Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Mar 1.
Published in final edited form as: Autism Res. 2022 Dec 28;16(3):630–641. doi: 10.1002/aur.2883

Fine motor skill and expressive language in minimally verbal and verbal school-aged autistic children

Lindsay K Butler 1,*, Helen Tager-Flusberg 1
PMCID: PMC10320849  NIHMSID: NIHMS1859697  PMID: 36578205

Abstract

Fine motor skill is associated with expressive language outcomes in infants who have an autistic sibling and in young autistic children. Fewer studies have focused on school-aged children even though around 80% have motor impairments and 30% remain minimally verbal (MV) into their school years. Moreover, expressive language is not a unitary construct, but it is made up of components such as speech production, structural language, and social-pragmatic language use. We used natural language sampling to investigate the relationship between fine motor and speech intelligibility, mean length of utterance and conversational turns in MV and verbal autistic children between the ages of 4 and 7 while controlling for age and adaptive behavior. Fine motor skill predicted speech production, measured by percent intelligible utterances. Fine motor skill and adaptive behavior predicted structural language, measured by mean length of utterance in morphemes. Adaptive behavior, but not fine motor skill, predicted social-pragmatic language use measured by number of conversational turns. Simple linear regressions by group corrected for multiple comparisons showed that fine motor skill predicted intelligibility for MV but not verbal children. Fine motor skill and adaptive behavior predicted mean length of utterance for both MV and verbal children. These findings suggest that future studies should explore whether MV children may benefit from interventions targeting fine motor along with speech and language into their school years.

Keywords: fine motor, minimally verbal, speech, language, social communication

Lay summary

Fine motor skill is related to expressive language in autistic children. In this study we examine the fine-motor expressive language relationship in 4–7-year-old autistic children include those who are minimally verbal. We examined three areas of expressive language–speech intelligibility, structural language, and social language use. We found that fine motor skill was related to speech intelligibility for minimally verbal children and related to structural language for minimally verbal and verbal children. The results suggest future studies should look at whether minimally verbal school-aged children may benefit from interventions that include fine motor skill along with expressive language.

Introduction

Autism is characterized by challenges in social communication and interaction as well as restricted interests and repetitive behaviors (American Psychiatric Association DSM-5 Task Force, 2013). Going beyond these primary diagnostic criteria, delays and disruptions in language and motor domains are common and persist into later childhood and adolescence for many autistic individuals. Estimates suggest that 50–74% of autistic preschool aged children are minimally verbal (MV) (Rose, Trembath, Keen, & Paynter, 2016; Thurm, Manwaring, Swineford, & Farmer, 2015), and 30% remain MV into their school years (Norrelgen et al., 2015; Rose et al., 2016; Smith, Mirenda, & Zaidman-Zait, 2007). Estimates suggest that 79–88% of autistic children have motor impairments (Bhat, 2021; Licari et al., 2019). Links between motor skill and language development are well-established (Iverson, 2010, for an overview). For infants who have an autistic older sibling, the quality of early motor development is related to expressive language into the toddler and preschool years (Bedford, Pickles, & Lord, 2016; Landa & Garrett-Mayer, 2006; LeBarton & Iverson, 2013; Leonard, Bedford, Pickles, Hill, & the BASIS Team, 2015; Stone & Yoder, 2001).

Several studies involving infants with older autistic siblings have reported associations between motor skill in early infancy and later expressive language outcomes. Choi, Leech, Tager-Flusberg, and Nelson (2018) investigated the fine motor–expressive language link in infant siblings using the Mullen Scales of Early Learning (MSEL; Mullen, 1995) Fine Motor, Visual Reception and Expressive Language subscales administered at 6, 12, 18 and 24 months. They found that infants with an eventual autism diagnosis had slower fine motor development trajectories than infants without an eventual diagnosis. In addition, they found that MSEL Fine Motor growth between 6 and 24 months predicted MSEL Expressive Language scores at age 3.

In a similar study, LeBarton and Landa (2019) investigated motor functioning in the first year of life and its relationship to later expressive language outcomes in infant siblings. Using the Peabody Developmental Motor Scales −2 (Folio & Fewell, 2000) Grasping, Visual-Motor integration, and Stationary subscales as well as the MSEL Expressive Language subscale, they found that motor skill at 6 months predicted expressive language at 30 and 36 months of age. These and numerous other studies point to the role of motor development in infancy as a contributing factor to pre-linguistic and expressive language outcomes in the second and third years of life (Bedford et al., 2016; Landa & Garrett-Mayer, 2006; LeBarton & Iverson, 2013; Leonard et al., 2015; Stone & Yoder, 2001).

Going beyond the early years, one recent study by Bal and colleagues (Bal et al., 2020) focused on preschool age predictors of expressive language at age 10, using the MSEL Fine Motor subscale, and the Vineland Adaptive Behavior Scales Expressive Language subscale. They found that the less delayed Fine Motor group (i.e., T-score > 20), compared to the extremely delayed fine motor group, showed greater gains in Vineland Expressive Language between age 4 and 10 even while controlling for initiation of joint attention, cognitive level, and age of walking.

To the best of our knowledge, only two studies have extended the focus to the relationship between motor functioning and expressive language in autistic children in their school years. Bhat (2021) examined a large sample from SPARK, a platform including data from thousands of individuals with ASD led by the Simons Foundation Autism Research Initiative (Feliciano et al., 2018). Bhat (2021) examined data from over 13,000 autistic children between the ages of 5 and 15. Bhat reported that 87% of the sample had signs of motor impairment. The relative risk ratio was 22 times greater than for the general population with risk increasing with increasing social communication and language impairments. In a follow-up study, Bhat and colleagues (Bhat, Boulton, & Tulsky, 2022) examined the motor–expressive language link in the same group of 5 to 15-year-old autistic children. Their goal was to understand which motor domains (gross-motor, including visuo-motor or multilimb coordination/planning, fine motor, or general coordination) best distinguished subgroups of school-aged autistic children and adolescents to predict core and co-occurring impairments, including social communication (measured by the Social Communication Questionnaire) and language delay (measured by parent reports), after accounting for age and sex. They found that visuomotor, fine motor and certain general coordination skills were best at predicting social communication impairments. All three motor dimensions explained variations in functional delays in social communication, severity of restricted and repetitive behaviors and cognitive delay. Visuomotor and fine motor skill best explained variation in language delay.

