Sociocultural Theory: Vygotsky's Ideas, With Examples

Go from storyboard to an interactive course in minutes

Start Free

Picture a new lasher at a container terminal, two weeks into the job. In the classroom session she matched every gear name to its picture and passed the ten-question quiz. Now she is on a ship's deck with a twistlock that will not lock, and on the radio she says, "The thing on the corner of the box near the back won't go." The signaller cannot work out which box she means, the crane waits, and her buddy has to walk over and point. She knew the words on paper. She had never used them with the people she works with.

Lev Vygotsky's sociocultural theory explains that gap. I'm an instructional designer at Coursebox, where we make instructional design software that turns a brief into a course, so I'll mention our product where I tested it. In this guide I explain the theory from Vygotsky's own texts, separate it from the wider sociocultural perspective in psychology, follow the later work of Barbara Rogoff, Alexei Leont'ev, Yrjö Engeström and James Wertsch, and compare it with Piaget. Then I use it to plan onboarding for a lashing gang, with three real cases, a product test and two free templates.

What is sociocultural theory?

Sociocultural theory, developed by the Soviet psychologist Lev Vygotsky between about 1924 and 1934, holds that people's higher mental abilities, such as voluntary attention, memory and reasoning, start in shared activity with other people and are shaped by cultural tools, above all language.

View the full-size diagram

Vygotsky stated the central claim most directly in a text translated in a 1981 collection edited by James Wertsch. He called it "the general genetic law of cultural development": "Any function in the child's cultural development appears twice, or on two planes. First it appears on the social plane, and then on the psychological plane" (p. 163). Mind in Society, the 1978 English compilation most readers meet first, puts the same idea as "first, between people (interpsychological), and then inside the child (intrapsychological)" (p. 57). His examples were voluntary attention, logical memory and the forming of concepts.

Two more ideas carry the rest of the theory. People do not act on the world, or on their own minds, directly. They act through tools and signs, and language is the most important sign system of all. And what starts between people is not copied inside a person; it is rebuilt, which Vygotsky called internalisation. In Vygotsky and the Social Formation of Mind (1985), Wertsch summed up the core of the framework as three themes: a genetic, or developmental, method; the claim that higher mental processes have their origin in social processes; and the claim that they can be understood only through "the tools and signs that mediate them" (pp. 14 to 15). That list is Wertsch's reading of Vygotsky, not one Vygotsky wrote, but it is a fair map.

Wertsch later gave the broad aim of the approach in one sentence, in Voices of the Mind (1991): "The basic goal of a sociocultural approach to mind is to create an account of human mental processes that recognizes the essential relationship between these processes and their cultural, historical, and institutional settings" (p. 6).

The sociocultural perspective in psychology

In an introductory psychology course, "sociocultural perspective" usually means something wider than Vygotsky. The APA Dictionary of Psychology defines it as "any viewpoint or approach to health, mental health, history, politics, economics, or any other area of human experience that emphasizes the environmental factors of society, culture, and social interaction." In that sense it sits beside the biological, cognitive and behavioural perspectives. It explains behaviour and mental processes through the groups, cultures and relationships a person lives among. A psychologist taking this view of a client asks how the client's culture and background shape the way they describe and cope with their symptoms.

The dictionary's second sense is the developmental one: "the view that cognitive development is guided by adults interacting with children, with the cultural context determining to a large extent how, where, and when these interactions take place." It names Vygotsky as "A major pioneer of this perspective". So Vygotsky did not found the whole sociocultural perspective. His sociocultural theory is the account of learning and development within it, and it is the subject of this guide.

Three points set the scope:

• It is mainly about how children's minds develop. Vygotsky studied children's thinking, speech and play. Applying the theory to adults learning a job is a later extension, and a useful one, as long as you treat it as an extension.
• It is a theory, not a teaching method. Vygotsky left no design procedure. Any workflow built on it, including the one in this guide, is an interpretation.
• It changes what you look at. Instead of one learner alone with some content, it looks at a person acting with other people and with tools, in a setting that has its own history and rules.

Here are the main ideas at a glance, with the question each one asks of your training.

The key ideas of sociocultural theory, worked through one job

Here is the hypothetical example I'll use from here on. A container terminal at an Australian port works three mid-size container ships a week. Eight new lashers are in their first six weeks in a lashing gang. Four came from warehousing, two from construction labouring and two from hospitality. All have done the terminal induction, the port safety induction and their PPE fit. None has worked on a ship, so none of their earlier jobs shows they can read a bay plan or work under a crane.

