Investigating the Impact of Individual Differences on Empathy

Abstract

Empathy is an integral element of social cognition that has more recently been conceptualised as a two-factor construct, consisting of a cognitive empathy dimension and an emotional reactivity dimension, was developed to measure the latent variable of empathy. Although older versions of the psychometric measure tend to consist of around 40 to 60 questions, more recent iterations have consisted of around 10 to 20 questions, such as the EQ-10 - though the validity of the measure and fit of the model has yet to be assessed when applied to a UK neurotypical healthy sample. Literature has also conceptualised relationships between individual differences as empathy, such as personality, interoception, and attachment dimensions.

The following thesis contains two studies: Study 1 uses a confirmatory factor analysis to determine the strength of loadings of survey items, indicating suitability to apply the model as an accurate measure for the cognitive and affective subscales of empathy. Following this, in Study 2 uses a stepwise regression to determine the impact of individual differences on the cognitive and emotional reactivity subscales. The studies conclude that the EQ-10 is a valid measure of suitable fit, and that individual differences significantly impact performance on the EQ-10 subscales.

The most impactful and significant predictors of the cognitive dimension of empathy were found to be interoception, agreeableness, and extraversion. Conversely, the most impactful and significant predictors of the emotional reactivity dimension of empathy were found to be neuroticism, agreeableness, and extraversion.

Results are discussed in relation to multidimensional theories of empathy, including the individuality of emotional identification and affect sharing processes, with future directions for studies surrounding restructuring of the EQ-10 items to improve overall model fit, and implementation of alexithymia directly as a predictor variable in future research.

Date of Award	9 Dec 2025
Original language	English
Awarding Institution	University of Bristol
Supervisor	Michael J Banissy (Supervisor) & Chris Kent (Supervisor)