Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/springer-journals/pre-school-skills-and-school-age-reading-comprehension-in-children-on-WZHGUjtCoA
- Publisher
- Springer Journals
- Copyright
- Copyright © The Author(s) 2023
- ISSN
- 0162-3257
- eISSN
- 1573-3432
- DOI
- 10.1007/s10803-023-05949-0
- Publisher site
- See Article on Publisher Site
Abstract
We explored reading comprehension development in children on the spectrum from pre-school to the first (YOS1) and third year of schooling (YOS3). Children were first assessed on meaning-related skills in pre-school. Forty-one children completed follow-up assessments of reading comprehension, reading accuracy, and listening comprehension in YOS1. Nineteen returned for assessments of reading accuracy, reading comprehension, and listening comprehension in YOS3. Children showed poorer reading comprehension than reading accuracy at both timepoints. Reading comprehension, reading accuracy, and listening comprehension were significantly concurrently correlated. Pre-school receptive vocabulary was a significant predictor of YOS3 reading comprehension. Results from this preliminary investigation highlight the potential for early identification of children on the spectrum at risk for reading comprehension difficulties. Keywords Longitudinal · Autistic · Students · Listening comprehension · Vocabulary Reading comprehension is a crucial skill acquired at school investigated pre-school predictors of decoding skills in chil- and is essential to learning across academic content areas. dren on the spectrum (Solari et al., 2022; Westerveld et al., School-age children on the autism spectrum at a group 2018), there is limited research investigating school-age level show significant vulnerability to reading comprehen- reading comprehension predictors from these pre-school sion impairments relative to their peers, and to their own emergent literacy skills in children on the spectrum. decoding skills (see recent meta-analysis, Sorenson Duncan Understanding how pre-school factors may predict later et al., 2021). At an individual level, significant variability in reading comprehension ability in children on the spectrum is school-age reading comprehension is found, with skills rang- important to identify children who may be at risk of reading ing from very low to above average (Sorenson Duncan et al., comprehension difficulties in the future and to inform devel- 2021). Reading development, however, commences before opment of early targeted interventions. This initial investi- children receive formal reading instruction. Therefore, gation addresses this need and follows up children in their recent studies have focused on the emergent literacy skills of third year of formal education, who were first assessed prior children on the spectrum during the pre-school years (for a to starting their formal education (pre-school) and again in review see Westerveld et al., 2016). These emergent literacy their first year of education. We explore concurrent associa- skills include print-related skills (e.g., alphabet knowledge) tions between reading comprehension, reading accuracy and related to later decoding skills and meaning-related skills listening comprehension during the first and third year of (e.g., vocabulary) related to later reading comprehension formal schooling. We also investigate whether pre-school skills in neurotypical development (National Early Literacy abilities predict later reading comprehension. Panel, 2008). Although an emerging body of research has * Jessica Paynter j.paynter@griffith.edu.au Griffith Institute for Educational Research, Griffith University, Brisbane, Australia The term “on the autism spectrum” is used in line with the prefer- School of Health Sciences and Social Work, Griffith ences of the autistic and autism community (Bury et al., 2020; Kenny University, Gold Coast, Australia et al., 2016). Vol.:(0123456789) 1 3 Journal of Autism and Developmental Disorders 54.6–71% (mixed reading disability), and 70–92% (non- Reading Development in School‑age specified). The finding that reading profiles are relatively Neurotypical Children stable during the early years of schooling, highlights the importance of early detection of predictors of reading Reading comprehension is a complex skill that relies on comprehension proficiency. a range of cognitive, language, and reading abilities such as non-verbal intellectual abilities, working memory, oral Emergent Literacy in Neurotypical Children language, word reading and comprehension monitoring as captured in theoretical models of reading such as the Emergent literacy refers to a set of skills developed prior to Direct and Indirect Effect Model of Reading (DIER; Kim, formal education, that are predictive of later reading skills 2017) and the Cognitive Foundations Framework (Tun- and include oral language (or ‘meaning’) related skills mer & Hoover, 2019). Given the preliminary nature of linked to the listening comprehension component and print- the current study into longitudinal preschool predictors of related skills associated with later word recognition. Chil- reading comprehension, and building on existing research dren generally develop their emergent literacy skills through in autism that has focused on the Simple View of Reading interactions with their parents, caregivers, or early childhood (SVR; Gough & Tunmer, 1986; Sorenson-Duncan et al., professionals, for example during shared book reading (Pen- 2021), we elected to use this model as the focus of the timonti et al., 2012). Print-related emergent literacy skills present study. According to the SVR, children who dem- that are predictive of later word recognition skills include onstrate adequate reading comprehension show proficiency alphabet knowledge, name writing, early developing phono- in two relatively independent components: word recogni- logical awareness, and print-concept knowledge (National tion skills (i.e., decoding) and listening comprehension Early Literacy Panel, 2008). Oral language-related skills (e.g., Lonigan et al., 2018). Based on the SVR, there are predictive of listening comprehension include vocabulary, four reading profiles: children with (1) good reading (ade- grammar, and oral narrative skills (Catts et al., 2015; Foor- quate/good word recognition and listening comprehen- man et al., 2015). Support for the importance of develop- sion skills); (2) poor reading with poor word recognition ing both print-related and oral-language related emergent skills, but adequate listening comprehension skills (i.e., literacy skills for successful reading comprehension was dyslexia); (3) poor reading with listening comprehension confirmed in a recent systematic review of 64 longitudinal difficulties, but adequate word recognition skills; and (4) studies investigating pre-school predictors of reading com- poor reading with both word recognition and listening prehension (Hjetland et al., 2017). Print-related skills in comprehension difficulties (i.e,. mixed reading difficulties) pre-school were indirectly related to later reading compre- (see Westerveld et al., 2020). However, the relative con- hension, via their impact on word recognition, whereas oral tributions of word-recognition and listening comprehen- language skills (e.g., vocabulary, grammar) were directly sion to reading comprehension change over time (Hjetland related to reading comprehension. Whether these results et al., 2017). generalize to children on the autism spectrum remains During the early years of schooling, children are in unclear. the ‘learning to read’ stage, with most of the variance in reading comprehension explained by word recognition. Emergent Literacy and Autism Once children develop accurate and fluent word recogni- tion skills (third or fourth year of schooling), listening Three major findings have been highlighted in research comprehension contributes more significantly to reading exploring the emergent literacy skills of children on the comprehension, with this stage referred to as the ‘learning autism spectrum. First, there is significant heterogeneity through reading’ stage (Chall, 1996). Of note is that read- in print- and oral language-related emergent literacy skills ing profiles are relatively stable over time. For example, across individuals in this population (Solari et al., 2022; Catts et al. (2003) followed a group of 604 children from Westerveld et al., 2016). Second, emergent literacy profi - kindergarten to grade 4, many of whom had been identified ciency is associated with children’s non-verbal cognition with a language impairment in kindergarten. In grade 2, and/or general language ability (Knight et al., 2019; Lanter 183 children showed poor reading (low reading compre- et al., 2012). This is consistent with the DIER model as hension). When tracking these children during the early applied to neurotypical development which outlines that school years, strong correlations were found (r > 0.70) domain general cognitive skills including non-verbal cog- between second and fourth grade word recognition, listen- nition both directly and indirectly via oral language, impact ing comprehension and reading comprehension. Further, literacy skills (Kim, 2017). Third, at a group level there is an depending on the cut-off used for classification, stability uneven profile with relative strengths in print-related skills