Introduction A large body of evidence indicates that retrieval practice (test-enhanced learning) and spaced repetition increase long-term information retention. Implementation of these strategies in medical curricula is unfortunately limited. However, students may choose to apply them autonomously when preparing for high-stakes, cumulative assessments, such as the United States Medical Licensing Examination Step 1. Methods Seventy-two medical students at one institution completed a survey concerning their use of user-generated (Anki) or commercially-available (Firecracker) flashcards intended for spaced repetition and of boards-style multiple-choice questions (MCQs). Other information collected included Step 1 score, past academic performance (Medical College Admission Test [MCAT] score, preclinical grades), and psychological factors that may have affected exam preparation or performance (feelings of depression, burnout, and test anxiety). Results All students reported using practice MCQs (mean 3870, SD 1472). Anki and Firecracker users comprised 31 and 49 % of respondents, respectively. In a multivariate regression model, significant independent predictors of Step 1 score included MCQs completed (unstandardized beta coefficient [B] = 2.2 × 10− 3, p < 0.001), unique Anki flashcards seen (B = 5.9 × 10− 4, p = 0.024), second-year honours (B = 1.198, p = 0.002), and MCAT score (B = 1.078, p = 0.003). Conclusions Medical students engage extensively in self-initiated retrieval practice, often with spaced repetition. These practices are associated with superior performance on a medical licensing examination and should be considered for formal support by educators.