Cognitive Apprenticeship

A definitional review. Citations follow IEEE style; see References.

Abstract

Cognitive apprenticeship is a model of instruction that adapts the methods of trade apprenticeship to subjects whose work is mental rather than manual. Its premise is a single observed asymmetry: in apprenticeship the processes of the activity are visible, whereas in schooling the processes of thinking are invisible to student and teacher alike. From that premise the authors derive six teaching methods and a four-dimensional framework for designing learning environments. This article sets out the definition, the lineage, the discriminating feature, the framework, what the empirical literature reports, and what it does not establish.

I. Definition

Cognitive apprenticeship is defined by its authors as a model of instruction that goes back to apprenticeship but incorporates elements of schooling, whose distinctive work is to make thinking visible [2].

The founding passage states the contrast the whole model rests on:

In apprenticeship, the processes of the activity are visible. In schooling, the processes of thinking are often invisible to both the students and the teacher. Cognitive apprenticeship is a model of instruction that works to make thinking visible. [2], quoted at p. 8 in [3]

The corollary is that in schooling the “practice” of problem solving, reading comprehension and writing “is not at all obvious — it is not necessarily observable to the student” [2], [3]. Where an apprentice watches a tradesman craft a cabinet and can copy what he sees, a student watching a teacher solve a problem cannot see the reasoning that produced the solution. Unless that reasoning is externalised, learners may acquire the answer as a trick learned by heart, and fail to transfer either the content or the strategy to a different context [3].

II. Origins

The model was first set out by Allan Collins, John Seely Brown and Susan E. Newman in a 1989 chapter in a volume honouring Robert Glaser [1]. That chapter could not be obtained for this article and is cited for attribution only; everything below rests on the authors’ own 1991 restatement in American Educator [2] and on a licensed reprint of its design framework [3].

The 1991 article identifies four critical aspects of traditional apprenticeship — modelling, scaffolding, fading and coaching — and argues that all four transfer to cognitive domains [2], [3]. In traditional apprenticeship “the expert shows the apprentice how to do a task, watches as the apprentice practices portions of the task, and then turns over more and more responsibility until the apprentice is proficient enough to accomplish the task independently” [2], quoted at p. 8 in [3]. Coaching is described as the “thread running through the entire apprenticeship experience”, with the expert diagnosing problems, giving feedback and overseeing the learning [2], quoted at p. 9 in [3].

Three translations are required to move that model into a classroom [2], [3]:

1. make the thinking visible — “by bringing these tacit processes into the open, students can observe, enact, and practice them with help from the teacher and from other students”;
2. situate the abstract — school curricula are “divorced from what students and most adults do in their lives”, so abstract tasks must be placed in contexts that make sense to students;
3. vary for transfer — teachers must “present a range of tasks, varying from systematic to diverse, and to encourage students to reflect on and articulate the elements that are common across tasks”, because school skills, unlike a carpenter’s, must generalise.

All three passages are quoted at p. 9 in [3].

The model’s own theoretical borrowings are usually given as the zone of proximal development, situatedness and scaffolding [5] — that is, cognitive apprenticeship inherits a Vygotskian account of development in which instruction targets the gap between what a learner can do alone and what they can do with help, and later applications name Vygotsky’s constructivism as the underlying basis [6]. The 1991 article itself does not present the model as a reading of Vygotsky.

III. What distinguishes it

Cognitive apprenticeship is not a synonym for hands-on, authentic or learning-by-doing. Its discriminating commitment is externalisation of process: the expert’s reasoning is spoken aloud while the task is performed, and the learner is then required to verbalise their own.

Two consequences follow that other activity-based pedagogies do not entail.

Modelling is a speech act, not a demonstration. In trade apprenticeship modelling means showing the behaviour; in cognitive apprenticeship it “is accompanied by experts explicitly explaining what they are thinking and why they are doing certain things while carrying out a task (i.e. thinking aloud)” [3].