ratings were 69.8–89% (dyslexia), 65–84% (hyperlexia), compared to oral-language related emergent literacy skills 1 3 Journal of Autism and Developmental Disorders (Davidson & Ellis Weismer, 2014; Westerveld et al., 2017) semantic processing tasks testing conceptual knowledge and specific weaknesses in discourse level-language skills, of underlying word meanings often pose difficulties (see such as narrative skills (Westerveld & Roberts, 2017). These Boucher, 2012, for a review). Taken together, considering early challenges in oral language may help explain the read- the importance of oral language proficiency for reading com- ing comprehension challenges many school-age children on prehension, one may thus expect children on the spectrum the spectrum demonstrate (Arciuli et al., 2013; Nation et al., to show reading comprehension difficulties. In addition, the 2006). impact of co-occurring intellectual impairments (approxi- mately 35%, Maenner et al., 2021), affecting learning in Reading Comprehension and Autism general may also influence reading development (see Kim, 2017). Finally core and associated features of autism may Sorenson Duncan et al. (2021) reviewed the cross-sectional influence interest and motivation in (aspects) of reading, literature into reading comprehension in autism draw- and cognitive features of autism (e.g., executive functioning ing from the SVR (Gough & Tunmer, 1986), focusing on impairments, local processing bias, and/or social cognition/ school-age children (6–18 years). This meta-analysis of theory of mind) may contribute to literacy learning chal- 26 studies highlighted the relative weaknesses at a group lenges (Solari et al., 2022; Westerveld & Paynter, 2021). level in reading comprehension, heterogeneity in skills at an individual level, and strong associations between read- Predictors of Reading Comprehension in Autism ing comprehension and oral language comprehension (M r = 0.65), and reading comprehension and word reading (M The limited body of research investigating the predictors r = 0.65), consistent with the SVR. However, some caution of reading ability in autism provides initial evidence that is warranted in generalizing results to the broader population reading-related earlier skills can predict later reading com- of individuals on the spectrum as the authors acknowledge prehension abilities (Åsberg Johnels et al., 2019; Kim et al., possible selection bias within individual studies, such as 2018; Knight et al., 2019). Knight et al. (2019) found that excluding children with lower intellectual functioning (e.g., decoding skills earlier in the school year predicted read- McIntyre et al., 2017) and potentially including individuals ing comprehension skills later in the school year within the with higher oral language and reading skills than the gen- same school grade (pre-kindergarten, kindergarten, grade 1, eral population of children on the spectrum. These exclu- grade 2), although children’s IQ significantly mediated the sions may reflect challenges in conducting valid assessments relationship between decoding and reading comprehension with children with oral language or intellectual impairments, in pre-kindergarten and kindergarten. Åsberg Johnels et al. however guidelines have been published that outline strate- (2019) found that early childhood (age 2:6 years) screening gies (e.g., selecting measures with lower or minimal verbal assessments were predictive of reading profile subgroups demands, using visuals and social stories) to support con- in the first or second grade of school in a longitudinal ret- ducting valid assessments of this population and demonstrat- rospective study. Reading profile groups included “skilled ing this is possible (e.g., Clendon et al., 2021; Paynter et al., readers” (adequate word reading and comprehension), 2016). Including children on the spectrum across a wide “poor readers” (low word reading and comprehension), and range of oral language and intellectual abilities is important “hyperlexic/poor comprehenders” (adequate/good word to better understand the reading comprehension skills of this reading with low reading comprehension). Skilled readers population across the full spectrum. scored significantly higher on parent-reported communica- Reading comprehension development in children on the tion measures and were more likely to be classified as verbal spectrum, from pre-school to school age may be influenced by clinicians in early childhood screening than children in by associated features of autism including oral language the other two groups. They also performed higher than the abilities, intellectual ability, and cognitive features (see hyperlexic/poor comprehenders on a standardized test of Westerveld & Paynter, 2021). At a group-level children on receptive language. However, groups did not differ on early the spectrum perform well below age-expectations on stand- cognitive ability, autism traits, nor parent-rated social skills ardized oral language assessments (Kwok et al., 2015), with in early childhood assessments. This retrospective study thus an estimated 30% not developing functional oral language provides initial evidence in line with the SVR (Gough & skills (Tager-Flusberg & Kasari, 2013). One of the early Tunmer, 1986) and research in typical development (Hjet- clinical signs of autism is the late onset of oral language land et al., 2017), that pre-school oral language difficulties development, with first words appearing after three years of identified prior to formal instruction may predict children age (Howlin, 2003a). Furthermore, evidence suggests diffi- at-risk of later reading difficulties. However, the impact of culties in grammar development, including use of complex cognitive ability or IQ is unclear with mixed findings across sentences (Eigsti et al., 2011). In contrast, relative strengths studies. have been observed in receptive vocabulary, although 1 3 Journal of Autism and Developmental Disorders To investigate if early challenges in aspects of emergent in terms of reading comprehension, no specific predictions literacy make children vulnerable to later reading compre- were made for change over time. Based on the recent school- hension challenges, Davidson and Ellis Weismer (2014) age meta-analysis (Sorenson Duncan et al., 2021), it was pre- followed a cohort of children on the spectrum from 2½ dicted at each time point that children on the spectrum would years to 5½ years of age. They found that nonverbal cogni- show lower reading comprehension than reading accuracy. tion, autism traits, social ability, and expressive language Based on the SVR, and a systematic review showing sup- at Time 1 accounted for 46% of the variance in children’s port for this population (Sorenson Duncan et al., 2021), it performance on a standardized assessment of comprehen- was predicted that reading comprehension at each time point sion of printed words, sentences and paragraphs at age 5. would show significant associations with reading accuracy Autism traits were not a significant independent predictor. and listening comprehension. Given previous research dem- In summary, Davidson and Ellis Weismer (2014) and Åsberg onstrating vocabulary was a significant early predictor of Johnels et al. (2019) provide evidence that pre-school skills later reading comprehension in typical development (Hjet- may predict later comprehension skills, at least within the land et al., 2017), it was tentatively hypothesized that vocab- first years of formal education, however whether pre-school ulary may significantly predict later reading comprehension. skills are predictive of later skills as children transition from learning to read to learning through reading in the third year of formal education has not yet been investigated. Method Current Study Design Our main aim in the present preliminary study was to under- A longitudinal cohort design was used. The study was not stand the development of reading comprehension from preregistered. Participants in YOS3 were recruited from a pre-school into the third year of education for children on cohort of students who had initially participated in a funded the autism spectrum. Children were first assessed prior to cross-sectional study into emergent literacy in autism (Wes- school-entry and were recruited for a follow-up assessment terveld et al., 2017), and had been recruited previously to in their first year of formal schooling, and a subsequent participate in an unplanned follow-up in YOS1 following cohort was recruited for a further assessment in their third securing additional funding, that investigated predictors year of schooling. At pre-school age, a range of meaning of word recognition skills (Westerveld et al., 2018). The and print-related emergent literacy skills were assessed that YOS3 component did not receive any specific funding. This are known predictors of reading acquisition and develop- study was performed in line with the principles of the 1964 ment in neurotypical children (selected based on National Declaration of Helsinki. Ethics approval was granted by the Early Literacy Panel, 2008). We recruited from this cohort Griffith University Human Ethics committee (AHS/13/14/ of young children on the spectrum who were assessed prior HREC). to school entry (Pre-school) to return for further assessment at two timepoints: in their first year of