Is there really no Theory of Mind deficit in autism? Part VI: problematic references
I spent my last five posts (besides the post announcing tomorrow and next Thursday’s FC conference) critiquing the arguments made by Gernsbacher and Yergeau in their 2019 article, “Empirical Failures of the Claim That Autistic People Lack a Theory of Mind.” Gernsbacher and Yergeau’s central argument is that ToM tests lack empirical validity. They argue, in particular, that (1) the original test results with autistic subjects have failed to replicate, (2) the tests themselves fail to converge on a meaningful psychological construct, and (3) the tests fail to predict autism-related traits, empathy and emotional understanding, and the ability to infer other people’s goals. In my earlier posts, I attempt to show how, to the extent that any of this holds true, it doesn’t override most of the empirical evidence for a Theory of Mind deficit that is fundamental to autism.
Morton A. Gernsbacher, Professor of Psychology, University of Wisconsin
In this final installment of my “Gernsbacher posts”, I turn to the bigger picture. That is, I’ll discuss how many of the articles that Gernsbacher and Yergeau cite to support their claims also report additional findings that undermine Gernbacher’s FC-friendly take on autism. This FC friendly take, which emerges from the eight Gernsbacher articles I’ve reviewed here and which we also see in other more explicitly FC-friendly articles, is that autism is not a socio-cognitive disorder, but something else: some sort of sensorimotor disorder and/or (at least in some cases) some subtype of language delay.
Ironically, many of the articles cited by Gernsbacher and Yergeau only add to the large body of evidence that autism is, in fact, fundamentally a social cognitive disorder.
Burnside et al. (2017), for example, found that their autistic participants, in contrast to their non-autistic counterparts, did not exhibit a preference for faces. Salter et al. (2008) note that, although their autistic participants used “mentalizing” language in the Triangle Animations as well as their non-autistic counterparts did, “the content of their descriptions was significantly less appropriate.” Colombi et al. (2009) found that children with autism responded less to bids for cooperative behavior than did children with other developmental disabilities. Peterson (2014) reports that her “[r]esults showed that children with ASD were significantly less empathic, according to their teachers, than typically developing children.”
Berger and Ingersoll (2014), in reporting that their autistic subjects attended less to social stimuli to distinguish between intentional and unintentional acts, note how this resonates with research showing a tendency in autism not to orient to social stimuli (they cite Dawson et al., 2004) and “provides support for the social motivation theory of ASD.” Li et al. (2019), in their discussion of moral judgments, found that autistic individuals were more upset when objects, as opposed to people, were the recipients of damaging acts.Even Kerr and Durkin (2004), who concluded from their problematic thought-bubble study (see my last post), that young autistic children are able to pass explicit, nonverbal ToM tests, add:
This is not to deny, of course, that children with autism do manifest profound social impairment, including characteristics associated with poor attention to the presence and interests of others (such as lack of eye gaze and pointing, joint attention problems) and impaired ability on many tests of theory of mind...
In one sense, Gernsbacher and Yergeau are correct. Children with autism do not uniformly and absolutely “lack a theory of mind.” Those autistic individuals who have participated in experiments—which generally exclude the most severely autistic individuals—have shown varying abilities to infer intent, at least when it comes to the kinds of instrumental intent involved in actions like reaching for objects or opening them up. Most are able to map basic emotions to facial expressions that don’t depend on the region around the eyes for interpretation: emotions like happy, sad, angry, scared. And many higher functioning autistic people are able to pass explicit Theory of Mind tests—though they:
first require significantly higher levels of linguistic mastery than non-autistic individuals require
still fail to automatically shift their gaze in the right direction in implicit versions of some of these ToM tests
still continue to experience ToM difficulties in more naturalistic, real-world settings.
In addition, some of the papers that Gernsbacher and Yergeau cite highlight problems with some of the ToM measurements—in particular the picture-sequencing test, which has failed replication; the Eyes test, in which vocabulary is the biggest factor; and the second-order ToM tests, which do not distinguish individuals with autism from those with ADHD. Furthermore, there appears to be some degree of disjunction between performance on socio-perceptual tasks like emotion recognition and socio-cognitive tasks like perspective taking.
