The use of voiceovers and review games in teaching immunology Authors: Emily Chou, Dylan Cooper, Pranshul Goel, Justin Tang, Garrett Henkle, Savannah Zemeida, Hannah Wolfson, Yennie L Shyu, Esther Chang, Alyssa Abram, Valerie A Gerriets, John K Cusick Published: 2026 ABSTRACT AbstractActive learning and flipped classrooms have gained popularity in higher education due to their ability to improve educational outcomes. This study evaluated whether combining voiceover lectures and review games could improve learner engagement and satisfaction while achieving learning outcomes comparable to those of traditional immunology didactic lectures. A total of 123 first- and second-year medical students were randomly assigned to groups that took formative quizzes on immunology material after exposure to either: (i) a traditional didactic lecture, (ii) a voiceover lecture followed by an in-class discussion of the voiceover and subsequent review game, or (iii) a blind control group that had no exposure to the material. Furthermore, students provided feedback on the pedagogy and their learning preferences. Students were very enthusiastic about the review games and preferred shorter voiceovers containing formative questions embedded within the voiceover. Importantly, there was no significant difference in quiz performance between the 2 pedagogies, and both groups performed significantly better than the blind control group. A slight but significant decrease in student preference for voiceovers was noted throughout the duration of this multiyear study, which occurred during and shortly after the COVID-19 pandemic. Faculty who observed this pedagogy expressed an increased willingness to use voiceovers and review games if they received training and information technology support. Collectively, this longitudinal study demonstrated support from both faculty and students for the combined use of voiceovers plus review games and captured a potential shift in medical student preferences for learning in recent years during the COVID-19 pandemic.Keywords: flipped classrooms, review games, voiceovers Introduction Traditional didactic lecturing has been a cornerstone of medical education since medical education’s inception.1 However, traditional lectures have limitations in terms of student engagement and attention, and attendance at pre-clerkship medical student lectures (ie the didactic lectures taken in the first 1.5 to 2 years of medical school before clinical rotations) has been trending downward nationally in recent years.2 Previous studies suggest the average attention span of a medical student peaks 10 to 15 min into a lecture and steadily declines afterward,3 and more recent studies indicate that the average attention span of current students may be even shorter than 15 min.4 There is some evidence that making lectures more dynamic and interactive can greatly increase student participation and engagement, and may help “restart the attention clock” during a lecture.5 The flipped classroom model has garnered significant attention in educational circles for its innovative approach to teaching and learning. Unlike traditional lecture-based instruction, the flipped classroom involves students engaging with preclass materials such as preassigned reading, podcasts, handouts, or prerecorded voiceover lectures (hereafter, voiceovers) prior to the scheduled class time.6,7 This preparatory work allows class time to be used more interactively, using methods like board games, small group discussions, review games, and various other activities to reinforce the learned material.8 Numerous studies have illustrated the benefits of this approach, reporting overwhelmingly positive responses from students, improvements in student motivation and attitudes, and notable improvements in test scores.6,7 The proposed mechanism behind the efficacy of the flipped classroom model is its promotion of active learning and higher levels of Bloom’s taxonomy in the classroom, such as application, analysis, and synthesis, leading to enhanced learning and retention of information.9 Entire new pedagogies such as team-based learning emerged in the 1970s to promote a more collaborative and active learning environment, encouraging students to take ownership of their education and engage more deeply with the subject matter.10 Medical students have demonstrated a strong preference for active learning pedagogies over traditional lectures in multiple studies,11,12 and a comprehensive meta-analysis of undergraduate courses across the science, technology, engineering, and mathematics disciplines demonstrated that active learning led to a significant improvement in student performance on exams while also significantly decreasing the rate of failing grades received.9 One mode of active learning is the use of review games, which offer the opportunity for students to participate in a friendly competition13 while enhancing the learning and memory of subject material.14 “Gamification” describes the use of game design elements in traditionally nongame contexts and has become increasingly popular in science education over the past several decades.15–17 Students competing as teams can collaboratively create learning communities that boost motivation, encourage continued involvement, and enhance learning.16,18 Studies in a medical school setting have indicated that gamification may increase motivation, engagement, and the overall effectiveness of educational experiences.17,19 The 2 review games primarily used in the present study were the Kahoot learning platform, which offers review games or “Kahoots,” and a game using PointSolutions (formerly