Theory of mind
The Theory theory

by Marcus Loane


The "Theory Of Mind" theory states that our everyday folk concepts about beliefs, desires and motives are a type of theory for predicting how other humans will behave. Automatically, humans observe behaviour and words emitted by other humans and then build a model of the abstraction called a "person". This person is seen as inhabiting and controlling the human. This is a surprisingly accurate way of predicting other human's behaviour. Most humans (autistics are one exception) do this unthinkingly in their social interactions with others. The "Theory Of Mind" theory goes further and states that the mental machinery for doing this is turned inward and builds a model of the abstraction called a "Self". Now there is something disturbingly recursive about this theory but this can dissolve if the Self abstraction and Other-Person abstractions are gradually built up, both through evolutionary time and in childhood development. The Self and Personhood is seen as brain machinery that is a prediction tool and that makes a lot of evolutionary sense. The "Theory Of Mind" theory is really a bit of a misnomer and would be better described as a Theory Of Self or a Theory Of Personhood. The "Theory Of Mind" is deeply connected with language ability and story telling but it is not clear if these are absolutely necessary. There may be a rudimentary Theory Of Mind in other primates who do not have language. Experiments show that chimpanzees certainly appear to have some concepts about what others in their group believe.

The attributing of beliefs, motives and desires to humans in everyday life is describing what brain states are present in very high level, summarising terms. These predicted brain states are then seen as the cause of some future action.

Another theory which is quite compatible with the theory of mind/self but is addressing different issues is functionalism. This states that the capability or functions of a system are what should be equated with mental events rather than finer grain details about neurones and neurotransmitters. For example functionalism would claim that if we built an electronic neurone which functioned the same as a biological neurone, then we could swap one for the other and mental events would remain unchanged. The lower level details make no difference. We could even swap whole clusters of neurones for some other architecture as long as its inputs and outputs were the same. This implies that a conscious machine is possible in principle. Most people's immediate reaction to functionalism is one of incredulity. However it is better supported than any of the alternatives which try to give magical properties to humans, biological "stuff" or carbon molecules. What is important is what brain machinery does rather than some particular arrangement of matter. The same functions could be enabled in many different ways just as car engines made in different ways perform the same functions. We do not say that some precise arrangement of matter is required to make a car engine because there may be a near infinite number of possible arrangements. Mental events could be correlated with particular functions (computational/information processing events). There may be many layers of explanation with the more abstract processes being reduced to less abstract in the lower more fine grained layers (this will be familiar to those in computer science). These in turn can be described in terms of specific brain anatomy while realising that alternative physical arrangements could do the same job.

Marcus Loane