The work is lashing. On deck, the gang removes lashing rods and turnbuckles before boxes are lifted off and fits them after loading, following the ship's lashing plan. On the quay apron, they remove and fit twistlocks as boxes are landed and check every lock is closed before a box goes up onto a stack. The gang has fixed roles: a leading hand who runs it and decides when a new lasher works without a buddy, experienced lashers who buddy one new starter each, and a signaller who is the only person who talks to the crane driver by radio.

The gang runs on a shared language, written for this example:

• The location code. Every container slot is bay, row and tier, read in that order, such as "bay 34, row 06, tier 84". Deck tiers start at 82.
• Four gear names. "Cone" means a twistlock, "SATL" a semi-automatic twistlock, "rod" a lashing rod and "bottle" a turnbuckle.
• Radio rules. Lashers never call the crane driver; they call the signaller on the gang channel. Anyone who sees danger calls "ALL STOP" and the signaller stops the crane. A fault is reported as location, then problem, then status: "Bay 34, row 06, tier 84, cone not locked, standing clear."
• One safety rule. Never stand under a suspended load, and wait for "clear" from the signaller before going back to a stack.

The gang's tools are the bay plan for the bay being worked, the ship's lashing plan and a pocket lashing card with the code, the four gear names with a picture of each, the fault-report order, the ALL STOP rule and a say-aloud check.

The old approach was a 90-minute classroom session with a slide of gear names and a 10-question multiple-choice quiz, then straight onto the quay. Everyone passed the quiz. On the radio, new lashers said "the thing on the corner" and "the box near the back", the signaller could not locate the fault, and twice in one month a new lasher walked back to a stack before "clear". The words were known on paper and missing in use.

The intended outcome is that by week 6 a new lasher can work one deck bay as part of the gang, use the code and gear names correctly on the radio, report a fault in the agreed order, call ALL STOP when needed and run the say-aloud check without prompting, with the leading hand's sign-off to work without a buddy. It is not a licence or a certificate of competence, and it gives nobody authority to talk to the crane driver.

The gang, terminology, schedule and radio rules are invented for this teaching example. They are not operating instructions. A terminal’s approved procedures determine equipment use, communication, supervision and readiness for independent work.

The example suits sociocultural theory because the job is held together by things the theory names. The words and the card are tools people think with on deck. Competence starts between people, at the buddy's side and on the radio, and moves inward. And the gang's rules and roles decide what a newcomer can learn and when.

1. Social origins: abilities start between people

Vygotsky's first claim is that higher abilities exist between people before they exist inside one person. "All the higher functions originate as actual relations between human individuals," he wrote (Mind in Society, p. 57). A new lasher's ability to report a fault starts as a joint act: she speaks, the buddy listens, corrects the order and confirms it with the signaller. Weeks later the same report comes out right with nobody correcting it. The skill was social first.

• Sample goal: By week 4, a new lasher makes fault reports on the gang channel in the order location, problem, status, with the buddy listening.
• How to assess it: The buddy listens to real reports on shift and notes the part that was missing; in the course, the learner writes each call exactly as it would be said.
• In the course: A task to read three bay plan locations aloud to the buddy each shift and log which one went wrong.

2. Mediation: people act through tools and signs

Vygotsky compared signs with ordinary tools. A tool "is externally oriented; it must lead to changes in objects". A sign "changes nothing in the object of a psychological operation. It is a means of internal activity aimed at mastering oneself" (Mind in Society, p. 55). A spanner tightens a turnbuckle. The word "bottle", the location code and the order of a fault report tighten nothing, but they organise what a lasher notices, says and does next.

In a 1930 talk he called such signs "psychological tools", and his list included "language; various systems for counting; mnemonic techniques; algebraic symbol systems; works of art; writing; schemes, diagrams, maps, and mechanical drawings; all sorts of conventional signs" (1981, p. 137). Calling a lashing card or a bay plan a psychological tool is my reading, but it follows his list closely. You will also see the phrase "tools of intellectual adaptation" given as Vygotsky's term; I did not find it in any of his texts I checked.

One line in Mind in Society changes how you treat job aids: "For many functions, the stage of external signs lasts forever, that is, it is their final stage of development" (p. 57). Some tools are not training wheels to remove. A lashing card that stays in the pocket is the tool working as intended.

• Sample goal: By week 4, a new lasher uses the location code and the four gear names correctly in every radio call.
• How to assess it: The buddy's record of calls made on shift, and written calls in the course compared with a model call.
• In the course: A words page that shows each word with the place it gets said out loud, followed by locations to write as radio calls, not a glossary to memorise.

Tools and rules are where training most often meets the rest of the workplace. A Brazilian hospital changed both, and the record of what happened is one of the clearest I found.

