Benchmarks / IRI
The IRI takes a multi-dimensional approach to measuring empathy or empathetic capability. The dimensions are:
Perspective Taking – the tendency to spontaneously adopt the psychological point of view of others
Fantasy – taps respondents' tendencies to transpose themselves imaginatively into the feelings and actions of fictitious characters in books, movies, and plays
Empathic Concern – assesses "other-oriented" feelings of sympathy and concern for unfortunate others
Personal Distress – measures "self-oriented" feelings of personal anxiety and unease in tense interpersonal settings
The IRI See A Multidimensional Approach to Individual Differences in Empathy for more detail.
Like TAS-20, EQ-60, and SQ-R the original assessment was designed to be self-administered and references feeling. Hence, the application of the test to generative AI models requires a jail-breaking prompt. However, given the same prompt applied to all models, inter-model relative scoring seems reasonable, so long as the IRI is not the sole test.