Most research on the motor-language relationship, however, has focused on the period between infancy and preschool when milestones in these domains are rapidly changing. Fewer studies have focused on autistic children in their early school years and even fewer have focused on MV children. Language and motor difficulties persist into later childhood, so it is imperative to ask to what degree the motor-language relationship is developmental rather than chronological. The earliest stages of language development are the preverbal stage (fewer than five different words) and the first words stage (up to 30 different words and mean length of utterance below 1.8) (Tager-Flusberg et al., 2009). While these stages correspond to 12–18 months of age in neurotypical development, autistic children who have language delays may be characterized by these initial stages of language development regardless of age (see, e.g., Butler et al., to appear, on 6–21-year-old autistic youth who are low and minimally verbal). The developmental approach to the motor-language link predicts that MV children (compared to those who are verbally fluent) in the age range of 4–7 will continue to show concurrent motor delays consistent with their language delays, while verbally fluent children will not, since they have surpassed key language development milestones. On the other hand, a chronological approach to the motor-language link predicts no differences between MV and verbal children by the age range of 4–7 because the motor-language relationship would only be relevant for very young children.

While Bhat and colleagues examined how different motor domains relate to social communication and language outcomes, no study has investigated different expressive language domains. Previous studies used broad measures of expressive language. Here, we use natural language sampling (NLS), a valid method for evaluating expressive communication with MV children (Barokova et al., 2020) in which a range of measures from various expressive communication domains can be derived (Barokova & Tager-Flusberg, 2018). We used percent intelligible utterances (%IUs) to measure speech production ability, mean length of utterance in morphemes (MLUm) as a measure of structural (or grammatical) language ability, and number of conversational turns (CTs) as a measure of social-pragmatic language use. We recruited 4–7-year-old children with heterogeneous language levels ranging from nonverbal to MV and verbal. We treated language level as a continuous variable based on number of different words (NDW), a measure of vocabulary development commonly used in the language development literature with typically developing children (e.g., Brown, 1973; Nelson, 1973) also used to categorize MV children (e.g., fewer than 20 different words in a 20-minute language sample, DiStefano, Shih, Kaiser, Landa, & Kasari, 2016; Kasari, Brady, Lord, & Tager-Flusberg, 2013). We used the Vineland Adaptive Behavior Scales (VABS-3) (Sparrow et al., 2016) as a developmental measure of adaptive behavior that is correlated with nonverbal cognitive ability (Alexander & Reynolds, 2020), since this study was carried out remotely and no measure of non-verbal intelligence was feasible. From the VABS-3, we used fine motor subdomain raw score as the fine motor measure. We focus on fine motor skill because it is the domain linked to social communication and expressive language in the recent studies by Bal and Bhat and colleagues (Bal et al., 2020; Bhat et al., 2022). Bal and colleagues found that fine motor rather than age of walking in 4.5-year-old autistic children predicted expressive communication outcomes at age 10.5. Similarly, Bhat and colleagues found that fine motor predicted both social communication and language impairment. Therefore, in this study, fine motor rather than gross motor, will be the focus.

In this study we addressed the following questions:

  1. Is fine motor skill related to a range of expressive language domains in 4–7-year-old autistic children while controlling for age and adaptive behavior (measured by VABS Standard Score) including:
    1. speech (measured by percent intelligible utterances)
    2. structural language (measured by mean length of utterance in morphemes)
    3. social-pragmatic language use (measured by number of conversational turns)?

  2. Do fine motor-expressive language relationships differ for MV children (NDW ≤ 20) compared to verbal children (NDW > 20)?

Methods

Participants

Participants included 90 autistic children (20 female) who were between the ages of 4 and 7 years (M=74.96 months, SD=12.85 months, Range=49–95 months) Families were recruited through social media advertising and the Simons Foundation Powering Autism Research for Knowledge (SPARK) research match registry (Feliciano et al., 2018). SPARK is a national ASD genotyping project that recruits families primarily through 31 U.S. academic medical centers with over 70,000 families enrolled. Once families enroll, they are offered the opportunity to continue hearing about and engaging in prospective research opportunities through their online research registry. SPARK has been shown to have high validity for autism diagnosis. Based on two different methods of confirming ASD diagnosis using electronic medical records Fombonne and colleagues (2021) found 98.8% agreement with SPARK cohort data. Written informed consent was obtained from all parents remotely using a webform with a virtual signature function. In addition, verbal assent from all children able to verbally assent was obtained prior to enrollment. This study was reviewed and approved by the Boston University IRB.

Measures

Vineland Adaptive Behavior Scales-Third Edition (VABS-3)

Parents completed the Vineland Adaptive Behavior Scales-Third Edition (VABS-3) (Sparrow et al., 2016) semi-structured interview, an individually administered measure of adaptive functioning used in the diagnosis of intellectual and developmental disabilities. The VABS-3 interview was administered remotely using Zoom by research-reliable technicians. The VABS-3 Standard Score is a composite of Communication, Socialization and Daily Living Domain scores representing a child’s overall adaptive behavior. The overall level of adaptive functioning is based on the Adaptive Behavioral Composite, the standard score (VABS-SS) (M=100; SD=15). Adaptive raw scores were computed at the subdomain level and converted to v-scale scores (M=15; SD=3). Table 1 shows the children’s VABS scores.

Table 1.

