Background: Postgraduate nursing education is a program which require knowledge and skills that guarantee the quality of professional competence. Many courses in nursing have learning outcomes including clinical reasoning (CR). Assessment is an important part of the learning process and should assess clinically relevant performance. Virtual Patients (VPs), have successfully been used in nursing education to promote learning and also been piloted for assessment of CR. Summary of work: However, it remains to solve how to score and grade VPs for assessment. Therefore, we need to determine the best practises that the assessment should be based on. This study investigates how clinically experienced specialist nurses apply CR for clinical decision making, to create the basis for a holistic assessment method based on VPs. Summary of results: VP cases representing different subdisciplines in paediatrics are implemented in a VP-system. 40 experienced nurses will work in pairs to solve the VP-cases, and their actions are logged in terms of what they do and in what order. Think Aloud is used to capture their thinking process. Conclusions: This study will give information about how experienced nurses solve clinical cases and thus which issues that should be assessed in VP-based assessment for nursing. Take-home messages: VP-based assessment for specialist nursing education is being developed.