Case study: Fewer episiotomies at a maternity hospital in Brazil

This was a long record of one hospital's practice, analysed with activity theory after most of the change had happened, so the theory cannot take the credit. The Women's Health Reference Center of Ribeirão Preto, in São Paulo state, is a public, low-risk maternity hospital and a teaching site for the University of São Paulo. In 2010 episiotomy was routine, and many staff believed that cutting back would "compromise residents' and students' training".

• What they did: From 2012 the hospital ran an annual symposium, wrote new protocols, made every episiotomy and its reason a required entry in the record, and fed audited rates back to teams and, from 2014 to 2017, to each professional. Team training ran twice a year, alongside a birth plan and new birthing rooms. From 2017 a pilot "applied CL principles", the Change Laboratory method from activity theory, to analyse the hospital's model of care. Personal feedback stopped once rates had stayed below 3% for three consecutive semesters.
• Evidence and result: In Lemes and colleagues' 2026 paper in Ciência & Saúde Coletiva, audited records covering 27,519 vaginal births from 2010 to 2022 show episiotomy among first-time mothers falling from 564 (66.67%) in 2010 to 3 (0.73%) in 2022, and among all vaginal births from 754 (34.54%) to 8 (0.50%). Recorded severe tears rose, which the authors read as better detection.
• Limitation: One hospital, no comparison site, national rates also falling and many changes at once. The fall began around 2012, years before the Change Laboratory work, and the authors note that management support meant little resistance.
• Lesson: Change the tools and rules people work with, not only the training. A required record field, a written protocol and personal rate feedback made the new practice visible on every shift, and the feedback was withdrawn on purpose once the practice held.

3. Speech that plans action, and inner speech

Vygotsky watched children talk to themselves while they solved practical problems and concluded that the talk was doing work. Children "solve practical tasks with the help of their speech, as well as their eyes and hands" (Mind in Society, p. 26). At first speech follows action. Later it "comes to precede action", and then "speech guides, determines, and dominates the course of action; the planning function of speech comes into being" (p. 28).

He called this self-directed talk egocentric speech, a term he took from Piaget, and argued that it does not fade but turns inward. In Thinking and Speech he wrote, "Inner speech is speech for oneself. External speech is speech for others" (p. 258). Inner speech is shortened: it drops the subject and keeps the point (p. 267). Researchers now call the spoken version private speech. Vygotsky did not set ages or stages for it, so be wary of charts that attach his name to them.

In this fictional gang, the spoken check is practised with a buddy listening. Shortening self-talk illustrates internalisation, but safety-critical read-backs and confirmation must remain audible where the approved procedure requires them. Move from prompted to independent performance when observation shows readiness, not simply because week 5 has arrived. The example illustrates a theory; its shorthand is not a complete lashing or crane procedure.

• Sample goal: By week 6, a new lasher runs the check at every box without prompting.
• How to assess it: The buddy observes whether every required check and communication is completed correctly, and records the help needed. Quiet speech alone is not evidence of competence or of a mental process.
• In the course: The check’s example wording, the buddy’s role and the readiness conditions for reducing prompts. Retain every audible confirmation required by the site procedure.

4. Internalisation: rebuilt, not copied

"We call the internal reconstruction of an external operation internalization," Vygotsky wrote (Mind in Society, p. 56). His example is a baby's failed reach for an object. When the mother responds, the reach becomes a point: "Pointing becomes a gesture for others", and its meaning "is established by others" (p. 56). The movement did not change. Its meaning came from the people around the child, and only then did the child use it on purpose.

Internalisation transforms activity over time rather than copying it inside a person. For this adult workplace example, look for independent use of the tools and language with less prompting. That is an application of the theory, not proof that a particular six-week schedule or quieter speech demonstrates internalisation.

• Sample goal: By week 6, a new lasher reads a bay plan location and calls it on the radio without first checking the card, while still carrying the card.
• How to assess it: The leading hand watches one bay per shift in weeks 5 and 6 and records the calls and checks.
• In the course: A card page that says plainly the card is a tool to keep, not a test to pass.

5. The zone of proximal development

Vygotsky defined the zone of proximal development as "the distance between the actual developmental level as determined by independent problem solving and the level of potential development as determined through problem solving under adult guidance or in collaboration with more capable peers" (Mind in Society, p. 86). His example was two children who both scored a mental age of 8 alone, but reached the 12-year-old and 9-year-old levels with help.

• Sample goal: By week 5, a new lasher works a bay with the buddy nearby, and the buddy steps in less each shift.
• How to assess it: The buddy records what the lasher did alone and the lowest help that worked.
• In the course: A weekly log asking what the learner did, one call they did not understand and who they asked.

