Why does Joint Attention look Atypical in Autism—and does it matter?

(Cross-posted at FacilitatedCommunication.org.)

This is the second in a series of occasional critiques of a series of articles co-authored by Morton Gernsbacher. Collectively, these articles attempt to present evidence for the 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.

Today’s article, Gernsbacher, Sauer, Geye, Schweigert, and Goldsmith (2008), is titled Why does Joint Attention look Atypical in Autism. Joint attention, discussed here, occurs when two people jointly attend to the same object or stimulus.

In this article, Gernsbacher et al. claim that even though autistic children don’t turn their heads to follow gazes as often as typical children do, they do engage in joint attention—just more covertly than others do—i.e., by subtly shifting just their eyes. Indeed, Gernsbacher et al. claim, autistic individuals do this more reliably than their non-autistic counterparts. As for why they don’t turn their heads to follow social stimuli more overtly, or initiate joint attention with others by pointing to things, the authors suggest this is because of… motor apraxia.

In terms of the first claim, the studies that Gernsbacher et al. cite on autistic responses to eye cues show, specifically, that autistic individuals

(1) Follow other people’s eye gaze when explicitly told to


(2) Do so reflexively (as opposed to deliberately) in controlled laboratory conditions (where they aren’t, say, distracted by an absorbing real-world activity)


(3) Do not necessarily process eyes as social stimuli

One of the more interesting findings, indeed, is that autistic individuals, unlike their non-autistic counterparts, respond to pointing eyes the same way they respond to pointing arrows.

Collectively, all this suggests that, to the extent that autistic individuals follow eye gaze on their own initiative, they are not doing so as part of a conscious, socially motivated process. That is, they are not doing what was involved in the Baron-Cohen et al. (1997) word learning experiment (discussed here). In this experiment, the participants had to, on their own initiative, disengage from what they were doing and follow eye gaze in reaction to another person engaging in the communicative act of labeling. If they didn’t do this, they didn’t learn the meanings of novel words.

In one of the studies that Gernsbacher et al. cite, Chawarska, Klin, & Volkmar (2003), participants were summoned by name and told “Look!” right before the speaker turned their face and eyes towards a mechanical toy. Experimenters noted thatalthough in naturalistic situations toddlers with autism do not follow the gaze of others” (which, we should note, contradicts Gernsbacher et al.’s first claim) “they are sensitive to directional cues inherent in eye movementand use these cues when explicitly told to.

Another study,Ristic et al. (2005), reports data suggesting that non-autistic individuals treat eyes as inherently socially salient, while high functioning autistics treat them as more instrumental and only worth paying attention to when they reliably serve an instrumental purpose. Ristic et al. found, specifically, that while non-autistics followed eye gaze even when it turned out to be nonfunctional/non-predictive (i.e. , when the correspondence between the direction of eyes on a screen and the position of a target that appeared on the screen was at chance), high functioning autistics did not shift their attention in response to non-predictive eye direction. Later in the paper, Gernsbacher et al. cited a paper by Bayliss, di Pelligrino, and Tipper (2005), who find the same thing with males and with those with high “Autism Quotients”.

Gernsbacher et al. make much of this finding, suggesting that it (1) accounts for reduced joint Attention in autism while (2) only reducing joint attention opportunities that are uninformative. Interestingly, however, another paper they cite, Pelphrey et al. (2005), finds something different—at least where brain activation is concerned. Pelphrey et al. find that the brain activations of autistic subjects, unlike that of non-autistics, showed no differences depending on whether or not the face’s gaze direction seemed purposeful (looking at a checkerboard) or not (looking at empty space).

It’s important to note that the kinds of eye-gaze shifting examined by Ristic et al. and other studies cited by Gernsbacher et al. (e.g., Kylliäinen & Hietanen, 2004) are about automatic shifting, not about deliberate shifting. Indeed, in many of these studies participants were told to ignore the eyes or the arrows, such that what was captured were the automatic responses to such stimuli that people have difficulty suppressing. Kylliäinen & Hietanen note similarities between autistic and non-autistic children in such automatic gaze shifts, as do Swettenham et al. (2003). But what’s more relevant to joint attention is deliberate shifting, or deliberately sustained attention once shifting has occurred. The above experiments do not measure this.

Furthermore, even where automatic shifting is concerned, Senju et al. (also cited by Gernsbacher et al.), find differences between autistics and non-autistics, specifically in their responses to pointing arrows vs. pointing eyes. While non-autistics were more responsive to eyes, “children with autism shifted their attention equally in response to eye gaze and arrow direction, failing to show preferential sensitivity to the social cue.” Vlamings et al. (2005) report similar findings, noting that “In participants with autism the overall visual orienting reflex was not different between arrows and eyes”, and that their gaze shifts may be governed, not by sensitivity to social stimuli, but by Vlamings et al. call a “Symbol Direction Detector.”