Age, VABS-3 and NLS scores of participants

Measure All participants N=90 Minimally verbal only N=48 Verbal only N=42
M SD Range M SD Range M SD Range U
Demographics Age in months 74.96 12.85 49–95 78.88 11.70 50–95 78.67 9.37 61–94 765.5
VABS-3 Adaptive Behavior Standard Score 58.21 13.51 31–84 49.29 12.17 31–76 66.03 10.12 30–94 217***
VABS-3 Fine motor raw score 53.36 21.37 20–94 39.15 17.13 20–79 65.67 16.44 30–94 211***
NLS Percent intelligible utterances 53.74 30.07 0–100 39.29 29.79 0–100 72.89 13.88 35–96 252.5***
NLS Mean length of utterance in morphemes 1.67 1.24 1–5.94 .87 .66 0–2.53 2.65 1.08 1.14–5.94 69.5***
NLS Number of conversational turns 58.59 42.45 0–170 29.07 28.95 0–112 93.17 29.66 33–170 111***
Open in a new tab

VABS-3 Fine Motor raw score (FM raw score)

We used the VABS-3 Motor Skills Domain Fine Motor Subdomain raw score. We used raw score as it is more sensitive for children with low fine motor skill. We included age in our statistical models to account for potential age-related differences. The fine motor subsection includes questions related to reaching, grasping, picking up and stacking objects. It also has questions related to drawing, coloring, cutting, writing, and erasing. Other items ask about skills for unwrapping, twisting, pouring, tying, manipulating small objects, and building complex toy structures. The Motor Skills Domain scores do not factor into the calculation of the VABS-3-Adaptive Behavior Composite Standard Score, and so they are independent.

Natural language samples

The natural language samples (NLS) were derived from 15-minute naturalistic parent-child interactions (Butler et al., 2022). The parent-child interactions were transcribed using the Systematic Analysis for Language Transcripts (SALT; (Miller & Iglesias, 2012)) procedures. In accordance with SALT procedures, utterances were segmented into communication units defined as an independent clause with its modifiers. A word was defined as a set of characters bound by spaces. Common phrases with co-occurring words that were spoken without pauses between them (e.g., “alldone”, “nothankyou”, “allgone”, “cleanup”, “gimme”, “kinda”) were transcribed as one word following transcription standards for children with ASD (La Valle, Plesa Skwerer, & Tager-Flusberg, 2020; Tager-Flusberg & Anderson, 1991; Tager-Flusberg et al., 2009). Words were transcribed using standard orthography to avoid erroneously increasing the number of different words used within and across transcripts. One researcher transcribed all utterances and marked bound morphemes according to SALT conventions. A second researcher then reviewed the file to proof the transcription. Transcription proofing involved reviewing the initial transcript while viewing the video of the parent-child interaction. Discrepancies were settled by the two researchers reaching a consensus in accordance with SALT conventions (Miller & Iglesias, 2012). In the rare case that a consensus could not be reached, the word or utterance in question was marked as unintelligible to avoid inflating the number of intelligible words produced.

All intelligible verbal utterances were included (including utterances that were interrupted or abandoned), since our focus was on number of different words rather than completeness of utterances. Unintelligible and nonverbal utterances were excluded. Following conventions for NLS with individuals with ASD (e.g., La Valle et al., 2020; Tager-Flusberg & Anderson, 1991), we did not include stereotyped language (e.g., echolalia, scripted recitation, and idiosyncratic language), signed language or alternative and augmentative communication (AAC) (e.g., speech generating devices). Established features of autistic speech, such as atypical prosody (e.g., Fusaroli et al., 2017) or pronoun reversals (e.g., Naigles et al., 2016), were not considered stereotyped unless they met the criteria described above.

Percent intelligible utterances (%IUs)

Percent intelligible utterances is a measure of how intelligible the child’s speech is to an unfamiliar listener (i.e., the trained transcriber). As previously mentioned, all utterances were given a first pass transcription by a trained transcriber and a second pass by a second trained transcriber. Discrepancies were settled by consensus, and the SALT guideline that no more than three passes should be given in determining intelligibility was followed. Percent intelligible utterances was derived through the SALT software, which automatically extracts the number of intelligible utterances over the total number of utterances. Only speech (versus non-speech) utterances were included. While intelligibility is a perceptual measure of speech production, as it is determined by an unfamiliar listener’s perception, measures of speech production are significantly correlated with measures of intelligibility (e.g., Chenausky, Gagné, Stipancic, Shield, & Green, 2022).

Mean length of utterance in morphemes (MLUm)

Mean length of utterance in morphemes (MLUm), a direct measure of grammatical complexity, was obtained via SALT. MLUm is calculated by dividing the total number of words (including morphemes) by the total number of utterances in a language sample (Barokova & Tager-Flusberg, 2018; Brown, 1973) For example, the word “play/ed” would count as a length of two morphemes: “play” and “/ed”. In the analysis of MLUm, only intelligible spontaneous speech utterances were included. MLUm was derived automatically using the SALT software.

Number of conversational turns (CTs)

A conversational turn is defined as an adult utterance followed by a child utterance (or vice versa). It is a single measure for the parent-child pair and is measured over the length of the language sample. CTs are a measure of the natural back-and-forth of conversation. A greater number of conversational turns means that the child is engaging in a high rate of back-and-forth conversation with the adult. A lower number of conversational turns means that one conversational partner (usually the adult) is doing most of the talking, and the other conversational partner (the child) is not frequently responding. If the adult talks, and the child never responds, the number of conversational turns is zero. In the analysis of conversational turns, only complete and intelligible speech utterances were included. CTs were derived automatically using the SALT software.

Defining minimally verbal using number of different words (NDW)

Number of different words (NDW) is a measure of vocabulary development. It includes only unique word mentions and includes only words roots (i.e., “shoe” and “shoes” are not counted as two separate words, but one word root– “shoe.”). NDW is frequently used to characterize MV children. e.g., NDW below 20 (DiStefano et al., 2016) or 30 (Tager-Flusberg et al., 2009) in a 20-minute sample. Definitions of language levels in autistic children vary (Bal, Katz, Bishop, & Krasileva, 2016; Koegel et al., 2020). While the use of the Autism Diagnostic Observation Scales (ADOS) (Lord et al., 2012) assignment to Module 1, preverbal to single words, Module 2, phrase speech, and Modules 3–4, verbally fluent, is the current best evidence (Bal et al., 2016), this study was conducted during the early stages of the COVID-19 pandemic when it was not possible for researchers to administer the ADOS-2. We base our characterization of language level on NDW in a 15-minute sample. As previously mentioned, NDW is typically collected during a 20-minute sample with an examiner. When autistic children are engaged with familiar caregivers in the home, however, they produce higher NDW (Barokova et al., 2020). This is particularly true for MV individuals. For these reasons, we based our measure of NDW on a 15-minute in-home parent child interaction to be closely comparable. As it related to the second research question, we defined minimally verbal children as those with NDW 20 or less during the parent-child interaction (N=48, M NDW = 5.21, SD = 6.13, Range = 0–20). The MV group includes children who were non-verbal, who did not produce any intelligible words during the 15-minute parent-child interaction. We defined verbal as those with NDW over 20 in the parent-child interaction (N=42, M NDW = 76.81, SD = 50.33, Range = 22–233). Groups did not differ significantly by age.