The learner’s articulation is the assessment channel. Because students repeatedly articulate what they see and do, “their thinking processes become visible, not only for themselves, but also for the teacher”, which is how the teacher learns what a student can do and where guidance is still needed [3].

The model also names the social environment as constitutive rather than incidental: a class gives learners continuous access to others at varying degrees of expertise, teaches them that more than one answer is often possible, and “encourages them to view learning as an incrementally staged process, while providing them with concrete benchmarks for their own progress” [2], quoted at p. 9 in [3].

IV. The framework

The 1991 article presents a design framework of four dimensions — content, method, sequencing and sociology — reproduced verbatim in [3] from Table 24.1 attributed to [2] at p. 43.

A. Content — types of knowledge required for expertise

• Domain knowledge — subject-matter-specific concepts, facts and procedures
• Heuristic strategies — generally applicable techniques for accomplishing tasks
• Control strategies — general approaches for directing one’s solution process
• Learning strategies — knowledge about how to learn new concepts, facts and procedures

B. Method — ways to promote the development of expertise

• Modelling — teacher performs a task so students can observe
• Coaching — teacher observes and facilitates while students perform a task
• Scaffolding — teacher provides supports to help the student perform a task
• Articulation — teacher encourages students to verbalise their knowledge and thinking
• Reflection — teacher enables students to compare their performance with others
• Exploration — teacher invites students to pose and solve their own problems

Methods “should be designed to give students the opportunity to observe, engage in, and invent or discover expert strategies in context” [2], quoted at p. 43 in [3]. Fading — the progressive withdrawal of support as competence grows — sits alongside scaffolding rather than in the list, and is what converts support into independence [3].

C. Sequencing — keys to ordering learning activities

• Global before local skills — conceptualise the whole task before executing the parts
• Increasing complexity — meaningful tasks gradually increasing in difficulty
• Increasing diversity — practice in a variety of situations to emphasise broad application

D. Sociology — social characteristics of learning environments

• Situated learning — students learn in the context of working on realistic tasks
• Community of practice — communication about different ways to accomplish meaningful tasks
• Intrinsic motivation — students set personal goals to seek skills and solutions
• Cooperation — students work together to accomplish their goals

The sociology dimension is where cognitive apprenticeship and the situated-learning literature touch: situated learning and community of practice are named here as design elements of a teaching environment, whereas in that other literature they are analytic descriptions of how learning occurs in practice. The two are contemporaneous and mutually citing, but a design element and an explanatory claim are not the same thing.

V. Scope of adoption

The 1991 article demonstrated the model through three worked cases from the research of the period: reciprocal teaching in reading, procedural facilitation in writing, and heuristic and control strategies in mathematics [2], [3].

Adoption since has been heaviest in professional and graduate education. A qualitative review of cognitive apprenticeship in STEM graduate education searched seven databases over 1989–2018, screened 95 publications and included 17 empirical studies, clustered into student research development, learning environments, advising methods and programme development [4]. The model also appears in library and health-professions instruction, where its four “core constructs” are sometimes restated as situatedness, peripheral participation, guided participation and community of practice [5] — a restatement that keeps the sociology dimension and drops most of the rest.

Coaching in challenge-based engineering education has been analysed explicitly within the cognitive apprenticeship framework, with observations of coaching sessions and interviews finding that the most frequent practices were asking open-ended questions, giving feedback on design progress, and encouraging exploration of alternatives — and that experienced coaches pushed students to reflect on their learning process and become self-regulated learners, whereas novice coaches concentrated on technical design [7].

The strongest single controlled result available is a randomised trial in critical-care residency training: 115 residents were allocated to a two-month cognitive apprenticeship course (n = 56) or to conventional teaching (n = 59) with identical contact hours, and the intervention group scored significantly higher afterwards on four critical-thinking dimensions and six clinical-competence dimensions (P < 0.05) [6].

