Do motor difficulties really explain speech/language difficulties in autism?

(Cross-posted at

This post is the fifth in a series of critiques of articles co-authored by Morton Gernsbacher. As I noted earlier, these articles collectively attempt to present evidence for the drastic redefinition of autism upon which the plausibility of FC depends: namely, the notion that autism is not (despite eight decades of research to the contrary) a socio-cognitive disorder, but rather a motor disorder. More specifically, autism is, purportedly, a disorder in which intentional motor movements, including speaking and pointing, are difficult or impossible to perform.

Today’s Gernsbacher article, “Infant and toddler oral- and manual-motor skills predict later speech fluency in autism,” relates more directly to the motor take on autism than the articles I have discussed already. 

Gernsbacher et al. start by surveying the previous research, beginning with the not-surprising finding that oral motor skills, along with other motor skills, may correlate with speech motor skills. But then the authors go further, citing research that, they claim, shows that other aspects of language, beyond speech articulation, are also linked to motor skills. In particular they claim, citing Thurm et al. (2007) and Stone and Yoder (2001), that autistic toddlers’ oral-motor skills and manual-motor skills are more predictive of their preschool speech fluency than are measures of social cognition, such as joint attention (JA).

And it is with these citations that the problems begin. That’s because neither Thurm et al. nor Stone and Yoder say quite what Gernsbacher et al. suggest.

Thum et al. found that the strongest predictors for language at age five were non-verbal cognitive ability at age two and communication scores at age three. They also state that, among those who had relatively strong non-verbal cognitive skills but did not develop language by age five, “Early joint attention as well as vocal and motor imitation skills were more impaired.” They add that “Certain specific skills stood out as potentially powerful predictors—imitation of sounds for expressive language, and responding to joint attention for receptive language.”

None of this is surprising. It is unsurprising that cognitive skills, communication skills, and early joint attention behaviors all contribute to language skills at age five. And it goes without saying that an inability to imitate sounds fatally undermines expressive language.

As for Stone and Yoder, they looked at predictors at age two of language levels at age four. But first they controlled for language skills at age two. Having done this, they found that, among child variables like play level, motor imitation ability and joint attention, and among environmental variables like socioeconomic status and hours of speech/language therapy, the only significant predictors were motor imitation and number of hours of speech/language therapy.

On closer inspection, however, these results do not support the claim that motor skills are a major player in speech/language development in autism.

For one thing, motor imitation skills, as Stone and Yoder note, are a function not just of motor skills, but also of socio-cognitive skills. As they point out, “motor imitation, like spoken language, is a multi-faceted behavior with social as well as representational components.” Indeed, it’s likely that (1) motor imitation skills improve with practice, and (2) how often kids imitate others is a function of social motivation.

As for Joint Attention (JA), Stone and Yoder do note that it becomes non-significant after language at age two is controlled for. But other studies indicate that JA, especially RJA (“responding to joint attention,” or attending to what other people are attending to), has a big effect on early word learning (see Baron-Cohen et al., 2007). Early word learning, of course, is a building block for future language development. Controlling for language at age two therefore may entail controlling for JA—and for subsequent language development. This, importantly, will mask any effects that JA may have on language development.

Furthermore, Stone and Yoder concede that the only component of JA they measured was IJA (“initiating joint attention,” or trying to get others to attend to what we’re attending to). They acknowledge that RJA (attending to what others are attending to) may play a more significant role in language development.

In short, Stone and Yoder do not find that motor skills, per se, predict language development, nor do they rule out a significant role for JA.

Let’s now turn to Gernsbacher et al.’s own study. Their results do indeed suggest that infant and toddler oral-motor and manual-motor skills (1) inter-correlated significantly, (2) distinguished autistic children from non-autistic children, and (3) correlate with levels of speech fluency.

But on closer inspection, these findings do not support a major role for motor skills in language development in autism.

The first question that arises is to what extent “fluency” inherently includes oral-motor skills. To the extent that it does, some of these correlations are unsurprising.

Fluency, Gernsbacher et al. say, was measured by impressions (by parents and others) of intelligibility and language level—specifically how close to age level a child’s speech seemed to be, how intelligible it was, and any atypicalities in rate, pitch volume, and echolalia. A “how close to age-level” criterion, potentially, could include any number of linguistic factors, but all the specific factors the authors explicitly enumerate relate more to the mechanics of speech production and, as such, to motor issues. As for other characteristics of speech like vocabulary and syntax, these were not specifically probed.