Scaffolding and the more knowledgeable other

Scaffolding is the temporary help a more skilled person gives so a learner can finish a task they cannot yet manage alone, withdrawn step by step as the learner takes over. The more knowledgeable other is whoever gives that help: a buddy, a supervisor or a peer who knows the task better. In the lashing gang, the buddy is the more capable peer, and the scaffold is the help the buddy gives on deck and takes away shift by shift.

Both terms are often credited to Vygotsky, but neither is his. "More knowledgeable other" does not appear in Mind in Society; Vygotsky wrote "more capable peers", and the longer phrase spread in education writing in the 1990s. Scaffolding comes from David Wood, Jerome Bruner and Gail Ross's 1976 paper on tutoring, which does not cite Vygotsky; he did not develop scaffolding as a teaching idea. Our guide to the zone of proximal development covers the zone in depth, including how to write help in steps and record the step each learner needed.

6. Culture and community: the gang's rules and roles

Vygotsky treated culture as the source of the tools and signs a person learns through; psychological tools, he wrote, are "social, not organic or individual" (1981, p. 137). Later writers, Rogoff and Engeström among them, added the rules and roles of a community. At work those are concrete: who may say what to whom, which mistakes get named out loud and what counts as a finished job. In the gang, only the signaller talks to the crane driver, only the leading hand decides when a new lasher works alone, and anyone may call ALL STOP. A new lasher learns those rules by taking part under them, and they shape what she can practise and when.

• Sample goal: From week 1, a new lasher calls the signaller, never the crane driver, and calls ALL STOP when she sees danger.
• How to assess it: The signaller and the buddy note any call made to the wrong person; in the course, the learner writes the call for a person walking under a lifted box.
• In the course: A "Before you start" section that names each role and what it means for the learner.

How the ideas fit together

Vygotsky's ideas are not a sequence of stages to run in order. Social origins, mediation and internalisation describe one process seen from three sides: an ability starts in activity with other people, it runs through shared tools and signs, and it is rebuilt inside the person over a long time. For the lashers, the words are learned in the talk with the buddy, and the talk is how the check moves inward.

The genetic method ties them together. Vygotsky called his approach "experimental-developmental" (Mind in Society, p. 61) and argued that "To study something historically means to study it in the process of change" (pp. 64 to 65). He wanted to watch higher abilities forming, because once a process has become automatic it turns into "fossilized behavior" (p. 63) and its origins are hard to see. Wertsch later organised the levels at which this change can be studied into four genetic domains: the evolution of the species, the history of a culture, an individual's development and the moment-to-moment formation of a single action, which he termed "microgenesis" (1985, pp. 54 to 56). Vygotsky and Luria had named the first three; the four-part list is Wertsch's.

For a designer, the practical consequence is to watch the learning while it happens. A week-6 test tells you whether a lasher can report a fault. A buddy's record across six weeks tells you how the report formed, where it stalled and what help moved it on.

Who was Vygotsky? Dates, texts and the name

Lev Vygotsky was born in 1896 in Orsha, now in Belarus, and died of tuberculosis in Moscow on 11 June 1934, aged 37. His psychology fits into about ten years, from the mid-1920s until his death, and his main book on language and thought, Thinking and Speech, came out a few months after he died.

English readers met him late and in pieces, which explains much of the confusion around his ideas:

• 1962, Thought and Language. The first English book, which its translators said "could be called an abridged version of the original". The chapter on Piaget lost the most.
• 1978, Mind in Society. Not one book by Vygotsky but a compilation. Its editors called it "not a literal translation of Vygotsky but rather our edited translation of Vygotsky", from which they "omitted material that seemed redundant" and to which they added material to make his points clearer. They admitted that "in tampering with the original we may have distorted history".
• 1986 and 1987, the full Thinking and Speech. Alex Kozulin's revised Thought and Language and Norris Minick's translation in volume 1 of the Collected Works.

So 1962, 1978 and 1987 are translation dates. Vygotsky's own texts are from about 1924 to 1934.

He did not call his work "sociocultural theory" either. The Soviet school he started was usually called the "sociohistorical" or "cultural-historical" approach, as Wertsch notes in his 1985 book (p. 30). The English word appears in the editors' introduction to Mind in Society ("Vygotsky's sociocultural theory of higher mental processes", p. 6), and Wertsch made it the standard label in 1991. "I use the term sociocultural because I want to understand how mental action is situated in cultural, historical, and institutional settings," he wrote, adding a warning about his own choice: "The danger in using sociocultural is that the historical dimension may get short shrift" (pp. 15 to 16).

