How Boundary Crossing Works in Learning | School, Work and Disciplines

eduKateSG Learning Node Series · 0131

A student can understand something perfectly inside one classroom and still fail when the same knowledge crosses a boundary.

In school, the variables are labelled. At work, the variables are hidden inside messy situations. In science, “evidence” may mean measurement and reproducibility. In history, evidence is partial testimony situated in time. In mathematics, proof has a different standard again. A teacher moving into leadership discovers that the same word—“success”—now belongs to more people, more systems and more time horizons.

Boundary crossing is the learning problem created when knowledge has to travel between practices that do not organise the world in exactly the same way.

Boundary crossing works when difference stops being only an obstacle and becomes information about how two worlds organise knowledge, action and responsibility differently.

The 50-Second Read

Boundary crossing is associated strongly with sociocultural learning research and with Sanne Akkerman and Arthur Bakker’s influential 2011 review.
A boundary is a meaningful discontinuity between practices, roles, disciplines, organisations or contexts.
The review identified four learning mechanisms at boundaries: identification, coordination, reflection and transformation.
Boundary objects are shared artefacts that different groups can use together without needing identical interpretations.
Examples include lesson plans, diagrams, shared data, rubrics, patient records, project briefs, prototypes and maps.
Boundary crossing is not the same as simple transfer. It focuses on what happens when practices themselves differ and have to interact.
The learner may need to translate vocabulary, reconcile standards, negotiate roles and recognise which assumptions belonged only to the original context.
Recent work continues to apply the framework to interdisciplinary education, teacher professional learning, school–work transitions and research–practice partnerships.
Differences can generate learning because they make tacit assumptions visible.
The destination is not the erasure of boundaries. It is the capability to coordinate across them intelligently.

Canonical Owner Boundary

This Learning Node owns learning that occurs when people and knowledge move across practices with different norms, tools, language, roles and assumptions. How Learning Transfer Distance Works owns how transfer becomes harder as a new task departs from practice conditions. How Studying Works | The School-to-World Handoff owns the learner-side transition from classroom capability to real-world use. How Cognitive Flexibility Theory Works owns advanced knowledge reorganisation within complex domains. This page owns the boundary itself as a learning site.

1. Boundaries Are Not Just Lines on an Organisation Chart

A boundary appears whenever two practices organise activity differently enough that smooth participation breaks.

The boundary may be between school and workplace, science and English, teacher and researcher, hospital ward and laboratory, home knowledge and classroom knowledge, software engineering and product design, or two cultural communities using the same words differently.

The important feature is discontinuity. Something that made sense on one side no longer fits cleanly on the other.

2. The Discontinuity Is Often the Learning Signal

When a familiar method fails after crossing a boundary, the instinct is to treat the boundary as friction to be removed.

Sometimes that is appropriate. But educational research on boundary crossing asks a different question: what can the mismatch reveal?

If an engineering student cannot explain a calculation to a client, the failure may expose assumptions about audience. If a science teacher and English teacher disagree about what counts as a strong explanation, the disagreement reveals disciplinary norms that had previously remained tacit.

Difference can become diagnostic.

3. The Foundational Review

Akkerman and Bakker’s 2011 Review of Educational Research article, Boundary Crossing and Boundary Objects, synthesised a large body of literature across work, education and collaboration.

The review treated boundaries as dialogical phenomena and identified four potential learning mechanisms: identification, coordination, reflection and transformation.

These mechanisms remain useful because they describe different things that can happen when practices meet. Sometimes the main gain is understanding who we are. Sometimes it is simply getting coordinated. Sometimes each side sees itself differently. Sometimes the practices themselves change.

4. Identification: What Is Us and What Is Them?

Crossing a boundary can clarify identity.

A mathematics teacher working with an art teacher may suddenly notice which habits they had assumed were universal: formal proof, symbolic compression, exact definitions. The art teacher may notice different assumptions about ambiguity, interpretation and material experimentation.

Identification is learning about the relation between practices—what belongs to each, where they overlap and what differences matter.

The boundary makes identity visible because contrast reveals what routine participation had hidden.

5. Coordination: Work Together Without Becoming the Same

Two groups do not need complete agreement to coordinate successfully.

A doctor and pharmacist can use the same medication record while bringing different expertise. A teacher and school psychologist can use the same student support plan while seeing different aspects of the learner. A software engineer and designer can work from one prototype while asking different questions.

Coordination creates enough shared structure for action while preserving legitimate difference.