Another question is how Gernsbacher et al. measured motor skills. These, too, they based largely on parental impressions of, as well as on video data on, a variety of motor activities. Included in these activities was pointing. Just as would be predicted by the diagnostic criteria for autism, Gernsbacher et al. find less pointing by autistic participants compared to typical participants. This was true, the authors note, both for on-demand pointing (pointing in response to someone asking you to point) and pointing in order to request something.

Deviating from the diagnostic criteria, and echoing Gernsbacher, Stevenson, Khandakar, and Goldsmith (2008) (discussed in an earlier post), Gernsbacher et al. claim that pointing is diminished in autism because it is motorically challenging for autistic individuals. As we noted in our earlier critique, however, pointing (even proto-imperative pointing, i.e., pointing for the sake of requesting) has a social dimension. Furthermore, failure to point “on demand” may reflect receptive language deficits and/or a lack of (social) motivation.

Many of the other motor skills Gernsbacher et al. probe in this study were also measured by whether they were performed “on request” or “in imitation,” or whether they were activities the child “likes to” do. This once again raises questions about comprehension (of the request), social motivation (for imitation), and/or will (desire to perform the action), on one hand, vs. ability (actual motor skill), on the other. Indeed, behaviors in which the language confound was eliminated, the social component was minimized, and the preferences were less likely to vary showed smaller percentage differences between “successful” autistics vs. non-autistics:

  • Grabs a dangling toy (69% ASD, vs. 98% typical)
  • Reaches into a cake (74% vs. 95%)

For similar behaviors that more directly involved another person and therefore included a social component, we see bigger differences:

  • Grabbing dangling earrings (47% vs. 76%)
  • Taking glasses off someone’s face (49% vs. 79%)

For clapping (54% to 98%), which is often a social activity, there were bigger differences still. Stacking blocks also turns up a big difference (39% to 81%), but stacking is as much of a cognitive skill as a motor skill. 

When it comes to cognitive factors, one thing that is notably absent from these various correlations is information about IQ scores. All Gernsbacher et al. provide is age and diagnosis, with subjects matched with non-autistic controls on age/gender. But it is well-known that motor scores correlate strongly with IQ.

When we consider all these factors, Gernsbacher et al.’s conclusion is unsurprising. In particular, speech fluency as they operationalize it has a significant oral-motor component; low language levels correlate with low non-verbal IQ, which in turn correlates with motor difficulties; and several of the “motor” activities they probe are relatively impaired in autistic vs. non-autistic participants in part as a function of how social they are.

Left out of Gernsbacher et al.’s study is receptive language, which, as Thurm et al. note, is relatively more impaired (and more correlated with JA) than expressive language is. Where language in general is concerned, as Thurm et al. put it, “ [m]ore children at age 5 had minimal receptive language than expressive language, indicating that deficits were not primarily oral-motor.”

Indeed, where language/language learning in general are concerned, none of Gernsbacher et al.’s findings support a major role for motor impairment, nor do they diminish the importance of joint attention.


Baron-Cohen, S., Baldwin, D. A., & Crowson, M. (1997). Do children with autism use the speaker’s direction of gaze strategy to crack the code of language?. Child development68(1), 48–57.

Gernsbacher, M. A., Sauer, E. A., Geye, H. M., Schweigert, E. K., & Goldsmith, H. H. (2008). Infant and toddler oral- and manual-motor skills predict later speech fluency in autism. Journal of Child Psychology and Psychiatry, 49, 43–50.

Gernsbacher, M. A., Stevenson, J. L., Khandakar, S., & Goldsmith, H. H. (2008). Why Does Joint Attention Look Atypical in Autism?. Child development perspectives2(1), 38–45.

Mundy, P., Sigman, M., & Kasari, C. (1990). A longitudinal study of joint attention and language development in autistic children. Journal of autism and developmental disorders20(1), 115–128.

Stone, W. L., & Yoder, P. J. (2001). Predicting spoken language level in children with autism spectrum disorders. Autism : the international journal of research and practice5(4), 341–361.

Thurm, A., Lord, C., Lee, L. C., & Newschaffer, C. (2007). Predictors of language acquisition in preschool children with autism spectrum disorders. Journal of autism and developmental disorders37(9), 1721–1734.

3 thoughts on “Do motor difficulties really explain speech/language difficulties in autism?

  1. the not-surprising finding that oral motor skills, along with other motor skills, may correlate with speech motor skills

    Is this a typo, or is there a difference between oral motor skills and speech motor skills?


  2. Not a typo: depending on who is using these terms, “oral motor” is either broader than “speech motor” or non-overlapping. Oral motor includes non-speech activities like sucking on a straw, chewing, swallowing, and whistling.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s