Analysis plan

Descriptive statistics

Shapiro tests determined that age, percent intelligible utterances, MLUm and CTs were normally distributed. VABS-3-SS and VABS-3 FM raw score were not, so we used a Spearman’s rho test to determine the correlations among variables with correction for multiple comparisons using the False Discovery Rate (Benjamini & Hochberg, 1995).

Question 1: Does fine motor skill relate to speech, structural language and social communication while controlling for age and adaptive behavior?

We used bidirectional stepwise regression to test the relationship between VABS-3 fine motor raw score and each outcome variable separately while controlling for the potential effects of age (in months) and adaptive behavior (VABS-SS). We conducted bidirectional stepwise regression for each outcome variable: 1) %IUs, 2) MLUm and 3) CTs). Due to correlation between VABS-3 fine motor raw score and VABS-SS, we transformed all predictor variables by centering around the mean to reduce multicollinearity. For the bidirectional stepwise regression, we started with a simple model including fine motor raw score only, the predictor variable of interest. Then, we added age in months. If age in months was not a statistically significant predictor, it was left out. If it was a statistically significant predictor, it was left in, given that it added significantly to the model in cross-validation. To cross-validate, the first and second models were compared using ANOVA. Given a statistically significant ANOVA, we retained the recently added variable as it added significantly to the variance explained by the model. Next, we added VABS-SS to the third model. If VABS-SS was not a statistically significant predictor, it was left out. To cross-validate, the first and third models were compared using ANOVA. Any predictor variable that added significantly to the variance of the model, as shown by model comparison using ANOVA, was retained. Statistical analyses were conducted using R (R Core Team, 2022).

Question 2: Do fine motor-expressive language relationships differ for minimally verbal children (NDW ≤ 20) compared to verbal children (NDW > 20)?

To investigate differences between MV and verbal children for those relationships that came out significant for Question 1, we used simple linear regression with fine motor raw score as the predictor and the relevant language sample variable (%IUs, MLUm) for the outcome with separate models for each group (MV, verbal) corrected for multiple comparisons using the False Discovery Rate (Benjamini & Hochberg, 1995).

Results

Participant scores on all measures are presented in Table 1 for the Vineland Adaptive Behavior Scales and the NLS measures. The correlations among all variables are shown in Table 2.

Table 2.

Spearman correlations among variables adjusted using the False Discovery Rate

Variable Age VABS-SS FM raw score %IUs MLUm CTs
Age - −.32** .17 .05 .01 −.19
VABS-SS −.32** - .62*** .38*** .68*** .60**
FM raw score .17 .62*** - .40*** .60*** .43**
%IUs .05 .38*** .40*** - .62*** .51**
MLUm .01 .68*** .60*** .62*** - .75**
CTs −.19 .60*** .43*** .51*** .75*** -
Open in a new tab

Note. Entries above the diagonal are adjusted for multiple tests using the False Discovery Rate.

*

Indicates p < .05.

**

Indicates p < .01.

***

Indicates p < .001.

Fine motor skill and %IUs

Table 3 presents the results of the stepwise regression analysis for %IUs. Fine motor raw score significantly predicted %IUs, while age and adaptive behavior did not add significantly to the model. Figure 1 shows the relationship between fine motor raw score and %IUs for MV compared to verbal children. In a simple regression model, fine motor skill significantly predicted %IUs for MV children (ß=.8, SE=.39, t=2.050, p<.05). In a simple regression model for verbal children, fine motor skill did not significantly predict %IUs (ß=−.049, SE=.219, t=−.224, p=.824). These results suggest that the relationship between fine motor skill and speech intelligibility is being driven by MV rather than verbal children (see Figure 1).

Table 3.

Stepwise regression results using percent intelligible utterances as the outcome

Step Predictor ß SE t Fit Model comparison
1 (Intercept) 54.31 *** 2.80 19.37 R2=.19
FM raw score 1.06 *** .23 4.53 Adj R2=.18
2 (Intercept) 54.31*** 2.82 19.27
FM raw score 1.08*** .24 4.48 R2=.19 Model 1 and Model 2:
Age in months −.08 .22 −.34 Adj R2=.17 F(87,88)=.11, p=.736
3 (Intercept) 54.31*** 2.76 19.68
FM raw score .69* .3 2.32 R2=.22 Model 1 and Model 3:
VABS-SS .51 .26 1.97 Adj R2=.21 F(87,88)=3.87,p=.052
Open in a new tab

Note.

*

Indicates p < .05.

**

indicates p < .01.

***

indicates p < .001.

The final model resulting from the stepwise regression analysis is shown in boldface type.

Figure 1.

Figure 1

Open in a new tab

Fine motor skill predicts percent intelligible utterances for MV children only

Fine motor skill and MLUm

Table 4 presents the results of the stepwise regression analysis for MLUm. Fine motor raw score significantly predicted MLUm and accounted for 32% of the variance in MLUm. In addition to fine motor raw score, adaptive behavior significantly predicted MLUm, and combined with fine motor raw score, accounted for 44% of the variance. Figure 2 shows the relationship between fine motor raw score and MLUm for MV and verbal children. In a simple regression model for MV children, fine motor significantly predicted MLUm (ß=.026, SE=.008, t=3.199, p<.01). Similarly, in a simple regression model for verbal children, fine motor significantly predicted MLUm (ß=.033, SE=.016, t=2.037, p<.05). These results suggest that fine motor skill is related to MLUm for both MV and verbal children.

