Mind blindness versus "trial and error"

I'm often puzzled by the evidence used to define "mind blindness" or Theory of Mind. The one most often cited is the Sally Anne experiment - that one with two dolls and a ball or marble. Sally puts the marble in her basket and goes out; while she is out, Anne moves the marble into her own box - the question is when she comes back where will Sally look for her marble?  Allegedly non autistic children can tell that Sally will look in her basket, whereas autistic children expect her to look in Anne's box.

The argument is that Autistic Children lack theory of mind, or are mind blind - they cannot anticipate what other people are thinking. It implies they don't have a sense that Sally will look for it where she left it, based on her prior knowledge. She wouldn't know that Anne had moved it.

The trouble is this experiment was carried out on pre-school children. Autism is for life. And it might not prove that people on the spectrum couldn't anticipate other people's thoughts, as the other option - 50% - is they considered whether they last saw it - not the same thing as mind blindness surely? How does this kind of analogy work for people's whole life experiences?

My thinking would rather put forward the door-handle problem. People on the spectrum are more driven by a certain kind of logic, which they tend to rely upon, and don't so readily use trial and error, which is a more natural response for non-autistic people.

Most people get caught out once in a while by door handles that don't turn logically away from the latch, but people on the spectrum get caught out often, and may actually be unable to open the door and keep trying to turn the handle the "right way". Non-autistic people are more ready to experiment - if it doesn't work one way, try the other - trial and error. Now there's no Theory of Mind involved here, but it is a phenomenon that catches out people on the spectrum.

How does this response differ from Mind Blindness experiments?  Well for all we know non-autistic children probably considered both options more or less simulteneously, and compared and thought through and arrived at an answer. But maybe the autistic children went for the "logical" option, which is where they last saw it placed, and being hooked on that concept, didn't readily consider the alternative.

Consequently, given the logic factor, its not that people on the spectrum have no theory of mind, just they aren't well disposed to trial and error.

Nuff said?

  • There are further issues with experiments of that ilk.  They often use rather complicated language to describe the "problem" involved and ask the relevant question.  Sometimes other tests, which seem to be testing similar Theory of Mind abilities, but are less language-based, appear to 'level the playing field' somewhat between autistic and NT youngsters, so that there is less difference between the two groups.  This would seem to suggest that some of the difficulties may be to do with communication problems inherent to the test rather than more general Theory of Mind deficits.

    An example of a less language-based test would be a more natural 'perspective-taking' test where a toy policeman is positioned behind or around various obstacles near a toy robber and the child is simply asked whether the policeman can see the robber (the child can clearly see both).

    Further, many autistic people who may not 'pass' the test at the expected age, may well go on to do so later.  Suggesting that whatever difference these tests are measuring may often be due to a delay rather than an absence of ability.