TurningPoint) technology. Both Kahoot and PointSolutions games permit students to compete using their cell phones or other personal electronic devices, and students can also participate in real time from their homes if lectures are delivered live by video capture or Zoom. Kahoot has been used extensively in the education of health sciences and has received praise for its appealing features, including music, leader-tracking features, appealing colorful graphics, and the ability of students to create their own team names. Multiple studies have shown that Kahoot is well received, with students believing that Kahoot strengthened the teaching content20 and increased their focus, understanding, and retainment of knowledge21 while promoting a fun and relaxing atmosphere.22,23 Review games using PointSolutions technology have also been used extensively, and previous reports have demonstrated that clickers using TurningPoint technology were well received.13 One advantage of PointSolutions is the ability to include questions asking students what percentage of points they wished to “wager” before a challenging question; a previous study reported that 74% of students surveyed found the wagering slides very or somewhat engaging.13 The use of prerecorded voiceovers that students watch on their own time is 1 method to facilitate flipping classrooms and has become more popular with the advance of video and auditory technology24; voiceovers have alternatively been termed “video podcasts” and “vodcasts.”25 High-quality video-based lectures can better engage medical students and help them overcome intimidating subjects,26 and medical students in 1 study indicated that videos are more engaging and facilitate better retention of information compared with textbooks.24 To our knowledge, no previous studies have compared the use of voiceovers in combination with review games versus traditional lectures. Previous studies have compared educational outcomes of students exposed to either a lecture or a review game (without voiceover)27–29 or compared whether students performed better based on the order of being exposed to a prerecorded lecture and review game.30 The influence of video-based flipped class instruction (ie use of voiceovers) on reading motivation has been assessed,31 as has the impact of using voiceovers combined with gamification in a flipped classroom model on student motivation and performance (yet no comparison with traditional lectures).32 There have been many studies on the impact of gamification in flipped learning, some of which used voiceovers as preassigned material.33 Considering the reported benefits of voiceovers and gamification strategies, it was of interest to compare their combined use with traditional lectures and assess the potential synergistic benefits to student engagement of combining voiceovers and review games. A big challenge for matriculating medical students is learning how to learn, because most have not experienced the rapid pace of learning they are required to master in medical school, with respect to both the depth and breadth of material, in what has been described through the metaphor “drinking from the firehose,” and many resort to purchasing external third-party resources to help prepare for exams. There are several proposed frameworks to describe different ways in how people learn, including the VARK model developed by Neil Fleming, which proposes students have 4 main learning preferences: visual, auditory, read/write, and kinesthetic (hence, VARK).34 We examined in the present study student preferences for learning by the VARK model, in addition to their preferences for learning by either attending traditional didactic lectures or participating in active-learning classroom pedagogies. A profound interest has emerged in identifying how students learn, because it has become apparent that the traditional lecture mode of teaching is not always the ideal delivery method for students,35 especially when considering that many medical students opt out of attending in-person learning sessions their considerable tuition already purchased to learn from third-party resources that require additional personal expense by the student.36–39 Collectively, in this longitudinal study, we evaluated whether combining voiceovers and review games could improve learner engagement and satisfaction in learning immunology, a complex subject that novice learners often find intimidating, while also maintaining learning outcomes comparable to those of traditional immunology didactic lectures. We obtained feedback from both students and faculty regarding the implementation of this pedagogy as well as feedback on student learning preferences during this multiyear study. Because this study coincided with the COVID-19 pandemic and the years shortly thereafter, it also provided an opportunity to examine whether student learning preferences shifted over this time. Materials and methods Participants Participants consisted of first-year (n = 102) and second-year (n = 41) medical students who matriculated into California Northstate University College of Medicine (CNUCOM) between 2019 and 2023. Participation was voluntary and signed consent was obtained from participating students. Participants were incentivized through entry into raffles for small prizes of school merchandise (eg university jackets, hats, and backpacks) upon completion of the surveys. This study was performed with first-year medical students during the fall semester and with second-year medical students (M2s) during the spring semester during the years 2021, 2022, and 2023. Faculty representing both the medical school and undergraduate college of