Table 4.

Stepwise regression results using mean length of utterance in morphemes as the outcome

Step Predictor ß SE t Fit Model comparison
1 (Intercept) 1.67*** .11 15.49 R2=..32
FM raw score .06*** .01 6.5 Adj R2=.32
2 (Intercept) 1.67*** .11 15.44
FM raw score .06*** .01 6.46 R2=.33 Model 1 and Model 2:
Age in months −.01 .01 −.62 Adj R2=.31 F(87,88)=.38, p=.537
3 (Intercept) 1.67 *** .1 17.07
FM raw score .03 ** .01 2.72 R 2 =.45 Model 1 and Model 3:
VABS-SS .04 *** .01 4.46 Adj R 2 =.44 F(87,88)=19.9,p<.001
Open in a new tab

Note.

*

Indicates p < .05.

**

indicates p < .01.

***

indicates p < .001.

The final model resulting from the stepwise regression analysis is shown in boldface type.

Figure 2.

Figure 2

Open in a new tab

Fine motor skill predicts mean length of utterance in morphemes for MV and verbal children

Minimally verbal is typically defined as those over the age of five (Tager-Flusberg & Kasari, 2013), while children under the age of five are considered “preverbal” as they may still go on to develop more advanced language skills. Our analyses included children ages 4–7. To verify that younger children were not driving these effects, we re-ran our analyses to include only 5–7-year-old children, and the same results were obtained (except for %IUs for MV children, which dropped to marginal significance). Fine motor raw score marginally predicted %IUs for MV children (ß =.83, SE=.41, t=2.00, p=.052) but not for verbal children (ß=.01, SE=.27, t=.03, p=.976) over the age of five. Similarly, fine motor raw score predicted MLUm for MV (ß =.02, SE=.01, t=2.97, p<.01) and verbal children (ß=.04, SE=.02, t=2.14, p<.05) over the age of five.

Fine motor skill and CTs

Table 5 presents the results of the stepwise regression analysis predicting number of conversational turns. Fine motor raw score alone accounted for 18% of the variance in CTs. Fine motor and age accounted for 24% of the variance, a significantly greater percentage. Fine motor, age and adaptive behavior accounted for 36% of the variance in CTs, a significantly greater percentage. Adaptive behavior was the only significant predictor of CTs in the final stepwise model. To ensure that caregiver talkativeness did not contribute to the number of conversational turns, we re-ran the model with caregiver words per minute included as a predictor. There was no statistical significance between the model without caregiver talkativeness and the model with, as determined by AVOVA (F(85,86)=1.771, p=.187).

Table 5.

Stepwise regression results using number of conversational turns as the outcome

Step Predictor ß SE t Fit Model comparison
1 (Intercept) 58.59*** 4.05 14.48 R2=..19
FM raw score 1.54*** .34 4.57 Adj R2=.18
2 (Intercept) 58.59*** 3.91 14.98
FM raw score 1.72*** .33 5.17 R2=.25 Model 1 and Model 2:
Age in months −.83 .31 −2.67 Adj R2=.24 F(87,88)=.7.14, p<.01
3 (Intercept) 58.59 *** 3.59 16.31
FM raw score .29 .46 .65 R 2 =.38 Model 2 and Model 3:
VABS-SS .1.72 *** .42 4.13 Adj R 2 =.36 F(86,87)=17.06,p<.001
Open in a new tab

Note.

*

Indicates p < .05.

**

indicates p < .01.

***

indicates p < .001.

The final model resulting from the stepwise regression analysis is shown in boldface type.

Discussion

The present study adds to the small but growing body of research on the relationship between fine motor skill and expressive language in school aged autistic children. Expressive language does not form a unitary skill, so one of the aims of this study was to examine three canonical areas of expressive language: speech (measured by %IUs), structural language (measured by MLUm) and social-pragmatic language use (measured by CTs). We expanded on previous work by examining these three domains of expressive language measured by NLS. Using NLS, we could compare across MV and verbal children with a single direct measure of expressive communication. We found that fine motor skill predicted speech intelligibility but only for MV children. We found that fine motor skill predicted MLUm for MV and verbal children. On the other hand, adaptive behavior predicted conversational turns beyond the effects of age and fine motor skill.

Fine motor is related to speech intelligibility only for MV children

Our results showed that fine motor skill predicted %IUs over the potential effects of adaptive behavior and age. In a simple regression analysis of fine motor, this effect was only significant for MV children, not verbal children. This finding suggests that fine motor is related to speech production, but the relationship for verbal children is less clear. As previously discussed, producing speech requires the coordination of complex oral motor movements. These findings align with studies of motor speech impairment in MV children. Estimates suggest that at least 25% of MV individuals have signs of childhood apraxia of speech, and another 25% show signs of motor speech impairment more broadly, even if they do not meet strict criteria for childhood apraxia of speech (Chenausky et al., 2019). This result suggests that the fine motor—speech relationship is developmental rather than chronological since there is evidence in MV children 4–7-year-old children but not in those who are more highly verbal. Children who are more highly verbal do not show the same fine motor—speech intelligibility relationship because their speech has developed past a point at which this relationship is central and/or they are not affected by motor speech impairments.

Fine motor skill and adaptive behavior are related to structural language

Our results showed that fine motor skill along with adaptive behavior predicted MLUm while age did not. In simple regression models looking at the fine motor—MLUm relationship in MV compared to verbal children, we found that fine motor predicted MLUm in both groups. Even in verbal children, fine motor skill was related to their abilities to produce complex language that incorporates longer sentences with morphological components. As MV individuals are often described as having no phrase speech (e.g., Chenausky et al., 2019; Lord et al., 2012), the question arises of what kind of multi-word utterances MV children are producing. While this question is out of the scope of the current study, Butler and colleagues (to appear) investigated the morphosyntactic features of spoken language in 49 6–21-year-old individuals who were MV (defined as those administered an ADOS Module 1) and low verbal (defined as those administered an ADOS Module 2). They reported that those with NDW between 1 and 20 had MLUm of 1.3. While they found little to no evidence of the use of early developing morphosyntactic forms (such as the plural –s or the present progressive –ing) in those with NDW below 60, there was evidence of multi-word utterances, such as “blue car” and “me now.”