But none of this contradicts the long-standing findings that autism is a socio-perceptual and socio-cognitive disorder involving diminished attention to social stimuli, diminished joint attention, diminished social motivation, and diminished automatic, intuitive perspective-taking.
True, to a large extent these differences are matters of degree, with significant variation across the spectrum and, often, significant improvement in the course of development. And there are additional complications. First, autism is not a smooth spectrum shading seamlessly from mild into severe. Within its multiple dimensions, the spectrum gets lumpy in places, showing evidence of distinct groups. For example, Lombardo et al. (2015) discovered of two discrete subgroups with respect to performance on the Eyes test. Wittke et al (2017) discovered two discrete subgroups among those who can put words together but struggle with syntax. Perhaps most importantly, individuals with high-functioning autism (HFA) differ significantly from those with severe autism. Indeed, a presentation at INSAR by Evdokia Anagnostou in Spring, 2022 suggests that the brains of HFA individuals have more in common with those of individuals with ADHD than with those of individuals with severe autism. Studies of HFA individuals—including those that Gernsbacher and Yergeau cite as evidence for intact ToM skills in autism—may well not generalize across the spectrum.
In addition, non-autism-related personality factors as well as gender surely play a role, perhaps interacting with autism traits like frequency of eye contact and joint attention, on one hand, and degree of social motivation, on the other. Furthermore, variations in executive functioning and linguistic mastery both affect—and are affected by—autism symptomology. Regarding language in joint attention and language ability (see Tomasello and Todd, 1983) “high-functioning” means being “at the high end of the joint attention skill spectrum.”
But despite all this variation and despite all these confounds, there emerges from the bulk of the research, including the bulk of the research cited here by Gernsbacher and Yergeau, a consistent theme.
Autism is a spectrum of diminished social motivation and of difficulty with social cognition.
And this, for those who wish to promote and/or to believe in FC, is an inconvenient truth.
REFERENCES:
Anagnostou, E. (2022). Exploring Heterogeneity in ASD: From Bench to Clinic to Stakeholder Priorities. Keynote address, International Society for Autism Research.
Berger, N. I., & Ingersoll, B. (2014). A further investigation of goal-directed intention understanding in young children with autism spectrum disorders. Journal of Autism and Developmental Disorders, 44, 3204–3214. https://doi.org/10.1007/s10803-014-2181-z
Burnside, K., Wright, K., & Poulin-Dubois, D. (2017). Social motivation and implicit theory of mind in children with autism spectrum disorder. Autism Research, 10, 1834–1844. https://doi.org/10.1002/aur.1836
Gernsbacher, M. A., & Yergeau, M. (2019). Empirical Failures of the Claim That Autistic People Lack a Theory of Mind. Archives of scientific psychology, 7(1), 102–118. https://doi.org/10.1037/arc0000067
Kerr, S., & Durkin, K. (2004). Understanding of thought bubbles as mental representations in children with autism: Implications for theory of mind. Journal of Autism and Developmental Disorders, 34, 637–648. https://doi.org/10.1007/s10803-004-5285-z
Li, T., Decety, J., Hu, X., Li, J., Lin, J., & Yi, L. (2019). Third-party sociomoral evaluations in children with autism spectrum disorder. Child Development. Advance online publication. https://doi.org/10.1111/cdev.13206
Tomasello, M., & Todd, J. (1983). Joint attention and lexical acquisition style. First Language, 4(12, Pt 3), 197–211. https://doi.org/10.1177/014272378300401202
Wittke, K., Mastergeorge, A. M., Ozonoff, S., Rogers, S. J., & Naigles, L. R. (2017). Grammatical language impairment in autism spectrum disorder: Exploring language phenotypes beyond standardized testing. Frontiers in psychology, 8, 532. https://doi.org/10.3389/fpsyg.2017.00532
