Recently, students' interest in science has been conceptualized as a network model: the science interest network model (SINM) in which affective, behavioural and cognitive components interact together; building on science interest being a dynamic relational construct. In the current study, we combine the Australian Programme for International Student Assessment (PISA) 2006 and Longitudinal Surveys of Australian Youth 2008 data (N = 4758) to investigate relationships between the network characteristics of Year 10 students with their decision to enrol in a science course in Year 12. Specifically, we identified indicators central to the SINM and tested whether they predicted chemistry, physics and biology course enrolment. Students' intentions to pursue a science-related study or career (future intentions in science) and their enjoyment of science (science enjoyment) were the most central indicators for all three science courses. Centrality was strongly related to course enrolment (r = 0.36-0.74), lending support to the validity of network theory in the context of science interest, as central indicators may play an influential role within the network.