Text-Based vs Audio-Based: What Differential Effects Do the Two Discussion Formats have on Scientific Reasoning?

Collaborative problem solving (CPS) is recognized as an effective method for developing students’ higher-order thinking. In this process, scientific reasoning plays a crucial role. Students work together, discussing ideas through oral and written, to engage in scientific reasoning and solve problems. By analyzing these discussions, we can gain insights into how students are thinking. In our study, we focused on 73 graduate students and analyzed their conversations during scientific reasoning tasks. We used a method called Ordered Network Analysis to visualize the directional connections between their cognitive processes. What we found was interesting: the text group tended to focus more on deep learning, while the audio group leaned towards surface learning. A statistically significant difference was observed between the two groups at the 0.05 alpha level in the first dimension, with an effect size of r  =  0.72. Thus, the text group generally had higher levels of thinking compared to the audio group. We also noticed a reversed sequence in how they processed information. The difference network further indicated significant differences in cognitive transfer between the two groups. Overall, these findings highlight the importance of considering different discussion formats in CPS.