The role of talk in science education has long been established; an essential part of learning science is for students to engage in scientific discourse. Nonetheless, productive science discussion is still rare in the classroom. The rarity can be partly attributed to the complexity of dialogic science teaching: teachers have to respond to the dynamic flow of student talk in the moment, orchestrate different voices towards a collective understanding, support the emergence of new ideas, ensure disciplinary rigour of scientific practice, and attend to the complex social relationships in the class. The construct of contingent responsiveness (CR) describes teachers’ adaptive expertise in responding to student ideas in the moment to promote collective sense-making and classroom equity.

This study used a design-based research method (DBR) to co-design a technology-enhanced professional development (PD) programme with teachers of students aged 5-12 years old in Pakistan, incorporating mixed-reality simulation technology (i.e. Mursion) over four iterations. The effectiveness of the PD programme in supporting CR was evident in the significant shift in teachers’ response patterns before and after the PD, shown by epistemic network analysis both visually and statistically. Furthermore, this study shed light on how to support teachers in developing CR using systematic conjecture mapping, tracing the path from design features to mediating processes, and then to the outcome. The conjecture map was refined over four iterations, which improved the design and learning theory over time. It was found that 1) adopting dialogic framings, 2) developing fluency with talk moves, 3) deploying flexible attention, 4) engaging in knowledge-based reasoning, and 5) experiencing metaphoric resonance could lead to CR. These processes were enabled by a combination of design features, i.e., mixed-reality simulations, talk moves, guided collaborative inquiry, case studies, and collective reflection.

This study achieved the dual goals of DBR, producing usable knowledge in the form of an effective PD programme and building a preliminary learning theory of CR. Furthermore, unpacking the mechanisms of the PD allows the design to be adapted and tested in other educational and cultural contexts, thus enhancing its adaptability, sustainability, and potential for scalability.