosis using EEG Signals. Lect. Notes Comput. Sci. Include. Subser. 11486 , 369–378 (2019). https://doi.org/10.1007/978-3-030-19591-5_38

H. Xue et al., Resting-state EEG reveals global network deficiency in dyslexic children. Neuropsychologia. 138 , 107343 (2020). https://doi.org/10.1016/j.neuropsychologia.2020.107343

P. Christodoulides et al., Classification of EEG signals from young adults with dyslexia combining a brain computer interface device and an interactive linguistic software tool. Biomed. Signal Process. Control. 76 (7), 103646 (2022). https://doi.org/10.1016/j.bspc.2022.103646

S.K. Parmar, O.A. Ramwala, and C.N. Paunwala, Performance evaluation of SVM with non-linear kernels for EEG-based dyslexia detection. IEEE Reg. 10 Humanit. Technol. Conf. R10-HTC . (2021). doi: https://doi.org/10.1109/R10-HTC53172.2021.9641696 .

N.J. Gallego-Molina, A. Ortiz, F.J. Martínez-Murcia, M.A. Formoso, A. Giménez, Complex network modeling of EEG band coupling in dyslexia: an exploratory analysis of auditory processing and diagnosis. Knowledge-Based Syst. 240 , 108098 (2022). https://doi.org/10.1016/j.knosys.2021.108098

A. Oliaee, M. Mohebbi, S. Shirani, R. Rostami, Extraction of discriminative features from EEG signals of dyslexic children; before and after the treatment. Cogn. Neurodyn. (2022). https://doi.org/10.1007/s11571-022-09794-2

G.P. Pralhad, A. Joshi, M. Chippa, G. Mishra, S. Kumar. Dyslexia prediction using machine learning. pp. 3–8 (2021)

G.D. Germano, B.P. de Alexandra, C. César, S.A. Capellini, Screening protocol for early identification of Brazilian children at risk for dyslexia. Front. Psychol. 8 , 1–13 (2017). https://doi.org/10.3389/fpsyg.2017.01763

H. Perera, M.F. Shiratuddin, K.W. Wong, A review of electroencephalogram-based analysis and classification frameworks for dyslexia. Lect. Notes Comput. Sci. including Subser. 9950 , 626–635 (2016). https://doi.org/10.1007/978-3-319-46681-1_74

N.A.M. Yuzaidey, N.C. Din, M. Ahmad, N. Ibrahim, R.A. Razak, D. Harun, Interventions for children with dyslexia: A review on current intervention methods. Med. J. Malaysia 73 (5), 311–320 (2018)

H. Perera, M.F. Shiratuddin, K.W. Wong, Review of EEG-based pattern classification frameworks for dyslexia. Brain Inform. (2018). https://doi.org/10.1186/s40708-018-0079-9

S. Mulakaluri, G.S. Girisha, “Review: Mass Screening framework for children with dyslexia using IOT and computing analysis. EAI Endorsed Trans. Internet Thing. 6 (21), 165504 (2020). https://doi.org/10.4108/eai.13-7-2018.165504

D. Aquino-Brítez et al., Optimization of deep architectures for eeg signal classification: An automl approach using evolutionary algorithms. Sensors 21 (6), 1–21 (2021). https://doi.org/10.3390/s21062096

S. Güven, N. Friedmann, “Even in predictable orthographies: Surface dyslexia in Turkish. Sci. Stud. Read. 26 , 1–25 (2022). https://doi.org/10.1080/10888438.2022.2058399

Download references

Acknowledgements

This work is financially supported by Department of Science & Technology, Government of India, under DST INSPIRE Fellowship Scheme bearing registration Number IF190563.

Author information

Authors and affiliations.

Department of Computer Science, University of Kashmir, Srinagar, J&K, India

Tabassum Gull Jan & Sajad Mohammad Khan

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Tabassum Gull Jan .

Ethics declarations

Conflict of interest.

The authors have no conflict of interest in this work.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Jan, T.G., Khan, S.M. A Systematic Review of Research Dimensions Towards Dyslexia Screening Using Machine Learning. J. Inst. Eng. India Ser. B 104 , 511–522 (2023). https://doi.org/10.1007/s40031-023-00853-8

Download citation

Received : 23 November 2021

Accepted : 01 January 2023

Published : 23 January 2023

Issue Date : April 2023

DOI : https://doi.org/10.1007/s40031-023-00853-8

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Machine learning
• Find a journal
• Publish with us
• Track your research

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals

Dyslexia articles from across Nature Portfolio

Dyslexia is a disorder characterized by an impaired ability to comprehend written and printed words or phrases despite intact vision. It can be developmental or acquired, but in either case it interferes with academic achievement or with activities of daily living that require reading skills.

Latest Research and Reviews

Action video games normalise the phonemic awareness in pre-readers at risk for developmental dyslexia

• Sara Bertoni
• Chiara Andreola
• Andrea Facoetti

Disrupted network interactions serve as a neural marker of dyslexia

Functional connectivity analyses demonstrate that individuals with dyslexia process written stimuli via a dorsal decoding route and show stronger reading-related interaction with the right cerebellum, while also providing evidence for aberrant task-related connectivity as a neural marker for dyslexia that directly impacts behavioural performance.

• Sabrina Turker
• Philipp Kuhnke
• Gesa Hartwigsen

Regular rhythmic primes improve sentence repetition in children with developmental language disorder

• Anna Fiveash
• Enikő Ladányi
• Barbara Tillmann

Cerebro-cerebellar networks facilitate learning through feedback decoupling

Behavioral feedback is critical for learning, but it is often not available. Here, the authors introduce a deep learning model in which the cerebellum provides the cerebrum with feedback predictions, thereby facilitating learning, reducing dysmetria, and making several experimental predictions.

