We followed two aims: testing the effects of databases' user interface interactivity (UII) on Scientific Behavior (SB) and exploring the flow experience (FE) as mediator between interface interactivity and SB, as well as self-efficacy (SE) role as an interferer. We used mixed method in this research. We made a SB questionnaire via a comparative literature study, FE and user UII through literature review. Faculty members and PhD students participated as scholars. Structural Equation Modeling was used for quantitative data analysis and interpretative approach to analyze qualitative data. The role of typological variables, such as gender, area of study, academic degree and English language skill level on SE, UII, FE and SB means are investigated. Finally, we tested the effects of databases' UII on SB and mediator role of FE and interfering role of SE. We found that the more self-efficient participants, the more they experience user interface interactivity and scientific behavior changes/adaptations. In other words, self-efficacy is an important characteristic to establish interactive search session and to upgrade scientific behavior in scholars. Also, we found those participants who experience more flow, have more chance to experience SB changes and adaptation in UII environment. So UII may have effect on researchers' SB. Results may be used in: 1) distance education or researcher training, since these areas are interested in developing, changing and adapting SB 2) Human-Computer Interaction field, because SB seems to be a new aspect of computer interaction effect on human 3) The Flow theory will be supported by this new implementation. We proposed a new theoretical framework for research