In a second line of argumentation that relates only indirectly to joint attention, Gernsbacher et al. claim that autistics have no trouble inferring the intentions of other people’s actions. Fatally, however, all their data on this concerns just one type of intentionality: instrumental/action-based intentionality, as opposed to deeper, psychological intentions like, say, greed or rivalry. In general, these studies looked at comprehension of instrumental gestures (like a palm thrust for “stop”, or a wiggling index-finger for “come here”). Or they looked at tasks like the so-called Meltzoff intentionality tasks—tasks like using a stick tool to push the button that activates a buzzer, or putting a plastic square with a hole in the center around a dowel—or similar tasks like using a rake or taking apart a toy. Where these tasks were concerned, the child’s understanding of intentionality was measured by their ability to successfully complete the task even when the person who demonstrated it failed—failed, e.g., to fit the plastic square around the dowel. In general, these studies find that young autistic children, like young non-autistic children, are able to infer these instrumental intentions, and thus complete the tasks correctly.

One study also looked at children’s ability to recognize an instrumental gesture (like reaching) or a conventional symbolic gesture (like saluting) when the hand is obscured. But again, these aren’t expressive gestures like the shoulder shrugs that indicate feelings of indifference, or clasped hands that indicate, say, enthusiasm.

Thus, as with the eye-following data, the data on inferring intentionality is more about instrumental understanding and behavior than about social understanding and behavior.

Gernsbacher et al. turn next to the gestural behaviors of individuals with autism and their connection to joint attention. Their claim is that “autistics are considerably challenged by gesture production” and that this is one reason why they don’t engage in more overt acts of joint attention like head turning and pointing. But the studies they cite do not support this.

One paper, Rogers, Bennetto, McEvoy, & Pennington (1996), finds difficulties not with gesture production in general (i.e., in motor initiation and basic motor coordination), but only with gesture imitation, as well as, interestingly, in pantomime (the subjects could imitate motor activities with actual objects, but had trouble doing the same actions via pantomime). Imitation skills have long been found deficient in autism, but difficulties with motor coordination are only one explanation; imitation is also a social activity.

Gernsbacher et al. cite several other papers as evidence that autistics have apraxia, or “atypically execute volitional action”. These articles, however, are not about difficulties with head turns and gestures (e.g., pointing), but about oral-motor control, toe-walking, gross motor, and, for fine motor, complex motor activities. Mandelbaum et al. (2006) note that low IQ autistic individuals failed to perform many of the activities in the assessment, but also note that they (the authors) were not able to differentiate between lack of ability and lack of willingness. (And, I would add, some participants may not have understood the directions).

Gernsbacher et al. also note that apraxia can include eye-motor control, a.k.a. “gaze apraxia”:  challenges with looking where another person is pointing, looking where another person asks you to look, and looking toward the source of an auditory stimulus.  But in none of the papers they cite is there any mention of gaze apraxia.

When it comes to initiating joint attention (IJA)—getting someone to look where you are looking—the key motor activity is pointing. And pointing, Gernsbacher et al. claim, is also affected by autism-related apraxia. Here, too, their case is unconvincing, but the story is more complex.

Pointing can be subdivided into protodeclarative pointing (pointing for the sake of sharing) and protoimperative pointing (pointing to request). Protodeclarative pointing has long been acknowledged a deficit in autism, appearing in the M-CHAT (Modified Checklist for Autism in Infants and Toddlers) as one of its critical screening items. But Gernsbacher et al. cite Baron-Cohen et al. (1996), who find that 90% of autistic toddlers who are unable to point protodeclaratively are also unable to point protoimperatively. Gernsbacher et al. also cite the discussion of the M-CHAT in Robins et al.(2001), which acknowledges that protoimperative pointing is also a good discriminator of autism: in their study, 71.8% of autistic children failed the protoimperative pointing item, vs. 82.1% who failed the protodeclarative pointing item.

Given all this, Gernsbacher et al. propose that

it is the core act of pointing and its underlying motor demands rather than any deficit in intentionality or desire to share experience that underlies autistic children’s lower frequency of pointing to initiate joint attention.

Gernsbacher et al. then cite the introductory page of an entire book, Shadmehr & Wise (2005), which discusses the physical calculations involved in pointing and the mechanical problems that a person’s nervous system must solve. But however much goes on in the nervous system when people point to things, pointing is one of the earliest motor behaviors performed by young children. And, while it may be diminished in frequency and/or delayed in autism, Ricks & Wing (1976) have observed that pointing is one of the earliest gestures learned by autistic children.  Its diminished frequency, even in the case of protoimperative pointing, is consistent with the standard social deficit model of autism: both kinds of points presuppose some awareness of and attention to others as volitional actors. Many young autistic kids who rarely if ever point are able to perform more complex motor activities like grabbing a caregiver’s hand and leading that person to a desired object out of reach—something that treats the person less as a volitional actor and more as a tool.