California Northstate University were invited to watch a voiceover lecture and attend the accompanying review game before providing their feedback on the pedagogy. Study design This study received approval from the California Northstate University Institutional Review Board. One faculty member (V.A.G.) was responsible for interfacing with the students regarding the research study, collecting the survey data, and ensuring the anonymity of student data. A different faculty member (J.K.C.) created the voiceover and review games and did the immunology teaching intervention. The other coauthors were medical students who participated in collecting and analyzing survey data on this project. Students were ensured that enrollment in the study was completely voluntary and would not affect their grades or educational experience. Students who were interested in enrolling in the study then gave their names and email addresses, and a random number generator was used to sort the participating students into 3 experimental groups that took an identical formative quiz before or after exposure to the subject material. Each participant was also given a unique 6-digit code to use for the surveys and only 1 researcher (V.A.G) had access to the key linking the names and codes. Although all students eventually had access to all the educational materials, the groups were instructed to do so in different sequences. Students assigned to the “Lecture” group were instructed to attend the assigned in-person lecture and take the formative quiz within 24 h of attending the lecture; these students were also instructed to complete the online formative quiz prior to watching the voiceover and/or attending the in-person review session and participating in the review game. The review session included a review game combined with clarification of concepts from the voiceover and was scheduled at least 24 h after the lecture. Students assigned to the “Voiceover + Review game” group were instructed to (i) not attend the lecture and to hold off on watching the video-captured in-person lecture; (ii) watch the voiceover in advance of attending and participating in the in-person review session, which included a review game; and (iii) to take the formative quiz within 24 h after participating in the review game. After participating in the voiceover and review game and taking the formative quiz, the students were free to view the lecture recording if they wished. A third group served as a blind control group (hereafter, Control). Students in this group took the formative quiz prior to exposure to either the lecture or voiceover plus review game pedagogies and were then free to attend all the educational sessions. All the participating students were also invited to complete a post-study survey about their views of the educational intervention because they had all been previously exposed to the use of voiceovers combined with review games multiple times for previous immunology topics. The 6-digit code given to each student was used to verify that the student took the formative quiz at the appropriate time for their assigned group. The formative quizzes were administered remotely using the platform SurveyMonkey. Prerecorded voiceovers were created using Camtasia recording software and were streamed through Mediasite. The voiceover contained the same slides and informational content that was present in the PowerPoint (PPT) file used for the lecture and contained audio narration by the same faculty member who gave the corresponding in-class lecture. The voiceovers ranged from 20 to 30 min in length and, when combined with a brief review of the voiceover in class, were approximately the same length as a 50-min lecture. A “script” in the form of a pdf was also provided to students along with the voiceover, enabling students to watch the voiceover without needing to take notes, because the script contained the exact text narrated by the faculty member to explain the slides in the voiceover. The content in the prerecorded and traditional lectures covered 3 subjects: (i) humoral immunity and (ii) MHC and antigen presentation, for first-year students; and (iii) the immunology of pregnancy, for second-year students. Both lectures for the first-year students occurred in their first medical school course, the Foundations of Clinical Medicine, whereas the lecture material presented to second-year students occurred in the Reproductive Course, the last didactic course CNUCOM students take before studying for the Step 1 exam, which they must pass prior to entering the third-year clinical clerkships. Kahoot and PointSolutions software were used as the source of review games in the final 2 years of the study. A Jeopardy game using team clickers and a team leaderboard40 was used in combination with Kahoot in the initial year of the study. The Jeopardy game was created using TurningPoint 5 software; sequential updates in the TurningPoint/PointSolutions software were incompatible, unfortunately, with the original Jeopardy game. Fortunately, PointSolutions began releasing updates to the platform that enabled the use of review games with a leaderboard; therefore, the Jeopardy game was replaced with PointSolutions review games in the second and third years of the study. Questions distinct from those used in the formative quiz were used for review games, and students were encouraged to participate as teams and discuss the questions as small groups before submitting an answer for each question. Explanatory PPT slides from lectures were incorporated into the review game after the correct answers were given, to help reinforce learning. Additionally, a “doc