Two stories about him need care. Vygotsky argued with Piaget's early books, but Piaget did not read Vygotsky in time to answer. In comments published in 1962, Piaget wrote: "I was never able to read his writings or to meet him in person." And the common line that Vygotsky's work was banned in the Soviet Union for decades after his death is less certain than it sounds. A 2016 volume edited by Anton Yasnitsky and René van der Veer, Revisionist Revolution in Vygotsky Studies, re-examines that story of Stalinist suppression along with other "historical and textual inaccuracies associated with received understandings of Vygotsky's work".

Vygotsky's books: which to read

• Mind in Society (1978). Read chapters 4 and 5 first. Chapter 4 holds the tool and sign comparison, internalisation and the "appears twice" passage (pp. 54 to 57), and chapter 5 sets out his method. Read it knowing the editors shaped it, and that its first four chapters come from a 1930 manuscript later printed under the names of Vygotsky and his colleague Alexander Luria.
• Thinking and Speech (1934), in Norris Minick's translation in the Collected Works (1987). Read chapter 7 for inner speech. It is the full text, including the argument with Piaget that the 1962 version cut most.
• James Wertsch, Vygotsky and the Social Formation of Mind (1985). Read it for the whole theory in one place. It is the clearest account of how social origins, mediation and the genetic method fit together, and it is careful to mark which terms are Wertsch's own.

After Vygotsky: Rogoff, Leont'ev, Engeström and Wertsch

Four later lines of work turned Vygotsky's ideas into tools for looking at learning in real settings, including workplaces.

Rogoff: three planes of sociocultural activity

Barbara Rogoff, a developmental psychologist, argued in a 1995 chapter that learning can be observed on three planes that are "inseparable, mutually constituting" (p. 139). Each has its own concept (p. 142):

• Community plane: apprenticeship. Individuals "participating with others in culturally organized activity" whose purpose includes "the development of mature participation in the activity by the less experienced people". For the lashers, that is the gang and the terminal.
• Interpersonal plane: guided participation. "The processes and systems of involvement between people as they communicate and coordinate efforts while participating in culturally valued activity." Rogoff counts observation as participation too. For the lashers, that is the buddy, the radio and the pre-shift talk.
• Personal plane: participatory appropriation. How individuals change through their involvement, "in the process becoming prepared for subsequent involvement in related activities". She called it "a process of becoming, rather than acquisition". For the lashers, that is how each one changes over the six weeks.

Her example was Girl Scout cookie sales in the United States, studied at all three planes at once. Rogoff's 1990 book is Apprenticeship in Thinking; she did not coin "cognitive apprenticeship", which comes from Collins, Brown and Newman in 1989 and is covered in our guide to cognitive apprenticeship. The community side connects with Jean Lave and Etienne Wenger's situated learning theory, which describes newcomers moving from peripheral to full participation.

Guided participation at work has been tried on purpose, and an Australian study shows how much depends on the workplace around it.

Case study: Guided learning trials in five Australian workplaces

These were six-month trials reported through learners' own accounts, with no comparison group and no learner count in the paper I read. The workplaces included a large food manufacturer, a textile maker and a recently corporatised power distributor, and the work was funded by the Australian National Training Authority. Billett started from the view that most learning at work happens through everyday work. When the trials ran in 1998, how far experienced co-workers could deliberately guide it had not been tested.

• What they did: Workplace mentors used "modelling, coaching, analogies, diagrams, questioning" for six months in each workplace. Learners were interviewed monthly about tasks they had recently done and about whom or what had helped them. Stephen Billett analysed the results with Rogoff's guided participation and Wertsch's ideas on how people make tools their own.
• Evidence and result: In Billett's 2001 paper to the AVETRA conference, the strategies "augmented learning through everyday work activities", but their use and value differed across the five enterprises, and reported learning was "generally higher" where the workplace supported it. At the textile maker one mentor gave high levels of support "against the norms of practice", which the two learners valued. At the food manufacturer, a reluctant recruit dismissed both the mentor and the strategies.
• Limitation: Learners' accounts only, no numbers in the text, and the researcher designed and evaluated the trials, so it shows a mechanism rather than an effect size.
• Lesson: Train the guides in a few moves, such as modelling, coaching and questioning, and give them time. Then check that the workplace lets newcomers take part, because the same strategies landed differently in different firms and the learner's own willingness counted as much as the guide.

Leont'ev: activity, action and operation

Alexei Leont'ev, Vygotsky's colleague, built the next layer in Activity, Consciousness, and Personality (1978 in English). He separated three levels. An activity is driven by a motive: "Activity does not exist without a motive." An action is a process "subordinated to a conscious purpose", a goal. And operations are the ways an action gets done: "Actions ... are related to goals, operations to conditions."

For the gang, the activity is turning the ship around safely. Lashing one bay is an action with a goal. Fitting a rod in rain at night is an operation that changes with the conditions. The three levels help in course design because conditions change from shift to shift. A course can teach the goal of each action, and the operations are better learned on deck, in rain, wind and darkness, with a buddy.