• Ellen Boven
• Joseph Pemberton
• Rui Ponte Costa

Visuo-spatial attention deficit in children with reading difficulties

• Sandro Franceschini

Spatial attention in encoding letter combinations

• Mahalakshmi Ramamurthy
• Alex L. White
• Jason D. Yeatman

News and Comment

Reforming learning disorder diagnosis following covid-19 educational disruption.

Current diagnostic criteria for learning disorders are insufficient because of ongoing COVID-19-related educational disruption. Diagnostic criteria for learning disorders should be modified to reduce the risk of misdiagnosis and ensure timely intervention.

• Mary K. Colvin
• Jennifer Reesman
• Tannahill Glen

Visual attention deficits in developmental dyslexia cannot be ascribed solely to poor reading experience

• Muriel Lobier
• Sylviane Valdois

Visual attention span deficits and assessing causality in developmental dyslexia

• Usha Goswami

Auditory thalamus is dysfunctional in dyslexia

• Man Tsuey Tse

Poor attention linked to dyslexia

Quick links.

• Explore articles by subject
• Guide to authors
• Editorial policies

• Skip to main content
• Keyboard shortcuts for audio player

'Our kids are not OK,' child psychiatrist Harold Koplewicz says

Terry Gross

The founder of the Child Mind Institute explains why young people are experiencing unprecedented levels of anxiety and depression — and what parents can do about it. His book is Scaffold Parenting.

TERRY GROSS, HOST:

This is FRESH AIR. I'm Terry Gross. There are so many reasons for children to be anxious today beyond all the standard childhood problems. There's the setbacks from the COVID lockdown, mass shootings in schools, feelings they're not measuring up to the great lives they see represented on social media, fears about the whole planet being in jeopardy. It's hardly unusual for parents to be unsure how to handle their child's anxiety, depression, learning problems, anger, tantrums. And it can be difficult for parents to evaluate whether their child should see a therapist or take medication.

My guest, child psychiatrist Harold Koplewicz, has dealt with these issues with many children and their parents. And there have been times he's been confounded about issues his own children faced. He's the founding president of the Child Mind Institute. Its stated mission is transforming the lives of children and families struggling with mental health and learning disorders by giving them the help they need to thrive. The institute also conducts related research.

From 1997 to 2009, he was the first director of the NYU Child Study Center. Koplewicz recently stepped down from his 25-year tenure as editor-in-chief of the Journal of Child and Adolescent Psychopharmacology. His latest book is titled "Scaffold Parenting: Raising Resilient, Self-Reliant And Secure Kids In An Age Of Anxiety." Dr. Harold Koplewicz, welcome to FRESH AIR. What are some of the problems and anxieties you're seeing now that you can connect to outside problems, like the COVID lockdown and its lingering aftereffects? How are you seeing that manifest in the children's anxieties that you're seeing?

HAROLD KOPLEWICZ: Well, I think our kids are not OK. And unfortunately, they weren't doing very well before COVID. But COVID has had a negative effect on all children. Children with mental health disorders and kids who are typically developing children being locked up for two years and living with fear that somebody close to you - someone near and dear - will die is very problematic. And we also know that 1 million Americans did die, which means that about 170,000 American children lost a caregiver or a parent.

And if we go back to 2001, after 9/11, we lost 3,000 Americans. And I can tell you that in New York, in certain pockets - Staten Island, where there were a lot of firemen, and Manhasset, where there were a lot of finance people who were in the building, and certainly people around ground zero - it was very hard to get kids to go back to school. Attendance rates didn't return to 9/10 - to September 10 levels for over a year and sometimes even longer.

So we do know that this kind of traumatic event is going to have lingering effects. And we have seen increases in anxiety disorders and in depression, particularly in girls but certainly even in boys. There are higher rates of kids trying to hurt themselves. And there is even an increase in the number of young people who have committed suicide. So there is no doubt that we had a problem before. And we have a greater problem now.

GROSS: The average child isn't necessarily, like, watching cable news or reading the newspaper. But you pick up a lot of this on social media. And it's also just in the air. Like, everybody's talking about these issues, like, environmental catastrophe, you know, political divisions. Is this the end of democracy? Is the planet burning? I mean, you're just - it's just in the air now.

KOPLEWICZ: Well, you know, there's something dramatically changed between 2010 and 2018. So the numbers start to jump when we started looking at children's mental health. There were higher rates of visits to emergency rooms by kids for suicidal thought and suicidal behavior. And the increase in the number of kids who died from suicide went from around 5,000 to 6,000. Now, just think about that. If it was diabetes, if it was cancer, that would have made the front page of every newspaper every single day. It would be on cable news 24/7. And somehow, we don't take mental health disorders as seriously as we take physical disorders.

And so, you know, what happened between 2010 and 2018 is that all of us started carrying a device with us that connected us to everybody on the planet 24/7. And that definitely had a negative effect on a certain percentage of the population. So I want to be clear that social media is not like smoking. It doesn't - it's not terrible for everyone. But it is particularly bad for kids who have mental health disorders. And we've really looked at this very carefully at the Child Mind Institute, where we had done a study before COVID that was looking for an objective test - a biological test. Psychiatry is the only discipline in medicine that doesn't have an objective test - doesn't have a chest X-ray or a blood test or a strep test. And therefore, that's the holy grail, right? We make the diagnosis with clinical information, which is how you start all diagnosis in every part of medicine. But you can confirm it with an EKG or with a brain scan. So psychiatry is missing that.

And so we started something called the Healthy Brain Network, where we offered any parent who was worried about their child - who was between the ages of 5 and 21 - a free psychiatric evaluation, free neuropsych testing, which looks for learning disabilities, a functional MRI and EEG, physical fitness, cardiovascular status, nutritional status. And this became the - and is still the largest collection of the developing brain of kids 5 to 21 that's ever been collected. And we share it with scientists around the world, who make an agreement with us that they won't try to find out who the subjects are.