cam” was used to draw on the projected screen periodically during either the review game or during traditional lectures to reinforce concepts and clarify any misconceptions. Attendance at both lectures and the review games was optional, as was participation in review games, because students were able to self-assemble into teams or compete individually if they chose to do so. Formative quizzes consisted of approximately 15 multiple-choice questions on material covered during the learning sessions; all students participating in the study received the same formative quiz. The 3 formative quizzes are included in Fig. S1. Student feedback and opinions regarding flipped classroom instruction were collected anonymously using SurveyMonkey to assess student impressions and feedback on prerecorded voiceovers, review games, as well as the learning preferences of students (Fig. S2). All participating students were encouraged to provide feedback on the post-study survey regardless of which group they were placed in, because voiceovers and review games were used periodically to teach immunology throughout the medical pre-clerkship curriculum, and students placed in either the Lecture or Control group, therefore, previously had been exposed to voiceovers and review games covering additional immunology subjects that were not part of this study. The length of the survey and questions used in the survey varied slightly with successive academic years due to changes and qualitative improvements in study questions and design. Faculty members (n = 9) were also invited to watch a voiceover lecture and attend the corresponding flipped classroom session and were asked to provide feedback, in a 27-question online survey, regarding this pedagogy (Fig. S3). Survey Monkey was used to administer the formative quizzes to students anonymously, as well as to collect impressions and feedback from both students and faculty anonymously. Faculty who mentored students in conducting this research project, in addition to the faculty member delivering the immunology content, did not submit answers on the Survey Monkey survey, to avoid a potential conflict of interest. Statistical analyses Single-factor ANOVA was used to assess the statistical significance of differences in quiz performance. This was followed by Tukey’s honestly significant difference test as a post hoc analysis to determine which groups, if any, demonstrated significant differences. Further statistical analysis on survey responses was conducted using R, version 4.3.2. Because the data were not normally distributed, appropriate nonparametric tests were used. For survey questions with responses recorded on a Likert scale, the Mann-Whitney U test was used to identify significant differences in survey responses by year of administration or matriculation year. For binary responses, Fisher’s exact test was used to determine associations between survey responses and these same factors. Simple linear regression analysis was used to assess trends in preferred learning modalities across classes. A P value < 0.05 was considered statistically significant for all tests. Results Quiz performance Students were divided into groups, and student performance after being exposed to the different teaching modalities was assessed through a formative quiz pertaining to the immunology subject material as described in Materials and Methods. The results were compiled throughout the multiyear study. The lecture group and voiceover/review game group scored an average of 63.8% (n = 47) and 61.7% (n = 47) correct on the post-intervention quiz, respectively (Fig. 1). Although performance between these 2 groups did not differ significantly (P = 0.89), both groups performed better than the negative control group, which averaged 30.3% (n = 29; P < 0.0001). Student performance on formative quizzes after varying teaching pedagogies. Students were divided into 3 groups and exposed to either (i) no prior exposure to learning material (n = 29), (ii) a traditional didactic lecture on class material (n = 47), or (iii) a short voiceover followed by a discussion and review game (n = 47). After exposure to different teaching pedagogies, students then took formative quizzes on lecture material. *P < 0.0001. Student feedback on use of voiceovers combined with review games Participants were surveyed for their preferences regarding flipped classroom pedagogy (Fig. 2A). Responses were scored on a 7-point Likert scale. Overall, students responded positively to the combined use of voiceovers and review games, because more students preferred voiceovers combined with review games over lectures (4.53 out of 7; n = 64). This score was higher than the preference for traditional lectures over the combined use of voiceovers and review games (3.95; n = 56). The enthusiasm for review games was greater than that of voiceovers, because more students wished to have review games more frequently (5.62; n = 73), versus having more voiceovers used for lectures (4.77; n = 83). Students felt the review game solidified their knowledge (5.80; n = 54) and cleared up any misconceptions they had regarding the material (5.63; n = 54), and they thought the voiceover lecture enhanced their performance in the subsequent review game (5.24; n = 59). Students reported higher energy levels and interest during review games in comparison with most traditional lectures (5.57; n = 58) and felt that the atmosphere during review games was friendly, yet not overly competitive (6.16; n = 58). They also preferred active learning modalities and felt these modalities helped them understand and retain material (5.24; n = 59), yet only a slight majority