Engeström: activity systems and expansive learning

Yrjö Engeström, a Finnish researcher, first set out expansive learning in Learning by Expanding in 1987, later reissued in a second edition by Cambridge University Press. In an introduction written for the 1999 editions, he described three generations of activity theory. "The first generation, centered around Vygotsky, created the idea of mediation." The second was "largely inspired by Leont'ev's work", and the third works with "minimally two interacting activity systems". His activity system adds rules, community and division of labour to Vygotsky's subject, tools and object, so you can see what else shapes an action besides the person doing it.

Engeström's main claim about change is that contradictions inside an activity system drive it. In the same introduction he wrote, "The expansive cycle begins with individual subjects questioning the accepted practice, and it gradually expands into a collective movement or institution." He is frank that this is rare: "The occurrence of a full-fledged expansive cycle is not common, and it typically requires long-term effort and deliberate interventions." His 2001 article Expansive learning at work applied the ideas to a hospital setting in Finland. The intervention method built on expansive learning is the Change Laboratory, in which staff look at data from their own work and redesign it.

Back at the terminal, suppose the same "cone not locked" fault keeps appearing on one berth's night shift. Course-first thinking says retrain the lashers. An activity system view looks at the tools (worn semi-automatic twistlocks, poor lighting), the rules (pressure to keep up the crane's moves per hour), the division of labour (the signaller also covering a second crane) and the object itself (turning the ship around fast against stacking it safely). The fix may be a roster change or a new rule, not another course.

A Danish hospital used this way of thinking to fix how its doctors learned on the job.

Case study: Paediatric doctors redesigning clinic training in Denmark

This was a Change Laboratory of five sessions with a follow-up three months later, reported through what participants said rather than measures of learning. At the paediatric outpatient clinic of Aalborg University Hospital, which had about 14,000 visits in 2014, records from 2006 to 2013 showed that residents' planned supervision was under-used and disrupted by missing supervisors, interruptions and delays.

• What they did: From December 2014 to June 2015, five sessions were merged into the residents' existing three-hour meetings. They used "Vygotsky's method of double stimulation": the clinic's own data as a mirror, and activity system models drawn on a whiteboard for residents and specialists to change. Each session had 4 to 11 residents and up to 12 specialists, and 21 residents took part over the study because of rotations.
• Evidence and result: In Skipper and colleagues' 2016 paper in BMC Medical Education, the sessions produced a joint action plan in three parts: before, during and after supervision. At three months participants agreed that several aims had been met, including "a change in the culture of attendance at the planned pre-supervision sessions". The planned restructure of the clinic did not happen.
• Limitation: The results are participants' reports, with no supervision logs, waiting times or patient outcomes. The first author facilitated and analysed the sessions, and participants may have found it hard to decline.
• Lesson: When people are not learning on the job, look at who supervises, when, with how much time and under which rules before you write more content. Show staff their own data, let them model the work, agree specific changes and check after three months what stuck.

Wertsch: mediated action

Wertsch's own contribution was to make the tool-using action the thing you study. "A sociocultural approach to mind begins with the assumption that action is mediated and that it cannot be separated from the milieu in which it is carried out," he wrote (Voices of the Mind, 1991, p. 18). Change the tool and you change the action. A new radio channel, a redesigned lashing card or a new bay plan format will change what lashers say and do, and the training has to follow the tool.

Here is how I would choose between the four lenses.

How to use sociocultural theory to design training

Vygotsky described how abilities form, not how to teach them, so this workflow is my own. It draws on his ideas of social origins, mediation and inner speech, Rogoff's guided participation, Engeström's activity system and the three cases, and each step uses the lashing gang.

1. Map the job's shared language and tools. List the words, codes, calls, signals and aids that experienced staff rely on, in their words. For the gang: the location code, four gear names, the fault-report order, ALL STOP, "clear", the bay plan and the card.
2. Find where each one is first heard and first used. For each item, note where a newcomer first hears it and the first real task where they must say or use it with someone listening. That task is where the learning happens, so plan it.
3. Plan guided participation on real tasks. Give each newcomer a named guide and an intended route: watch and help, work beside, work watched. The six-week plan is illustrative; the leading hand changes it according to observed readiness, with safety controls and release authority retained.
4. Make talk part of the task. Build in the read-backs, spoken checks and radio reports the work requires. Reduce prompts as learners become independent, while keeping required audible communication.
5. Keep the job aid available at work. Assess correct performance with the aid where that reflects the job, rather than requiring unnecessary memorisation. People can still be assessed on using the aid correctly.
6. Use the course for what a course can do. It can give the words and the card, prompt the talk with a real person and give a place to log real tasks. Ask for written calls in open text, not multiple-choice recall.
7. Watch the process, not only the end. The buddy's record across six weeks and the leading hand's observation of one bay per shift show how competence is forming.
8. When the same fault keeps recurring, look at the system before retraining. Check the tools, rules, roles and competing goals first, as in the night-shift cone fault.