GROSS: Wait. So is the point of this to figure out, is there a - like, a biological diagnosis you can make? Does the cohort of people who have, like, depression or anxiety or whatever share certain biological markers? Is that the point?

KOPLEWICZ: That would be the point. The real trick is, can you tell the difference between one atypical child and another? Not the difference between a typical developing child and someone who may have a mental health disorder or a learning disorder but the difference between Terry, who has anxiety, and Harold, who has depression. And is there something on the EEG or on the functional MRI? Can we find a definitive objective test? But the good news here is that when you collect all this data - and it turns out that 9% of the 7,000 kids that participated did not have a disorder. They had symptoms, but they didn't meet psychiatric criteria for a diagnosis. You now have described, very accurately and very specifically, phenotypically what these kids look like. And then you get COVID. And you find that their use of social media jumps. They are using the internet six to eight hours a day. And a...

GROSS: All the kids in the study?

KOPLEWICZ: No, no, no. Just a large percentage of them. And we start defining that as problematic internet usage. Not only are you using it a lot, but when you force them to stop, they get distressed. It almost feels like an addiction, right? And we do know that - it turns out for the 9%, who are typically developing kids, that when you use the internet more than six to eight hours a day, you will sleep less. You will exercise less. And you'll have less interactions in real life. All three of them are important for healthy brain development, but you don't become mentally ill. However, if you have a mental health disorder and you start behaving that way, your symptoms get worse. It's almost like a toxic agent. It turns out that the internet usage of over six to eight hours a day can make your symptoms of depression, your symptoms of ADHD significantly worse, which is a really important phenomena.

GROSS: Why do you think that is?

KOPLEWICZ: Well, it's a very good question. Why? Our guess is that for these kids, someone who has depression, they're already socially more isolated than the average person, and they start losing their skill set and their ambition to interact with the rest of the world. Kids with ADHD can get very hyper-focused with certain activities and at times feel very lost, very impulsive, feel very often like a failure when they can't pay attention in school or are missing things that everyone else is picking up. So what's important about this is that if you're a parent and you know your child has one of these disorders, you have to be very aware that their usage of social media, it could potentially be toxic and it has to be controlled. It can't be unlimited. Not that it's good for anyone to have unlimited, but it's particularly bad for those kids.

So we know that social media was out there between 2010 and 2018. And unfortunately, there's no regulation on it. And it means that parents have to be more aware. I mean, I think of it as, you know, a jungle, right? The jungle is an exciting place, very nutritious fruit and vegetables and lots of terrific stuff. Maybe medicines even can get discovered in the jungle. But it also has snakes, it also has dangerous plants that can kill you, it also has animals. And therefore, if you're going to let your child participate, you should be a very active participant in that permission.

GROSS: So if you think that social media can be very harmful to certain children, how would you suggest parents try to limit their time on social media? That's something that is really hard to do.

KOPLEWICZ: I think it is challenging, but I think it's very doable. We also have some good data. We know that parents who are using the internet in a, you know, problematic way are more likely to have kids that are doing it. Parents have to model this. They have to have periods where, we're putting the phone away at nighttime, and you're not allowed to look at it because we want you to sleep. We do want to look and see how much time you're spending on it, and we want you to be aware of how much time you're spending on it. So it's not, you know, punitive. It's a collaboration, especially if they're a teenager or a pre-teen. But I also think that, you know, it's time for us to get much more sophisticated about this.

GROSS: I want to talk with you a little bit about suicide since you brought it up. And I want to ask you first - just in terms of our show, we always give warnings when we're going to be talking about suicide. And we always give the suicide prevention hotline number, the idea being that hearing talk about suicide can almost be encouraging to someone who has had suicidal ideation. So do you think that's helpful?

KOPLEWICZ: Well, I think it's important to recognize that even if it's a small percentage, to give people that information - that lifeline is very important - and also to let them know that they're not alone. So I think the way to think about this is, why are teenagers so much more at risk than you or me? And the way to think about a teenager is, they feel everything. They're boiling. They're freezing. I hate you. I love you. You know, what happened to I'm warm or it's a little cool in here? That doesn't happen. And in some ways, it's really kind of terrific because they are so creative and they see opportunity everywhere. And they don't recognize risk very well.

I mean, there's some really interesting studies of a teenage boy who goes and picks up a friend to come into his car. And the teenager driver is wearing a seatbelt, and the teenage male who sits down next to him doesn't put a seatbelt on, and the teenage driver takes his seatbelt off. He goes and picks up a girl, and the girl gets into the car and she puts her seatbelt on, and the teenage driver now puts his seatbelt on.

So they're very easily moved by their peer group in a way that they hadn't been before. And parents should note this, that even though the peer group becomes significantly more influential when you're a teenager, parents are still the most influential factor in a kid's life. But it's important that parents keep talking, keep sharing their viewpoint, keep listening to their kid's viewpoint and not back off because their kids say, well, everyone's doing it.

GROSS: A child comes into your office, let's say a teenager comes into your office. You think that the possibility of this teenager attempting suicide is real. What do you do to try to prevent that from happening?

KOPLEWICZ: Well, it really depends on how serious they are about the attempt. Do they have a plan? Have they been thinking about it a long time? Have they stopped doing their usual pleasurable experiences? They no longer are hanging out with friends or not eating the food that they love. And you have to really recognize that if they are very serious about it, you have to intervene. You have to save their lives. You have to either say to them, I don't feel you're safe, or ask them if they feel safe, and then sometimes make the decision that they have to be in an environment where they'll be watched, in a hospital. Or you'll talk to their parents and see can they watch them until this mood and this ideation actually passes.