of students claimed they learned more from their teammates during review games in comparison with most lectures (4.12; n = 57). Students also indicated they prefer active-learning modalities even if they were required to complete preassigned activities (4.71; n = 59), although few students stated they regularly use textbooks or complete the readings suggested by faculty (1.81; n = 58). Additional student feedback obtained from survey is listed in Table S1. Student survey responses regarding preferences for learning methodologies. From January 2021 to August 2023, students were surveyed about their preferences for learning methodologies. Responses were recorded on a 7-point Likert scale for compiled survey responses across all years (A), selected survey responses stratified by academic year (B), and selected survey responses stratified by matriculation year (C). For (A) and (B), the P values of select comparisons of data points are indicated, asterisks denote statistically significant differences of P < 0.05. Interestingly, when comparing student responses obtained from successive academic years, students from the 2023–24 academic year noted a statistically significant decrease in satisfaction with voiceovers compared with student feedback obtained from the 2021–22 academic year (P = 0.039) (Fig. 2B). We also stratified student responses by year of matriculation, and a similar trend was obtained, because students who matriculated in 2023 expressed a statistically significant decrease in satisfaction with voiceovers compared with students who matriculated in 2020 (P = 0.0079) (Fig. 2C). Similarly, students who matriculated in 2023 expressed a decreased preference for voiceovers combined with review games versus traditional lectures when compared with students who matriculated in 2019 (P = 0.047) (Fig. 2C). Participants were also surveyed about their views regarding voiceover and review game logistics (Fig. 3). Students expressed a strong preference for formative quizzes contained within the voiceovers (6.07; n = 55) and preferred short voiceover lectures that only focus on a single high-yield topic (4.71; n = 63). Students preferred being provided a “script” of the lectures (5.13; n = 40) and few indicated having a script of the voiceover would dissuade them from watching the voiceover (1.72; n = 50). Students voiced their favor for faculty to use alternative preassignments, such as short reading assignments, YouTube videos, or third-party resources as preassigned material before review games if faculty are unfamiliar with the technology required to make voiceovers (5.47; n = 59). Student survey responses regarding preferences for logistics of voiceover lectures and review games. From January 2021 to August 2023, students were surveyed about their preferences for logistics of voiceover lectures and review games. Responses were recorded on a 7-point Likert scale and compiled across all years. Regarding the implementation of review games, participants felt that a brief transitionary question-and-answer review of the voiceover prior to the review game helped them understand the material (5.80; n = 64). The brief reviews were conducted through Socratic exchange, in which the professor would ask thought-provoking questions to promote discussion among the students. This review was similar to an interactive “chalk talk,” except that drawings were conducted on a doc cam instead of a chalkboard. Students expressed a slight preference for not being able to use notes during the game (4.87; n = 82), because enabling students to use notes or PPT files might tempt them to look up the answers rather than to discuss the question with their colleagues. They felt slightly incentivized to attend more review games by the potential to earn small prizes for the winning team (4.41; n = 90) and preferred a variety of review games to be used (4.98; n = 58). Students did not strongly support awarding extra credit to the winning teams of review games (2.97; n = 59), because the potential for extra credit being awarded to the winning teams might cause students to be preoccupied with their grades rather than learning the material (5.75; n = 59). Review games from PointSolutions offer student teams the opportunity to wager points for a question, similar to the traditional televised Jeopardy! game. Students overwhelmingly thought this option helped keep them engaged, especially for students not listed on the leaderboard periodically shown throughout the review game (5.68; n = 56), because wager questions provide a better opportunity for poorly performing teams to make it back onto the leaderboard prior to the conclusion of the game. Students did not have a strong preference for the ability to use personal phones versus clickers in the review game (3.44; n = 55), yet they strongly objected to the idea of preassigned teams in review games (1.89; n = 86). Faculty feedback on the use of voiceovers combined with review games Nine faculty members representing both the medical school and accompanying undergraduate campus who witnessed the pedagogy were also surveyed regarding their opinions (Fig. 4). Faculty voiced support that voiceovers combined with review games was more beneficial than traditional lectures (5.63 out of 7; n = 9 for all faculty responses). Few faculty thought students regularly used the textbook to complete the recommended readings (2.44), and faculty were in strong agreement that students who did not attend class missed out on critical information important for becoming good physicians (6.00). Faculty reported they would be more willing to make voiceovers (6.22) and incorporate review games into