The alternative (and standard) account—that the issue with pointing in autism is social rather than motoric—is consistent with a paper that Gernsbacher et al. themselves cite: Attwood, Frith, & Hermelin (1988). Attwood et al. looked at autistic adolescents with varying degrees of intellectual disability (compared with non-autistic controls). They found that the ability to initiate simple instrumental gestures on verbal request was most impaired in the more intellectually impaired autistic groups—which might suggest that the issue was intellectual impairment rather than autism. However, spontaneously on the playground, the story was different.

On the playground, relative to how frequently they interacted with others (which was diminished compared to the other groups), the autistic group used nonverbal instrumental gestures as a means of communication to the same extent as the other groups—though these gestures were disproportionately gestures used to reduce or terminate social contact (“Go away” or “Be quiet”). When it came to expressive gestures, however, Attwood et al. note that “no autistic adolescent ever used expressive gestures”. Their conclusion: “autistic children have specific problems with expressive but not instrumental gestures.” Again, as with pointing, the deficit appears to be social rather than motoric.

Gernsbacher et al. cite as a final reason for atypical Joint Attention, specifically atypical responses to bids for joint attention (RJA), what they call an “atypical resistance to distraction.” This noncontroversial observation, indeed, is reflected in the diagnostic criteria for autism. Gernsbacher et al. propose that resistance to distraction leads to good focused attention and less attention to non-predictive or uninformative social cues. On closer inspection, however, this mostly uncontroversial conclusion, undermines what seems to be the authors’ implicit message: that joint attention deficits aren’t a problem in autism. The all-encompassing interests and resistance to social cues that characterize autism diminish opportunities for social learning and language acquisition. When it comes to spontaneous learning opportunities, we should note, it’s hard to predict ahead of time which social cues will turn out to be uninformative.

Finally, Gernsbacher et al. link autism to an unusually strong ability in parallel perception, citing a study that suggests that autistics are, in Gernbacher et al.’s words “more facile than nonautistics in spreading their covert attention across a wider visual angle” (Rutherford, et al., 2007). Presumably their point here is to suggest that, at the same time that autistic individuals are more focused and less prone to distraction, they are also less tuned out of social cues than they appear to be. Rutherford et al., however, looked at visual cues like letters and shapes; not social cues.

Gernsbacher et al. sum things up as follows:

[A]utistic challenges in volitional action execution, in addition to their ability to attend covertly, interpret gestures, search the environment fluidly, and resist interruption and distraction, contribute to their atypical responses to conventional bids for their joint attention.

Gernsbacher et al. conclude (without evidence) that autistic individuals initiate joint attention but (because of their purported pointing difficulties) do so in atypical and unconventional ways. Tellingly, while the authors discuss how atypical joint attention works in the case of blind individuals, they do not propose anything specific about how atypical joint attention might work with autistic individuals.

Using the same references as Gernsbacher et al., my conclusions are different. What the data point to is that

1. Automatic gaze following is intact in autism, but eyes are processed more like symbols (arrows) than like social stimuli.

2. Where gesture interpretation is concerned, interpreting instrumental gestures and conventional symbolic gestures is intact in autism

3. Where expressive gestures and deeper psychological intents are concerned, autistic individuals have difficulty

4. Where performance is concerned, there are frequent gross motor and complex fine motor challenges in autism

5. Where gestures like pointing and head turning are concerned, the motor skills are intact. It is, rather, the social motivation, especially for expressive gestures and for joint attention, that may be lacking.

The fundamental issues underlying joint attention in autism, in other words, uniformly involve social deficits, not deficits with instrumental phenomena or motor control.

But regardless of the underlying reasons, diminished eye-contact and diminished joint attention have serious consequences for linguistic and socio-cognitive development. Where language is concerned, treating eyes as socially salient is crucial–especially for basic word learning. In the Baron-Cohen et al. (1997) word learning experiment, for example, participants had to disengage from what they were doing and look up at the speaker in order to learn novel words. Where learning more generally in concerned, including social learning, it’s useful, and often essential, to follow the eyes of speakers/teachers over to what they’re looking at. To follow people’s eyes one generally has to look at them first.

Therefore, even if joint attention is atypical in autism for reasons that have nothing to do with socio-cognition, its atypicality nonetheless sends autistic individuals down a path of diminished socio-cognitive development.


