It surveys how this term is defined in the published literature — what the sources agree on, where they diverge and what remains contested — and carries its own numbered bibliography. It does not describe what we run. Programme policy is on the core pages.
Our own use of the word is often narrower than general usage; the short definition and the boundary are in the glossary, under Tinkering. The review date below matters, because the literature moves.
A definitional review. Citations follow IEEE style; see References.
Abstract
Tinkering names an improvisational, exploratory mode of making in which the learner authors the goal rather than receiving it. It is associated above all with the Exploratorium’s Tinkering Studio and, in a parallel strand, with the MIT Media Lab. This article sets out the competing definitions, the constructionist lineage, the feature that separates tinkering from designed instruction and from adjacent modes of making, the design heuristic most often invoked to operationalise it, what the empirical literature reports, and the central unresolved objection — that producing an artefact is not evidence of having understood anything.
I. Definition
There is no single definition, and the disagreement is substantive rather than terminological.
The most-quoted characterisation treats tinkering as a mindset: Martinez and Stager describe making, tinkering and engineering as three “ways of knowing” within constructionist practice, where making is active construction working towards a planned product, tinkering is “a mindset involving a playful approach to solving problems through ‘direct experience, experimentation, and discovery’”, and engineering extracts “principles from direct experience” (as reported in [1]). On this reading tinkering is not an activity type at all but a disposition that can cut across intentional building and disciplinary work [1].
Much of the literature disagrees, representing “tinkering as a kind of making activity, not just a mindset” [1]. On that reading tinkering is a sub-category of making characterised by improvisational, creative problem-solving [1].
A third framing is operational and comes from the Exploratorium’s Tinkering Studio: “the process of becoming stuck and then ‘unstuck’ is at the heart of tinkering” (Petrich, Wilkinson and Bevan, as reported in [1] and [2]). Struggle is definitional here, not incidental — which is why the tradition treats a “rickety or lopsided” artefact that nonetheless resolves the problem as the compelling outcome [1].
II. Origins
Tinkering has two proximate homes and one deeper lineage.
The museum strand. The Exploratorium in San Francisco houses the Tinkering Studio, and the research literature on tinkering pedagogy is largely generated from it and from its partnership work — notably the Tinkering Afterschool Program run with the Boys & Girls Clubs of San Francisco, in which adult and youth educators join elementary-aged children in weekly workshops building stop-motion films, wooden pinball machines, wearable circuits and musical instruments. The programme was explicitly designed around equity and has been under sustained ethnographic research since its inception in 2012 [1].
The computational strand. In parallel, “the word ‘tinkering’ is most associated with the MIT Media Lab and, in particular, Resnick’s Lifelong Kindergarten group” [2]. Resnick and Rosenbaum contrast tinkering with more planful, engineering-oriented approaches and connect it to Lévi-Strauss’s notion of bricolage (as reported in [2]).
The lineage. Both strands sit inside constructionism. Papert extended Piaget’s constructivism with the claim that learners’ interactions are more robust when they are engaged in building public, shareable artefacts, and coined instructionism for the transmission model he opposed [5], [3]. Tinkering inherits the primacy of the object and adds the refusal of a pre-specified outcome. Beyond that the antecedents are the practical, playful modes of inquiry advanced by Dewey, Froebel, Montessori and Papert, so that making as educative practice is “largely a return to earlier learner-driven, inquiry-oriented pedagogies” rather than a new invention [1].
One caution attaches to the origin story: tinkering’s public face is culturally narrow. The resurgence has been described as a middle-class rediscovery, since “working-class folk have not had the luxury of discovering making and tinkering; they’ve been doing it all their lives to survive” (Rose, as quoted in [8]) — and “inventing, making, tinkering, designing, are indigenous practices, that is, practices that originate and occur naturally in particular ecologies” [8]. A definition that identifies tinkering with a specific set of tools and venues excludes most of its practice.
III. What distinguishes tinkering
A. The learner authors the goal
The discriminating feature is who sets the problem and the constraints. In Tinkering Studio activities, learners “author” the goals and constraints active in the activity, so that “the learners’ goals, interests, and sense of what is fun and cool are primary.” This is set explicitly against hands-on engineering or problem-based activities such as robotics competitions, “where the goals and constraints are externally determined” (Petrich et al., as reported in [2]). Working inside external constraints is an important skill, the same authors argue, but not an ideal starting point for engagement [2].