The shared-language plan for the lashing gang

Here is step 2 for the gang's main words and calls.

The same plan works for any job with its own language, such as handover terms on a hospital ward or call codes in a contact centre: list the words, then plan where each one gets said out loud and who hears it.

What Coursebox course requests say about language and culture

Sociocultural theory says the job's words and tools are learned by using them with other people. To see how often a course brief asks for that, I looked at course requests typed into Coursebox in the 90 days to 7 October 2026: one per person, from English-language browsers, with no uploaded file. Of 266 prompts, 104 were course requests, and about 30 of those were for people at work (28 to 32, depending on the reader). Two independent AI readers each classified all 104 against written definitions, and I re-read every positive request in full. Counts use the union of the two readers’ classifications after my review. Treat them as classification estimates, with uncertainty around borderline and missed cases, not statistical upper bounds.

The one exception was a course for interior designers that turned construction terms into five questions to ask before sending drawings to a builder, such as "Can it be priced?" and "Can the trades understand it?". The other requests asked for terminology sections with a quiz or a final test. Four workplace requests asked the course to teach workplace culture or norms, and each listed it as a topic to cover, such as "workplace culture and role expectations" with "a quiz in each module". About 1 in 10 of all 104 named the learners' language or cultural context, such as Tamil, Hindi, Portuguese or an Australian setting. Culture showed up as the language of delivery, rarely as the practices of the team the learner was joining.

The sample is small, it covers what people typed rather than finished courses, and teams may well teach their words on shift without saying so in a prompt. Still, a brief that names only the terms leaves the course free to treat them as reading and a quiz. If you want new starters to use the team's words with the team, write where each word gets said, and who listens, into the brief yourself.

Sociocultural theory vs Piaget and other learning theories

Sociocultural theory is often set against Piaget's theory and confused with several neighbours. The table is my synthesis, not a ranking.

The sharpest contrast with Piaget is over children's self-talk. In Thinking and Speech, Vygotsky put it this way: "In Piaget's view, egocentric speech arises from the inadequate socialization of what is initially an individual form of speech. In our view, it arises from the inadequate individualization of an initially social speech" (p. 263). For Piaget, development ran from the individual towards the social. For Vygotsky, "The actual movement in the development of the child's thinking occurs not from the individual to some state of socialization but from the social to the individual" (p. 72). When Piaget finally read the argument in 1962, he accepted Vygotsky's point that egocentric speech turns into inner speech: "I find myself in complete agreement with these hypotheses."

The two also differ over which relationships help most. In a 2012 comparison, Orlando Lourenço argued that the key difference concerns "relationships among equal peers vs. those based on authorities": Piaget valued exchange between equals, while Vygotsky gave more weight to guidance from more capable others. Both are counted as constructivists, Piaget in the cognitive strand and Vygotsky in the social strand, although the label "social constructivism" was applied to Vygotsky later. Our guide to constructivism covers both strands, and our comparison of behaviourism, cognitivism and constructivism places them in the wider history.

For the gang, I would use Vygotsky and Rogoff for the route and the language, and Bandura's social cognitive theory for the demonstrations, when a new lasher needs to see a bay lashed and believe she can do it.

Criticisms of sociocultural theory

Sociocultural theory is widely cited and often misread, and its critics point at both the texts and the evidence.

• The texts are edited, abridged and partly uncertain. Some passages in Mind in Society "have been taken from the work of Vygotsky's students or collaborators", its editors say. Margaret Gredler argued in 2012 that progress was held back by "the unavailability of accurate translations of his complete theory for several years". Revisionist historians now question who wrote some of what is attributed to him.
• It gives designers little method. Vygotsky described how abilities form, not how to teach them. In a 2003 chapter on the zone of proximal development, Seth Chaiklin wrote that Vygotsky "does not seem to have any systematic principles, methods or techniques" for how collaboration should be conducted, and Engeström admits that a full expansive cycle "is not common" and needs "deliberate interventions". Every design step in this guide is an interpretation.
• It can underplay the individual. Critics such as Lourenço argue that the theory's weight on guidance and social influence leaves less room for the learner's own reconstruction and for learning among equals, the side Piaget stressed.
• Popular versions add ideas he did not write. "More knowledgeable other", scaffolding as his concept, fixed ages for private speech and lists of "three ways of learning" all come from later writers. They are not always wrong as practice, but they are not Vygotsky.
• The workplace evidence is thin. Most of it is accounts, plans and process data, and no study I found isolates sociocultural theory and links it to a hard outcome with a comparison group.

