Objective: The aim of this study was to explore the relationship between cognitive and behavioural empathy in medical students.

Methods: Fourteen 4th year medical students recruited on the basis of their scores on the self-reported Jefferson Scale of Empathy (JSE-S) were divided into two groups: low JSE-S scorers (n = 8) (M = 96.75, SD = 10.3) and high JSE-S scorers (n = 6) (M = 121.3, SD = 2.94). They were discreetly videotaped while taking history with an incognito standardized patient. Students' behavioural empathy was measured using the Verona Coding System (VR-CoDES-P) and rating of non-verbal behaviour.

Results: Patients expressed the same number of concerns per encounter in both groups but gave more cues to high-scorers (p = 0.029). However, students of both groups demonstrated the same amount of verbal empathy (high: 16% vs low: 15% p = 1.00). High JSE-S scorers' non-verbal communication tended to be rated slightly higher than low JSE-S-scorers with a higher use of facial expression (p = 0.008).

Conclusion: This study did not reveal any differences of students' verbal empathy to patients' cues and concerns between low and high JSE_S scorers.

Practice implications: The VR-CoDES_P is a useful tool to assess medical students and physicians empathic behaviour, allowing to disentangle the different components of empathy.