[1] EQ tests the empathizing–systemizing theory, a theory that places individuals in different brain-type categories based on their tendencies toward empathy and system creation, and that was intended to determine clinically the role of lack of empathy in psychopathology, and in particular to screen for autism spectrum disorder.
This cognitive theory attempts to account for two aspects of autism disorder: the social and communication barriers and the narrow interest and attention to detail.
That theory is based on the finding that males score significantly higher on the systemizing quotient[3] and lower on the empathy quotient than females in both adult and child populations,[1][6] and that scores of the autistic population were also higher on systemizing and lower on empathizing but to an extreme.
Chapman et al. found that male children who had been exposed to more prenatal testosterone scored lower on the EQ, indicating that there is not only a sex difference in empathy but also a difference within the male population which is correlated with prenatal testosterone.
The authors give many possible explanations for this finding which are contrary to the extreme male brain theory of autism, for example it is possible that the psychometric properties of the quotients are lacking or that the theory itself is incorrect and the difference in autistic brains is not an extreme of normal functioning but of a different structure altogether.
In initial testing, the EQ was examined by a panel of six experimental psychologists, who were asked to rate the match of the items in the measure to the following definition of empathy: "Empathy is the drive or ability to attribute mental states to another person/animal, and entails an appropriate affective response in the observer to the other person’s mental state.
"[1] This definition is based on Baron-Cohen's theory of empathy which includes both a cognitive and affective response to another individual's emotions.
[2] The original authors did not create those divisions because they considered it impossible to separate the cognitive from the emotional aspects of empathy.
Principal component analysis indicated that this shortened questionnaire was able to measure empathy reliably.
The reliability of the Serbian 40-question scale was lower than the English version, and the original theory that empathy is a one-factor concept was not confirmed.
[10] A study of the Dutch version also confirmed the three-factor model, and indicated good cross-cultural validity in Europe and the US, but less so in Asia.