GROSS: So I just want to pause here and give the national Suicide and Crisis hotline number. And this is the number to call or to text. It's 988, so it's a simple number. Just three numbers, 988, to either call or text the national Suicide and Crisis hotline. So if you are having thoughts of suicide, please get some help. Well, let me reintroduce you. If you're just joining us, my guest is Harold Koplewicz and he is a child psychiatrist, the founding president of the Child Mind Institute. His books include "Scaffold Parenting: Raising Resilient, Self-Reliant, And Secure Kids In An Age Of Anxiety." We have to take a short break here, and we'll be right back. This is FRESH AIR.

(SOUNDBITE OF MUSIC)

GROSS: This is FRESH AIR. Let's get back to my interview with Dr. Harold Koplewicz. He is a child psychiatrist, he's the founding president of the Child Mind Institute, and his books include "Scaffold Parenting: Raising Resilient, Self-Reliant, And Secure Kids In An Age Of Anxiety."

You specialize in ADHD - attention-deficit/hyperactivity disorder. Why don't you define what the symptoms are and how to recognize it?

KOPLEWICZ: So it's a challenge for lots of people to think about it because they think, oh, aren't we all hyperactive at some time? But the difference here is a deficit in attention toward what's normal developmentally. So the attention span of a 5-year-old is very different than the attention span of a 10-year-old. But any individual who has ADHD is chronically less attentive, tends to be more impulsive, and if they have hyperactivity, they're moving around more. They can get themselves into physical problems because they basically have ants in their pants. They're constantly in motion. The diagnosis when you have hyperactivity is much easier to make than when you just have ADD without H. But it's a chronic illness, and therefore, it may change over time. Your symptoms might lessen. Hyperactivity might go away when you become a teenager. But you are always going to have a shorter attention span and going to be more impulsive than the average person your age.

GROSS: I think this is one of the problems in which brain imaging is starting to be used - fMRIs, where you can see, like, which parts of the brain light up in different situations and different thoughts. How are fMRIs being used in ADHD?

KOPLEWICZ: Right. It's the holy grail for us to find that objective test. One of the things we've discovered at the Child Mind Institute is that the way your brain connects to itself while a child's at rest turns out to be diagnostic. It's called connectomes. So does the front of the brain connect to the side of the brain or to the back of the brain?

And what's been very interesting is that we took a few hundred scans and sent them to a group of people who were statisticians, who were electrical engineers, and asked them if they could group those different scans in different buckets. And we found the group that actually won this competition were statisticians from Hopkins. And they said, well, these 150 scans go together, and these 50 scans go together, and these hundred scans go together. And these are individuals who have never seen a patient with psychiatric disorder. But what's really interesting - in bucket one, the overwhelming majority of those patients had ADHD. In the group of 50, they had autism. And the group of a hundred, they had both ADHD and autism. So we're really excited by the fact that we have found something that might lead us to a definitive, objective test.

Now, the important part for everyone to remember - it's not just one child. It's not a strep test - yes, you're positive or someone else is negative. It's a group difference. But that's the way we're going to get closer and closer to making a definitive diagnosis.

GROSS: So in a study like the fMRI study that you were referring to, how do you know whether the brain is reflecting the behavior or whether the behavior is predetermined by the brain? Do you know what I mean?

KOPLEWICZ: Sure. Well, it's...

GROSS: It's, like, if I move my left arm - if I say, I'm going to move my left arm right now, and I'm doing it with intent, it's going to register on an fMRI, probably. But it's not like I have a disorder that's moving my left arm. It's, like, I've decided to behave this way, and it's registering in my brain.

KOPLEWICZ: So, you know, let's think about this for a second. This is exactly where the field of research in functional MRI has gone to. You know, they used to give a trigger to a kid. You know, pay attention to this while you're in the machine, or we're going to show you scary faces and see what happened to the brain. It turns out that the most powerful way of doing this is just letting kids rest or sleep in the functional MRI. And your brain is incredibly active while you're at rest or sleeping. And that's when you're going to see most of these connections. So in the case of the study, we weren't triggering them. We weren't saying, you know, this clearly should be what makes the - you know, we'll catch them being inattentive, and then we'll look at the MRI. We're just looking at their brains at rest.

GROSS: Oh, that's really interesting. So has this affected your treatment at all?

KOPLEWICZ: So we're not there yet. You know, it's not ready for prime time. I wish it - you know, I could say, oh, we're going to give everyone EEGs, because they're only 60 bucks, and an MRI is 500, and we found some correlation. That's what I'm hoping for. But, you know, science has to wait for real data. So at this moment, we still have to rely on clinical diagnosis. You're asking parents what they think. You're asking teachers and report cards, because this is not something that just pops up when you're about to apply to college or because you didn't make partner at the law firm. This is a lifelong illness. And you can document that by looking at things from a longitudinal basis.

And then you have to examine the child. The child basically confirms the diagnosis or doesn't. I think it's fascinating when you do give a kid meds, and they do significantly better, that a young child will tell you the medicine's not working. And you say, really? What's changed? He said, my teacher is much nicer. I said, that's really interesting.

GROSS: (Laughter).

KOPLEWICZ: You take a pill and your kid - your teacher's much nicer. That really is absolutely amazing. And they said, yeah. You know, you're 8 years old. OK.

GROSS: Well, we need to take another break here, so let me reintroduce you. If you're just joining us, my guest is Dr. Harold Koplewicz. He's a child psychiatrist. He's the founding president of the Child Mind Institute. And his books include "Scaffold Parenting: Raising Resilient, Self-Reliant And Secure Kids In An Age Of Anxiety." We'll be right back after a short break. I'm Terry Gross, and this is FRESH AIR.

