Is there really nothing inherently atypical about language development in autism?

(Cross-posted at

In 2015 and 2016, Gernsbacher, Morson and Grace published a pair of articles on language development in autism. In these papers, the authors try to make a case that there’s nothing inherently atypical about language development in autism as compared to (1) typical language development and/or (2) development in non-autistic children with general language delays. The notion that there is no autism-specific language deficit, if true, would lend support to the notion that there’s nothing suspect about the linguistic skills outputted by individuals being subjected to FC/RPM/S2C.

Today I continue my critique of Gernsbacher’s FC-friendly papers with the 2015 paper, Language Development in Autism.

Morton A. Gernsbacher, Professor of Psychology, University of Wisconsin

Gernsbacher, Morson & Grace begin by acknowledging that some studies do show atypical language development in autism. However, they continue, other studies “have not shown that autistic language development differed from typical language development.”

The problem is that none of the studies they cite support this claim.

Gernsbacher et al. begin by citing Goodwin et al. (2012). This paper, they claim, shows that autistic toddlers do not “differ from typically developing toddlers in the number of words that they produced.” The focus of Goodwin et al.’s paper, however, is WH-questions. Their results found their autistic participants to be delayed in their comprehension of WH-questions “compared to age-matched TD [typically developing] children, but not when matched on overall language levels.” The only time that “number of words produced” comes up is in the preliminary measurements conducted. These included two measures relating to number of words produced. On number of words produced out of a set of 396, the TD group produced an average of 118.8, while the ASD group produced an average of 106.6, which looks relatively close. But on a different test involving a 100-word set, the TD group averaged 81.94 while the ASD group averaged only 48.21.

Skipping over this inconvenient result, Gernsbacher et al. then cite Åsberg (2010), Henderson et al. (2011) & Paul et al. (2005). These studies, Gernsbacher et al. claim, show that “autistic teenagers did not differ from typically developing teenagers in the number of words that they understood.” But do they?

First, let’s look at Åsberg et al. Åsberg et al. compared an autistic group ranging from 10 years 9 months to 15 years 8 months to a control group, matched accordingly to Performance IQ, ranging from 7 years 7 months to 14 years 8 months.  Their results, indeed, did not show significant group differences in oral receptive vocabulary or reception of grammar. But they did find “significantly lower abilities in narrative discourse comprehension for the ASD group.” While this supports Gernsbacher et al.’s claims about words understood, it does not support their main claim—i.e., that language development/skills in general in autistic individuals are similar to that of other groups.

As for Henderson et al., these researchers specifically chose their subjects so that their vocabularies would match those of the TD control group: “The groups were pairwise matched on chronological age, receptive vocabulary”. This, of course, guarantees that the groups would “not differ” in “the number of words they understood”—and tells us nothing about whether receptive vocabulary skills in autistics vs. non-autistics are similar outside the lab. Furthermore, their findings, like Åsberg’s, indicate higher-level linguistic challenges:

Children with ASD showed intact access to semantic information early in the time course of processing; however, they showed impairments in the selection of semantic representations later in processing.

Finally, Paul et al. (2005), as its title (Perception and Production of Prosody by Speakers with Autism Spectrum Disorders) suggests, has nothing to do with vocabulary. Its focus, rather, is on speech prosody, specifically:

stress (emphasis on syllables and words), intonation (pitch changes over the course of phrases and sentences), and phrasing (the pattern of rate and pausing within utterances).

Nor does Paul et al. contain a single statement about the numbers of words understood by the autistic vs. the non-autistic participants. Why Gernbacher et al. cited it as such a source on this question is left a mystery.

Instead, moving on, Gernsbacher et al. cite Troyb (2011). Troyb (2011), Gernsbacher et al. claim, shows that:

autistic children, teens, and adults did not differ from typically developing participants in the quality or quantity of their written language production, be it number of words, length of words, length of sentences, or complexity of sentences.

