Paper-based versus digital-based learning among undergraduate medical, nursing and pharmaceutical students in Japan: a cross-sectional study

Discussion

We examined the format preference for academic performance and eyestrain between DBL and PBL among undergraduate medical, nursing and pharmaceutical students during the COVID-19 pandemic. These students were born in the digital era. We demonstrated that PBL was better than DBL regarding memory retention and absorption for most students, and the majority of students responded that they experienced more eyestrain during DBL. Moreover, although our study showed a positive association between daily time spent using DBL and students’ digital preference only in the comprehension domain, the association was not seen in memory retention, absorption and eyestrain. This means that even those students who used DBL for a long duration did not evaluate DBL favourably (table 2). To the best of our knowledge, this study is the first to demonstrate an association between the duration of DBL use and the digital preference for academic performance among university students during the COVID-19 pandemic.

In psychology, there are three memory processes: encoding information (learning it by perceiving it and relating it to previous knowledge), storing it (maintaining it over time, memory retention) and then retrieving it (accessing the information when needed).25 Comprehension in our study corresponded to the encoding information phase, and students’ preferences for it were evenly distributed among PBL, equivalent and DBL (figure 1). Compared with the memory retention and absorption results, not a small number (one-third) of students reported that DBL was better. This may be attributed to digital textbooks offering a wide range of text and graphical information, containing hyperlinks to specific content and allowing individualised learning. For example, manipulating a three-dimensional figure in digital text in an anatomy class may help students understand the complex structure of the human body. In addition, students who used DBL for a longer duration significantly evaluated it better (table 2). A plausible explanation is that students became proficient at using useful functions such as writing and underlining on PDFs or browsing text swiftly, leading to better understanding. However, caution is warranted when interpreting the results of students’ evaluation of digital learning. Several studies have reported greater overconfidence, meaning a wider difference between the predictive and actual academic performance scores in DBL than in PBL.14 15 Overconfidence was considered to be caused by mind wandering, distraction and poorer judgement, which frequently occurs in DBL owing to less focus in reading the text.12 26

Memory retention, the second memory process25 and absorption (ie, concentration) are crucial components of the learning process. Without absorption, deeply memorising complex and time-consuming information is impossible. In our study, contrary to our estimation in this digitally transformed society, most students preferred PBL for memory retention and absorption (figures 2–3). This finding aligned with previous studies showing the phenomenon of ‘screen inferiority’.11–14 27 In a previous study of 10 293 tertiary students worldwide, Mizurachi et al reported that 72.3% agreed that information retention was more enhanced when reading printed pages.24 Several plausible factors may explain the superiority of PBL over DBL. The paper’s constant (unchanged) tactile substrate with physical sizes and spatial locations can serve as a deepening memory cue, and note-taking on paper may enable students to unconsciously summarise and reframe information in their own words.13 28 Another explanation is that digital media such as e-mail or online communication may only stimulate superficial reading strategies, including scanning, skimming and browsing, at the expense of a more time-consuming process that allows for deeper comprehension.19 29 Umejima et al performed a biophysical study using functional magnetic resonance imaging among university students in Japan. They observed brain activation in the hippocampus and precuneus, which are related to memory function during the deeper encoding and retrieval of specific information. They found that the activation in these areas was stronger in groups using a paper notebook than in those using digital devices,30 indicating that using paper notebooks promotes active brain function. Digital devices can also encode information equally to paper; however, it is inferior to paper in storing information.

Eyestrain is one of the symptoms of computer vision syndrome, along with other symptoms such as irritation, burning sensation, dry eyes and blurred vision.31 32 These symptoms occur more frequently with longer use of visual displays, viewing from a closer distance to the screen and glaring background luminance.32 33 A previous study reported that the symptoms of eye discomfort following computer use were significantly worse than those after hard-copy viewing.34 Consistently, the majority of students (84.6%, in online supplemental figure 1) experienced more eyestrain during DBL. These results can renew our knowledge: digital-native students born or raised in the digital technology age were as susceptible to eyestrain during DBL as the older generation. Government reports showed a drastic increase in internet use in Japan after the COVID-19 pandemic. The average time of internet use on a weekday among people in their 20s was 177 min in 2019 and it increased to 275 min in 2021.35 Therefore, our study emphasised the significance of raising public awareness about the possible occurrence of eyestrain caused by the use of digital devices among younger generations.

Our study reported no association between daily time spent using DBL at home and their digital preference for memory retention, absorption and eyestrain (table 2). Students who were used to DBL were expected to be more accepting of DBL; however, the results of students’ preferences seemed to contradict this expectation. Our data demonstrated that although students experienced screen inferiority in academic performance and eyestrain, they studied digitally because most academic texts provided by the university were only available digitally. To the best of our knowledge, no other study has assessed the association between students’ daily time spent using DBL and their preference after the drastic transformation of the learning environment due to the COVID-19 pandemic. Our findings show the possibility that students’ academic performance will not increase in highly digitalised study environments. Therefore, further prospective studies are needed to clarify the association between DBL and academic performance among students in the future.

Among the characteristics, different trends were observed in sex and students’ majors: medical, nursing and pharmaceutical courses. In our study, women were more likely to prefer PBL than men (figures 2 and 3). This finding aligned with previous studies,3 4 showing that more women were dissatisfied with online-based education. Although the difference was small in the students’ majors, medical students preferred DBL for academic performance compared with other students (figures 1–3). This may be because medical students need to earn more credits (subjects) and carry heavier and more printed textbooks every day. Therefore, more medical students may prefer digital text for easy accessibility and portability.12 24

Our study has some limitations. First, it is based on students’ evaluations (preference), not empirical or objective data. However, in a study using a computer-based achievement test conducted among medical students in Japan, those who preferred face-to-face classes had significantly higher scores than those who preferred online classes.3 Meta-analyses of empirical studies also reported inferior results with on-screen learning.14 15 Therefore, our subjective study is not largely different from studies using objective data. Second, this study was only conducted among medical, nursing and pharmaceutical students in one university; therefore, our results cannot be fully generalised to other university students. However, our study was conducted when all students were forced to adapt to DBL in Japan because of the COVID-19 pandemic. Therefore, our findings seem more generalisable than previous reports, ^14,15 which were conducted before the COVID-19 pandemic. Third, the response rate was less than 40%. Thus, our data might be subject to selection bias. However, we considered that students who preferred digital devices were more likely to answer our online questionnaires. Therefore, PBL preference rate may be higher among the general university student population than in our study. Finally, our study did not evaluate other potential factors, such as the online teaching skills of faculty members, socioeconomic status (affordability) and home environments. All faculty members prepared digital textbooks and videos in the chaotic situation of the COVID-19 pandemic36; thus, improvised online textbooks might not have been well organised compared with print textbooks. In the future, these potential factors should be included in studies to explore the factors influencing students’ academic performance.