6. Reflection: Seeing Your Own Practice Through Another Practice

Crossing a boundary can create perspective.

A researcher working closely with teachers may realise that a clean experimental variable becomes entangled with timetables, attendance and classroom relationships. A teacher working with researchers may discover that a familiar classroom judgment can be examined more systematically.

Reflection is not merely learning about the other side. It is returning with a changed view of your own side.

7. Transformation: The Boundary Produces a New Practice

Sometimes coordination is not enough.

The interaction exposes a contradiction that requires both sides to change. A new interdisciplinary curriculum appears. A shared workflow is redesigned. A school–industry programme invents a role that did not previously exist.

Transformation is the strongest mechanism because the boundary produces a new way of working rather than merely better translation between old ways.

8. Boundary Objects: Shared Things That Travel

Boundary objects are artefacts that different communities can use together while interpreting them through their own practices.

A lesson plan can be a pedagogical object to a teacher, a research design element to a researcher and a schedule to an administrator. A prototype can be a user experience to a designer, an implementation problem to an engineer and a cost question to finance.

The object helps coordination because everyone can point to something shared even when their meanings are not identical.

9. Good Boundary Objects Are Stable and Flexible at Once

A useful boundary object is stable enough that groups are genuinely working on the same thing and flexible enough to support different local uses.

If the object is too vague, it coordinates nothing. If it is so rigid that only one group’s interpretation fits, it becomes an instrument of dominance rather than collaboration.

Lesson rubrics, maps, student work samples, shared dashboards and prototypes can all function well or badly depending on how much common structure they create.

10. Brokers Carry Meaning Across Boundaries

Some people become boundary brokers.

They know enough of two communities to translate language, anticipate misunderstandings and recognise which practices can be connected.

A teacher with industry experience may translate workplace expectations into curriculum. A bilingual student can sometimes mediate not only language but cultural assumptions. A data specialist inside a school may translate technical metrics into questions teachers can use.

Brokerage is a capability, not merely a position title.

11. Boundary Crossing Is Not the Same as Transfer

Transfer asks whether learning acquired in one context can be used in another.

Boundary crossing asks what happens when the two contexts are themselves organised differently and the learner must negotiate the discontinuity.

Transfer can be treated as movement of knowledge. Boundary crossing pays more attention to interaction among practices, identities, tools and norms.

The two ideas overlap, but boundary crossing gives us a richer vocabulary for school-to-work, interdisciplinary and interprofessional learning.

12. School to Work: The Same Mathematics Has Different Jobs

In school, a calculation may be performed because the question asks for it.

At work, nobody may tell you which calculation is needed. The mathematical problem is embedded inside a production target, budget, tolerance, customer complaint or safety limit.

Crossing the boundary requires recognising where school mathematics lives inside workplace activity and how workplace constraints change what counts as a useful answer.

13. School to University: Independence Is a Boundary Too

A learner may arrive at university with strong grades yet struggle because the organisation of learning changes.

Reading is less prescribed. Deadlines are longer. Feedback is less frequent. Knowledge must be integrated across lectures, labs and independent study.

The difficulty may not be intellectual capacity. It may be a boundary-crossing problem between two educational practices.

14. Primary to Secondary: A Familiar Subject Can Become a Different Practice

Students sometimes think they are continuing the same subject when the epistemic rules have changed.

Science becomes more model-driven and quantitative. Mathematics becomes more symbolic. English may demand more analytical control. Several teachers replace one classroom teacher.

Transition support should therefore explain not only harder content but the changed practice of learning the subject.

15. Interdisciplinary Learning Begins With Productive Misunderstanding

When two disciplines collaborate, they often discover that shared words are not actually shared.

“Model,” “validity,” “structure,” “evidence,” “significance,” “function” and “critical” can carry different disciplinary meanings.

These misunderstandings are useful when surfaced. They reveal the conceptual architecture each discipline usually leaves implicit.

16. English and Science: The Word “Explanation” Crosses a Boundary

In English, explanation may involve interpretation of language and textual evidence. In science, explanation often requires a mechanism linking cause and effect.

A student who has learned that a longer answer sounds “more explained” can struggle because the disciplinary standard changed.

Teaching the boundary means naming how explanation differs across subjects while preserving the general idea of making a claim intelligible through reasons and evidence.

17. Mathematics and Physics: Symbolic Fluency Is Not Enough

A mathematically competent student may manipulate equations correctly in physics yet miss the physical meaning of variables, constraints or signs.

