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

(Cross-posted at

Following up on my previous post on the FC-friendly claim in Gernsbacher, Morson & Grace that there is nothing inherently atypical about language development in autism I now turn to Gernsbacher, Morson & Grace’s 2016 paper.

This paper looks at three linguistic factors that have been claimed to be characteristic of autistic speech: pronoun reversals, echolalia, and a smaller-than-to-be-expected lag between comprehension and production. Gernsbacher et al. argue that none of these is unique to autism, and therefore that “none can or should serve as diagnostic of autism.” They furthermore suggest that, while they also occur in typical development, these factors are only seen as pathological in autism because of the pathological way in which we view autistic people.  

As an example of this, Gernsbacher et al. opens with a behavior that is often linked to autism: “using another person’s hand to indicate wants and needs”. Indeed, it has often been observed that individuals with autism sometimes use other people’s hands as tools, or lead people across the room to something they want. Gernsbacher et al. argue that this sort of behavior, when carried out by non-autistic people “is considered a highly adaptive form of communication that is as sophisticated as (if not more sophisticated than) pointing, looking, and other mechanisms for requesting joint attention.” But what they overlook is the different ways in which “using another person’s hand” can occur. Is the child looking up at the adult while leading him or her to an object of interest or desire, or is the child simply dragging the adult, or the adult’s hand, to the desired object? While it may be hard to tell the degree to which a child is communicating with the adult as an intentional agent, or merely treating the adult as a tool, these are, nonetheless, two distinct behaviors, and the latter is more worrying (and more suggestive of autism) than the former.

Turning, now, to the three linguistic phenomena under consideration, Gernsbacher et al. first address pronoun reversal. They begin, quite reasonably, by dismissing the psychoanalytic account of why autistic children reverse “you” and “I” (the notion that autistic children lack awareness of their own identity). But they do not consider the alternative account: namely, that pronoun reversal suggests a difficulty with perspective taking: the same sort of difficulty that arises with the false-belief tests.

Instead, Gernsbacher et al. cite a variety of papers that show that pronoun reversal is not unique to autism and occurs in a variety of non-autistic populations, including precocious neurotypicals and in children with various other disabilities: Evans & Demuth (2012), Schiff-Myers (1983), Clark (1978), Petito (1987), Perez Pereira (1999), and Yoder et al. (1998). These findings are non-controversial.

What some researchers have cited as specific to autism, however, is a relatively high frequency of pronoun reversals—something that Gernsbacher et al. do not directly address. Here, the studies are inconsistent. Naigles et al. (2016) comparing 18 TD (typically developing) toddlers and 15 children with ASD at similar language levels, found low rates of pronoun reversals in each group, but slightly higher rates in the ASD group. Furthermore, they found that “In the ASD group, early language and joint attention scores contributed significantly and independently to the incidence of reversal”—which suggests that pronoun reversals in autism are explained, in part, by autism-specific phenomena (i.e., Joint Attention deficits). Barokova & Tager-Flusberg (2020), on the other hand, looked at language samples from autistic children at 2, 3, 4, years, a developmental period during which they used more and more first and second person pronouns, and, yet “very rarely” reversed their pronouns. Barokova & Tager-Flusberg’s conclusion: “This lack of pronoun errors suggests that pronoun difficulty in autism might not occur for long periods of time throughout development and might not be as prevalent in autism as previously thought.”

The only mention Gernsbacher et al. make of frequency of pronoun errors appears to be in reference to pronoun errors in general, not pronoun reversals in particular. First, they state that “when autistic children are carefully matched with non-autistic children in their language production skills, autistic children do not produce more pronoun errors than non-autistic children.” Their reference for this is Norbury & Bishop (2003). But the pronoun errors surveyed by Norbury & Bishop do not include pronoun reversal errors, and, as for the types of pronoun errors they do discuss, they in fact do find autism-specific difficulties:

In  only  one  area  did  the  clinical groups appear to differ and that was with respect to referencing, in particular in  producing  ambiguous  nouns  and  pronouns.  These  types  of  errors  were more common in children with a diagnosis of autism.

Second, Gernsbacher et al. state that “when autistic children are carefully matched with non-autistic children in their language comprehension skills, autistic children do not comprehend pronouns more poorly than non-autistic children”.  (Here their reference is a conference paper,  Edelson et al. 2011). But it is hardly surprising that pronoun comprehension is tied to language comprehension. Furthermore, as we saw, comprehension has consistently turned up as an autism-related language weakness.