GROSS: This is FRESH AIR. I'm Terry Gross. Let's get back to my interview with Dr. Harold Koplewicz. He's a child psychiatrist and the founding president of the Child Mind Institute. His books include "Scaffold Parenting: Raising Resilient, Self-Reliant, And Secure Kids In An Age Of Anxiety." Your middle son has dyslexia. No one realized it at first. How did you discover what it was? Because this is, I assume, one of the issues that you treat as a child psychiatrist.

KOPLEWICZ: Right. Well, I think dyslexia has very often been put into another category until about the last 25 years, so that it was something educational experts did and not necessarily child psychiatrists. And brain scans and functional MRIs have changed that, but in the case of our family, he was 4 years old. We were visiting his grandmother for a day, and she was a pretty neutral individual, very careful not to say anything critical or even say anything overly praising, and she mentioned that she had trouble understanding him. And we said, Well, you know, his articulation isn't great. And she said, no, his stories are out of sequence, and I don't think he knows my name. I love listening to him, but everything seems a little mixed up. And I was the one who said, you know, we should listen to grandma here.

And we went on this journey to figure out what was wrong. We had him tested, and the tester said, oh, he's very bright. Well, at 4 years old, you don't read on these kinds of neuropsych testing, but she did mention that he had some word retrieval problems and that he couldn't name certain things, and she was concerned that maybe he didn't recognize the alphabet. And I remember at the time saying, What do we do about it? And she said, well, it's going to cost a lot of money, and it's going to take a lot of time. And I thought, well, he's 4 years old. He has a lot of time. And for this, we'll find a lot of money. We'll mortgage our house. He's got to learn how to read.

So we went on this journey, and it turned out that we wasted a lot of time. There were a lot of dead ends where we thought we were doing well, but it turns out he was memorizing words, that he couldn't decode the language. He couldn't tell the difference between Sally, Susan, and Sarah. And it was only by third grade, when math turned into word problems, that we saw how frustrated he was and how he recognized that he was ahead of kids in math but way behind them in reading, and he knew this was happening. And that gave us the moment to reevaluate and figure out a more evidence-based approach.

GROSS: Which was?

KOPLEWICZ: Well, it turns out that a multisensory approach to learning how to read, teaching kids the sounds of the language, brother sounds, what your lips look like, and there was a program called Lindamood-Bell, which is now in 50 states, and it's very intensive. You spend 4 hours a day doing these exercises with a different tutor every hour, and then you do another 30 minutes or 40 minutes of homework. And they basically teach you how to break the code, that the rest of us are learning how to read with only one side of our brain, and they are teaching you how to read thinking that, you know, your brain thinks it's Italian or Spanish, that you're learning a foreign language. It's really a remarkable intervention.

GROSS: I'm going to stop you. What do you mean by that, that your brain thinks you're learning a foreign language?

KOPLEWICZ: When we learn a foreign language, we use both sides of our brain. When we're learning...

GROSS: We do?

KOPLEWICZ: Yes.

GROSS: (Laughter). I hadn't heard that before.

KOPLEWICZ: Right. So when we're learning our native language, we activate one side of the brain, and kids with dyslexia underactivate that brain. And so when you teach them a new way of learning, it's like teaching them a foreign language, and so when you check what's going on in a functional MRI, they're lighting up both sides of their brain. The thing that's really awful about dyslexia, as far as I'm concerned, is what it does to kids' self-esteem. You know, once a year, I get to interview someone who's struggled with it, whether it's Orlando Bloom or Ari Emanuel or, you know, Lorraine Bracco, and you hear how bad they felt about themselves.

Think about it. Every day, you go to work, and every day, you feel like a failure, so it's not surprising that you think you're stupid or that you don't want to be there. And what they - all these successful people have in common is a great mom. You know, Barbara Corcoran has it, and she told me that the nuns were really giving her a tough time in parochial school. And her mother said, don't listen to the nuns. You are not stupid. They're stupid. Well, having a mom who's telling you you're still bright or Orlando Bloom's mom, who said, let's do poetry - you know, those are these great moms who are basically saying, I'm on your side, and we're going to figure this out. But for those who don't have those moms, school is impossible. There's high school dropout rates. There are high rates of getting involved in the juvenile justice system because you're not in school. And when we look at the juvenile justice system, we see that 70% of the inmates have dyslexia.

GROSS: You know, I used to think that dyslexia was a problem with, like, reversing words, so you'd have to read slowly 'cause words would get reversed in your mind, but it's much more profound than that. Can you give us, like, the latest understanding of what dyslexia is?

KOPLEWICZ: Sure. So that's a myth, you know, the d, the b. What it really is, first of all, it's a brain-based disorder, and there's two major symptoms. One is alphabet recognition, being able to look at the A and knowing it's an A and looking at the D and knowing it's a D. And we all learned that, kind of, you know, very easily. And the other (inaudible) awareness, hearing the sounds of the language, being able to say to yourself or let your brain recognize that S-L-O-W comes out slow and S-H-O-W comes out show. And so you have to be able to hear those two different phonemes. And I will tell you that now that America is recognizing that this kind of evidence-based learning is really important, that we have to teach kids phonemes, we have to teach them how to read no matter who they are, we are really addressing this in an earlier age, so kids who have dyslexia will be picked up sooner and will be able to get interventions that are more effective, again, before it affects their self-esteem.

GROSS: So let me reintroduce you here. If you're just joining us, my guest is Doctor Harold Koplewicz. He's a child psychiatrist, author of the book "Scaffold Parenting: Raising Resilient, Self-Reliant And Secure Kids In An Age of Anxiety" and founding president of the Child Mind Institute. We'll be right back. This is FRESH AIR.

