Is there really no Theory of Mind deficit in autism? Part IV: Do Theory of Mind tests fail to predict empathy and emotional understanding?

In recent posts, cross-posted at, I critiqued the arguments in Gernsbacher and Yergeau’s 2019 article (“Empirical Failures of the Claim That Autistic People Lack a Theory of Mind”) that ToM tests lack empirical validity—in particular, that the original test results with autistic subjects have failed to be replicated, and that the tests themselves fail to converge on a meaningful psychological construct and fail to predict autism-related traits.

As the article continues, Gernsbacher and Yergeau turn to the extent to which ToM tests predict empathy and emotional understanding. Here, too, their arguments are problematic. The ToM tests the authors consider here are:

  • The Eyes Test, which measures the ability to recognize emotions from facial information that is restricted to a rectangular region around the eyes
  • The false-belief tests, which measures the ability to make inferences about the beliefs held by (or about) individuals who are missing key pieces of information

First, as before, some of Gernsbacher and Yergeau citations do not include participants with autism, and so their relevance here is unclear. In addition, once again, some of these results are not surprising, given what specific ToM tests measure. Gökçen et al. (2016) found “non-significant correlations between measures tapping cognitive and affective empathy.” Wilson (2014) reported similar results. Peterson (2014) found that false-belief tests were not correlated with teacher reports of affective empathy behaviors like “offering comfort to a child who is in distress”. And Melchers et al. (2015) found that the Eyes test “barely” correlated with participants’ self-reported assessments of empathy.

But there’s no reason to think that the cognitive perspective-taking skills measured by the false-belief tests would relate directly to emotional empathy and emotional understanding, or that facial expression reading skills are the only determining factor in empathy.

In addition, as I discussed in my previous post, there is the issue of the heterogeneity of the autism spectrum. Lombardo et al., (2015), recall, identify two discrete groups, only one of which had difficulty with Eyes test. Yet both groups, they report, were similar on measures of empathy. Relatedly, there are Baron-Cohen et al. (2015)’s gender-specific results, which also pertain to his Empathy Quotient test. Baron-Cohen et al. found a tight correlation between high performance on Eyes test and high Empathy Quotient scores, but only in autistic females.

Finally, one of the papers cited by Gernsbacher and Yergeau, namely Olderbak et al. (2015), while underscoring the language confound in the Eyes test (discussed in my previous post), still notes that this test is also correlated, if not as strongly as language levels are, with cognitive empathy and emotion perception.

Gernsbacher and Yergeau’s next claim is that ToM tests fail to correlate with everyday social skills. Most of the studies they cite here pertain to non-autistic groups—in particular, individuals with schizophrenia and epilepsy. Two studies—Frith et al. (1994) and Sparrevohn & Howie (1995)—while they do address autism, do not address social behavior. Several smaller studies, however, do address autism and social behavior.

Bennett et al. (2013) found that performance on the Eyes test was correlated with language and adaptive functioning in the communication domain, but not in the social domain. Fombonne et al. (1994) found that, after controlling for language, there was no correlation between performance on false-belief tasks and social behavior. Similarly, Joseph & Tager-Flusberg (2004) found that false-belief tests do not correlate with reciprocal social interaction.  Finally, Prior et al. (1990) found that performance on an emotion recognition task and on false-belief tests were only weakly related to parent- reported social behavior.

In line with what we discussed earlier, however, the fact that the cognitive perspective-taking skills measured by the false-belief tests do not relate directly to social behavior, or that facial expression reading skills are not a significant factor in social functioning, may simply mean that these tests are picking up different aspects of the social weaknesses underlying autism, which may vary from individual to individual.

Moreover, one of the tests, Hughes et al. (1997), reaches the opposite conclusion about whether ToM tests correlate with everyday social skills:

This study offers a (small) cross-cultural replication of recent work suggesting that differences in the mentalising skills of children with autism are reflected in the everyday social behaviour of this group.

Next, Gernsbacher and Yergeau turn to the correlation—or, they claim, lack thereof—between ToM tests and social attention, cooperation, anticipation, persuasion, deception, and avoidance. Of the studies with autistic subjects, one focuses instead on autobiographical memory, whose relationship to these other phenomena is unclear. Four other studies are relevant. Angus et al. (2015) found no correlation in autism between passing second-order false-belief tests and being able to prepare oneself for a social interaction by anticipating its likely form and content. Burnside et al. (2017) found no correlation between performance on an implicit false-belief test and measures of social motivation. And Chin and Bernard-Opitz (2000) found that teaching conversation skills does not improve performance on first and second-order false-belief tests. The common theme here, however, may simply be that cognitive perspective-taking is largely independent from other social skills.

The fourth study that Gernsbacher and Yergeau cite, however, does not support their claim. Peterson, Slaughter, & Wellman (2018) find that false-belief performance predicts skills at persuasion.

Finally, Gernsbacher and Yergeau turn to the correlation—or lack thereof—between ToM tests and peer relations and pro-social behavior. Here, two studies involve autistic children.  

Begeer et al. (2010) used a novel cognitive perspective-taking test—an interactive task in which participants had to pay attention to which objects were visible to their partners. Begeer et al. compared performance on this task to performance on another task in which participants heard a narrative about an interaction and then were asked to retell it, explaining behavior like trembling hands or a sudden departure from a room. The authors found that despite performing as well as typically developing controls on the cognitive perspective-taking task, they tended not to use any words pertaining to mental states or emotions in their story retellings. Once again, we see a disconnect between cognitive perspective-taking skill and other aspects of psychosocial awareness.