Attwood, A., Frith, U., & Hermelin, B. (1988). The understanding and use of interpersonal gestures by autistic and Down’s syndrome children. Journal of autism and developmental disorders18(2), 241–257. https://doi.org/10.1007/BF02211950

Baron-Cohen, S., Cox, A., Baird, G., Swettenham, J., Nightingale, N., Morgan, K., Drew, A., & Charman, T. (1996). Psychological markers in the detection of autism in infancy in a large population. The British journal of psychiatry : the journal of mental science168(2), 158–163. https://doi.org/10.1192/bjp.168.2.158

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.

Bayliss, A. P., di Pellegrino, G., & Tipper, S. P. (2005). Sex differences in eye gaze and symbolic cueing of attention. The Quarterly journal of experimental psychology. A, Human experimental psychology58(4), 631–650. https://doi.org/10.1080/02724980443000124

Chawarska, K., Klin, A., & Volkmar, F. (2003). Automatic attention cueing through eye movement in 2-year-old children with autism. Child development, 74(4), 1108–1122. https://doi.org/10.1111/1467-8624.00595

Gernsbacher, Sauer, Geye, Schweigert, and Goldsmith (2008), Why does joint attention look atypical in autism. Child Development Perspectives 2(1), 38-45. doi/full/10.1111/j.1750-8606.2008.00039.x

Kylliäinen, A., & Hietanen, J. K. (2004). Attention orienting by another’s gaze direction in children with autism. Journal of child psychology and psychiatry, and allied disciplines45(3), 435–444. https://doi.org/10.1111/j.1469-7610.2004.00235.x

Mandelbaum, D. E., Stevens, M., Rosenberg, E., Wiznitzer, M., Steinschneider, M., Filipek, P., & Rapin, I. (2006). Sensorimotor performance in school-age children with autism, developmental language disorder, or low IQ. Developmental medicine and child neurology48(1), 33–39. https://doi.org/10.1017/S0012162206000089

Pelphrey, K. A., Morris, J. P., & McCarthy, G. (2005). Neural basis of eye gaze processing deficits in autism. Brain : a journal of neurology128(Pt 5), 1038–1048. https://doi.org/10.1093/brain/awh404

Ricks, D. M., & Wing, L. (1975). Language, Communication, and the Use of Symbols in Normal and Autistic Children. Journal of Autism & Childhood Schizophrenia, 5, 191-221. https://doi.org/10.1007/BF01538152

Ristic, J., Mottron, L., Friesen, C. K., Iarocci, G., Burack, J. A., & Kingstone, A. (2005). Eyes are special but not for everyone: the case of autism. Brain research. Cognitive brain research, 24(3), 715–718. https://doi.org/10.1016/j.cogbrainres.2005.02.007

Robins, D. L., Fein, D., Barton, M. L., & Green, J. A. (2001). The Modified Checklist for Autism in Toddlers: an initial study investigating the early detection of autism and pervasive developmental disorders. Journal of autism and developmental disorders31(2), 131–144. https://doi.org/10.1023/a:1010738829569

Rogers, S. J., Bennetto, L., McEvoy, R., & Pennington, B. F. (1996). Imitation and pantomime in high-functioning adolescents with autism spectrum disorders. Child development67(5), 2060–2073.

Rutherford, M. D., Richards, E. D., Moldes, V., & Sekuler, A. B. (2007). Evidence of a divided-attention advantage in autism. Cognitive neuropsychology24(5), 505–515. https://doi.org/10.1080/02643290701508224

Senju, A., Tojo, Y., Dairoku, H., & Hasegawa, T. (2004). Reflexive orienting in response to eye gaze and an arrow in children with and without autism. Journal of child psychology and psychiatry, and allied disciplines45(3), 445–458. https://doi.org/10.1111/j.1469-7610.2004.00236.x

Shadmehr R. & Wise, S. P.(2005). The Computational Neurobiology of Reaching and Pointing A Foundation for Motor Learning, MIT Press.

Swettenham, J., Condie, S., Campbell, R., Milne, E., & Coleman, M. (2003). Does the perception of moving eyes trigger reflexive visual orienting in autism?. Philosophical transactions of the Royal Society of London. Series B, Biological sciences358(1430), 325–334. https://doi.org/10.1098/rstb.2002.1203

Vlamings, P. H., Stauder, J. E., van Son, I. A., & Mottron, L. (2005). Atypical visual orienting to gaze- and arrow-cues in adults with high functioning autism. Journal of autism and developmental disorders35(3), 267–277. https://doi.org/10.1007/s10803-005-3289-y

Leave a Reply

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

WordPress.com Logo

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

Facebook photo

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

Connecting to %s