their teaching (5.22) if they had access to training and additional information technology (IT) support (6.22). Faculty felt students spent too much time using third-party resources and an insufficient amount of time studying materials provided in class (5.25), yet they also empathized with the increased amount of knowledge that medical students are required to master currently in comparison with when faculty members were medical students (5.38). Faculty agreed that incorporating active-learning pedagogies would lead to better educational outcomes compared with traditional didactic lectures (6.22). Additional faculty feedback obtained is listed in Table S2. Faculty feedback on pedagogy methodologies. Medical school faculty (n = 9) were invited to observe a review game lecture and were surveyed for their feedback and impressions. Responses were recorded on a 7-point Likert scale. Student feedback on learning preferences Students were surveyed regarding their preferences in learning modalities. Results were stratified by participants’ matriculation year (Fig. 5). Most students surveyed expressed a preference for visual learning, including 100% of participants in the 2021 and 2022 matriculating classes, and 75.6% in the 2023 matriculating class. The next most preferred learning modality among participants was auditory in the 2021 matriculating class (85.7%), and reading and writing in the 2022 and 2023 matriculating classes (75.0% and 65.9%, respectively). Simple linear regression analysis indicated a statistically significant decreasing trend in preferences for visual and auditory learning during the study. Matriculation year explained 7.8% of the variation in preference for visual learning and 9.3% of the variation in preference for auditory learning (P = 0.030 and 0.018, respectively), indicating that learning preferences for students participating in the study changed over time throughout the duration of this study. Student learning preferences. From January 2021 to August 2023, students were surveyed about their preferences for various methods of learning. Results were stratified by matriculation year including 2021, 2022, and 2023 (n = 7, 12, and 41 respectively). Simple linear regression analysis was used to assess trends across matriculation years for visual (R2 = 0.078, P = 0.030), auditory (R2 = 0.093, P = 0.018), reading and writing (R2 = 0.0092, P = 0.47), and kinesthetic (R2 = 0.0012, P = 0.80) learning. Discussion In this study, we evaluated whether a flipped classroom model combining voiceover lectures and review games could improve medical student learner engagement and satisfaction while achieving learning outcomes comparable to those of traditional immunology didactic lectures. The combined use of voiceovers and review games to teach immunology did not differ significantly compared with a traditional lecture as determined by formative quizzes; however, students and faculty observers both expressed enthusiasm for the active-learning pedagogy, and valuable feedback regarding the use of review games and voiceovers was obtained. Prerecorded lectures provide students the flexibility to watch the videos on their own schedule, and the ability to deliver more information through technology outside the classroom permits more time for active learning in the classroom.35 Students were supportive of the use of voiceovers in this study (Fig. 2A) and approximately 75% of students in a separate study preferred watching video lectures compared with attending traditional lectures.41 The students in our study voiced a strong preference for shorter voiceovers, similar to what was observed in previous studies,4,41,42 in addition to formative quizzes embedded within the voiceovers (Fig. 3). The students in this study clearly indicated much stronger enthusiasm for review games in comparison with voiceovers, yet flipping the classrooms through voiceovers, or an alternative method of delivery such as reader chapters or handouts, enables much more class time to be devoted to review games rather than didactic lectures. It should be noted that both faculty (Fig. 4) and students in our study (Fig. 2A) agreed that students rarely use textbooks to complete preassigned readings. Likewise, a previous study found that faculty significantly overestimated how much students use textbooks.37 Therefore, if preassigned readings are used to increase the amount of time available for in-class review games, student compliance for preparatory work would be best achieved by using either short handouts or very brief assigned textbook readings. Lastly, the effectiveness of voiceovers could be improved by combining them with online assessments done prior to class, because this would help faculty ascertain student understanding of varying topics prior to beginning in-person flipped classroom activities.31 Students voiced strong support for review games and preferred the use of a variety of review games (Fig. 2A), which supports our use of 2 different review games (ie Kahoot and Jeopardy for first year of study, Kahoot and PointSolutions technology for last 2 years of study). Students in our study strongly opposed preassigned teams for the review games (Fig. 3); the ability for students to choose their teams, and choose whether to compete alone or to only passively observe the game, increased the perceived autonomy by students, and thus helped drive their intrinsic motivation for learning.16 Students voiced their clear preference for using review games (Fig. 2A), yet it is important that review games be efficiently delivered and that gamification should never be used to salvage poor instruction.16 An additional mechanism