Travis et al. (2001), meanwhile, found a disconnect between performance on both false-belief tests and affective (or emotional) perspective taking, on the one hand, and level of engagement in peers on the playground and pro-social behavior in a structured laboratory test, on the other. Affective perspective taking was measured by the child’s ability to match a depiction of one of the four basic emotions—happy, sad, angry, scared—along with its verbal label, to a character in a vignette who experienced one of these emotions. Again, these disconnects between measurements may simply reflect that different measurements are measuring different, largely unrelated aspects of social awareness/behavior. Pro-social behavior may be largely unrelated to performance on false-belief tests and the understanding of basic emotions because it is, perhaps, largely a function of social motivation.

To sum up:

  • It’s true that certain ToM tests—particularly the tests measuring facial expression reading or deductions about false beliefs—have failed to predict emotional empathy and emotional understanding.
  • But this may simply reflect that some social phenomena in autism (e.g., pro-social behavior like peer-engagement, conversational skills, and offering comfort to those in distress) may be more a function of social motivation than of facial-expression reading skills or of cognitive perspective-taking ability.

Autism is a heterogeneous disorder that affects individuals who, as individuals, also vary in ways that are unrelated to autism. It’s not surprising, therefore, that we find variations in social motivation and facial-expression reading skills that aren’t perfectly correlated, one with the other. After all, the same sorts of variation occur in the non-autistic population.


Angus, D. J., de Rosnay, M., Lunenburg, P., Meerum Terwogt, M., & Begeer, S. (2015). Limitations in social anticipation are independent of imaginative and Theory of Mind abilities in children with autism but not in typically developing children. Autism, 19, 604–612.

Baron-Cohen, S., Bowen, D. C., Holt, R. J., Allison, C., Auyeung, B., Lombardo, M. V., . . . Lai, M. C. (2015). The “Reading the Mind in the Eyes” Test: Complete absence of typical sex difference in ∼400 men and women with autism. PLoS ONE, 10, Article e0136521.

Begeer, S., Malle, B. F., Nieuwland, M. S., & Keysar, B. (2010). Using theory of mind to represent and take part in social interactions: Comparing individuals with high-functioning autism and typically developing controls. European Journal of Developmental Psychology, 7, 104–122.

Bennett, T. A., Szatmari, P., Bryson, S., Duku, E., Vaccarella, L., & Tuff, L. (2013). Theory of mind, language and adaptive functioning in ASD: A neuroconstructivist perspective. Journal of the Canadian Academy of Child and Adolescent Psychiatry, 22, 13–19. Retrieved from

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.

Fombonne, E., Siddons, F., Achard, S., Frith, U., & Happé, F. (1994). Adaptive behaviour and theory of mind in autism. European Child & Adolescent Psychiatry, 3, 176–186.

Frith, U., & Happé, F. (1994). Autism: Beyond “theory of mind.”Cognition, 50, 115–132.

Gernsbacher, M. A., & Yergeau, M. (2019). Empirical Failures of the Claim That Autistic People Lack a Theory of Mind. Archives of scientific psychology7(1), 102–118.

Gökçen, E., Frederickson, N., & Petrides, K. V. (2016). Theory of mind and executive control deficits in typically developing adults and adolescents with high levels of autism traits. Journal of Autism and Developmental Disorders, 46, 2072–2087.

Hughes, C., Soares-Boucaud, I., Hochmann, J., & Frith, U. (1997). Social behaviour in pervasive developmental disorders: Effects of informant, group and “theory-of-mind.”European Child & Adolescent Psychiatry, 6, 191–198.

Joseph, R. M., & Tager-Flusberg, H. (2004). The relationship of theory of mind and executive functions to symptom type and severity in children with autism. Development and Psychopathology, 16, 137–155.

Lombardo, M. V., Lai, M. C., Auyeung, B., Holt, R. J., Allison, C., . . . Baron-Cohen, S. (2015). Enhancing the precision of our understanding about mentalizing in adults with autism. BioRxiv.

Melchers, M., Montag, C., Markett, S., & Reuter, M. (2015). Assessment of empathy via self-report and behavioural paradigms: Data on convergent and discriminant validity. Cognitive Neuropsychiatry, 20, 157–171.

Olderbak, S., Wilhelm, O., Olaru, G., Geiger, M., Brenneman, M. W., & Roberts, R. D. (2015). A psychometric analysis of the reading the mind in the eyes test: Toward a brief form for research and applied settings. Frontiers in Psychology, 6, Article 1503.

Peterson, C. (2014). Theory of mind understanding and empathic behavior in children with autism spectrum disorders. International Journal of Developmental Neuroscience, 39, 16–21.

Peterson, C. C., Slaughter, V., & Wellman, H. M. (2018). Nimble negotiators: How theory of mind (ToM) interconnects with persuasion skills in children with and without ToM delay. Developmental Psychology, 54, 494–509.

Prior, M., Dahlstrom, B., & Squires, T.-L. (1990). Autistic children’s knowledge of thinking and feeling states in other people. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 31, 587–601.

Sparrevohn, R., & Howie, P. M. (1995). Theory of mind in children with autistic disorder: Evidence of developmental progression and the role of verbal ability. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 36, 249–263.

Travis, L., Sigman, M., & Ruskin, E. (2001). Links between social understanding and social behavior in verbally able children with autism. Journal of Autism and Developmental Disorders, 31, 119–130.

Wilson, C. E., Happé, F., Wheelwright, S. J., Ecker, C., Lombardo, M. V., Johnston, P., . . . the MRC AIMS Consortium. (2014). The neuropsychology of male adults with high-functioning autism or Asperger syndrome. Autism Research, 7, 568–581.

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