Recently, faculty at Pritzker School of Medicine, The University of Chicago, have made efforts to improve the depth of radiological anatomy knowledge that students have, but no insights exist as to student and resident opinions of how clinically helpful deep anatomical understanding is. A single-institution survey of second- and fourth-year medical students and postgraduate year 1-4 residents from 11 specialties, composed of five-point Likert questions, sample examination questions, and narrative response questions, was distributed in 2015. One hundred seventy-seven of the 466 potential respondents replied (71 residents and 106 students), response rate 38.0%. No nonresponse bias was present in two separate analyses. Respondents generally favored a superficial "identification" question as more relevant to clinical practice, which was positively associated with increasing clinical experience ρ = 0.357, P < 0.001 by point-biserial correlation. Students and residents most commonly used self-directed methods to learn medical imaging during their medical anatomy courses (72.6 and 57.7%, respectively). Small group education was least commonly used by students and residents (45.3 and 39.4%, respectively), but most commonly recommended (62.3 and 69%, respectively). A total of 56.6 and 64.8% of students and residents, respectively, reported that having multiple learning methods was "quite" or "extremely" important. Respondents with more clinical experience were more likely to report that a superficial identification question was more clinically relevant than a question testing deeper radiological anatomy knowledge. Small group learning was preferred among students and residents but was the least commonly employed method of instruction. Both findings contrast starkly with current radiological anatomy instructional understanding and practices. Anat Sci Educ 11: 25-31. © 2017 American Association of Anatomists.