Embedded in this is a claim about cognition, not just motivation: goal development and problem finding are treated as cognitively rich activities in their own right, and as the source of ownership [1].
B. Tinkering against designed instruction
The contrast that matters pedagogically is with the scripted activity. Making “can sometimes be implemented as step-by-step, recipe-like, construction activities, running counter to the inquiry-based explorations that the maker movement espouses” (Resnick and Rosenbaum, as reported in [1]); it is precisely to name this failure mode that some researchers reserve tinkering for the improvisational case [1]. A related distinction separates making from assembling: “making seems to be cognitively and socially richer than assembling as it involves more active testing and fitting and less routine following of directions” (Espinoza, as reported in [1]).
The same worry recurs when making enters formal settings. Maker activities that become too prescriptive are compared to school science labs “designed for rigorous, disciplined, and scripted experiences in which students are guided towards the re-discovery of a unifying principle” (Blikstein and Krannich, as quoted in [9]); without deliberate pursuit of creativity as an outcome, makerspaces are warned to risk becoming “imagination ghettos” where “students are tasked with cookie cutter activities and trivial projects to complete” (Crichton and Carter, as quoted in [9]).
C. Not a synonym for unsupported discovery
Tinkering is frequently misread as the absence of teaching. The literature does not support that reading. Facilitators in the Exploratorium’s Tinkering Studio were found to make three kinds of move: sparking (orienting learners and establishing the safety needed to take risks), sustaining (offering tools and suggestions, re-engaging waning interest, revoicing ideas), and deepening (fostering reflection, challenging learners to complexify their work) (Gutwill and colleagues, as reported in [1]). Researchers of after-school tinkering settings go further and warn against the field’s drift from “teaching” to “facilitation”: “minimizing the role of the teacher can shortchange the many generative aspects of pedagogical talk and interaction” [1].
IV. The design heuristic: low floor, wide walls, high ceiling
The phrase most often used to specify what a tinkerable environment requires comes from Resnick, drawing on Papert, and names three independent properties of a tool or activity: low floor (easy to get started), high ceiling (opportunities to create sophisticated projects) and wide walls (supporting many different types of projects) — Resnick, as reported in [1].
The third is the one most often dropped. A low floor with a low ceiling produces a demonstration; a high ceiling behind a high floor excludes novices; narrow walls produce near-identical artefacts and so remove the authorship that §III identifies as definitional. The same source records the complementary obligation to support complexification over time, “enabling the act of creating to evolve with increasing levels of user sophistication,” and warns that digital fabrication machines “might generate aesthetically-pleasing products with little effort,” so educators “should shy away from quick demonstration projects and push students towards more complex endeavors” (Ackermann and colleagues, and Blikstein, as reported in [1]).
V. What the evidence reports
The literature is overwhelmingly qualitative — ethnographic, case-study, interview and descriptive — with a smaller number of surveys and pre-post measures [1]. Within those limits the findings cluster in three places.
Disposition and engagement. Regular participants showed “a dispositional shift,” repeatedly reporting that they were thinking about and doing things they had never previously considered; in a survey of 25 maker programmes, young people increased their “activation” towards STEM — a construct covering fascination with and valuing of STEM, competency belief, problem-solving and creative thinking (Sheridan and colleagues; Dorph and Cannady, as reported in [1]).
Failure and iteration. An emphasis on drafts reframed “mistakes” and “failed attempts” as moments in the process of creation, with students shifting their relationship to problems over time and coming to embrace iteration [1]. A dissenting note within the same tradition cautions against confusing iteration with failure, since an iterative design cycle “is about continuous improvement, keeping what works, and improving what doesn’t” — “this is learning, not failure” (Martinez and Stager, as reported in [1]).
Conceptual gains, where explicitly designed for. Studying simple computational circuits built with e-textiles, students significantly increased their understanding of circuitry concepts including current flow, circuit polarity and connectivity (Peppler, as reported in [1]). The mechanism claimed is transparency of materials, not tinkering as such.
VI. Limitations
No demonstrated long-term effect. The most direct statement in the review literature is that “the literature has yet to provide evidence for the influence of making/tinkering experiences on young people’s long-term trajectories”; most evidence is observed or self-reported during or soon after the experience, and focuses on those who continued participating rather than on those who did not [1].
