Some difficulties behind the concept of the 'Extreme male brain' in autism research. A theoretical review

The idea that autistic symptoms are produced by an 'extreme male brain' warrants further examination. The concept of the 'extreme male brain' derives from scores on the Autism Spectrum Questionnaire (AQ) i.e. it is defined behaviourally rather than anatomically. But if the concept of the 'extreme male brain' is to explain rather than describe autistic behaviour then evidence must come from a non-behavioural (e.g. physiological, biochemical or anatomical) source to avoid circularity of argument. The lack of a cognitive intervening variable linking autistic behavioural traits to brain activity casts doubt on the existence of a uni-dimensional spectrum of 'brain gender' for which autism is to be found at the 'extreme male' end. The inappropriate conflation of the dependent variable (score on the AQ) and the independent variable (brain anatomy of respondent) has led to the claim that a person with autism, even if female, has an 'extreme male brain'. This is comparable to the claim that, because on average men are taller than women, extremely tall women have 'extreme male height'. This stereotypical view of gender fails to recognize the overlapping diversity of cognitive styles found in males and females.