Abstract
Objective: As user-centered information systems, digital libraries play a decisive role in shaping users' search and browsing behavior. Browsing behavior, as an essential complement to information seeking, not only offers an alternative path to accessing information but also serves as a key mechanism for serendipitous discovery. Accordingly, this study identifies various dimensions of browsing behavior within the MedlinePlus digital library to extend and validate Chang's browsing theory in digital information environments.

Methods: This applied study employed a mixed-methods approach with a sequential exploratory design. Qualitative data were first collected and analyzed to develop an initial model, which was subsequently validated using quantitative data (Delphi method). The study population consisted of 60 master's and doctoral students in medical sciences from four Iranian universities of medical sciences (Isfahan, Tehran, Shiraz, and Mashhad), selected through a two-stage sampling approach (initial cluster sampling of universities followed by random selection of 15 participants per university). Various dimensions of users' browsing behavior, based on Chang's browsing theory, were gathered and analyzed using three simultaneous methods: direct observation and screen recording, think-aloud protocols, and semi-structured interviews. An initial model of information browsing in the digital library was developed, designed as a structured questionnaire, and sent to Delphi panel members over three consecutive rounds. The process continued until sufficient consensus was achieved (mean score ≥ 4 and Kendall's coefficient ≥ 0.7) for all model components, leading to the approval of the final model of information browsing in the digital library.

Findings: Five dimensions of user information browsing were identified for the MedlinePlus website of the U.S. National Library of Medicine. These dimensions integrated the four dimensions of Chang's browsing theory with a "digital" dimension derived from the empirical findings of this study. The findings indicated that the behavioral dimension of browsing in digital environments relies heavily on physical movements (vertical and horizontal scrolling). Additionally, new motivations arising from the dynamics and specific capabilities of the digital platform were identified. Approximately 87% of users entered the browsing process with relatively specific cognitive intentions. This finding, along with the high frequency of accidental browsing, introduces the concept of purposeful discovery.

Conclusion: This study demonstrated that information browsing in digital libraries is a far more complex and multidimensional phenomenon than can be explained solely by traditional models. The digital platform is not merely a context for passive information-seeking behavior but also an active agent shaping browsing behavior.