GROSS: This is FRESH AIR. Let's get back to my interview with Dr. Harold Koplewicz, the founding president of the Child Mind Institute, author of the book "Scaffold Parenting: Raising Resilient, Self-Reliant And Secure Kids In An Age Of Anxiety." He's a child psychiatrist and has been one for how many years?

KOPLEWICZ: Almost 40.

GROSS: OK. So this is a kind of personal question, but knowing what you know now - and there's so much more research that's been done into childhood, you know, behavioral problems and mental health disorders - do you think you had any undiagnosed problems as a child?

KOPLEWICZ: I don't think so, but...

GROSS: And I don't mean that, like...

KOPLEWICZ: No, no, no.

GROSS: ...I recognize symptoms in (laughter)...

KOPLEWICZ: No, no, no. Yeah.

GROSS: ...In how you're behaving...

KOPLEWICZ: I - but I know - but I would tell you that I clearly became much more of a student when I was in college than I was in high school. I had Eastern European parents. I had parents who survived the Holocaust and got to the United States in 1949. And they didn't believe that education was a journey. It was a destination. And they couldn't wait until, you know, you graduated and go to college. And so I was two years younger than everyone in elementary school. And I think that was most probably not a great idea - that most boys developed, you know, late. And so that was a problem.

And I would also tell you that, you know, the parents that I had when I was growing up were much more traumatized by the Holocaust than the parents I had later on in life, when they were in their 80s and 90s and were less anxious and the nightmares had stopped and they felt more comfortable in the United States - and also comfortable that, you know, I was going to be successful. I had graduated medical school. I had children. They - I was married. And that seemed to really calm them down.

But I do recognize that they were overly invested in my being successful because they were trying to recreate stuff that they lost. My parents were both - by the way, my father had graduated law school in 1936, and my mother was in law school in 1938. And neither one of them ever practiced law. They came to this country as immigrants. They had to start all fresh again. My father started a business. I think they struggled financially. My mother eventually went back to school and got a B.A. and then an MSW. But there was this idea of what could have been if there wouldn't have been the Holocaust. And therefore, my sister and I had to carry, you know, that weight, which is, you know, understandable but was very unpleasant when it was happening.

GROSS: Were your parents in camps?

KOPLEWICZ: My father was literally in 14 concentration camps and the Warsaw Ghetto. And how is that possible? Well, at the very first camp, they asked, who knows how to make airplanes? And my father raised his hand. And when asked about that, he said, well, they had already killed the lawyers. And he figured, well, I know how to use a screwdriver. I'll figure it out. And he went from camp to camp. And he was with one other man who kept being moved with him, and they got a little piece of metal. And the other guy was very artistic, and he engraved a sailboat and a horn of plenty. And on the other side, every time they moved from one camp to another, my father inscribed the date and the name of the camp. And they were hoping that it would be at least a record, that what they were experiencing would be recorded and documented. And that piece of metal, by the way, is at the U.S. Holocaust Museum in New York - I'm sorry, in D.C., in Washington...

GROSS: Yeah.

KOPLEWICZ: ...D.C.

GROSS: Yeah. So you mentioned your - so your mother was in camps, too?

KOPLEWICZ: No. My mother got papers as a Catholic and - false papers as a Catholic and walked out of the ghetto. And in some ways, it was more trying for her in the respect that - think about it. You have fake papers, and if the Gestapo stops you and starts really examining the papers and starts asking you questions like, what is your mother and father's name? Oh, they're dead. OK. And what was your priest's name? And where are you from? It wouldn't take very long.

So she moved around 16 different villages, outside of Warsaw, working as a maid. And she was a terrible housekeeper, so it is really amazing how she managed to do that, because she - you know, she really had a very tough time and was very isolated and just basically, you know, surviving from day to day. And it was, I think, a little more than two years where she was moving around. The war ended first in Poland. And so my father didn't come and find her until several months later.

GROSS: Oh, they were married before the war started.

KOPLEWICZ: Well, I wish I could tell you that's true, and that was the story I was told. But it turns out that when my then-12-year-old son was doing a - my wife insisted that if he was going to be bar mitzvahed, it had to be intergenerational. So he kept asking my mother her life story and recording it. And at a certain point, my son said, I don't understand, Grandma. Where was the infrastructure in the ghetto for you to get married? And my mother said, oh, you know, in the Jewish religion, you can get married and become the stars and the moon. And my son said, I don't think that's true. I think you need a contract.

KOPLEWICZ: And she said, well, August 12. It was the day I lost my virginity with your grandfather. And he came home and said, I don't know if Grandma and Grandpa ever were married. I think they're celebrating the day they had sex. So I called my mother and said, I don't understand. Why did you tell him that? She said, I never slept with anybody else, and I thought, enough. And he asked much better questions than you ever did.

KOPLEWICZ: So I think they got married when they were leaving Poland to go to a displaced persons camp in Germany. But - and I have to tell you as an example, their people - my mother was madly in love with my father before the war. You know, she lusted for him. He was very attractive, and he was a lawyer already. And then after the war, when he returned, he was skin and bones. And, you know, he was truly a different person. And she was a different person. She was no longer a bit of a princess. She was a survivor. She knew hard (ph) - and she - he came and found her. And she said, I'm going to let you come in, but I'm leaving. I've got papers to go either to Palestine or to Australia or Canada or the United States. I'm not staying here. And he said, well, I am staying here. I'm a lawyer, and we're going to make a lot of money. And she said, that's OK.

The idea that they lived together for three months and she got the papers and he decided to go with her - it's really a very romantic story that they fell in love again. And my father, every year on their anniversary, would give my mother - if they had money, he gave her a red rose for every year they were together and three white roses for the three years they weren't together with the same note - life had no color without you. So they really rediscovered each other and I think gave - that bond was so close. In some ways, my sister and I sometimes felt out of it because they were such a partnership that that's what carried them through later on.

