Due to the substantial emphasis on engineering in K-12 science education efforts in the USA, it is important for teachers to develop a deeper understanding of the nature of engineering (NOE) and the relationship between engineering and science. A deep understanding is characterized not only by the presence or absence of ideas but also by the interconnections among the ideas. There are few ways to measure the interconnections among the ideas associated with a deep understanding of NOE. The present proof-of-concept study addresses this need by providing a preliminary testing of card sort epistemic network analysis (cENA) and network models' potential to extend what can be learned about how preservice teachers conceptualize NOE. To test the potential of cENA, we used cENA with 52 preservice elementary teachers enrolled in a course emphasizing science and engineering practices. Our findings indicated a shift within the participant community's epistemic frame for NOE toward more expert-like views of NOE, including some emphasis on the application of science in engineering. The findings from this study suggest cENA may be a productive and fruitful methodology for capturing changes in students' understandings of NOE. cENA has the potential to guide science teacher educators' approaches in designing evidence-based learning experiences and interventions to improve learners' depth of understanding about NOE. Rigorous validation of cENA is now warranted.