When we looked across increasingly larger chunks of expressive communication, from the intelligibility of speech sounds to the length of utterance to the number of conversational turns throughout the language sample, we observed a gradient effect in which the influence of fine motor waned as the influence of adaptive behavior increased. Fine motor skill was related to speech intelligibility, but only for 4–7-year-old children who were MV. Fine motor skill and adaptive behavior predicted structural language (MLUm), and fine motor predicted MLUm in 4–7-year-old children who were verbal and MV. When we looked at conversational turns throughout the language sample, adaptive behavior was the only significant predictor. These findings raise the question of to what extent speech intelligibility contributes to adaptive behavior (see e.g., Baghdadli et al., 2012), as children with less intelligible speech face high frustration when they are not understood and unable to communicate their wants and needs. While a mediation analysis with the present dataset is feasible, future studies should be designed to directly target the question (Rohrer et al., 2022) of to what extent speech production ability mediates the relationship between fine motor skill and other domains of expressive language.

Profiles of minimally verbal children

Furthermore, it is important to note the differences between MV and verbal children in areas other than the fine motor–expressive language association, as profiles of MV individuals may be fundamentally different. In terms of speech production, Chenausky and colleagues (Chenausky, Brignell, Morgan, and Tager-Flusberg, 2019) examined the relationship between expressive language and signs of motor speech impairment in MV and low verbal autistic children. They found that speech production skill predicted expressive language, but only for the MV group. In the low verbal group that had higher expressive language, receptive language was the best predictor.

In the area of word learning, Joseph and colleagues (Joseph, Plesa Skwerer, Eggleston, Meyer, and Tager-Flusberg, 2019) presented an experimental novel word learning task to MV children and adolescents. They had participants learn novel words and tested them for retention 2 hours later. They found that language level, in particular expressive vocabulary level, differentiated between participants were and were not successful at novel word retention, while controlling for differences in nonverbal IQ.

In a pragmatic analysis of language samples from verbal and MV children, adolescents and young adults, La Valle et al. (2020) found that MV youth had significantly different profiles of pragmatic language use compared to those who were verbal. Minimally verbal youth used agree/acknowledge/disagree, responding to a question, and requesting functions. For verbal youth, commenting was the primary pragmatic function.

Along similar lines, in the present study, we found that fine motor skill affects expressive language in the domain of speech production but only for MV children. Taken together, these findings suggest that MV youth have fundamentally different profiles than verbal youth across the domains of speech, word learning, pragmatics, and the fine motor–expressive communication relationship.

Clinical Implications

In terms of clinical practice, these results suggest that future studies should explore whether MV school-aged children benefit from fine motor skill incorporated into assessments and interventions targeting expressive communication. Considering that motor impairments in autistic children are under-diagnosed, clinicians (i.e., pediatricians, psychologists, speech-language pathologists) should be attentive to potential motor impairments in autistic children and refer to occupational and physical therapists for further evaluation of fine motor impairments. Bhat et al. (2022) found that 87% of children with ASD in their sample showed signs of motor impairments, but only 15% had a co-diagnosis of developmental coordination disorder suggesting that motor impairments are not only under-researched but also under-treated in clinical practice. Understanding to what extent the link between fine motor and expressive language is associated with motor deficits that more specifically affect speech production (i.e., oral motor skill) will be particularly important for informing specific intervention targets.

For MV children, evidence is also lacking in terms of best practice for communication interventions (Brignell, Chenausky, Song, Suo, & Morgan, 2018; Koegel, Bryan, Su, Vaidya, & Camarata, 2019). A recent study of preschool aged autistic children suggests that fine motor skill plays a role in response to a social communication intervention (JASPER; Joint Attention, Symbolic Play, Engagement and Regulation) in MV children. Those with lower fine motor skills were more likely to respond more slowly to the social communication intervention as measured by change scores from expressive language age equivalents on the Mullen Scales of Early Learning (Panganiban and Kasari, 2022). While more research is necessary to identify more specific links between fine motor skill and expressive communication that translate into assessment and intervention, co-treatment between occupational therapists and speech-language pathologists to simultaneously target fine motor and expressive language goals could have reinforcing benefits for autistic children.

Limitations and future directions

One of the aims of this study was to use NLS to obtain direct observational measures of naturalistic communication from which we derived measures of speech, structural language, and social-pragmatic language use. Our fine motor measure, however, came from parent report–the Vineland Adaptive Behavior Scales. Future work would benefit from direct observational or experimental measures of motor skill in autistic children to identify the specific motor skills associated with producing intelligible speech and increasingly complex language.

Conclusions

In this study, we used natural language sampling to understand the relationship between fine motor skill and three domains of expressive language–speech intelligibility, mean length of utterance and conversational turns–in MV and verbal autistic children between the ages of 4 and 7 while controlling for age and adaptive behavior. We found that fine motor skill predicted speech intelligibility but only for MV children. Fine motor and adaptive behavior predicted structural language for both MV and verbal children. These findings suggest that future studies should explore whether MV children may benefit from interventions targeting fine motor and speech and language skill into their school years. Altogether, our results suggest that the fine motor–speech/language link is developmental, and more attention to the relationship between motor and language skills in school aged is warranted to improve assessment and intervention, particularly for children who are MV.

Acknowledgements

The authors would like to thank the following individuals for their roles in data collection and transcription of the parent-child interactions for this study: Lue Shen, Chelsea La Valle, Sophie Schwartz, Joseph Palana, Judith Licht, Hazel Harvey-Barker, Cerelia Liu, Natalie Peterman, Maria Ayoub. Thanks to Nick Wagner for comments on the statistical analysis plans.

Funding statement

This research was funded by NIH NIDCD P50DC18006 (MPI: Tager-Flusberg/Kasari).

Footnotes

Conflict of Interest Statement

The authors have no conflicts of interest related to this publication to report.