GROSS: What impact do you think it had on you as a child to know that they were having these nightmares, these concentration camp...

KOPLEWICZ: Oh, it was nightmares.

GROSS: ...Or posing-as-a-Catholic kind of nightmares? Did they tell you about that? Could you sense it? And in the same mode there, like, did they let on what they had experienced and how traumatic it was?

KOPLEWICZ: So the stories were never consistent or chronological, so you only got bits and pieces. You know, something about the showers - right? - that one of my grandmothers died in the showers. You know, I hate to tell you that they didn't explain the camps to me, but you also knew that they were so upset by it that you didn't pursue it. You didn't ask them a lot of things. And I can certainly tell you that since they weren't very Jewish by education or training, the holidays were just terrible. I mean, you know, most people light a memorial candle for all the people that have died in their family. Well, all their brothers and sisters and their parents and their cousins, so that, you know, there were, like, 10 yahrzeit candles, these memorial candles, but they weren't kept in the kitchen. Like, all the ghosts were on the dining room table for Rosh Hashanah or Yom Kippur. And it's kind of amazing the transformation they made over time - that, you know, they were literally able to become more stable and calmer and more effective as adults.

GROSS: You must have grown up with a very dark view of life.

KOPLEWICZ: At times, yes. You know, at time, yes. And yet the amazing part about my parents were they couldn't care less about material things. So other kids would live in houses near us in Queens and then in Nassau County, where the slip covers, plastic slip covers were put on the furniture, and my mother would say, What are they waiting for? You know, they'll be dead. If something broke in our house, my mother never cared about it. I mean, it really gave them a whole different attitude about what was important, and certainly, material things were not important to my parents. They had lost a lot because they stayed. You know, I used to say, why didn't you come to America, for God's sakes? and my mother would say, Al Capone and peasants - they came to America, not - you know, not the intelligentsia. You know, she's kind of snobby about, you know, her academic credentials and who her family was, but, you know, they lost a lot because they didn't want to leave property or whatever it was or the life that they had.

GROSS: So one last question. You know, some parents really want to be their children's best friend, and some parents really want to maintain their stature as the authority figure, not the best friend. And in terms of being a parent yourself, I'm curious where you fit on that spectrum, if you are comfortable talking about that.

KOPLEWICZ: Sure. Well, I'm not my kid's best friend, you know, and that's OK, because even though they're all adults now, which is a whole different kind of relationship... I mean, my children are 41, 37, and 35. It really is frightening...

KOPLEWICZ: ...To see that one of my kids has gray hair. You know, it's like, how did this happen? Because I'm still 35. You know, but all along, I think there is this pull that you certainly want your kids to love you because you love them so much, but it's OK for them not to like you because you do have to protect them. And when you protect them, there are certain things that they want to do that you know are dangerous for them or are just not good for them or not healthy for them. And so I think it's very hard, if not impossible, to be a friend, which is be a peer - right? - and share the same point of view and not have control. Your friend does not have control over you. It's much more equal. And I don't think that's possible as a parent. I think the best type of parenting, by the way, is an authoritative parent who has a lot of warmth but has a lot of control. So both the kid and the parent know, at the end of the day, the parent is going to make the decision, maybe with input from the child. But at the end of the day, it's not a democracy. It's going to be the parent who has the responsibility to make those decisions.

GROSS: Dr. Harold Koplewicz, thank you so much for talking with us.

KOPLEWICZ: Oh, it's been a pleasure, Terry.

GROSS: Dr. Harold Koplewicz is the founding president of the Child Mind Institute. His latest book is titled "Scaffold Parenting." After we take a short break, Justin Chang reviews what he describes as a marvelous new movie. This is FRESH AIR.

(SOUNDBITE OF BABO VALDES TRIO'S "LAMENTO CUBANO")

Copyright © 2024 NPR. All rights reserved. Visit our website terms of use and permissions pages at www.npr.org for further information.

NPR transcripts are created on a rush deadline by an NPR contractor. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.

Featured Topics

Featured series.

A series of random questions answered by Harvard experts.

Explore the Gazette

Read the latest.

Getting ahead of dyslexia

How did you get that frog to float?

Lifting a few with my chatbot

Why ai fairness conversations must include disabled people.

Naomi Saphra, Lawrence Weru, Maitreya Shah.

Photos by Jon Chase and Kris Snibbe/Harvard Staff Photographers and courtesy of Maitreya Shah; photo illustration by Judy Blomquist/Harvard Staff

Eileen O’Grady

Harvard Staff Writer

Tech offers promise to help yet too often perpetuates ableism, say researchers. It doesn’t have to be this way.

Third in a four-part series on non-apparent disabilities.

AI researcher Naomi Saphra faced “a programmer’s worst nightmare” in 2015. After a decade of coding and just as she was about to start a Ph.D. program in Scotland, neuropathy in her hands rendered typing too painful.

Seeking a solution, Saphra turned to the very technology she was studying. She began the long process of teaching herself how to code using voice-to-text dictation technologies. Today, a system called Talon, which Saphra has heavily customized to complete specific tasks, allows her to code and to write papers for her research on language models.

“I rely on it completely,” said the research fellow at the Kempner Institute for the Study of Natural and Artificial Intelligence . “I would absolutely not have a career if we didn’t have AI for speech-to-text. These days it’s pretty hard to exist in the world if you’re not able to use a computer much. And as things have advanced, it’s been very important that the word error rate has gotten lower over the years.”

AI technology can be a powerful assistive tool for people like Saphra with non-apparent disabilities — physical or mental conditions that are not immediately obvious to others. But disability advocates say these tools have a long way to go to become truly accessible. Experts say including disabled people in conversations on AI fairness and the development process is key.

