Improving Clinical Education Through the Use of Virtual Patient-based Computer Simulations.

The term Virtual Patient (VP) refers to the use of virtual characters which embody patients in a virtual environment. They are implemented in computer simulations to create realistic clinical encounters. VPs have been used successfully in health education to promote and foster clinical communication skills. Additionally, computer simulations offer the advantage of being standardized, safe, repeatable, and do not require as much resources as role-play simulations which rely on actors. This thesis addresses the design and evaluation of a VP-based system aimed for clinical trainees, and uses the field of audiology as a case study. The system is designed to simulate real client encounters and allows students to practice using a standard set of procedures that they have to master in their profession. A wide range of VPs have been implemented for this purpose. The system was evaluated with audiology students, reinforcing the ecological validity of the research. The design of the system was guided by an iterative process of implementation, usability testing, and experiments focusing on students' learning outcomes. The Clinical Audiology Simulator (CAS) was evaluated during five experiments, assessing students learning gains following exposure to the CAS. Learning gains have been assessed through the use of role-play simulations and paper assessments. The procedures evaluated are clinical history taking, pure tone audiometry, and speech audiometry. A further experiment assessed the impact of additional formative feedback on students learning gains, using the pure tone audiometry procedure as an example. The results of these experiments suggest that the system has a great potential to foster students learning, with measurable gains in some of these procedures. They also indicate that feedback and its delivery take an important role in this process. This thesis elaborates how VP-based simulations can reinforce young clinicians' ability to learn procedural skills. I highlight some of the challenges a researcher faces in designing and evaluating such systems, focusing on the implementation of interaction scripts for the VPs, the assessment of learning gains and transfer of skills, and the evaluation of computer simulations as part of a curriculum. VPs have the potential to promote clinical trainees' learning of skills, and to provide students with more opportunities for safe practice in a field where beginning trainees often have few opportunities for actual hands on experience