References

  1. Alexander RA, & Reynolds MR (2020). Intelligence and adaptive behavior: A meta-analysis. School Psychology Review, 49(2), 85–110. [Google Scholar]
  2. American Psychiatric Association DSM-5 Task Force. (2013). Diagnostic and statistical manual of mental disorders: DSM-5™ (5th ed.). [Google Scholar]
  3. Baghdadli A, Assouline B, Sonié S, Pernon E, Darrou C, Michelon C, Picot M-C, Aussilloux C, & Pry R (2012). Developmental Trajectories of Adaptive Behaviors from Early Childhood to Adolescence in a Cohort of 152 Children with Autism Spectrum Disorders. Journal of Autism and Developmental Disorders 42, 1314–1325. [DOI] [PubMed] [Google Scholar]
  4. Bal VH, Fok M, Lord C, Smith IM, Mirenda P, Szatmari P, ... Zaidman-Zait A (2020). Predictors of longer-term development of expressive language in two independent longitudinal cohorts of language-delayed preschoolers with autism spectrum disorder. Journal of Child Psychology and Psychiatry, 61(7), 826–835. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bal VH, Katz T, Bishop SL, & Krasileva K (2016). Understanding definitions of minimally verbal across instruments: evidence for subgroups within minimally verbal children and adolescents with autism spectrum disorder. Journal of Child Psychology and Psychiatry, 57(12), 1424–1433. [DOI] [PubMed] [Google Scholar]
  6. Barokova M, La Valle C, Hassan S, Lee C, Xu M, McKechnie R, ... Tager-Flusberg H (2020). Eliciting language samples for analysis (ELSA): A new protocol for assessing expressive language and communication in autism. Autism Research, 14(1), 112–126. [DOI] [PubMed] [Google Scholar]
  7. Barokova M, & Tager-Flusberg H (2018). Commentary: Measuring language change through natural language samples. Journal of Autism and Developmental Disorders, 50(7), 2287–2306. [DOI] [PubMed] [Google Scholar]
  8. Bedford R, Pickles A, & Lord C (2016). Early gross motor skills predict the subsequent development of language in children with autism spectrum disorder. Autism Research, 9(9), 993–1001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Benjamini Y, & Hochberg Y (1995). Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society Series B, 57, 289–300. [Google Scholar]
  10. Bhat AN (2021). Motor impairment increases in children with autism spectrum disorder as a function of social communication, cognitive and functional impairment, repetitive behavior severity, and comorbid diagnoses: A SPARK study report. Autism Research, 14(1), 202–219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Bhat AN, Boulton AJ, & Tulsky DS (2022). A further study of relations between motor impairment and social communication, cognitive, language, functional impairments, and repetitive behavior severity in children with ASD using the SPARK study dataset. Autism Research, 15(6), 1156–1178. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Brignell A, Chenausky KV, Song H, Suo C, & Morgan AT (2018). Communication interventions for autism spectrum disorder in minimally verbal children. Cochrane Database Systematic Reviews, 5(11). [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Brown R (1973). A first language: The early stages. Cambridge, MA: Harvard University Press. [Google Scholar]
  14. Butler LK, La Valle C, Schwartz S, Palana JB, Liu C, Peterman N, ... Tager-Flusberg H (2022). Remote natural language sampling of parents and children with autism spectrum disorder: Role of activity and language level. Frontiers in Communication, 7. [Google Scholar]
  15. Butler LK, Shen L, Chenausky KV, La Valle C, Schwartz S, & Tager-Flusberg H (To appear). Lexical and morphosyntactic profiles of autistic youth with minimal or low spoken language skills. American Journal of Speech-Language Pathology. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Chenausky K, Brignell A, Morgan A, & Tager-Flusberg H (2019). Motor speech impairment predicts expressive language in minimally verbal, but not low verbal, individuals with autism spectrum disorder. Autism and Developmental Language Impairments, 4, 1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Chenausky KV, Gagné D, Stipancic KL, Shield A, & Green JR (2022). The relationship between single-word speech severity and intelligibility in childhood apraxia of speech. Journal of Speech, Language and Hearing Research, 65(3), 843–857. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Choi AB, Leech KA, Tager-Flusberg H, & Nelson CA (2018). Development of fine motor skills is associated with expressive language outcomes in infants at high and low risk for autism spectrum disorder. Journal of Neurodevelopmental Disorders, 10(14). [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. DiStefano C, Shih W, Kaiser A, Landa R, & Kasari C (2016). Communication growth in minimally verbal children with ASD: The importance of interaction. Autism Research, 9(10), 1093–1102. [DOI] [PubMed] [Google Scholar]
  20. Feliciano P, Daniels AM, Green Snyder L, Beaumont A, Camba A, Esler A, ... The SPARK Consortium (2018). SPARK: A US cohort of 50,000 families to accelerate autism research. Neuron, 97(3), 488–493. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Folio MR, & Fewell RR (2000). Pdms-2: Peabody developmental motor scales. Austin, TX: Pro-Ed. [Google Scholar]
  22. Fombonne E, Coppola L, Mastel S, & O’Roak BJ (2021). Validation of autism diagnosis and clinical data in the SPARK cohort. Journal of Autism and Developmental Disorders, 52(8), 3383–3398. [DOI] [PubMed] [Google Scholar]
  23. Fusaroli R, Lambrechts A, Bang D, Bowler DM, Gaigg SB (2017) Is voice a marker for Autism spectrum disorder? A systematic review and meta-analysis. Autism Research, 10, 384–407. [DOI] [PubMed] [Google Scholar]
  24. Iverson J (2010). Developing language in a developing body: The relationship between motor development and language development. Journal of Child Language, 37, 229–261. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Joseph RM, Plesa Skwerer D, Eggleston B, Meyer SR, & Tager-Flusberg H (2019). An experimental study of word learning in minimally verbal children and adolescents with autism spectrum disorder. Autism and Developmental Language Impairments, 4, 1–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kasari C, Brady N, Lord C, & Tager-Flusberg H (2013). Assessing the minimally verbal school-aged child with autism spectrum disorder. Autism Research, 6(6), 479–493. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Koegel LK, Bryan KM, Su PL, Vaidya M, & Camarata S (2019). Intervention for non-verbal and minimally-verbal individuals with autism: A systematic review. International Journal of Pediatric Research, 5(2). [Google Scholar]
