Personality Colors Reveal Surprising Gaps in Medical Student Exam Performance

Can a simple four-color personality test predict how well a medical student will perform on a grueling anatomy practical or a high-stakes neuroscience board exam? A new retrospective study from Mayo Clinic Alix School of Medicine suggests the answer is more nuanced than educators might hope. In research published in BMC Medical Education, a team led by Aimen Vanood of the Department of Neurology at Mayo Clinic in Phoenix, Arizona, examined whether personality characteristics, measured by the Real Colors instrument, were associated with performance across a battery of summative assessments in two of the most demanding foundational courses in the medical curriculum. The findings, drawn from nearly four hundred student-course records collected over four years, reveal statistically significant differences between certain personality groups on some examinations, while other measures of learning showed no relationship with personality at all.

The Real Colors instrument, administered to students during their first two months of school by a certified facilitator, sorts individuals into four temperament types designated by color: Blue, Green, Gold, and Orange. The tool is designed to help people understand human behavior, uncover the motivators specific to each group of personality characteristics, and improve communication with others. Blue types are typically described as relationship-oriented and empathetic, Green types as analytical and intellectually driven, Gold types as organized and duty-focused, and Orange types as action-oriented and spontaneous. The framework shares conceptual roots with other temperament models used in medical education research, including the Myers-Briggs Type Indicator and the HEXACO model, which have been correlated in previous studies with medical student performance and specialty preference, albeit with inconsistent results across the literature.

To test whether these temperamental categories mattered for actual course outcomes, the researchers linked personality data to a comprehensive set of assessments. Performance was measured through scores on National Board of Medical Examiners anatomy and neuroscience examinations, an institutionally developed neuroscience written exam, neuroscience quizzes, an anatomy laboratory practical examination, and a radiologic anatomy exam. Personality data were available for 193 students in the neuroscience course and 196 students in the anatomy course, spanning four academic years at the school’s Arizona campus. The study received ethical approval from the Mayo Clinic Institutional Review Board under protocol number 21-012951, with a waiver of consent granted, and was conducted in accordance with the principles of the Declaration of Helsinki.

The first striking result was the distribution of personality types within the student population. Blue was by far the most common color among these future physicians, accounting for 43.0 percent of students in the neuroscience cohort and 45.4 percent in the anatomy cohort. Green came second at 29.0 and 28.6 percent respectively, followed by Gold at 16.0 and 14.8 percent, and Orange at just 11.9 and 11.2 percent. This skew toward Blue and Green temperaments offers a quantitative portrait of who applies to and matriculates at a highly selective medical school, and it echoes long-standing observations that certain personality profiles are overrepresented in the physician workforce compared with the general population.

When the researchers applied the Kruskal-Wallis test, a nonparametric statistical method for comparing outcomes across multiple independent groups, they found significant differences between color groups on three assessments: the neuroscience NBME exam, the anatomy NBME exam, and the anatomy laboratory practical examination. These are among the most consequential measures in the preclinical curriculum, since NBME subject exams are standardized national tests and laboratory practicals demand hands-on mastery of three-dimensional anatomical structure. The fact that personality group membership surfaced as a statistically significant factor on precisely these assessments, and not on the institutionally written neuroscience exam, the quiz series, or the radiologic anatomy exam, hints that temperament may interact with specific formats and demands of testing rather than with general academic ability.

Post-hoc analysis using the Dunn-Bonferroni correction, a conservative method that guards against false positives when making multiple pairwise comparisons, sharpened the picture considerably. The significant differences were consistently located between the Green and Orange groups. In other words, the analytical, intellectually oriented Green students and the action-oriented, spontaneous Orange students differed reliably on those three high-stakes assessments. Given that Orange students made up only about 11 to 12 percent of each cohort, they represent a small minority whose experience of the anatomy and neuroscience curriculum may differ in measurable ways from that of their Green classmates. The direction and magnitude of these differences matter for educators trying to ensure that no temperament group is systematically disadvantaged by course design.

The study also took advantage of a natural experiment created by the COVID-19 era: the abrupt shift between virtual and in-person learning environments. A subset of the data was used to examine whether personality characteristics were associated with success in one modality versus the other. The sub-group analysis found no differences in course performance between color groups when comparing virtual with in-person environments. This null result is notable in itself. It suggests that whatever advantages or disadvantages personality confers on particular assessment formats, those effects were not amplified or erased by whether instruction was delivered face-to-face or through a screen, at least for the courses and cohorts examined.

The authors are careful about interpretation, and their conclusions lean toward caution rather than dramatic claims. They suggest that the discrepancies detected in the analysis may be due to learning preferences and preferences for particular learning environments that are potentially related to personality characteristics. They also point out that other measures of performance, such as standardized patient encounters, may enhance understanding of how personality characteristics apply to medical education. Clinical skills assessments, which evaluate communication, empathy, and interpersonal conduct, are precisely the domains where Blue and Orange temperaments might be expected to shine, and where Green and Gold students might face different challenges. Without such measures, the study captures only the cognitive and technical dimensions of medical training, leaving the interpersonal half of the profession unexamined.

Perhaps the most practical takeaway from the research is its recommendation about how personality instruments should actually be used in medical schools. The authors conclude that these tools may be best utilized not to predict grades or gatekeep admissions, but to foster effective communication and teamwork skills and to help educators develop instructional methods and materials that appeal to learners with diverse preferences. In a profession where physicians must collaborate across specialties, communicate with patients from every walk of life, and function in high-stress teams, understanding one’s own temperament and that of colleagues could deliver benefits that far outweigh any marginal correlation with an exam score. The finding that nearly half of one class shares a single dominant color also underscores the risk of homogeneous teams and the value of deliberately mixing temperaments in small-group learning.

For the broader field of medical education research, the study adds a rigorous data point to a literature long criticized for inconsistency. Previous efforts to correlate personality with performance have produced contradictory findings, partly because of varied instruments, small samples, and differing outcome measures. By using a validated facilitator-administered instrument, a large multi-year cohort, standardized national exam scores, and appropriate nonparametric statistics with corrected post-hoc comparisons, the Mayo team has modeled a methodologically sound approach. The results confirm that personality is not irrelevant to medical student outcomes, but neither is it destiny. The effects are selective, appearing on some assessments and not others, and they did not interact with the pandemic-driven shift to virtual learning. As medical schools continue to refine admissions, curriculum design, and learner support, this study suggests that personality data can inform how educators teach and how students work together, while grades on the next anatomy practical will still depend, above all, on the hours a student puts into the dissection lab.

Subject of Research: The relationship between medical students' personality characteristics and their academic performance in anatomy and neuroscience courses

Article Title: Do personality characteristics impact medical student learner outcomes in anatomy and neuroscience courses?: a retrospective observational study

Article References: Vanood, A., Langley, N. R., Hoerth, M. T., Chhabra, N., Drazkowski, J., Bright, R., Cannon, K., Thorn, P. M., & Shah, A. (2026). Do personality characteristics impact medical student learner outcomes in anatomy and neuroscience courses?: a retrospective observational study. BMC Medical Education. https://doi.org/10.1186/s12909-026-10476-4

Image Credits: AI Generated

DOI: 10.1186/s12909-026-10476-4

Keywords: medical education, personality, Real Colors instrument, anatomy, neuroscience, NBME exams, Mayo Clinic, psychometrics, learning environments, medical students, temperament, retrospective study