to implement gamification is in the preassigned work for flipped classrooms, because gamification has been used in combination with online learning management systems,33,43 and these online review games could be conducted after students watched instructional voiceovers, providing increased incentive for students to show up to the in-person flipped classrooms adequately prepared. Perceptions of faculty who witnessed the teaching pedagogy Faculty surveyed as part of this study strongly believed that active-learning pedagogies would lead to better educational outcomes in comparison with traditional lectures (Fig. 4). Faculty who witnessed the pedagogy thought it was more beneficial than traditional lectures, and faculty expressed an increased willingness to make voiceovers and review games if they received training and IT support (Fig. 4). The time needed for training instructors, in addition to the need for resources and IT support, can be a major barrier to incorporation of more flipped classrooms and active learning,44 as can be traditionally designed lecture halls that resemble theaters and are not always conducive to flipped classrooms. Alignment of promotion and tenure policies with pedagogical innovation may incentivize faculty engagement to incorporate more flipped classrooms.42 A possible solution to lessen the impact on faculty would be to assign material from outside sources for students to learn before in-class review games, which students in our study supported (Fig. 3). In addition to YouTube videos that offer high-quality short videos worthy of introducing science topics to medical students, there has been a rapid expansion of massive open online courses such as Khan Academy to provide additional resources for faculty to provide preassigned learning material for flipped classrooms. Although faculty surveyed in this study thought students spend too much time studying outside third-party resources (Fig. 4), there exists an opportunity to use videos from third-party resources as a preassignment work for a flipped classroom that subsequently provides more in-classroom instruction by the faculty member. For example, some medical schools have provided their students a subscription to Boards and Beyond,39 and these short videos could be used as a preclass assignment to supplement faculty lectures yet also enable more active learning in the classroom, including review games. Changing student preferences during the longitudinal study Interestingly, when student survey responses on the pedagogy were stratified by either year of survey administration (Fig. 2B) or year of matriculation (Fig. 2C), a statistically significant decline was observed in more recent classes’ preference for voiceovers. These differences are intriguing because students from more recent matriculating classes experienced much of their undergraduate education through virtual classrooms during the COVID-19 pandemic, in contrast to previous cohorts. It is well documented that the pandemic severely affected higher education, characterized by reduced opportunities for collaboration, decreased access to academic support, fewer hands-on experiences, and decreased sense of connectiveness of students.45–47 Effective online education requires careful planning of course content and preparation for productive digital interactions,48 yet the rushed conversion to online formats during the pandemic may have caused some students to prefer the “normalcy” of traditional passive lectures in lecture halls.47 Furthermore, it is conceivable that a portion of students who spent a significant portion of their undergraduate education in a virtual environment might prefer the social connectiveness experienced by learning in the classroom versus watching a voiceover by themselves. We also noted a change in the self-reported learning preferences of students throughout the duration of this study. In the most recent matriculating class, there was a significant decline in participants who self-identified as visual learners (Fig. 5). Furthermore, across matriculating classes, there was a statistically significant trend of decreasing popularity of auditory learning. It is possible that the decreasing popularity of visual and auditory learning among the most recent matriculating class could be attributed to the effects of remote learning during the COVID-19 pandemic, because some students may not have enjoyed the virtual lectures as much as the in-person learning activities they had received before the pandemic. Conversely, we noted a trend toward increased preference for reading and writing during this study, and it is possible that some students became more accustomed to reading textbooks during the pandemic to attain information for their undergraduate courses. Although students undoubtedly have preferred methods of learning, the presence of evidence to support the VARK model of distinct student learning styles has been questioned, and it is also not clear that teaching in a manner distinct from a student’s preferred way of learning always leads to a disadvantage in learning.49 It is perhaps more important that the instructional style matches the nature of the content,49 for example, visual or kinesthetic teaching for anatomy, and the use of either reading and writing or auditory for teaching how to take a patient’s personal history. We were intrigued by the possibility that our observed differences in self-reported learning preferences by students (Figs. 2B and C, and 5) could have been influenced by the COVID-19 pandemic, and we next sought to compare our survey results regarding medical student learning preferences with national data