Troyb (2011), however, is listed in Gernsbacher et al.’s references as an unpublished master’s thesis. Fortunately, however, there is also a published paper with the same title with Troyb as first author: Troyb et al. (2014). But Troyb et al. (2014) specifically focuses on high functioning autistics, including a group that no longer met the criteria for autism.

According to Troyb et al. those who still met the criteria for autism scored significantly lower than both the TD (typically developing) group and the group that lost its diagnosis on verbal IQ and on subtests of reading comprehension. Not surprisingly, the group that lost its diagnosis was very similar on language and other measures to the TD group. Furthermore, where linguistic conventions like number of words, length of words, length of sentences, and complexity of sentences are concerned, it has long been observed that individuals with high functioning autism have close-to-normal abilities. But neither this—nor Troyb et al.’s study conclusions—generalize to the linguistic skills of individuals with autism across the spectrum.

Gernsbacher et al. do acknowledge that delayed language development is common, if not universal, in autism. At the same time, however, they claim that “there is good evidence that language develops independently from autistic traits.” Here they cite a study involving 3000 pairs of twins (Taylor, Charman, Robinson, Hayiou-Thomas, Happé, Dale, & Ronald, 2014). Taylor et al. do report that the correlation between language skills and autistic traits is “weak.” At the same time, however, their study does not contradict findings that language development in autism is, overall, atypical, or that autistic traits like joint attention deficits affect language development.

Turning to the developmental trajectory of language in autism, Gernsbacher et al. cite Howlin (2003). This study, Gernsbacher et al. state, finds that a group of autistic individuals whom they followed from childhood through adulthood, regardless of whether their language was initially delayed, “did not differ in either their expressive or receptive vocabulary.” But, like Troyb, Howlin was not looking at autistic individuals across the spectrum; only at individuals with High Functioning Autism and Asperger’s. Furthermore, Howlin found that “in both groups language abilities were well below chronological age level.”

Gernsbacher et al. go on to claim that there are linguistic similarities between autistic toddlers with “late talkers.”  Citing Ellis Weismer et al. (2011), Gernsbacher et al. report that

the two groups are just as likely to produce words from a range of grammatical and semantic categories, including emotion terms, and the autistic toddlers did not differ from the non-autistic late talkers in the complexity of their grammatical utterances.

But Weismer et al. included in their study only those autistic toddlers who would produced at least 15 different words. As Weismer et al. note:

[i]t is unclear the extent to which the results would generalize to autistic children with more severely delayed early language development.

Second, the number of emotion terms produced was quite low in both the autistic and the late talking groups. Third, their study only looked at a snapshot in time and said nothing about how language unfolds in these two groups. As Weismer et al. caution:

These findings may not generalize to older age ranges when children on the autism spectrum have been reported to develop more highly specialized vocabularies related to their specific areas of interest and children without autism routinely begin to include a larger variety of psychological state terms, including cognitive terms, in their conversations… The lack of group differences at an early point in development, does not rule out the possibility of later differences. (p. 1074)

Finally, Gernsbacher et al. claim that

When autistic preschool-age children are compared with non-autistic preschool-age children who have other developmental disabilities, the autistic children do not differ from the non-autistic children in their correct production of noun phrases, sentence structures, plurals, singulars, or past-tense inflections.

Here they cite Park et al. (2012). But they leave out one of Park et al.’s key findings. There is, Park et al. find, one specific subskill in which the autism group showed greater weaknesses than the developmentally disabled non-autistic group: namely, the use of verb phrases. Verb phrases, Park et al. point out, contain a high proportion of morpho-syntactic phenomenon. Accordingly, Park et al. report as their main finding that:

children with autism have unevenly developed morphological and syntactic sub-skills; they have skills which are a combination of intact, delayed, and atypical.

In addition, Park et al. note that, in terms of the Index of Productive Syntax (IPSyn) “[t]he pattern of results was of lower performance of the group with autism” compared to the developmentally delayed group.