The same algebraic operation now lives inside a disciplinary practice where units, models and physical plausibility matter.

Boundary crossing explains why “they know the maths” does not guarantee successful physics.

18. Research and Practice: Evidence Has Different Timelines

Researchers may want clean measures and controlled comparison. Practitioners may need a decision by Monday morning.

Neither timeline is inherently wrong. The practices optimise for different responsibilities.

Research–practice partnerships work when boundary objects and brokers allow evidence to remain rigorous while becoming usable.

A 2024 Frontiers in Education study of a continuous-improvement research–practice partnership described driver diagrams, network meetings and translating roles as boundary infrastructure supporting coordination across researchers, educators and improvement practitioners.

19. A Recent Singapore-Linked Example: Interdisciplinary Teacher Learning

A 2024 doctoral study by Edith Koh examined interdisciplinary teacher design teams in three Singapore secondary schools. English and science teachers collaborated to design lessons around critically reading science news articles.

The study used the four boundary-crossing mechanisms—identification, coordination, reflection and transformation—to analyse learning opportunities. Science news articles, pedagogical frameworks and lesson plans functioned as important shared artefacts in the collaboration.

The example is useful because the learning did not require either discipline to disappear. Learning came from working productively across the difference.

20. Interdisciplinary Student Projects Need Boundary Skills, Not Just Mixed Teams

Putting an engineer, writer and business student at the same table does not automatically create interdisciplinary learning.

They may divide the work into separate pieces and never integrate their practices.

A 2024 study in the Journal of Management Education examined postgraduate students working with industry clients and described boundary-spanning skills developed through observation, rehearsal and reflection.

Interdisciplinarity must be designed as interaction, not merely co-location.

21. Coordination Can Work Before Deep Understanding

Sometimes people must work together before they fully understand one another.

A common template, checklist, schedule or protocol can create operational coordination while deeper reflection develops later.

This is important in schools. Shared rubrics, moderation samples and curriculum maps can allow cross-class consistency even when teachers retain different styles and interpretations.

22. But Coordination Without Reflection Can Freeze Bad Assumptions

A shared form can make collaboration look successful while hiding disagreement.

Everyone completes the same template, but each person means something different by the fields.

Periodic reflection is needed: what do we each mean by this category? What does the other practice see that ours misses? Which difference should remain and which should be redesigned?

23. Boundaries Can Protect Expertise

Not every boundary should be dissolved.

Medicine, engineering, law and teaching contain specialised standards for good reasons. Removing every boundary can create vague generalism.

The educational objective is not “everyone does everything.” It is “specialists can coordinate, translate and understand enough of adjacent practices to work intelligently together.”

24. Boundaries Can Also Hide Responsibility

“That belongs to another department” is a classic boundary failure.

A student’s learning problem crosses subject lines. A safety issue sits between contractor and owner. A data problem sits between IT and curriculum. Everyone owns one piece and nobody owns the interface.

Boundary-crossing capability includes responsibility for the handoff.

25. Boundary Objects Can Expose the Interface

A shared student work sample can bring English and science teachers into the same conversation. A prototype can force design and engineering constraints to meet. A process map can reveal where one department hands incomplete information to another.

The object turns an abstract boundary into something inspectable.

26. Students Need Practice Translating Standards

Ask a student to explain how “evidence” differs in science, history and English.

Ask how “model” differs in mathematics, physics and geography. Ask what “critical” means in critical reading versus critical value in physics or statistics.

These exercises teach that vocabulary is embedded in practice rather than stored in one universal dictionary.

27. Make the Handoff Visible

Learning often fails at transitions because both sides assume the other will perform the translation.

Primary teachers assume secondary teachers will rebuild context. Secondary teachers assume learners already understand new subject norms. School assumes university will teach independence. University assumes students arrived independent.

A strong system names the handoff explicitly: what changes here, what stays invariant and what new responsibility begins?

28. Boundary Crossing and Representational Competence

Representations often become boundary objects.

A graph can allow scientists, policymakers and journalists to discuss the same data while asking different questions. A diagram can let teacher and student inspect the same mechanism. A project brief can coordinate specialists.

This connects naturally with How Representational Competence Works: crossing boundaries frequently requires reading how the same object changes meaning across communities.

29. Boundary Crossing and AI

AI systems increasingly sit at boundaries between disciplines, users and data systems.

A generated explanation may be linguistically fluent but violate disciplinary standards. A student may ask a general model for medical, legal or scientific guidance without recognising that the practice has specialised evidential rules.