  28. Koegel LK, Bryan KM, Su PL, Vaidya M, & Camarata S (2020). Definitions of nonverbal and minimally verbal in research for autism: A systematic review of the literature. Journal of autism and developmental disorders, 50(8), 2957–2972. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. La Valle C, Plesa-Skwerer D, & Tager-Flusberg H (2020). Comparing the pragmatic speech profiles of minimally verbaland verbally fluent individuals with autism spectrum disorder. Journal of Autism and Developmental Disorders, 50(10), 3699–3713. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Landa R, & Garrett-Mayer E (2006). Development in infants with autism spectrum disorders: a prospective study. Journal of child psychology and psychiatry, and allied disciplines, 47(6), 629–638. [DOI] [PubMed] [Google Scholar]
  31. LeBarton ES, & Iverson JM (2013). Fine motor skill predicts expressive language in infant siblings of children with autism. Developmental Science, 16(6), 815–827. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. LeBarton ES, & Landa RJ (2019). Infant motor skill predicts later expressive language and autism spectrum disorder diagnosis. Infant Behavior and Development, 54, 37–47. [DOI] [PubMed] [Google Scholar]
  33. Leonard HC, Bedford R, Pickles A, Hill EL, & the BASIS Team. (2015). Predicting the rate of language development from early motor skills in at-risk infants who develop autism spectrum disorder. Research in Autism Spectrum Disorders, 13–14, 15–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Licari MK, Alvares GA, Varcin K, Evans KL, Cleary D, Reid SL, ... Whitehouse A (2019). Prevalence of motor difficulties in autism spectrum disorder: Analysis of a population-based cohort. Autism Research, 13(2), 298–306. [DOI] [PubMed] [Google Scholar]
  35. Lord C, Rutter M, DiLavore PC, Risi S, Gotham K, & Bishop SL (2012). Autism diagnostic observation schedule (Second ed.). Torrance, CA: Western Psychological Services. [Google Scholar]
  36. Miller J, & Iglesias A (2012). Systematic analysis of language transcripts (SALT) (Computer software). SALT Software. (Research Version 2012) [Google Scholar]
  37. Mullen EM (1995). Mullen scales of early learning [Computer software manual]. Circle Pines, MN. [Google Scholar]
  38. Naigles LR, Cheng M, Rattansone NX, Tek S, Khetrapal N, Fein D, & Demuth K (2016). “You’re telling me!” The prevalence and predictors of pronoun reversals in children with autism spectrum disorder and typical development. Research in Autism Spectrum Disorder, 27, 11–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Nelson K (1973). Structure and strategy in learning to talk. Monographs of the Society for Research in Child Development, 38(1–2, Serial No 149), 136. [Google Scholar]
  40. Norrelgen F, Fernell E, Eriksson M, Hedvall A, Persson C, Sjölin M, & Kjellmer L (2015). Children with autism spectrum disorders who do not develop phrase speech in the preschool years. Autism, 19(8), 934–943. [DOI] [PubMed] [Google Scholar]
  41. Panganiban J, & Kasari C (2022). Super responders: Predicting language gains from JASPER among limited language children with autism spectrum disorder. Autism Research, 15(8), 1565– 1575. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Pecukonis M, Plesa Skwerer D, Eggleston B, Meyer S, & Tager-Flusberg H (2019). Concurrent social communication predictors of expressive language in minimally verbal adolescents with autism spectrum disorder. Journal of Autism and Developmental Disorders, 49, 3767–3785. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. R Core Team. (2022). R: A language and environment for statistical computing (Tech. Rep.). Vienna, Austria: R Foundation for Statistical Computing. [Google Scholar]
  44. Rohrer JM, Hünermund P, Arslan RC, & Elson M (2022). That’s a Lot to Process! Pitfalls of Popular Path Models. International Journal of Distributed Sensor Networks. 10.1177/15501477211053997 [DOI] [Google Scholar]
  45. Rose V, Trembath D, Keen D, & Paynter J (2016). The proportion of minimally verbal children with autism spectrum disorder in a community-based early intervention programme. Journal of Intellectual Disability Research, 60(5), 464–477. [DOI] [PubMed] [Google Scholar]
  46. Smith V, Mirenda P, & Zaidman-Zait A (2007). Predictors of expressive vocabulary growth in children with autism. Journal of Speech, Language, and Hearing Research, 50, 149–160. [DOI] [PubMed] [Google Scholar]
  47. SPARK Consortium (2018). SPARK: A US Cohort of 50,000 Families to Accelerate Autism Research. Neuron, 97(3), 488–493. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Sparrow SS, Cicchetti DV, & Saulnier CA (2016). Vineland Adaptive Behavior Scales (Third ed.) San Antonio, TX. [Google Scholar]
  49. Stiegler LN (2015). Examining the echolalia literature: Where do speech-language pathologists stand? American Journal of Speech-Language Pathology, 24, 750–762. [DOI] [PubMed] [Google Scholar]
  50. Stone WL, & Yoder PJ (2001). Predicting spoken language level in children with autism spectrum disorders. Autism, 5(4), 341–361. [DOI] [PubMed] [Google Scholar]
  51. Tager-Flusberg H, & Anderson M (1991). The development of contingent discourse ability in autistic children. Journal of Child Psychology and Psychiatry, 32(7), 1123–1134. [DOI] [PubMed] [Google Scholar]
  52. Tager-Flusberg H, & Kasari C (2013). Minimally verbal school-aged children with autism spectrum disorders: The neglected end of the spectrum. Autism Research, 6(468–478). [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Tager-Flusberg H, Rogers S, Cooper J, Landa R, Lord C, Paul R, ... Yoder P (2009). Defining spoken language benchmarks and selecting measures of expressive language development for young children with autism spectrum disorders. Journal of Speech, Language & Hearing Research, 52(3), 643–656. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Thurm A, Manwaring SS, Swineford L, & Farmer C (2015). Longitudinal study of symptom severity and language in minimally verbal children with autism. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 56(1), 97–104. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES