The Factor Structure of Cognitive and Emotional Empathy in Individuals with Autism, Their Parents and General Population Controls

Background: Empathy is a vital component for social understanding as it allows an individual to make sense of and predict the behaviour of another.  It is posited that some of the core features of autism spectrum conditions (ASC) can be accounted for by a deficiency in empathising, with individuals with autism and their family members scoring lower on questionnaire and cognitive tasks involving empathy. Several studies suggest that empathy is multidimensional, with at least two components: cognitive and emotional empathy. Cognitive empathy is the drive to attribute mental states to oneself and others; emotional empathy refers to the drive to respond with an appropriate emotion to another’s mental state. The Empathy Quotient (EQ), a self-report measure of empathising has been studied in detail across three studies. Two studies highlighted a three dimensional structure including cognitive empathy, emotional empathy and social skills, the third showing these dimensions can be accounted for by a higher order empathy factor. However, the relationship between the behavioural subscales of the EQ and other cognitive tasks of empathy has not been assessed to date. Moreover, it is unclear whether the structure is similar across general population and clinical samples.

Objectives: The current project examined empathy using both behavioural and cognitive task data from measures including the ‘Reading the Mind in the Eyes’ Test and the Karolinska Directed Emotional Faces Task; tasks assessing the ability to identify emotion in another individual. The research aimed to assess whether the factor structure of empathy differs across samples stratified by genetic vulnerability.

Methods: Factor analyses were conducted to assess the latent structure of empathy across samples with a high (individuals on the spectrum), medium (first-degree relatives) or low (general population controls) estimated genetic vulnerability to autism.

Results: Analysis of the EQ highlighted a three factor model, confirming a cognitive empathy, emotional empathy and social skills factor across all the three groups. Conversely, the cognitive tasks did not primarily load on cognitive empathy, but were nearly equally related to each behavioural empathy subfactor across each group. However, the cognitive tasks were more related to emotional empathy in parents and individuals with autism compared with controls.

Conclusions: This study provides insights into the latent structure of empathy in individuals with high, medium and low genetic risk for autism. Results highlight that empathy as a behavioural trait shows evidence of multidimensionality, in which three factors can be distinguished irrespective of genetic vulnerability. However, performance measures of tasks assessing empathy were related in almost equal magnitude to all three components, rather than solely to cognitive empathy. These tasks may be more strongly related to emotional empathy in individuals with a medium to high genetic vulnerability to autism. Impairments on these cognitive tasks are likely to have wider implications on the behavioural level for emotional empathy and social skills.