A new tool on its own does not change practice either. In a Copenhagen emergency department, a project analysed through activity theory introduced a screening tool for older patients at risk of coming back into hospital.

In their 2019 analysis, Kirk, Andersen and Petersen found the screening tool "did not mediate a general and sustained transformation" in how the hospital and the community services worked together. The department had a "flow culture" in which tasks that did not speed patients through were "flow stoppers". Where it gave nurses and GPs shared terms, though, a GP noticed the difference: "now I receive many concrete inquiries." The lashing card is no different. Tools, goals and workplace rules can support or obstruct changed practice; one tool does not establish a lasting change. The evidence for sociocultural theory at work shows how tools, rules, talk and guidance shape learning; it does not give an effect size for the theory.

Using sociocultural theory with an AI course builder

On 11 October 2026 I briefed Coursebox to build the lashing gang course. Coursebox is our product, so weigh my account with that in mind. The brief gave the gang's words, radio rules, say-aloud check and card word for word, the roles and the six-week buddy route, and asked for about 40 minutes in eight sections, in this order:

1. A short "Before you start".
2. "The words we use on deck", with bay plan locations to write as radio calls.
3. "Talking on the radio", with two calls to write in open text.
4. "Week 1: watching and helping".
5. "Weeks 2 to 4: working beside your buddy".
6. "Weeks 5 and 6: working a bay".
7. "Your lashing card".
8. A wrap-up as the very last section.

I asked for open-text answers and logs instead of multiple-choice quizzes, and for every section to send the learner to say, do or check something with a real person.

After one revision, the outline came back as briefed. The first version merged the six weeks into one section and turned the radio calls and the final review into quizzes. One message naming the fix brought back the structure: the second outline had all eight sections in order, 11 pages, no quizzes and 5 open-text assignments, with the wrap-up last. The card page kept the point of a job aid, calling it "a tool in your pocket, not a test to pass", and said the course supports the buddy and the leading hand rather than replacing them. The builder added no theory claims: no "more knowledgeable other" and no learning styles.

Then I opened the rendered pages one by one, and the words page had come back as the thing the brief said not to build. "Bay Row Tier Practice" was a three-column glossary table of word, meaning and picture, and it had blurred the meanings I supplied. A cone became "a guide or locking piece used during lashing tasks", SATL "a standard lashing gear name used on deck" and a bottle "a lashing jack or tightening device", when the brief said a twistlock, a semi-automatic twistlock and a turnbuckle. The assignment after it told learners to "refer to the three provided bay-plan locations", and the page gave none. That is the sociocultural problem in miniature: the course turned the gang's working words into vocabulary to look at, and loosened what they meant.

Chat edits after generation did not change the pages, so I rewrote the page in the editor. It now has a table of each word on deck, what it is and where you will use it out loud, with the four meanings as briefed and the bay, row and tier code with its example. Under it are three bay plan locations to write as radio calls and the shift task of reading three locations aloud to the buddy. I reloaded the learner view and checked it: "A twistlock" and "A turnbuckle" are there, the locations are there, and "standard lashing gear name" is gone. Two assignments had also asked for essays of at least 500 words; I cut them back to the two exact radio calls and a three-question review.

I checked the outline and the rendered pages, not a run with lashers on a ship. The outline estimated about 1 hour 4 minutes against the 40 I asked for, so it needs trimming. A log on a page does not prove that anyone says the check on deck; that depends on the buddy and the leading hand, and no course builder can create the gang, the radio channel or the access to real work. With the fix in place, the words page now asks new lashers to say the gang's words the way they will use them on the radio, with the meanings the gang actually uses.

The other half of the work happens on the quay and on deck, where the buddy and the leading hand watch the route and the talk.

Putting sociocultural theory to work

Sociocultural theory answers a plain question: where do a new starter's abilities come from? Vygotsky's answer is that they start between people, run through shared tools and words, and are rebuilt inside the person over time. At work, that means the team's language, job aids, rules and guides are not the setting for the training. They are what the new starter learns through. A course can supply the words, the card and the prompts to use them, but the learning happens when the words are said to someone who listens.

Use the theory to plan where each word and tool gets used with another person, and to look at the work system when the same mistake keeps coming back after training. Do not treat it as a teaching method or a promise of measured results; the evidence shows how learning at work forms, not how much a sociocultural design adds.

Your next step: pick one job you train new starters for and list five words, codes or calls that experienced staff use every shift. For each, write the first task where a new starter has to say it out loud, and who will be listening.