Cognitive Apprenticeship: Teaching the Crafts of Reading, Writing, and Mathematics

Allan Collins and John Seely Brown were both established figures in cognitive science and artificial intelligence when they wrote this, and both were working at research labs rather than universities. Collins was at Bolt Beranek and Newman, the Cambridge firm where much early work on intelligent tutoring systems was done; he had earlier worked on semantic memory and plausible reasoning, was a founding editor of the journal Cognitive Science and the first chair of the Cognitive Science Society, and later moved to a chair at Northwestern. Brown was at Xerox's Palo Alto Research Center, which he would go on to direct, and is associated with the situated-learning argument — the same year, with Collins and Paul Duguid, he co-wrote "Situated Cognition and the Culture of Learning". Susan E. Newman was the third author. The two senior names carry a track record on how knowledge is represented and how it is taught.

The piece is a chapter, not a journal article — pages 453 to 494 of Knowing, Learning, and Instruction: Essays in Honor of Robert Glaser, an edited volume published by Lawrence Erlbaum Associates in 1989. Lauren B. Resnick edited the collection as a festschrift for Glaser, the University of Pittsburgh learning researcher who founded its Learning Research and Development Center. The editorial filter of a festschrift chapter is lighter than that of a refereed journal: chapters in such volumes are commissioned, read and edited by the volume editor and her referees rather than passed through anonymous journal peer review, and they are written with more room for synthesis and argument than a data paper allows.

The chapter is a secondary, synthesising source — it advances a model rather than reporting an experiment. Its argument generalises traditional craft apprenticeship to cognitive skills that have no visible product to copy: where a tailor's apprentice can watch and imitate, a reader or a mathematician thinks invisibly, so the chapter's programme is to make that thinking visible and then teachable. It names a sequence of methods — modelling the expert performance aloud, coaching the learner through attempts, scaffolding that support and then fading it as competence grows — and illustrates them from existing instructional research in reading, writing and mathematics, including Palincsar and Brown's reciprocal teaching and Scardamalia and Bereiter's work on writing. Those illustrative programmes are cited and can be chased to their own papers. Within this set the chapter sits beside the other acquisition-side anchors as a third route to the same territory: where Bloom's 1984 "2 Sigma Problem" reaches the feedback loop through controlled comparison and Sweller's 1988 cognitive-load paper reaches it through working-memory theory, this chapter reaches a structured coach-and-fade sequence through the analogy to the trades.

The authors' stake is intellectual and reputational. No product was sold; the chapter is a contribution to a scholarly volume. Collins and Brown were senior researchers placing a framework into a high-visibility collection edited in honour of a leading figure in their field, and the term "cognitive apprenticeship" became a durable piece of instructional-design vocabulary — a reputational return of the ordinary kind for a synthesis that gives a field a usable name.