schooling (YOS1) and then in the third year of formal schooling (YOS3) and Participants investigated: Participants were recruited for the initial study (Westerveld A) How does the reading comprehension skills of students et al., 2017) via early childhood services for children on on the spectrum in YOS3 compare to their performance the spectrum, private speech pathology clinics, a children’s in YOS1? hospital, flyers on parent support websites, and via profes- B) How does reading comprehension performance compare sional networks of the authors. Participants for the initial to reading accuracy performance in YOS1 and YOS3? pre-school assessment study were recruited from Queens- C) What are the concurrent associations between reading land and New South Wales in Australia. Inclusion criteria comprehension, reading accuracy, and listening compre- included a formal autism diagnosis verified with the Social hension within YOS1 and YOS3 of schooling? Communication Questionnaire (SCQ; Rutter et al., 2003) or D) What are the longitudinal associations between pre- the Autism Diagnostic Observation Schedule—2 (ADOS) school variables (listening comprehension, vocabulary, (Lord et al., 2012); 48 months or older; not commenced autism traits, and non-verbal IQ) and reading compre- formal education; ability to speak in short sentences; and hension in YOS3? the ability to participate in preschool activities. The initial cohort included 57 children, see Table 1. All parents com- Given the paucity of research tracking children on the pleted the SCQ Lifetime Form and for 25 children ADOS spectrum from the first year of formal schooling to the third results were available from previous assessments at their 1 3 Journal of Autism and Developmental Disorders Table 1 Participant characteristics Pre-school YOS1 subset YOS3 subset (n = 57) (n = 41*) (n = 19) Variable Mean Range Mean Range Mean (SD) Range (SD) (SD) Pre-school Gender: m/f 48/9 – 35/6 – 16/3 – Age in months 57.60 (6.11) 48–70 57.63 (5.72) 49–70 57.00 (5.26) 50–68 NVIQ 79.11 (19.53) 44–119 78.10 (20.56) 44.00–119.23 79.24 (19.25) 44.85–110.53 Vocabulary 90.00 (16.30) 64–127 89.22 (16.05) 64–127 89.11 (14.98) 64–121 Autism traits 15.79 (5.75) 5–32 15.83 (6.02) 5–32 15.32 (7.00) 5–32 YOS1 Age in months – – 73.61 (4.62) 66–81 72.47 (5.16) 67–81 Months of schooling – – 9.17 (2.01) 4–12 9.05 (1.87) 6–12 Oral Language ability – – 75.58 (20.50) 45–122 78.29 45–110 (19.62) n = 41 for all except Language Ability (n = 38). Gender is a proportion NVIQ = Mullen Scales of Early Learning Developmental Quotient; Vocabulary = Peabody Picture Vocabulary Test Standard Score; Autism nd Traits = Social Communication Questionnaire Total Score; Language Ability = Clinical Evaluation of Language Fundamentals-Preschool 2 Edition Core Language Score early intervention centers which were provided. The SCQ third year of formal education. A sample of 19 children was was used in the first instance to verify diagnosis to reduce recruited, see Table 1. Non-recruitment was predominantly burden on child participants. For children without an ADOS due to being unable to contact participants from YOS1 verifying diagnosis already, four children scored below the (n = 12) or geographical distance which precluded assess- clinical cut-off on the SCQ with two excluded, and two com- ment (n = 7). Only three families who were located within pleted an ADOS with one of these children scoring above travel distance of the authors and were able to be contacted the clinical cut-off and subsequently included in the final declined (no reason, n = 2; onset of medical condition, sample. n = 1). We confirmed representativeness of YOS3 partici- Following entry to formal education, participants were pants by comparing included/not recruited participants on invited to return for an assessment during their first year of key variables hypothesized to relate to reading comprehen- formal education with 41 participants recruited, see Table 1. sion (receptive vocabulary, autism traits, and non-verbal Inclusion criteria were having participated in the initial developmental quotient) and found no significant differences study (Westerveld et al., 2017) and having commenced (all p > 0.05). The 19 participants tested at YOS3 had a mean their first year of formal schooling. This commences after age of approximately 8 years (M = 95.89 months, SD = 5.57, children turn five in Australia (with differing month of the range 87–106 months). year cut-offs by state/territory), with the school year January to December. Declines to invitations to participate in this Procedure phase, were due to change of location (n = 3), availability (n = 3), significant increases in challenging behavior limiting Initial pre-school assessment was completed over two ses- capacity to complete the assessment (n = 2), no reason given sions of up to 90 min by a certified practicing speech-lan- (n = 2), or unable to contact (n = 6). We confirmed repre- guage pathologist. Parents were then invited to participate sentativeness of YOS1 participants by comparing returning in the follow-up study, following their child starting their participants to those who did not, on key variables from first year (4–12 months) of formal schooling to complete the pre-school hypothesized to relate to reading comprehension YOS1 assessment (time between pre-school to YOS1 assess- (receptive vocabulary, autism traits, and non-verbal devel- ments, M = 15.59 months, SD = 4.44, range 8–30 months). opmental quotient) and found no significant differences (all YOS1 assessment was completed in one session of approxi- p > 0.05, for further details see Westerveld et al., 2018). mately two hours by one of four research assistants (three Families were then approached to participate at YOS3. certified practicing speech-language pathologists and a Inclusion criteria were that participants had consented and psychology PhD candidate). Parents were then invited to completed the YOS1 assessment; were in the local area of participate in the three-year follow-up and YOS3 assess- the researchers due to practical constraints; and were in their ment approximately two years after their YOS1 assessment 1 3 Journal of Autism and Developmental Disorders (M = 23.05 months, SD = 2.17, range = 20–27 months). dividing by the child’s chronological age, and multiplying Assessment at YOS3 was completed in one session of by 100. The use of a developmental quotient was selected as approximately 1.5–2 h, by a qualified certified practicing children on the spectrum may score too low for calculation speech-language pathologist. Assessment location (school/ of meaningful standard scores, and this process has been early learning setting, home, or clinic) for each timepoint established in prior literacy research in this population (e.g., was selected on parent preference. All examiners were pro- Davidson & Ellis Weismer, 2014). vided training and supervision by the authors across each time-point who are a clinical psychologist (author 1) and Listening Comprehension a certified practicing speech pathologist (author 3) both of whom have > 15 years’ experience in the assessment of The Profile of Oral Narrative Ability (PONA; Westerveld young children and teach their respective areas at postgradu- & Gillon, 2010) comprehension component was used ate levels. as a measure of listening comprehension at pre-school and YOS1. In this task, children listen to an unfamiliar story while viewing story book pictures on a computer Measures screen and then answer eight open response comprehen- sion questions, with one point for each question accurately Data to address research questions were extracted from data answered. Raw scores (0–8) were used for analyses (as per collected at the pre-school assessment (Westerveld et al., Westerveld & Gillon, 2010). 2017) and YOS1 (Westerveld et al., 2018) and are outlined The Clinical Evaluation of Language Fundamentals − 4 in brief below. Pre-school measures extracted to address the Understanding Spoken Paragraphs subtest (CELF-4 USP; research questions included autism traits, receptive vocabu- Semel, 2006) was administered to assess children’s listen- lary, non-verbal cognition, and listening comprehension. ing comprehension at YOS3. As per the manual, children YOS1 measures included oral language, listening compre- were asked to listen to spoken paragraphs and then answer hension, and passage reading accuracy and comprehension. open questions about the passages. Scaled scores (M = 10, YOS3 measures completed for the present study were listen- SD = 3) were used for analyses. ing comprehension, passage reading accuracy and passage reading comprehension. Oral Language Autism Traits The Clinical Evaluation of Language Fundamentals-Pre- school-2nd Edition (CELF-P2, Wiig et al., 2004) Core The Social Communication Questionnaire Lifetime Form Language Subtests (sentence structure, word structure, (Rutter et al., 2003) was completed by primary caregivers and expressive vocabulary) were administered to describe at the pre-school assessment. Total raw score (maximum the sample and as a measure of oral language at YOS1. 