Natural semantic networks (NSNs) provide an approach to understanding concepts in terms of their density and strength. The objective of this study was to conceptualize mistreatment and good treatment among health science students in Chile, according to gender and training cycle. Methods: A mixed-methods study with a relational scope was conducted, involving 994 health science students. The present study is exploratory and combines both qualitative and quantitative analysis. We utilized the NSN technique, applying a questionnaire via SurveyMonkey after obtaining informed consent. Results: A total of 4386 defining words for mistreatment were identified, with the most relevant being ‘aggression’, ‘abuse’, and ‘violence’. In the case of good treatment, a total of 4146 defining words were obtained, with ‘respect’, ‘empathy’, and ‘kindness’ standing out. For good treatment, a statistically significant relationship was observed between the defining words and the variables of gender (χ2 = 41.158; p < 0.05) and training cycle (χ2 = 28.698; p < 0.05). For mistreatment, a statistically significant relationship was found only with training cycle (χ2 = 35.858; p < 0.05). Conclusions: Exploring the meanings associated with mistreatment and good treatment has a theoretical implication in the way in which two highly polysemic aspects of the educational environment can be understood.