No experimental base. A scoping review screening 180 publications on digital making with children and adolescents found most quantitative work used one-group posttest-only designs and reports that the authors “could not identify even a single study in the corpus with a fully-fledged experimental design” [7]. A parallel review concludes that experimental research demonstrating the efficacy of makerspaces against their stated learning goals “has not yet been conducted” [4]. The systematic review of school-based studies, working from 22 empirical studies of which 17 were wholly qualitative, states that “we cannot, at present, contribute to broader theory about how students learn in these spaces” [6].
Success is not understanding. The strongest theoretical objection is that tinkering to a working artefact does not establish conceptual learning. Drawing on Piaget’s distinction between succeeding and understanding, Valente and Blikstein observe that “through trial and error, a product can be successfully constructed without the learner necessarily being able to understand all the concepts involved in the process,” and conclude that “the learner’s having produced something is not enough to ensure that she has constructed knowledge” [3]. Their remedy is not more tinkering but mediation — an educator who elicits and formalises the concepts and strategies the learner actually used, using methods such as Piaget’s clinical interview [3]. The encyclopaedic review reaches a compatible practical conclusion: across differing pedagogical positions, “there is general agreement that provision of some scaffolding is more productive than random tinkering” [4].
Implicitness can reproduce inequity. The equity argument runs the same way and is uncomfortable for the purest reading of tinkering: engaging with the big ideas of science and engineering “without making those ideas transparent or explicit can reproduce existing inequities,” and in the same setting a number of children drew a sharp line between play and science — the more fun an activity was, the less scientific they judged it to be [1]. The recommendation is to make STEM concepts and practices explicit within the playful context, not to choose between them [1].
The tools are not the mechanism. Finally, a persistent misreading treats the equipment as the active ingredient. That conceptualisation is “seductive, but fatally flawed”; a tool-centric adoption “will neglect the critical elements of community and mindset,” and when the truncated version fails to produce change “it will be labeled a failed experiment” [2].
Note on sourcing
Several works central to this article — Martinez and Stager’s Invent to Learn (2013), Petrich, Wilkinson and Bevan’s and Resnick and Rosenbaum’s chapters in Design, Make, Play (2013), Gutwill and colleagues’ facilitation study (2014), and Resnick’s formulation of low floor / wide walls / high ceiling (2011) — were not read in full for this article. They are cited here as reported, with quoted wording, in the two reviews that were read in full [1], [2], and are marked as such at every point of use. They should be obtained and read before being leaned on further.
References
[1] S. Vossoughi and B. Bevan, “Making and Tinkering: A Review of the Literature,” commissioned paper, Committee on Successful Out-of-School STEM Learning, National Research Council, Washington, DC, USA, 2014.
[2] L. Martin, “The Promise of the Maker Movement for Education,” Journal of Pre-College Engineering Education Research (J-PEER), vol. 5, no. 1, art. 4, pp. 30–39, 2015, doi: 10.7771/2157-9288.1099.
[3] J. A. Valente and P. Blikstein, “Maker Education: Where Is the Knowledge Construction?” Constructivist Foundations, vol. 14, no. 3, pp. 252–262, 2019.
[4] G. Bull, J. Rutter, J. Garofalo, and M. Littman, “Maker Education: A Historical Perspective,” 2022, doi: 10.4324/9781138609877-REE82-1.
[5] P. Blikstein, “Maker Movement in Education: History and Prospects,” in Handbook of Technology Education, M. J. de Vries, Ed. Cham, Switzerland: Springer, 2018, pp. 419–437, doi: 10.1007/978-3-319-44687-5_33.
[6] R. Rouse and A. G. Rouse, “Taking the maker movement to school: A systematic review of preK-12 school-based makerspace research,” Educational Research Review, vol. 35, art. 100413, 2022, doi: 10.1016/j.edurev.2021.100413.
[7] K. Smolarczyk and S. Kröner, “Two Decades in the Making: A Scoping Review on Research on Digital Making and its Potential for Digital Empowerment in Non-formal Settings,” 2021, doi: 10.1080/15391523.2021.1974987.
[8] S. Vossoughi, P. K. Hooper, and M. Escudé, “Making Through the Lens of Culture and Power: Toward Transformative Visions for Educational Equity,” Harvard Educational Review, vol. 86, no. 2, pp. 206–232, 2016.
[9] R. Vuorikari, A. Ferrari, and Y. Punie, Makerspaces for Education and Training: Exploring Future Implications for Europe, EUR 29819 EN. Luxembourg: Publications Office of the European Union, 2019, doi: 10.2760/946996.