Student interaction at a whole-class level is tied to learning outcomes, but encouraging students to engage with their peers at a whole-class level is difficult. While little research has sought to promote student engagement with one another at this level, one factor that has been shown to increase student interaction in pairs and small groups is student--student interdependency. Yet, to date, no studies have attempted to examine the impact of prior interdependency at this smaller scale on students' interactions as a whole-class. The current study tested this relationship in an undergraduate class learning science content (n = 19) through structuring interdependency (task and resource) during science learning. Students completed six science learning sessions each of which included a science learning activity followed by a whole-class concept mapping task. Students were audio and video recorded during the whole-class task, and then individual student behavior was coded in 10 s intervals during the first 10 min of each session. Generalized regression results showed that adding small group interdependency during learning predicted significantly more student science discussion (R[superscript 2] = 0.05) and supportive interactions (R[superscript 2] = 0.16), while predicting decreased expressions of science misunderstanding (R[superscript 2] = 0.07) during the whole-class task. Additionally, the combination of task and resource interdependency predicted a further decrease in student expressions of misunderstanding (R[superscript 2] = 0.07), but also increased student social distraction in comparison to task interdependency alone (R[superscript 2] = 0.28). Together these findings suggest that prior interdependency can be leveraged to increase student behavioral engagement at the whole-class level.