Traditional Cadaver vs Virtual Dissection: Preferences and Effectiveness in Sri Lankan Medical Students

Introduction: Anatomy education traditionally relies on cadaver dissection for immersive, tactile learning. However, shortages of body donations, ethical concerns, and logistical constraints have prompted integration of virtual dissection tools worldwide. In Sri Lanka, cadaver-based instruction remains primary method of learning anatomy, but limited availability and emerging digital alternatives raise questions about student preferences and perceived effectiveness. This study compared traditional cadaver and virtual dissection tools among Sri Lankan medical students, focusing on preferences, perceived learning effectiveness, and ethical considerations. Methods: A cross-sectional survey was conducted among 150 undergraduate medical students from three government medical faculties in the Western Province. Data were collected via a structured, expert-validated online questionnaire. Descriptive statistics, chi-square tests, t-tests, and ANOVA explored associations between demographics, preferences, and perceptions. Results: Nearly half of participants (46.7%) preferred cadaver dissection, 34.7% a combined approach, and 18.7% chose virtual tools alone. Confidence was higher among students favoring traditional methods (p=0.001, d=0.65). The combined approach was rated most effective for learning (M=4.08, SD=1.08). Ethical acceptance of cadaver use was high (90.7%), whereas only one-third supported virtual dissection as a replacement. Comfort with cadavers was strongly associated with method preference (χ²=11.83, p=0.003). Mid-program students most strongly endorsed hybrid learning, reflecting recognition of complementary strengths. Conclusion: Cadaver dissection remains highly valued in Sri Lankan anatomy education, but virtual tools offer supplementary advantages. Blended learning integrating both methods may optimize experiential depth and flexibility, guiding curriculum development and resource allocation in culturally and resource-sensitive contexts.