Skipping over these details, Gernsbacher et al., citing Norbury and Bishop (2003) and Young et al. (2005), go on to claim that there is no specific deficit in figurative language in autism. This is a claim we addressed in the last post—the one on Gernsbacher and Pripas-Kapit (2012). As we noted there, while Gernsbacher and Pripas-Kapit’s claims hold up for idioms and metaphors, they do not hold up for inferences.

Gernsbacher et al. conclude their paper by claiming that “autistic children do not differ from language-matched non-autistic children in their sequential achievement of important language comprehension processes.” As we’ve seen, there are numerous problems with this claim. The biggest one of all concerns comprehension. In a number of the studies Gernsbacher et al. cite—Åsberg (2010), Henderson et al. (2011), Troyb et al. (2014), and Norbury & Bishop (2003)—comprehension comes up repeatedly as being a specific area of weakness in autism. This finding, unsurprising as it is, dates back to some of the earliest clinical observations and empirical research concerning language development in autism. And nothing in Gernsbacher et al. overturns these conclusions.


Åsberg, J., & Dahlgren Sandberg, A. (2012). Dyslexic, delayed, precocious, or just normal? Word reading skills of children with autism spectrum disorders. Journal of Research in Reading, 35, 20-31.

Ellis Weismer, S., Gernsbacher, M., A., Stronach, S., Karasinski, K., Eernisse, E. R., Venker, C. E., & Sindberg, H. (2011). Lexical and grammatical skills in toddlers on the autism spectrum compared to late talking toddlers. Journal of Autism and Developmental Disorders, 41, 1065-1075.

Gernsbacher, M.A., Morton, E. M., & Grace, E. J. (2015). Language Development in Autism. In G. Hickok & S. Small (Eds.), Neurobiology of Language, pp. 879-886.

Gernsbacher, M. A., & Pripas-Kapit, S. (2012). Who’s missing the point? A commentary on claims that autistic persons have a specific deficit in figurative language comprehension. Metaphor & Symbol, 27, 93– 105.

Goodwin, A., Fein, D., & Naigles, L. R. (2012). Comprehension of wh-questions precedes their production in typical development and autism spectrum disorders. Autism Research, 5, 109-123.

Howlin, P. (2003). Outcome in high-functioning adults with autism with and without early language delays: Implications for the differentiation between autism and Asperger Syndrome. Journal of Autism and Developmental Disorders, 33, 3-13.

Norbury, C. F., & Bishop, D. V. M. (2002). Inferential processing and story recall in children with communication problems: A comparison of specific language impairment, pragmatic language impairment and high-functioning autism. International Journal of Language Communication Disorders, 37, 227–251.

Park, C. J., Yelland, G. W., Taffe, J. R., & Gray, K. M. (2012). Morphological and syntactic skills in language samples of preschool aged children with autism: Atypical development? International Journal of Speech-Language Pathology, 14, 95-108.

Paul, R., Augustyn, A., Klin, A., & Volkmar, F. R. (2005). Perception and production of prosody by speakers with autism spectrum disorders. Journal of Autism and Developmental Disabilities, 35, 205-220.

Taylor, M. J., Charman, T., Robinson, E. B., Hayiou-Thomas, M. E., Happé, F., Dale, P. S., & Ronald, A. (2014). Language and traits of autism spectrum conditions: Evidence of limited phenotypic and etiological overlap. American Journal of Medical Genetics, Part B: Neuropsychiatric Genetics, 165, 587-595

Troyb E., Orinstein, A., Tyson, K., Helt, M., Eigsti, I. M., Stevens, M., & Fein, D. (2014). Academic abilities in children and adolescents with a history of autism spectrum disorders who have achieved optimal outcomes. Autism : the international journal of research and practice18(3), 233–243.

Young, E. C., Diehl, J. J., Morris, D., Hyman, S. L., & Bennetto, L. (2005). The use of two language tests to identify pragmatic language problems in children with autism spectrum disorders. Language, Speech, and Hearing Services in Schools, 36, 62-72.

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