Empowering interactive agents with empathetic capabilitiesleads, on the human side, to more trust and increased engagement. Thisarticle focuses on modeling the empathetic behavior of virtual agentsinteracting with humans employing behavior trees which we further enhancewith neuro-symbolic capabilities. This allows the specification ofvarious kinds of empathy for the agents to be deployed for user supportin critical fields such as digital forensics. This paper is a follow-up to ourwork in the COST Action DigForASP.