40) was used as a measure of autism traits as per previous Standard Scores (M = 100, SD = 15 in neurotypical devel- research (e.g., Fulton et al., 2017). opment) from the manual were used. Receptive Vocabulary Passage Reading and Passage Comprehension The Peabody Picture Vocabulary Test- Fourth Edition The York Assessment of Reading for Comprehension Pri- (PPVT-4; Dunn & Dunn, 2007) was administered with chil- mary (YARC; Snowling et al., 2012) was administered to dren as a measure of receptive vocabulary with standard evaluate passage reading accuracy (based on the number of scores (M = 100, SD = 15 in neurotypical norms from the reading errors) and passage reading comprehension (based manual) used for analyses. on the number of open response questions answered cor- rectly) at YOS1 and YOS3. This measure was chosen due to having Australian norms, and using open-ended ques- Non‑verbal Cognition tions, as opposed to a cloze measure, to reflect text-level comprehension (Westerveld, 2009). In this test, children The Mullen Scales of Early Learning visual reception and are required to read passages aloud, then answer questions fine motor subtests were administered with children to calcu- following the reading. As per the manual, at YOS1, chil- late a non-verbal developmental quotient for a non-verbal IQ dren were first asked to read aloud the beginner passage. score (NVIQ) by averaging age equivalents across subtests, Only upon successful completion of the beginner passage 1 3 Journal of Autism and Developmental Disorders (i.e., < 15 reading errors), the child is asked to read aloud as missing data for this measure (i.e., no score given). At the next passage (passage 1). As per the administration YOS1, one child did not complete the listening comprehen- guidelines, children need to be able to read aloud two pas- sion task. At YOS3, five children did not show required sages (at a level suitable to the child’s reading ability) for skills to complete beyond the beginner passage on the YARC a standard score to be computed for reading accuracy and (thus no standard score). As described above, 22 children did reading comprehension. These standard scores (M = 100, not return for the YOS3 assessment and analysis included SD = 15) for reading accuracy and reading comprehension only those who completed YOS3 to answer the research were each used for analyses. questions. Data were deleted listwise by analysis to use the full data available for each analysis. Data were screened for Data Analysis assumptions of analyses including normality, outliers, nor- mality, independence of residuals, linearity, homoscedastic- To address our first research question regarding stability of ity, and collinearity. Assumptions were met. performance over time, two analyses were conducted. First YARC comprehension standard scores were categorized into Reading Comprehension Outcomes at YOS1 within (± 1 standard deviation [SD]) or below average range and YOS3 (< − 1SD) using an age-standardized score (SS) of 85 as a cut-off, with those unable to complete the task (e.g., could At YOS1 of the 41 children assessed, only 21 could accu- not progress beyond the beginner passage of the YARC) rately read two grade level passages yielding standard scores assigned to below average, consistent with our previous for passage and reading comprehension, see Table 2. The research (Westerveld et al., 2018) to evaluate differences 20 children who did not receive a standard score due to not between children showing skills within the average range reading two grade level passages, showed developmental for their age, versus those showing skills below the average quotients ranging from 44 to 108.69 (M = 70.37, SD = 18) range or no showing these abilities. Children who did not and receptive vocabulary ranging from standard scores of receive a valid score were included only for group compari- 66.00–102 (M = 81.6, SD = 9.68) at pre-school. Of the 21 sons and were excluded listwise by analysis for correlations. children who received scores, 18 showed passage read- The proportion of participants in each category was com- ing accuracy within the average range (standard scores of pared using an exact McNemar’s test. Second, to compare 85–115) for their age, while only eight showed reading those who achieved interpretable scores only, within groups comprehension within the average range. At an individual paired t-tests of YOS1 and YOS3 scores were conducted on level, of the 21 children who could read passages, only eight standard scores. To address research question two of rela- children showed both accuracy and comprehension within tive performance on reading accuracy and comprehension the average range; 10 showed average reading accuracy but at each time point paired t-tests were conducted. To address reading comprehension below average; and three children research question three and four of concurrent and longi- showed both reading accuracy and comprehension below tudinal associations correlations were conducted includ- average. The thirteen children whose comprehension stand- ing only participants with longitudinal data (YOS1 n = 41; ard scores were in the below the average range (range 70–83) YOS3 n = 19). Given the exploratory nature of the study, showed varying non-verbal abilities with developmental potential Type 2 errors were deemed of more concern than quotients from 44.85 to 99.15 (M = 76.17, SD = 18.20), and Type 1, as such no control for multiple comparisons was varying receptive vocabulary with standard scores ranging implemented. Effect sizes were interpreted using conven - from 64.00 to 103.00 (M = 86.54, SD = 12.78) at pre-school. tions for φ (0.1 = small; 0.3 = medium;0.5 = large), Cohen’s The eight children who showed comprehension standard d (0.2 = small; 0.5 = medium, 0.8 = large) and Pearson’s r scores within the average range showed varying non-verbal (0.1 = small, 0.3 = medium, and 0.50 = large) (Cohen, 1988). abilities from below average to above average ranges (DQ 76.42–119.23, M = 100.58, SD = 14.52), whereas receptive vocabulary at pre-school was within the average range for Results all children, with a mean above average (SS range 91–127, M = 112.63, SD = 11.53). Data Screening YOS 3 At YOS1 two children were unable to complete the YARC, and 18 did not show required reading skills to complete At YOS3, in the third year of schooling of the 19 children beyond the beginner passage (i.e., a standard score could not assessed, 14 could read passages and could subsequently be computed). Three children did not complete the CELF- be assessed on reading comprehension. The five children P2 due to non-compliance with the task and were treated who did not receive a score showed varying abilities at 1 3 Journal of Autism and Developmental Disorders Table 2 Listening Full sample Paired valid YARC participants comprehension, reading accuracy and comprehension Measure First year of Third year of First year of school Third year in the first and third year of school (YOS1) school (YOS3) (YOS1) of school schooling (YOS3) n 41 19 11 11 Passage Reading Accuracy (SS) n 21 14 11 11 Mean 102.95 99.50 98.91 103.36 (SD) (17.33) (14.66) (16.70) (13.73) Range 74–130 70–127 74–130 70–127 WNL 18 12 9 10 (% WNL of total n at timepoint) (43.9%) (63.2%) (47.4%) (90.9%) Reading Comprehension (SS) n 21 14 11 11 Mean 88.90 83.64 82.27 87.00 (SD) (20.00) (14.91) (18.29) (15.17) Range 70–123 70–110 70–123 70–110 WNL 8 7 2 7 (% WNL of total n at timepoint) (19.5%) (36.8%) (10.5%) (36.84%) Listening Comprehension PONA (rs) CELF (Scs) PONA (rs) CELF (ScS) n 40 19 19 11 Mean 2.78 4.74 2.68 5.91 (SD) (2.24) (3.53) (2.19) (3.83) Range 0–7 1–13 0–7 1–13 WNL – 5 – 4 (% of n) (26.3%) (36.4%) Listening Comprehension- Profile of Oral Narrative Ability (PONA, T2) raw score or CELF-4 Understand- ing Spoken Paragraphs Scaled Score (ScS with 7–13 WNL, T3); Reading Comprehension: York Assess- ment of Reading for Comprehension (YARC) SS; Passage Reading: YARC SS, Within Normal Limits (WNL) = SS 85–115. Full sample is used for concurrent analyses (outcomes, comprehension vs. accuracy); paired valid YARC at each time point is used for evaluation of stability over time pre-school (non-verbal DQ range 57.76–88.89, M = 74.75, Reading Performance Over Time SD = 12.91; receptive vocabulary SS range 75.00–102.00, M = 82.00, SD = 11.45). Of these 14 children, 12 showed Eleven children of the 19 children assessed at YOS3, showed passage reading accuracy in the average range, but only adequate passage reading to complete the comprehension seven showed reading comprehension in the average range. task at both YOS1 and YOS3. Of these children, six (54.5%) At an individual level, seven children showed both accu- remained in the same reading comprehension category at racy and comprehension in the average range; five showed YOS3 (two in the average and four in the below average reading accuracy in the average range but reading compre- range), and five moved from below to within the average hension below average; and two showed both accuracy and range (45.5%); the proportion of children in each group did comprehension below average. The seven children with not significantly differ between time points using an exact reading comprehension below the average range (scores McNemar’s test, exact p = 0.063, φ = 0.36. For these 11 chil- 70–74) showed varying non-verbal (DQ 44.85–107.69, dren with reading comprehension standard scores at both M = 72.88, SD = 25.03) and receptive vocabulary abilities YOS1 (M = 82.27, SD = 18.29, range = 70–123) and YOS3 (SS 64.00–100.00, M = 81.13, SD = 11.89). The seven chil- (M = 87.00, SD = 15.17, range = 70 -110), non-significant dren with reading comprehension scores within or above increases over time with a medium effect were found, t average scores for their age showed varying non-verbal abili- (10) = 1.07, p = 0.31, d = 0.37. ties at pre-school from within the low to high average ranges (DQ 66.96–110.53, M = 88.81, SD = 14.26), but receptive Concurrent Reading Comprehension vs. Reading Accuracy language abilities close to the average range or above aver- age (SS 84.00–121.00, M = 102.14, SD = 11.42). Twenty-one children completed the reading comprehen- sion task at YOS1 and 14 at YOS3. Note these numbers are 1 3 Journal of Autism and Developmental Disorders higher than the matched comparisons above as three children Pre‑school Correlations to Reading Comprehension in YOS3 at YOS3 were able to complete the tasks who had not com- pleted it at YOS1. At YOS1, of those with sufficient reading Performance on reading comprehension in YOS3 was sig- accuracy to complete the reading comprehension component nificantly correlated, with a large effect, with pre-school (n = 21), participants performed significantly lower, as pre- receptive vocabulary (r = 0.76, p = 0.001), see Table 3. Non- dicted, on reading comprehension (M = 88.90, SD = 20.00, significant medium effects were found in correlations with range = 70–123) than on reading accuracy (M = 102.95, preschool listening comprehension (r = 0.38, p = 0.18) and SD = 17.33, range = 74–130) with a medium effect, preschool NVIQ (r = 0.40, p = 0.09). Autism traits showed a t(20) = 3.23, p = 0.004, d = 0.70. The average difference in small non-significant correlation (r = 0.11, p = 0.65). scores was 14.05 points. Likewise, at YOS3, participants (n = 14) performed significantly lower, as predicted, on read- ing comprehension (M = 83.64, SD = 14.91, range = 70–110) Discussion than on reading accuracy (M = 99.50, SD = 14.66, range 70–127), with a large effect, t (13) = 4.42, p = 0.001, d = 1.18. While a growing body of research has explored reading At YOS3, the average difference in scores was 15.86 points. in children and adolescents on the autism spectrum, lim- ited research has followed children longitudinally from Concurrent Correlations in YOS1 pre-school into the early school years. This is a valuable area for exploration given it would enable identifica- For the children who were able to complete the reading com- tion of students at risk prior to experiencing challenges, prehension task at YOS1 (n = 21), scores were significantly enabling the possibility of earlier interventions prior to correlated to concurrent passage reading accuracy with a school entry, which may ease the transition to school for medium effect (r = 0.44, p = 0.048), listening comprehen- this group that demonstrates high rates of educational sion (PONA) with a large effect (r = 0.83, p < 0.001), and challenges (Australian Bureau of Statistics., 2019). This oral language (CELF-P2 Core Language score) with a large preliminary longitudinal study complements and extends effect ( r = 0.77, p < 0.001), see Table 3. Listening compre- our research into pre-school predictors of reading accuracy hension and oral language showed large correlations with (Westerveld et al., 2018) when children were in their first each other (r = 0.76, p < 0.001). Reading comprehension year of schooling. At that stage, only 21 of the 41 children was not significantly correlated with age at assessment, nor were able to complete a norm-referenced reading assess- months of schooling. ment, due to reading accuracy difficulties. In the current study, we followed 19 of the original 41 children when Concurrent Associations in YOS3 they attended their third year of schooling. Key findings in terms of reading comprehension performance, relative Performance on reading comprehension in the third year of reading accuracy vs. comprehension performance, and schooling showed a large significant concurrent correlation concurrent and longitudinal associations between pre- with reading accuracy at the passage level (r = 0.59, p = 0.03) school variables and school-age reading comprehension and listening comprehension (CELF-4, Understanding Spo- are discussed. ken Paragraphs) (r = 0.66, p = 0.01), see Table 3. Pre‑school Correlations to Reading Comprehension in YOS1 Reading Comprehension Performance Performance on reading comprehension in YOS1 was Consistent with previous research (Arciuli et al., 2013; significantly correlated with large effects with pre-school Nation et al., 2006), a substantial proportion of children NVIQ (r = 0.60, p = 0.004), receptive vocabulary (r = 0.78, showed significant challenges in reading comprehen- p < 0.001), and listening comprehension (PONA) (r = 0.69, sion on a norm-referenced test of reading ability dur- p = 0.001), see Table 3. NVIQ, VIQ, and listening compre- ing the early years of schooling. In YOS1, 19.5% (8/41) hension showed large significant correlations with each other showed reading comprehension within the average range (all r > 0.50). Autism traits at pre-school did not significantly (SS 85–115) for their age. These eight children showed correlate with YOS1 school-age reading comprehension good reading (both accuracy and comprehension in the (r = 0.20, p = 0.40). average range), however 20 could not read, three showed mixed reading difficulties (below average on both read- ing accuracy and reading comprehension), and 10 showed poor reading comprehension (below average with reading accuracy within the average range). Comparing only those 1 3 Journal of Autism and Developmental Disorders 1 3 Table 3 Correlations between pre-school (PS), YOS1, and YOS3 reading variables for children with longitudinal data (n) Autism NVIQ (PS) Vocab (PS) Listening Listening Oral School Age (YOS1) Reading Reading Listening Reading Traits Comp. Comp. Language months Accuracy Comp. Comp Accuracy (PS) (PS) (YOS1) (YOS1) (YOS1) (YOS1) (YOS1) (YOS3) (YOS3) NVIQ (PS) .03 (41) Vocabulary (PS) .02 .72** (41) (41) Listening Comprehension .23 .61** .72** PONA (PS) (41) (41) (41) Listening Comp PONA .06 .64** .67** .79** (YOS1) (40) (40) (40) (40) Oral Language (YOS1) − .04 .74** .85** .74** .76** (38) (38) (38) (38) (37) Number of months at school − .08 .30 .08 .13 .28 .22 (YOS1) (41) (41) (41) (41) (40) (38) Age at assessment (YOS1) .12 − .26 − .33* − .12 − .09 − .38* .03 (41) (41) (41) (41) (40) (38) (41) Reading Accuracy (YOS1) − .11 .45* .48* − .01 .32 .42 .29 − .10 (21) (21) (21) (21) (21) (19) (21) (21) Reading Comprehension .20 .60** .78** .69** .83** .77** .18 − .23 .44* (YOS1) (21) (21) (21) (21) (21) (19) (21) (21) (21) Listening Comprehension .12 .35 .81** .53* .72** .72** .04 .05 .59 .60 CELF-4 USP (YOS3) (19) (19) (19) (19) (19) (17) (19) (19) (11) (11) Reading Accuracy (YOS3) − .43 .31 .50 .08 .38 .48 .35 − .27 .73* .43 .52 (14) (14) (14) (14) (14) (12) (14) (14) (11) (11) (14) Reading Comprehension − .07 .40 .76** .38 .51 .82** .19 − .35 .42 .63* .66* .59* (YOS3) (14) (14) (14) (14) (14) (12) (14) (14) (11) (11) (14) (14) p < .05; **p < .01; NVIQ = non-verbal intellectual quotient; PONA = Profile of Oral Narrative Ability; CELF-4 USP = Clinical Evaluation of Language Fundamentals-4 Understanding Spoken Paragraphs. (n) The number of participants for each correlation is in brackets with missing data deleted listwise Journal of Autism and Developmental Disorders children who completed assessments at both time points, the eligibility criteria, with many previous studies excluding in YOS3, the proportion of children who could read with children based on intellectual ability (e.g., McIntyre et al., comprehension improved significantly, with 36.8% (7/19) 2017) which was not part of our criteria. showing reading comprehension (compared to only two of the same 19 students or 10.5% at YOS1) within the average Concurrent Associations Between Reading Measures range for their age, defined as ± one standard deviation of and Oral Language the mean of the measure. Comparing categorizations of performance over time As expected, reading comprehension was significantly corre- into below/within average range, categorizations were lated to reading accuracy as measured on the YARC at YOS1 stable or improved for the group with valid scores at each (r = 0.44) and YOS3 (r = 0.59). Furthermore, strong asso- timepoint. A promising observation was that no child ciations were found between listening comprehension and showed a change from performing within the average reading comprehension at both times (r = 0.83 and r = 0.66 range to dropping below this. Children either stayed within respectively). These correlations are consistent with previ- the same category (54.5%) or moved from below to within ous research in autism investigating associations between the average range (45.5%). Of concern, however, was the word recognition and reading comprehension as summarized finding that a substantive proportion of children contin - by Sorenson Duncan et al. (2021) and underpin the impor- ued to show performance below age expectations in read- tance of word recognition and listening comprehension for ing comprehension (12/19; 63.2%) in YOS3, consistent reading comprehension in autism, as per the SVR. with previous research (Arciuli et al., 2013; Nation et al., 2006). At YOS3, five children were unable to read pas - Longitudinal Associations Between Pre‑school sages (26%), two children showed mixed reading