VI. Limitations

The authors do not claim general applicability. The 1991 article states that the model can be a useful tool at certain moments but “certainly does not suit all forms of instruction and learning” — reading a book or watching a documentary may be a better route to factual knowledge [3]. It is also explicitly not a recipe: cognitive apprenticeship “is not a model of teaching that gives teachers a packaged formula for instruction” [2].

Implementation is partial and systematically skewed. The STEM review found that across 17 studies, coaching and scaffolding accounted for 15.3% of all applied codes, while reflection, articulation and exploration together accounted for about 9.9% (3.5%, 3.5% and 2.9% respectively) [4]. Sequencing was the least discussed dimension of the four — appearing in 13 of 17 studies but only 5.3% of codes, and there largely “in vague language” [4]. Since articulation and reflection are precisely the mechanisms that make thinking visible, an implementation that keeps coaching and drops them retains the label and loses the premise.

The evidence base is thin and its reviews do not appraise quality. The STEM review states that it did not assess study rigour, that discipline-level differences were not accounted for, and that it excluded work using the framework’s elements without naming it — including work framed as communities of practice or situated learning [4]. It reaches a favourable conclusion about the framework’s usefulness, but that conclusion rests on 17 unappraised studies, and its own qualifier is that the result “does not occur by magic; instead, it requires deliberate action by faculty supervisors and students” [4]. Controlled evidence exists [6] but is single-site, small and confined to one clinical setting.

Citation of the founding text is unstable. The 1989 chapter is variously cited as 1987, 1989 and 1991, with differing titles and page ranges across the secondary literature — the health-sciences application cited here attributes it to 1987 [5], and one review attributes the four dimensions to a different 1989 paper by Brown, Collins and Duguid [4]. Anyone building on the framework should verify which text they mean; this article could not obtain the 1989 chapter and does not quote it.

References

[1] A. Collins, J. S. Brown, and S. E. Newman, “Cognitive apprenticeship: teaching the crafts of reading, writing, and mathematics,” in Knowing, Learning, and Instruction: Essays in Honor of Robert Glaser, L. B. Resnick, Ed. Hillsdale, NJ, USA: Lawrence Erlbaum Associates, 1989, pp. 453–494. Not obtained; cited for attribution only.

[2] A. Collins, J. S. Brown, and A. Holum, “Cognitive Apprenticeship: Making Thinking Visible,” American Educator, vol. 15, no. 3, pp. 6–11, 38–46, Winter 1991. [Online]. Available: https://www.aft.org/ae/winter1991/collins_brown_holum

[3] P. A. Kirschner and C. Hendrick, “‘Cognitive Apprenticeship’ Revisited: Shining a Light on the Processes of Thinking to Understand Learning,” American Educator, pp. 37–40, 50, Fall 2020. Adapted from How Learning Happens: Seminal Works in Educational Psychology and What They Mean in Practice. Abingdon, U.K.: Routledge, 2020, ch. 24. (Reproduces Table 24.1 and page-attributed quotations from [2].)

[4] L. M. Minshew, A. A. Olsen, and J. E. McLaughlin, “Cognitive Apprenticeship in STEM Graduate Education: A Qualitative Review of the Literature,” AERA Open, vol. 7, 2021, doi: 10.1177/23328584211052044.

[5] H. Jasmin, K. Hohmeier, and C. Spivey, “Implementation of the cognitive apprenticeship model for enhancement of advanced searching skills in a pharmacy academia rotation,” Journal of the Medical Library Association, vol. 110, no. 1, pp. 119–125, 2022, doi: 10.5195/jmla.2022.1108.

[6] Y. Zhang, X. Xia, J. Zeng, L. Zhang, and C. Guo, “Application of the Cognitive Apprenticeship Teaching Model in the Standardized Training of Critical Care Medicine Resident Physicians: A Randomized Clinical Trial,” Advances in Medical Education and Practice, vol. 16, 2025, doi: 10.2147/AMEP.S540031.

[7] S. M. Gómez Puente, K. Doulougeri, and M. Bruns, “Coaching Practices in Challenge-Based Learning: Characteristics in Students’ Projects,” 2022.