Allan Collins
American cognitive scientist (Allan M. Collins), long associated with the educational-research firm BBN (Bolt, Beranek and Newman) in Cambridge, Massachusetts, and later with the School of Education and Social Policy at Northwestern University, where he is Professor Emeritus of Education and Social Policy. Two programmes of work define the career: the spreading-activation model of semantic memory developed with Elizabeth Loftus in the early 1970s, which became the cognitive-science account of how meaning is retrieved from associative networks; and the cognitive apprenticeship programme developed with John Seely Brown and Susan Newman through the 1980s, which is the work that anchors his presence in this topic.
Stake§
Collins's stake in the 1989 cognitive-apprenticeship paper is the long-term one of a cognitive scientist trying to bring craft-pedagogy methods into the formal classroom. The paper was written for a Festschrift honouring Robert Glaser; it is a programmatic statement rather than an empirical study. The argument's reception has been largely friendly — cognitive apprenticeship sits inside most contemporary teacher-education curricula — but the structural implementation in K-12 schools has been limited by the same classroom-time constraints that limit mastery learning.
The 1989 paper Cognitive Apprenticeship: Teaching the Crafts of Reading, Writing, and Mathematics is Collins's principal contribution to the skill-formation literature. The paper's organising move is to notice that traditional craft apprenticeship — a novice working alongside a master, watching the work done, doing it under supervision, getting feedback — produces skilled performers reliably in domains where formal education does not. The question the paper asks is whether the methods that make apprenticeship work in carpentry, tailoring, or surgery can be adapted to teach the cognitive crafts of reading, writing, and mathematics in classrooms where the actual work is invisible — happening inside the practitioner's head.
The paper's answer is the six-step framework that became the operating definition of cognitive apprenticeship: modelling (the expert demonstrates the cognitive process out loud, so the novice can see what ordinarily happens silently); coaching (the novice attempts the work while the expert observes and intervenes); scaffolding (temporary support that fades as the novice gains capability); articulation (the novice is asked to put the reasoning into words); reflection (the novice compares their own performance to the expert's); exploration (the novice tries the work in new domains). The framework had implementation case studies in mathematics (Alan Schoenfeld's problem-solving instruction), writing (Scardamalia and Bereiter's procedural facilitation), and reading (Palincsar and Brown's reciprocal teaching).
Collins's later work extended into educational technology and the design of learning environments. Rethinking Education in the Age of Technology (Teachers College Press, 2009, with Richard Halverson) argued that the institutional form of twentieth-century schooling — age-grouped cohorts moving in lockstep through a fixed curriculum — was a poor fit for what the cognitive-science evidence said about how skill is built, and that the technological capacity for self-paced, mastery- oriented instruction would eventually erode the form. The book sits in conversation with Bloom's Two Sigma argument and with the ungrading literature; it has been less absorbed into mainstream education policy.