difficul- Abilities and Reading Comprehension ties (below average on both reading accuracy and reading comprehension), and five children showed poor reading We were particularly interested in longitudinal associations comprehension (below average) with reading accuracy in from pre-school into the early years of schooling. Although the average range. no significant associations were found between autism traits Taken together, reading comprehension findings are in and reading comprehension, significant large effects were line with Solari et al. (2019) who investigated reading pro- found between pre-school NVIQ, receptive vocabulary, files of 8–16-year-old children on the spectrum, 30 months and listening comprehension, and reading comprehension apart who similarly found stable or improving group mem- in the first year of schooling. These results show associa - bership. Our results suggest young children who perform in tions between oral language proficiency at word- (receptive the average range for reading comprehension are also likely vocabulary) and text-level (listening comprehension) prior to remain in the average reader group. Furthermore, some to school-entry and reading comprehension in the first year children transition into the average reader group over time. of formal schooling. These findings are in line with those Whether these potential developments or increases reflect conducted with children without an autism diagnosis (Catts broader changes in social cognition during the preschool et al., 2015; Kendeou et al., 2009) and lend further support period to early school years (Ricketts et al., 2013), the to the notion that the high incidence of language impair- impact of teaching, supports or intervention (e.g., an empha- ment in children on the autism spectrum (Kwok et al., 2015) sis on comprehension strategies during reading instruction), contributes to their increased risk of reading comprehension or other factors (e.g., oral language intervention or other difficulties. Moreover, the strong correlations between NVIQ therapies participants may have received) is an important and reading performance in the first year of formal school- question for future longitudinal research. ing, poses a potential risk factor for reading challenges. Of note however, is that while NVIQ and receptive vocabulary Reading Accuracy vs. Comprehension were predictive, they did not fully explain children’s ability or inability to read, as significant variability was observed Consistent with meta-analysis findings by Sorenson Dun- within groups of children who could not read, who could can et al. (2021), at group level we found children per- read but showed below average comprehension, and who formed better on reading accuracy than comprehension. read within the average range. For example, some children We observed differences between skills of just under one with average or above average NVIQ or receptive vocabulary standard deviation (SD = 15) at YOS1 (14 point difference) could not read paragraphs, while at least one child with a low and just over one standard deviation at YOS3 (M = 15.86), NVIQ (< 70) in pre-school was able to read with comprehen- which is within the range of differences found by Sorenson sion in the third year of schooling. It may be, consistent with Duncan et al. (2021) of 2.78–25.5 (M = 8.96). This varia- the DIET model (Kim, 2017) that NVIQ has both direct and tion in average mean differences between studies may reflect indirect (e.g., via oral language abilities) effects on reading, 1 3 Journal of Autism and Developmental Disorders and that understanding these related skills may be important Medium associations were found between pre-school lis- for determining areas of strengths and needs in this popula- tening comprehension and reading comprehension at YOS3, tion. Further research understanding what factors facilitate as well as pre-school NVIQ and reading comprehension at or impede development of reading abilities across verbal and YOS3. However, these were non-significant, which may non-verbal intellectual levels is needed given the frequent have been due to insufficient power to detect these smaller exclusion of children at lower levels from research to date. effects. Alternatively, listening comprehension and NVIQ No significant associations were found between pre- may show stronger effects in the short term, with significant school autism traits and any of the reading outcome meas- associations found in the first year of schooling as outlined ures. This is in line with previous research that found that above. Given mixed results in previous autism research in pre-school autism traits did not predict later reading com- terms of NVIQ (Åsberg Johnels et al., 2019; Davidson & prehension skills (Åsberg Johnels et al., 2019; Knight et al., Ellis Weismer, 2014; Knight et al., 2019) there is a need 2019), and that autism traits did not link to a measure of for further longitudinal research into these broader potential comprehension of printed words, sentences and paragraphs influences on reading comprehension. in young children on the spectrum (Davidson & Ellis Weis- mer, 2014). However, it contrasts with findings that con- Limitations and Future Directions current autism traits (Åsberg Johnels et al., 2019; McIntyre et al., 2017) were significantly higher in groups with sig- Our initial exploratory study provides important direc- nificant mixed reading disabilities (i.e., below expectations tions to inform future research, expanding on the paucity of on both reading accuracy and reading comprehension) in research tracking literacy development from pre-school in school-age children. It may be that associations are stronger autism. Nevertheless, a number of limitations are acknowl- with current abilities, or that an omnibus measure (i.e. total edged. First, our study was an initial investigation, build- score) may reflect a different constellation of traits for each ing on our initial funded cross-sectional study (Westerveld child yielding differing results across groups of children. et al., 2017) which meant that due to challenges contacting It may be that specific traits or associated autism features past participants and geographical distance for this unfunded may have differing influences on reading comprehension. follow-up, our small sample size constrained the complexity Initial evidence for this possibility may be seen in recent of analyses that were possible and impacted power to detect research using the Autism Quotient (Auyeung et al., 2008) smaller effects such as medium associations. Participants subscales in predicting different types of listening compre- however were representative of the full sample in terms of hension (literal vs. inferential) skills in Chinese children on key variables hypothesized to be important for reading com- the autism spectrum (Zhao et al., 2021). Given the strong prehension at each time point (NVIQ, receptive vocabulary, association between listening comprehension and reading and autism traits). While we deliberately selected a small comprehension, it may be hypothesized different subtypes of number of variables to address our research questions drawn autism traits or features may likewise show differing associa- from the SVR, we acknowledge the importance of fluency, tions with reading comprehension. Future research looking and broader decoding skills (e.g., phonological awareness, at subtypes of autism traits or features, both over time and alphabet knowledge) to reading comprehension and high- concurrently, and reading comprehension may be of value in light the inclusion of a broader range of variables would understanding mixed findings in research to date. be of value in future research for more fine-grained under - Receptive vocabulary, measured using the PPVT in pre- standing of the interplay of these in development of reading school, significantly correlated with reading comprehension in autism. in YOS3, consistent with previous research in neurotypical Our findings provide a foundation for future research, development (Hjetland et al., 2017). Further, these results highlighting that even within this small group, pre-school are also consistent with the significant parameter estimates predictors (vocabulary) may be able to predict later read- found by Catts et al. (2015) who used a composite measure ing comprehension beyond the first two years of formal of oral language that included the PPVT to predict reading schooling which have been the focus of research to date. comprehension in the third year of schooling in a sample of Future research should include a wider range of cognitive children without autism. In terms of the potential mecha- and linguistic measures to test the applicability of theoretical nisms of vocabulary impacting reading comprehension, we models (e.g., DIER, Kim, 2017) of reading to autism. Future observed large correlations between pre-school receptive studies should also include more fine-grained measures of vocabulary (PPVT) and YOS3 listening comprehension, and vocabulary to investigate vocabulary depth (e.g., word defi- YOS3 listening comprehension and reading comprehension, nitions) in line with theoretical word learning models (see consistent with the SVR and indicative of potential indirect Hadley & Dickinson, 2020, for a review) and acknowledg- effects of receptive vocabulary on reading comprehension ing that children on the spectrum often show difficulties in via listening comprehension. higher order semantic processing skills (Eigsti et al., 2011). 