AI literacy therefore includes boundary awareness: which practice owns the final judgment, what standard governs the answer and which parts of a generic response require expert translation?

30. Cross-Domain Comparison: International Airports

An airport is a designed boundary system.

Passengers cross jurisdictions, languages, security regimes, airlines and transport networks. Successful movement depends on shared documents, signs, codes, protocols and handoff points.

Education has similar interfaces. Certificates, rubrics, transcripts, portfolios and shared representations allow knowledge and identity to travel—but only when both sides can interpret them.

31. Cross-Domain Comparison: APIs

Software systems do not need identical internal architectures to communicate. They need a stable interface contract.

Boundary objects play a similar role in human collaboration. They define enough shared structure that different practices can exchange useful information without becoming the same system.

When the interface is ambiguous, integration fails even when both internal systems work perfectly.

32. Cross-Domain Comparison: Translation

Good translation does not replace every word mechanically.

It preserves function and meaning across systems with different conventions.

Boundary crossing in learning often requires the same skill: what is the equivalent move in this practice, and what has no exact equivalent at all?

33. A Practical Boundary-Crossing Protocol

Name the boundary: which practices, roles or contexts are meeting?
Find the discontinuity: where does action stop being smooth?
Map both sides: goals, language, tools, evidence standards and responsibilities.
Identify shared objects: what artefact can both sides inspect and work through?
Clarify terms: which shared words hide different meanings?
Coordinate first if necessary: build a minimum viable interface for action.
Reflect: what does the other practice reveal about ours?
Use brokers: involve people who can translate across communities.
Make handoffs explicit: what information and responsibility cross the boundary?
Test the interface: can both sides use the shared object correctly?
Transform when needed: redesign the practice if recurring contradictions cannot be coordinated away.
Preserve useful difference: do not erase specialist standards merely to make collaboration feel easy.

34. Failure Mode: Interdisciplinary Means “Everyone Does Their Own Part”

The project is divided cleanly and no genuine crossing occurs.

Repair: require shared interpretation, cross-review and a final product that depends on integrated decisions.

35. Failure Mode: One Discipline Colonises the Others

One practice defines the language, criteria and problem so completely that others become service providers.

Repair: make standards and contributions explicit. Ask which decisions genuinely require multiple forms of expertise.

36. Failure Mode: Shared Vocabulary Creates False Agreement

Everyone uses “evidence,” “quality” or “model” while meaning different things.

Repair: ask each practice to define the term through examples and counterexamples.

37. Failure Mode: The Boundary Object Becomes Paperwork

A template exists, but nobody uses it to think together.

Repair: anchor conversation and decisions to the artefact. If it does not coordinate work, redesign or remove it.

38. Failure Mode: Translation Work Is Invisible

One person quietly translates between teams and the organisation assumes coordination happens naturally.

Repair: recognise brokerage as real work, document interfaces and spread translation capability.

39. Failure Mode: The School Teaches Content but Not the New Practice

Students entering a new stage are told what topics become harder but not how the rules of participation change.

Repair: teach the boundary explicitly—new evidence standards, independence expectations, tools, genres and forms of reasoning.

40. The Missing-Node Scan

If students perform well in class but not in internships, if interdisciplinary projects become divided labour, if departments repeatedly misunderstand one another, if the same word carries incompatible meanings, or if transitions force learners to rediscover how a subject works, the missing node may be boundary-crossing capability.

Look for interface failures: “That is not how we do it here,” “I knew this in school but did not recognise it at work,” “we thought you meant something else,” “the handoff looked complete but the next team could not use it,” or “the student can do the skill only in the original subject.”

The knowledge may not be missing. The crossing mechanism may be.

41. The Return Path

Return to the student who knew the content in one classroom and lost it at the boundary.

The first interpretation is that the learning did not transfer.

Sometimes that is true. But another possibility is that the new world changed the rules of use.

The student may need to identify the new practice, coordinate with its tools and language, reflect on what the old practice assumed, and eventually transform how the knowledge is organised.

The boundary is not empty space between two places. It is a place where learning can happen.

Boundary crossing works when learners can carry knowledge into a different practice without pretending the practice is the same: identify the difference, build an interface, reflect across it and redesign the route when necessary.

Research and Further Reading

eduKateSG Learning Node Series · 0131 · Previous: 0130 — How Threshold Concepts Work.

How Boundary Crossing Works in Learning | Knowledge Changes When It Moves Between School, Work, Disciplines and Worlds