Long, E. L., Catmur, C., & Bird, G. (2025). The theory of mind hypothesis of autism: A critical evaluation of the status quo. Psychological Review. doi.org/10.1037/rev0000532
For several years the Theory of Mind (ToM) hypothesis has been suggested as an underlying mechanism for understanding autism. This paper suggests that no existing Theory of Mind ideas fully explain ToM in autism. So, the team introduce the Mind-space framework.
Problems with the current Theory of Mind hypothesis:
- Describing and explaining ToM differences in autism is quite unclear.
- There are oversimplified claims, such as autistic people “lacking ToM,” which contradicts existing evidence… and the stories of autistic people
- There seem to be limited success of interventions based on the ToM hypothesis.
ToM testing issues:
- Many tests fail to measure true ToM abilities as they do not really measure what they set out to measure. They may mix up spotting emotions identification with guessing mental states, so group differences can be explained by alexithymia rather than autism. Or conflate perspective taking tasks with spatial reasoning skills.
- fMRI-based brain scans cannot be reliably interpreted – some have come up with no difference in brain patterns between groups, and those with a difference sometimes don’t find any behavioural differences.
- Tests of inference accuracy (coming up with a good ToM) often lack a clear “correct” answer.
- Tests of propensity (trying to measure if ToM is used in other contexts) may be measuring language use, rather than a ToM. And to add to that, if your test is using animated shapes, it only really tells you about the minds of animated shapes.
The Mind-Space framework
The mind-space theory moves away from trying to represent single desires or beliefs about others in traditional ToMs. Instead, it thinks about how we are all entire complex minds. To work out what is going on in another mind we need to know how other people’s minds might vary from each other and vary from ours. They might be more reckless, more trusting, have a better or worse memory, be a problem-solver or pattern recogniser. They may be creative, persistent or agreeable. Or they may not. We need to know this before we can guess what they might do in any given circumstance.
This Mind-space framework – thinking about all the different kinds of people – accounts for the double empathy problem, the majority might often miss how autistic minds are thinking and vice versa. This is difference, not deficit. With this model it is easy to see why some neurodivergent folk might really click with other neurodivergent folk, and – sadly – the problems caused when the majority group of thinkers misunderstand any minority group.
Testing the Mind-space framework
We need Theory of Mind tests that take all these ways of being human into account. As a start, the authors consider the Interview Task, which has a video of a practice job interview… which wasn’t done by actors! And the scores are based on reports of those who took part in the interviews, not some strange group consensus. We need more tasks like this that use a natural situation to talk about the complexity of the Mind-spaces we all create.