1 3 Journal of Autism and Developmental Disorders We also acknowledge the importance of reading fluency for reading comprehension utilizing a broader theoretical model proficient reading comprehension and future research should of reading (e.g., DIER, Kim, 2017) that includes oral lan- include evaluation of the relative contributions of both accu- guage as well as broader neurodevelopmental features (e.g., racy and fluency to the reading comprehension process. autism trait subscales, social cognition) would contribute Taken together, future prospective longitudinal research to theory and inform early identification of those at risk of that includes ongoing contact/updating contact details and/or reading difficulties. Findings would inform development of secondary contacts to enable tracking participants over time, supports to strengthen skills before challenges emerge sup- with large samples would enable exploration of the relative porting literacy success for children on the autism spectrum. contributions of a wider range of pre-school skills, autism- Acknowledgments We are grateful to the children on the autism spec- specific predictors (e.g., social-cognitive skills) as well as trum and their families for giving their time to support this research pathways (e.g., mediation vs. direct effects) to school-age study. The authors acknowledge the financial support for the pre-school and YOS1 components, of the Cooperative Research Centre for Living reading comprehension. with Autism (Autism CRC), established and supported under the Aus- Second, our measure selection for assessment of reading tralian Government's Cooperative Research Centres Program. accuracy and comprehension, the YARC, impacted which children were able to receive a score for each component. Funding Open Access funding enabled and organized by CAUL and its Member Institutions. The disadvantage of this test is that no reading comprehen- sion score is obtained if children make too many reading Declarations accuracy errors on a given passage. It is possible some chil- dren would have been able to answer comprehension ques- Conflicts of interest The authors acknowledge the financial support for tions in response to more advanced passages despite exceed- the pre-school and YOS1 components, of the Cooperative Research ing the maximum number of reading errors. Future research Centre for Living with Autism (Autism CRC), established and sup- ported under the Australian Government's Cooperative Research Cen- that uses a measure with separate measures for reading accu- tres Program. The authors did not receive support from any organiza- racy and comprehension may enable more fine-grained data tion for the YOS3 component. to be collected particularly for individuals with discrepant profiles. Ethics Approval This study was performed in line with the principles of the 1964 Declaration of Helsinki. This study received ethical approval from the University Human Research Ethics Committee (blinded for peer review). Implications Consent to Participate Informed consent was obtained from parents of all individual participants included in the study and assent (verbal or Until recently the academic needs of students on the autism non-verbal) was obtained from children. spectrum received limited attention relative to their social- emotional and behavioral needs and priorities. While in need Open Access This article is licensed under a Creative Commons Attri- of replication with larger samples, our preliminary research bution 4.0 International License, which permits use, sharing, adapta- highlights that as early as the first year of formal schooling tion, distribution and reproduction in any medium or format, as long many children on the autism spectrum are showing chal- as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes lenges in learning to read with comprehension. Of note is were made. The images or other third party material in this article are the fact that many of these children showed adequate to included in the article's Creative Commons licence, unless indicated good skills in pre-school emergent literacy skills (see Wes- otherwise in a credit line to the material. If material is not included in terveld et al., 2017 evaluation of pre-school emergent lit- the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will eracy skills). Challenges in reading comprehension relative need to obtain permission directly from the copyright holder. To view a to reading accuracy observed in the first year of schooling copy of this licence, visit http://cr eativ ecommons. or g/licen ses/ b y/4.0/ . continue into the third year of schooling, with evidence of an increasing gap over time. This widening gap in skills over time, emphasizes the need for earlier intervention as References children move from learning to read to learning through reading, with reading comprehension increasingly impor- Arciuli, J., Stevens, K., Trembath, D., & Simpson, I. C. (2013). The tant for participation and success as children move through relationship between parent report of adaptive behavior and direct school. The finding that pre-school receptive vocabulary assessment of reading ability in children with autism spectrum disorder. Journal of Speech, Language, and Hearing Research, was a significant predictor of reading comprehension in the 56(6), 1837–1844. https:// doi. or g/ 10. 1044/ 1092- 4388(2013/ third year of schooling is a key finding and highlights the 12- 0034) potential for identifying children at potential risk before Åsberg Johnels, J., Carlsson, E., Norbury, C., Gillberg, C., & Minis- they learn to read and experience difficulties. Future, more calco, C. (2019). Current profiles and early predictors of reading skills in school-age children with autism spectrum disorders: A fine-grained investigations of preschool predictors of later 1 3 Journal of Autism and Developmental Disorders longitudinal, retrospective population study. Autism, 23(6), 1449– Howlin, P. (2003). Outcome in high-functioning adults with autism 1459. https:// doi. org/ 10. 1177/ 13623 61318 811153 with and without early language delays: Implications for the dif- Australian Bureau of Statistics. (2019). Autism in Australia. Retrieved ferentiation between autism and Asperger syndrome. Journal of 8/09/2020, from https:// www. abs. gov. au/ AUSST ATS/ abs@. nsf/ Autism and Developmental Disorders, 33(1), 3–13. https:// doi. Lookup/ 4430. 0Main+ Featu res10 2018org/ 10. 1023/A: 10222 70118 899 Auyeung, B., Baron-Cohen, S., Wheelwright, S., & Allison, C. (2008). Kendeou, P., van den Broek, P., White, M. J., & Lynch, J. S. (2009). The Autism Spectrum Quotient: Children’s version (AQ-Child). Predicting reading comprehension in early elementary school: The Journal of Autism and Developmental Disorders, 38(7), 1230– independent constributions of oral language and decoding skills. 1240. https:// doi. org/ 10. 1007/ s10803- 007- 0504-z Journal of Educational Psychology, 101(4), 765–778. Boucher, J. (2012). Research review: Structural language in autistic Kenny, L., Hattersley, C., Molins, B., Buckley, C., Povey, C., & Pel- spectrum disorder – characteristics and causes. Journal of Child licano, E. (2016). Which terms should be used to describe autism? Psychology and Psychiatry, 53(3), 219–233. https:// doi. org/ 10. Perspectives from the UK autism community. Autism, 20(4), 442– 1111/j. 1469- 7610. 2011. 02508.x 462. https:// doi. org/ 10. 1177/ 13623 61315 588200 Bury, S. M., Jellett, R., Spoor, J. R., & Hedley, D. (2020). “It Defines Kim, Y.-S.G. (2017). Why the Simple Vew of Reading is not simplistic: Who I Am” or “It’s Something I Have”: What language do [Autis- Unpacking component skills of reading using a Direct and Indirect tic] Australian adults [on the Autism Spectrum] prefer? Journal Effect Model of Reading (DIER). Scientific Studies of Reading, of Autism and Developmental Disorders. https://doi. or g/10. 1007/ 21(4), 310–333. https://doi. or g/10. 1080/ 10888 438. 2017. 12916 43 s10803- 020- 04425-3 Kim, S. H., Bal, V. H., & Lord, C. (2018). Longitudinal follow-up Catts, H. W., Hogan, T. P., & Fey, M. E. (2003). Subgrouping poor of academic achievement in children with autism from age 2 to readers on the basis of individual differences in reading-related 18. Journal of Child Psychology and Psychiatry, 59(3), 258–267. abilities. Journal of Learning Disabilities, 36(2), 151–164. https:// https:// doi. org/ 10. 1111/ jcpp. 12808 doi. org/ 10. 1177/ 00222 19403 03600 208 Knight, E., Blacher, J., & Eisenhower, A. (2019). Predicting reading Catts, H. W., Herrera, S., Nielsen, D. C., & Bridges, M. S. (2015). comprehension in young children with autism spectrum disorder. Early prediction of reading comprehension within the Simple School Psychology, 34(2), 168–177. https://doi. or g/10. 1037/ spq00 View framework. Reading and Writing: An Interdisciplinary Jour- 00277 nal, 28, 1407–1425. https:// doi. org/ 10. 1007/ s11145- 015- 9576-x Kwok, E. Y. L., Brown, H. M., Smyth, R. E., & Oram Cardy, J. (2015). Chall, J. S. (1996). Stages of reading development. Harcourt Brace Meta-analysis of receptive and expressive language skills in College Publishers. autism spectrum disorder. Research in Autism Spectrum Disor- Clendon, S., Paynter, J., Walker, S., Bowen, R., & Westerveld, M. ders, 9, 202–222. https:// doi. org/ 10. 