Understanding how learners engage with learning technologies, and its relation with their learning, is crucial to the design of effective learning interventions. Assessing the learners’ state however, is non-trivial. Research suggests that performance is not always a good indicator of learning, especially with open-ended constructivist activities. In this paper, we describe a combined multi-modal learning analytics and interaction analysis method which uses video, audio and log data to identify multi-modal collaborative learning behavioral profiles of 32 dyads as they work on an open-ended task around interactive tabletops with a robot mediator. These profiles, that we name Expressive Explorers, Calm Tinkerers, and Silent Wanderers, confirm previous findings that in a collaborative setting, the amount of speech interaction and the overlap of speech between a pair of learners are highly discriminating behaviors between learning and non-learning pairs, signifying that overlapping speech while turn-taking can indicate engagement that is conducive to learning. However, additionally considering learner affect and actions during the task helps us identify that there exist multiple behavioural profiles exhibited even among those who learn. Specifically, we discover that those who learn vary in their behaviors along the two dimensions of problem solving strategy (actions) and emotional expressivity (affect), suggesting that there is a relation between problem solving strategy and emotional behaviour; one strategy leads to more frustration compared to another. These findings have implications for the design of real-time learning interventions that support productive collaborative learning in open-ended tasks.