reported by the American Association of Medical Colleges (AAMC) All Schools Summary Report. National surveys indicate that in-person student attendance for pre-clerkship courses and lectures was consistently decreasing prior to the onset of the pandemic, because the percentage of M2s who self-reported attending in-person pre-clerkship lectures most of the time decreased sequentially each year from 39.4% in 2014 (n = 9,920) to 25.3% in 2020 (n = 12,816). Conversely, the percentage of M2s reporting attending in-person courses or lectures almost never increased sequentially each year from 16.8% in 2014 (n = 9,920) to 37.0% in 2020 (n = 12,816).2 However, after the onset of the COVID-19 pandemic, the trends reversed: the percentage of M2s reporting attending in-person lectures most of the time increased from 25.3% in 2020 each year thereafter to 36.9% (n = 10,551) in 2024, whereas the percentage of M2s reporting attending almost never decreased sequentially each year from 37.0% in 2020 to 19.6% in 2024 (n = 10,551).2 The successive decline of in-person class attendance prior to 2020 was noted in selected findings of annual AAMC reports, as was the subsequent increase of in-person class attendance after 2020,2 yet we are not aware of any other reports that correlated this reversal of trends with the timing of the COVID-19 pandemic. Study limitations This study has several limitations. The self-selection of participants may have introduced a bias, because those who opted to participate might have different learning preferences and motivations compared with the general student population. This was a single-institution study, thus demographic and institutional characteristics of the participant population may differ from those of medical students in other settings, thus potentially affecting the generalizability of the findings. This focus of the study was on evaluating the short-term performance of students, with postintervention quizzes administered shortly after exposure to the material. Although this provides insights into immediate learning outcomes, it does not capture the long-term retention and understanding of the material. Additionally, the impact of students knowing they will be taking a formative quiz immediately after a learning session, whether a traditional lecture or the combination of voiceover plus review game, could be expected to influence the attention of students during the learning activities to which they were assigned. Lastly, the participation of students in groups during a review game may provide additional benefits relating to communication, professionalism, and other soft skills that were not assessed in this study. Conclusion Our study found that post-intervention quiz performance of students exposed to the combined use of voiceovers and review games did not differ significantly from students exposed to traditional lectures, and that both students and faculty expressed positive attitudes toward this flipped classroom pedagogy and collectively preferred it over traditional lectures. Although some active learning and flipped classroom studies have demonstrated improved test scores,7,9 other reports have demonstrated a preference for flipped classroom modalities over traditional lectures by both students and faculty despite no observed improvement in test scores or exams.11 It is important to note that some of the added benefits of interactive discussions with fellow students and faculty provided by flipped classroom modalities may be demonstrated in interpersonal skills versus multiple-choice exams.50 There has been a growing effort to replace traditional lectures with active learning and flipped classrooms to promote deeper understanding and retention of material.6,35,42 Yet student enthusiasm for flipped classrooms is not universal7; some students have expressed a desire to learn from the instructor who is an expert rather than from a colleague during a flipped classroom activity.51 Flipping classrooms does not need to be an “all or nothing” approach.52 A previous survey of medical students revealed that most students desired a mixture of lectures and flipped classrooms,53 and we interspersed our flipped learning sessions with traditional lectures throughout this study. It is important to note that student preferences for learning are likely to change over time, and although the reasons for the statistically significant changes in learning preferences throughout the duration of this study may not be ascertainable, student preferences for learning have been dramatically shifting over the years since the advent of the internet and household computers. Millennials were raised in an entertainment-focused multimedia environment and, as a result, may have shorter attention spans and have lower tolerance for boredom, crave instant gratification, and want learning to be connected to technology more than previous generations did.53,54 These changes in student preferences are only likely to continue with Generation Z and Generation Alpha. Their familiarity with technology may help explain the attraction of current medical students for third-party resources37 that can adversely affect attendance, even when classrooms are flipped to incorporate active learning,55 and both students and faculty have previously recognized the negative impact of poor student attendance on faculty enthusiasm for teaching.56 Providing short, “bite-sized” voiceovers as preassignment material, followed by conducting engaging and informative in-class review games after a brief review of the voiceover, can be 1 way to engage future generations and address these emerging trends. Supplementary Material