1016/j. rasd. 2014. 10. 008 F. (2021). Emergent literacy assessment in children with autism Lanter, E., Watson, L. R., Erickson, K. A., & Freeman, D. (2012). spectrum disorder who have limited verbal communication skills: Emergent literacy in children with autism: An exploration of A tutorial. Language, Speech, and Hearing Services in Schools, developmental and contextual dynamic processes. Language 52(1), 165–180. https:// doi. org/ 10. 1044/ 2020_ LSHSS- 20- 00030 Speech and Hearing Services in Schools, 43(3), 308–324. https:// Cohen, J. (1988). Statistical power analysis for the social sciences (2nd doi. org/ 10. 1044/ 0161- 1461(2012/ 10- 0083) Edition). Erlbaum. Lonigan, C. J., Burgess, S. R., & Schatschneider, C. (2018). Examin- Davidson, M. M., & Ellis Weismer, S. (2014). Characterization and ing the Simple View of Reading with elementary school children: prediction of early reading abilities in children on the autism spec- Still simple after all these years. Remedial and Special Education, trum. Journal of Autism and Developmental Disorders, 24(4), 39(5), 260–273. https:// doi. org/ 10. 1177/ 07419 32518 764833 828–845. https:// doi. org/ 10. 1007/ s10803- 013- 1936-2 Lord, C., Rutter, M., DiLavore, P. S., Risi, S., Gotham, K., & Bishop, Dunn, L. M., & Dunn, L. M. (2007). Peabody Picture Vocabulary Test- S. (2012). Autism Diagnostic Observation Schedule (2nd ed.). 4. Circle Pines, MN: American Guidance Service. Western Psychological Services. Eigsti, I. M., de Marchena, A. B., Schuh, J. M., et al. (2011). Language Maenner, M. J., Shaw, K. A., Bakian, A. V., Bilder, D. A., Durkin, M. acquisition in autism spectrum disorders: A developmental review. S., Esler, A., Furnier, S. M., Hallas, L., Hall-Lande, J., Hudson, Research in Autism Spectrum Disorders, 5(2), 681–691. https:// A., & Hughes, M. M. (2021). Prevalence and characteristics of doi. org/ 10. 1016/j. rasd. 2010. 09. 001 autism spectrum disorder among children aged 8 years—Autism Foorman, B. R., Herrera, S., Petscher, Y., Mitchell, A., & Trucken- and Developmental Disabilities Monitoring Network, 11 sites, miller, A. (2015). The structure of oral language and reading and United States, 2018. MMWR Surveillance Summaries, 70(11), their relation to comprehension in kindergarten through grade 2. 1–16. Reading and Writing, 28(5), 655–681. https:// doi. org/ 10. 1007/ McIntyre, N. S., Solari, E. J., Grimm, R. P., Lerro, E., & L., E. Gonza- s11145- 015- 9544-5 les, J., & Mundy, P. C. (2017). A comprehensive examination of Fulton, A. M., Paynter, J. M., & Trembath, D. (2017). Gender compari- reading heterogeneity in students with high functioning autism: sons in children with ASD entering early intervention. Research in Distinct reading profiles and their relation to autism symptom Developmental Disabilities, 68, 27–34. https://doi. or g/10. 1016/j. severity. Journal of Autism and Developmental Disorders, 47(4), ridd. 2017. 07. 009 1086–1101. https:// doi. org/ 10. 1007/ s10803- 017- 3029-0 Gough, P. B., & Tunmer, W. E. (1986). Decoding, reading, and reading Nation, K., Clarke, P., Wright, B., & Williams, C. (2006). Patterns of disability. Remedial and Special Education, 7(1), 6–10. https:// reading ability in children with autism spectrum disorder. Journal doi. org/ 10. 1177/ 07419 32586 00700 104 of Autism and Developmental Disorders, 36(7), 911–919. https:// Hadley, E. B., & Dickinson, D. K. (2020). Measuring young children’s doi. org/ 10. 1007/ s10803- 006- 0130-1 word knowledge: A conceptual review. Journal of Early Child- National Early Literacy Panel. (2008). Developing Early Literacy: hood Literacy, 20(2), 223–251. ht t ps :/ / d oi . o rg / 1 0. 1 17 7 / 1 46 87 Report of the National Early Literacy Panel. National Institute 98417 753713 for Literacy. Hjetland, H. N., Brinchmann, E. I., Scherer, R., & Melby-Lervåg, M. Paynter, J., Westerveld, M. F., & Trembath, D. (2016). Reading assess- (2017). Preschool predictors of later reading comprehension abil- ment in children with autism spectrum disorder. Journal of Psy- ity: A systematic review. Campbell Systematic Reviews, 13(1), chologists and Counsellors in Schools, 26(2), 205–217. https:// 1–155. https:// doi. org/ 10. 4073/ csr. 2017. 14doi. org/ 10. 1017/ jgc. 2016. 15 1 3 Journal of Autism and Developmental Disorders Pentimonti, J. M., Zucker, T. A., Justice, L. M., Petscher, Y., Piasta, approach to guide assessment and Intervention. Springer Open. S. B., & Kaderavek, J. N. (2012). A standardized tool for assess-https:// doi. org/ 10. 1007/ 978- 981- 15- 3492-8. pdf ing the quality of classroom-based shared reading: Systematic Westerveld, M. F., & Gillon, G. T. (2010). Profiling oral narrative Assessment of Book Reading (SABR). Early Childhood Research ability in young school-aged children. International Journal of Quarterly, 27(3), 512–528. https://d oi.o rg/1 0.1 016/j.e cresq.2 011. Speech-Language Pathology, 12(3), 178–189. https:// doi. org/ 10. 12. 0073109/ 17549 50090 31941 25 Ricketts, J., Jones, C. G., Happé, F., & Charman, T. (2013). Reading Westerveld, M. F., Paynter, J., Trembath, D., Webster, A. A., Hodge, comprehension in autism spectrum disorders: The role of oral A., & Roberts, J. (2017). The emergent literacy skills of preschool language and social functioning. Journal of Autism and Devel- children with autism spectrum disorder. Journal of Autism and opmental Disorders, 43(4), 807–816. https:// doi. or g/ 10. 1007/ Developmental Disorders, 47(2), 424–438. https:// doi. or g/ 10. s10803- 012- 1619-41007/ s10803- 016- 2964-5 Rutter, M., Bailey, A., & Lord, C. (2003). Social Communication Ques- Westerveld, M. F., & Roberts, J. M. A. (2017). The oral narrative tionnaire. Western Psychological Services. comprehension and production abilities of verbal preschoolers Semel, E., Wiig, E. H., & Secord, W. A. (2006). Clinical Evaluation on the autism spectrum. Language, Speech, and Hearing Ser- of Language Fundamentals - Fourth Edition - Australian Stand- vices in Schools 48, 260–272. https:// doi. or g/ 10. 1044/ 2017_ ardised. Harcourt Assessment.LSHSS- 17- 0003 Snowling, M. J., Stothard, S. E., Clarke, P., Bowyer-Crane, C., Har- Westerveld, M. F., Paynter, J., O’Leary, K., & Trembath, D. (2018). rington, A., Truelove, E., & Hulme, C. (2012). York assessment of Preschool predictors of reading ability in the first year of school- reading for comprehension (YARC). GL Assessment. ing in children with ASD. Autism Research, 11(10), 1332–1344. Solari, E. J., Grimm, R. P., McIntyre, N. S., Zajic, M., & Mundy, P. C. https:// doi. org/ 10. 1002/ aur. 1999 (2019). Longitudinal stability of reading profiles in individuals Westerveld, M. F., & Paynter, J. (2021). Introduction to the forum: with higher functioning autism. Autism, 23(8), 1911–1926. https:// Literacy in autism - across the spectrum. Language, Speech, and doi. org/ 10. 1177/ 13623 61318 812423 Hearing Services in Schools, 52(1), 149–152. https:// doi. org/ 10. Solari, E. J., Henry, A. R., Grimm, R. P., Zajic, M. C., & McGinty, 1044/ 2020_ LSHSS- 20- 00124 A. (2022). Code-related literacy profiles of kindergarten students Westerveld, M. F., Trembath, D., Shellshear, L., & Paynter, J. (2016). with autism. Autism, 26(1), 230–242. ht tps :/ / d oi . o rg / 1 0. 1 17 7/ A systematic review of the literature on emergent literacy skills 13623 61321 10259 04 of preschool children with autism spectrum disorder. The Journal Sorenson Duncan, T., Karkada, M., Deacon, S. H., & Smith, I. M. of Special Education, 50(1), 37–48. https://doi. or g/10. 1177/ 00224 (2021). Building meaning: Meta-analysis of component skills 66915 613593 supporting reading comprehension in children with autism spec- Wiig, E. H., Secord, W., & Semel, E. (2004). Clinical Evaluation of trum disorder. Autism Research, 14(5), 840–858. https:// doi. org/ Language Fundamentals - Preschool - Second Edition (CELF 10. 1002/ aur. 2483 preschool-2). The Psychological Corporation. Tager-Flusberg, H., & Kasari, C. (2013). Minimally verbal school- Zhao, J., Gao, Z., Lai, J., & Joshi, R. M. (2021). The relationship aged children with autism spectrum disorder: The neglected end between autism traits and listening comprehension among Chinese of the spectrum. Autism Research, 6(6), 468–478. https://doi. or g/ preschool children with autism spectrum disorder. Reading and 10. 1002/ aur. 1329 Writing. https:// doi. org/ 10. 1007/ s11145- 021- 10216-8 Tunmer, W. E., & Hoover, W. A. (2019). The cognitive foundations of learning to read: A framework for preventing and remediating Publisher's Note Springer Nature remains neutral with regard to reading difficulties. Australian Journal of Learning Difficulties, jurisdictional claims in published maps and institutional affiliations. 24(1), 75–93. https:// doi. org/ 10. 1080/ 19404 158. 2019. 16140 81 Westerveld, M. (2009). Measuring reading comprehension ability in children: Factors influencing test performance. Acquiring Knowl- edge in Speech, Language and Hearing, 11(2), 81–84. Westerveld, M. F., Armstrong, R., & Barton, G. (2020). Reading suc- cess in the primary years: An evidence-based interdisciplinary 1 3
Journal
Journal of Autism and Developmental Disorders – Springer Journals
Published: Mar 17, 2023
Keywords: Longitudinal; Autistic; Students; Listening comprehension; Vocabulary