“If we’re creating tools that we know are fed with information that can bias against certain groups, and we integrate those into very crucial aspects of our lives, what’s going to be the impact of that?” asks Lawrence Weru, an associate in biomedical informatics at Harvard Medical School.

Kris Snibbe/Harvard Staff Photographer

Lawrence Weru, an associate in biomedical informatics at Harvard Medical School , was initially excited when voice-activated AI tools such as Siri and Alexa were released in the early 2010s. As someone who learned to code from a young age on public library computers before personal computers were common, he has long been fascinated by advances in digital technology. But Weru, who has a stutter, quickly found voice-activated technology more frustrating than helpful. When asking Siri for directions, the digital assistant would not understand the question if he stuttered.

While this was hardly a new experience — before AI, Weru remembers the frustration of trying to contact his bank and not being able to get past the automated phone system — it was disappointing to realize that the AI likely had not been trained on data from people with disabilities like his.

“People create things and people always have a vision in mind of who is going to be using their thing,” Weru said. “Sometimes not everybody is included in those personas.”

His experience with Siri makes Weru concerned about the future of voice-activated AI technology, envisioning a world in which critical tasks — making doctor appointments, applying for jobs, accessing education — are powered not by humans, but by technologies that can’t be used by everyone.

“If we’re creating tools that we know are fed with information that can bias against certain groups, and we integrate those into very crucial aspects of our lives, what’s going to be the impact of that?” Weru said. “That’s a concern that I hope people would be having enough foresight to try to address in advance, but historically accessibility is usually something that’s treated as an afterthought.”

“I would absolutely not have a career if we didn’t have AI for speech-to-text,” said Naomi Saphra, a research fellow at the Kempner Institute.

Jon Chase/Harvard Staff Photographer

Maitreya Shah, a fellow at the Berkman Klein Center for Internet and Society , recently launched a research project analyzing different schools of thought on “AI fairness,” or movements seeking to mitigate AI bias against people in marginalized groups. Shah , a blind lawyer and researcher, wants to go beyond conversations about accessibility and examine what he believes is the root of the issue: People with disabilities are not being included in conversations about AI, even in conversations about AI fairness.

“A lot of research so far has focused on how AI technologies discriminate against people with disabilities, how algorithms harm people with disabilities,” Shah said. “My aim for this project is to talk about how even the conversation on AI fairness, which was purportedly commenced to fix AI systems and to mitigate harms, also does not adequately account for the rights, challenges, and lived experiences of people with disabilities.”

For his research, he’s interviewing scholars who have studied the issue and evaluating frameworks designed to maintain AI fairness proposed by governments and the AI industry.

Shah said developers often consider disability data to be “outlier data,” or data that differs greatly from the overall pattern and is sometimes excluded. But even when it’s included, there are some disabilities — like non-apparent disabilities — that are overlooked more than others. If an AI is trained on a narrow “definition” of disability (like if data from people who stutter is not used to train a voice-activated AI tool) the outcome will be that the tool is not accessible.

“There is a paradox,” Shah said. “If you don’t incorporate disability data, your algorithms would be open to discriminating against people with disabilities because they don’t fit the normative ideas of your algorithms. If you incorporate the data, a lot of people with disabilities would still be missed out because inherently, the way you incorporate datasets, you divide data on the axes of identity.”

“Do people with autism or other disabilities even want these technologies? No one asks them.” Maitreya Shah

In his own life, Shah uses some AI technologies as assistive tools including “Be My AI,” which describes images, and “Seeing AI,” which provides users with visual information such as text, color, light, and scenery. Blind people were very involved in the development and testing process for both those tools.

But Shah said too often people with disabilities are not included in the high-level decision-making and development processes for AI that is purported to benefit them. He cited, as an example, technology designed to diagnose autism or address learning disabilities.

“The question is: Do people with autism or other disabilities even want these technologies? No one asks them,” Shah said.

In his research, Shah proposes adopting perspectives from disability justice principles, such as participation.

“Let people with disabilities participate in the development and the deployment of technologies. Let them decide what is good for them, let them decide how they want to define or shape their own identities.”

Saphra agrees, which is why she believes any developer creating assistive AI should make it easily customizable, not just by AI experts or coders, but by people who may not be tech experts. That way, users can set up the system to perform specific, essential tasks like Saphra did for writing code.

“It’s very important to make sure that everything you release is hackable, everything is open-source, and that everything has an accessible interface to start with,” Saphra said. “These things are going to make it more useful for the greatest number of disabled people.”

Share this article

• University Disability Resources serves as the central resource for disability-related information, procedures, and services for the Harvard community.
• Students who wish to request accommodations should contact their School’s Local Disability Coordinator.
• The 24/7 mental health support line for students is 617-495-2042. Deaf or hard-of-hearing students can dial 711 to reach a Telecommunications Relay Service in their local area.
• Harvard Law School Project on Disability

Also in this series

Navigating Harvard with a non-apparent disability

4 students with conditions ranging from diabetes to narcolepsy describe daily challenges that may not be obvious to their classmates and professors

Anticipate, accommodate, empower

How to ensure students with disabilities have an equal chance to succeed?

Harvard lab’s research suggests at-risk kids can be identified before they ever struggle in school

You might like

Ever-creative, Nobel laureate in physics Andre Geim extols fun, fanciful side of very serious science

Sociologist Sherry Turkle warns against growing trend of turning to AI for companionship, counsel

Forget ‘doomers.’ Warming can be stopped, top climate scientist says

Michael Mann points to prehistoric catastrophes, modern environmental victories

College accepts 1,937 to Class of 2028

Students represent 94 countries, all 50 states

Yes, it’s exciting. Just don’t look at the sun.

Lab, telescope specialist details Harvard eclipse-viewing party, offers safety tips