<p>Currently, practices in science education are hotly contested with calls for inquiry approaches to be replaced by ‘explicit teaching’ or ‘direct instruction’ and ‘knowledge rich’ curricula. This move is associated with an increasing emphasis on accountability represented by comprehensive testing regimes, critique of teacher education expertise, and questioning of teachers’ capabilities. Justification for explicit teaching is often framed around the development of a ‘science of learning’ based largely on cognitive load theory and controlled trials purporting to represent compelling evidence for explicit instruction, explained in terms of a model of working memory. We argue that this narrow advocacy reflects an unfortunate focus on one branch of science representing one aspect of learning. We draw on instances of learning in science to demonstrate how ‘explicit teaching’ can be differently framed to support richer views of learning that recognise the need to bridge between students’ everyday and disciplinary languages and experience. We argue that student learning of foundational knowledge and practice needs to go beyond careful teacher framing of abstracted content. We suggest that education, as a social science, is better served by research drawing on a range of methodologies and epistemologies that acknowledge the complexity of classroom practice, the multiple purposes of science education, and varied forms of evidence grounded in both scientific and sociocultural theories and methods.</p>