Similar issues come up in Gernsbacher et al.’s discussion of echolalia. They emphasize that it isn’t unique to autism, but again, this is noncontroversial. They mention that echolalia can be communicatively functional and that it fades away as language skills improve, but these, too, are noncontroversial. The interesting question, rather, is whether echolalia is more frequent in autism. Citing van Santen et al. (2013), Gernsbacher et al. claim that it is not.

What’s complicated about Gernsbacher et al.’s claim is that it includes as instances of echolalia what they call “self-echoing.” However, since most people consider this particular phenomenon merely repetition (or self-repetition) rather than echolalia, it is not generally included in claims about echolalia in autism. As far as the common understanding of echolalia goes, Gernsbacher et al. claim, citing van Santen et al., that “autistic children are no more likely than non-autistic children with language delay to echo another person.” But van Santen et al., while they find that self-repetition is not more frequent in autism, state that their results “confirm previous findings that children with ASD repeat the language of others more than other populations of children.”

In particular, van Santen et al., find that rates of echolalia not only reliably distinguish autistic children from TD (typically developing) children, but also reliably distinguish the subset of autistic children without language delays from TD children. Indeed, their two autistic subgroups, those with and without language delays, “had nearly identical rates of echolalia despite significant differences in overall language ability”—suggesting that echolia, in autism, is independent of language delay. (They also note, interestingly, that echolalia appears to be independent of number/severity of autistic traits—e.g., ADOS scores). While van Santen et al. do find that the participants with Specific Language Impairment (SLI) had similar levels of echolia to the ASD groups, they “cautiously conclude that it [the SLI group] was in-between the ASD groups and the TD group.”

Finally, Gernbacher et al. turn to the issue of whether there is a smaller-than-expected lag between comprehension and production in language in autism. Here they cite Kwok et al., a meta-analysis that fails to find such a lag in comprehension of vocabulary and syntax. However, as Kwok et al. themselves note, these findings do not rule out an expressive-over-receptive advantage at higher levels of language. Higher levels include not just complex sentences, but also the level of discourse: dialogs, paragraphs, narratives. Indeed, indicators of comprehension difficulties in autism, as we discussed earlier, appear in a number of the papers Gernsbacher et al. cite.

And this, for all the arguments that Gernsbacher et al. attempt to make against autism-specific language difficulties, takes us back to what is perhaps the most significant area of difficulty—beyond vocabulary, beyond basic syntax, beyond idioms and metaphors—namely, overall comprehension.


Barokova, M., & Tager-Flusberg, H. (2020). Person-reference in autism spectrum disorder: Developmental trends and the role of linguistic input. Autism research : official journal of the International Society for Autism Research13(6), 959–969.

Clark E. V. (1978). From gesture to word: on the natural acquisition. In Human Growth and Development: Wolfson College Lectures, ed. J. S. Bruner, A. Garton, pp. 85–120. Oxford, UK: Oxford Univ. Press

Edelson L.R., Meyer A.T., Tager-Flusberg H. (2011). Cues to pronominal reference resolution in children with and without autism spectrum disorders. Presented at Int. Meet. Autism Res., San Diego.

Evans, K.E., & Demuth K. (2012). Individual differences in pronoun reversal: evidence from two longitudinal case studies. Journal of Child Language, 39, 162–91

Gernsbacher, M. A., Morson, E. M., & Grace, E. J. (2016). Language and Speech in Autism. Annual review of linguistics2, 413–425.

Henderson, L. M., Clarke, P. J., & Snowling, M. J. (2011). Accessing and selecting word meaning in autism spectrum disorder. Journal of Child Psychology and Psychiatry, 52, 964-973.

Naigles, L. R., Cheng, M., Rattansone, N. X., 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 Disorders and Typical Development. Research in autism spectrum disorders27, 11–20.

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.

Perez Pereira M. (1999). Deixis, personal reference, and the use of pronouns by blind children. ´ Journal of Child Language, 26, 655–80.

Petito L. A. (1987). On the autonomy of language and gesture: evidence from the acquisition of personal pronouns in American Sign Language. Cognition 27, 1–52.

Schiff-Myers N. B. (1983). From pronoun reversals to correct pronoun usage: a case study of a normally developing child. Journal of Speech and Hearing Disorders, 48, 394–402

van Santen, J. P., Sproat, R. W., & Hill, A. P. (2013). Quantifying repetitive speech in autism spectrum disorders and language impairment. Autism research : official journal of the International Society for Autism Research, 6(5), 372–383.

Yoder, P. J. , Warren, S. F., McCathren,  R. B. (1998). Determining spoken language prognosis in children with developmental disabilities. American Journal of Speech Language Pathology, 7, 77–87.

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