Transfer of Learning: Why Skills Trained in One Context Often Fail in Another

A candidate completes a flawless precision landing on the training session obstacle. Two weeks later, on a different surface, in a different environment, with slightly different lighting and slightly different footing, the same movement falls apart.

This is not unusual. It is, in fact, one of the most common and most instructive phenomena in coaching. And it has a name: the transfer problem.

What transfer of learning means

Transfer of learning refers to the degree to which a skill trained in one context can be applied successfully in a different context. High transfer means the skill works broadly, across varied conditions. Low transfer means it works reliably only in the conditions in which it was trained.

The transfer problem is this: practice conditions in coaching tend to be consistent and controlled, because consistency makes skill acquisition faster and more measurable. But the environments in which skills need to actually work are varied and unpredictable. The gap between these two things is where transfer lives, and where most coaching falls short.

If you train a skill under identical conditions every time, you produce a practitioner who is excellent at that skill in those conditions. You do not necessarily produce a practitioner who can apply it in the world. These are different things, and conflating them is one of the most common errors in coaching design.

Specific and general transfer

Motor learning researchers distinguish between two types of transfer.

Specific transfer occurs when skills trained in one context apply directly to a closely related context. A practitioner who has trained precision landings on a specific wall in a specific gym will transfer some of that skill to another wall of similar height, surface, and dimension. The more similar the new context is to the training context, the more transfer occurs.

General transfer occurs when the underlying capabilities developed through training apply across a wide range of contexts. Strength, coordination, balance, spatial awareness, the capacity to read a movement problem and select an appropriate response: these are general capabilities that transfer broadly because they are not tied to specific conditions.

The most durable and valuable thing a parkour coach can develop in participants is not proficiency in specific movements under specific conditions. It is general movement capability: the motor literacy, the perceptual skill, the adaptive capacity that allows a practitioner to encounter an unfamiliar surface or an unfamiliar challenge and engage with it intelligently.

This is also, not coincidentally, the definition of physical literacy in practice.

Variable practice: the most important tool coaches are not using enough

The research on transfer is clear and consistent on one point: variable practice produces better transfer than constant practice, even though it produces slower skill acquisition during the practice period itself.

Variable practice means deliberately changing the conditions under which a skill is trained: different surfaces, different heights, different approach angles, different lighting, different environments, different movement combinations. The learner who has trained precision landings on ten different surfaces in five different environments is less polished on any single surface than the one who has trained only on one surface, but they are significantly more capable in the world.

This finding is robust enough that it should change how coaches structure sessions at a fundamental level. The goal of a coaching session is not to produce clean performance on a specific obstacle in a controlled environment. The goal is to develop capability that transfers. And transfer is developed by variability, not by repetition of identical conditions.

This does not mean that repetition on a single obstacle has no place. Early skill acquisition, particularly for complex or safety-critical movements, benefits from controlled conditions that allow the motor pattern to form. But once a movement pattern has been established, the priority should shift to variable practice. The coach who keeps returning to the same obstacle under the same conditions for weeks is not deepening the skill. They are narrowing it.

Contextual interference: making practice harder to make learning better

A related concept is contextual interference: the finding that practising multiple skills in a mixed or random order produces better long-term retention and transfer than practising each skill in a blocked sequence, even though performance during practice is worse.

In a blocked practice structure, a group trains precision landings for twenty minutes, then vaults for twenty minutes, then rolls for twenty minutes. In a mixed practice structure, the same skills are trained in a varied, interleaved order throughout the session. Performance during the session is better in the blocked condition. Retention and transfer are better in the mixed condition.

The mechanism is thought to involve the increased cognitive effort required in mixed practice: because the learner cannot rely on the momentum of repeating the same movement, they must reconstruct the motor programme each time. This effortful reconstruction strengthens the underlying representation of the skill in a way that blocked repetition does not.

The practical implication is that sessions which feel harder and less polished, because movements are varied and interleaved rather than blocked and repeated, may be producing better learning than sessions that look clean and controlled. Coach satisfaction and learner development are not always aligned. The clean session is not always the effective one.

The environment as a training variable

One of the most powerful and most underused variables in parkour coaching is the environment itself. Most coaching sessions happen in a fixed space with a fixed set of obstacles. The variations that occur are variations within that space: different movement combinations, different starting points, different skill targets. The space itself remains constant.

Taking sessions to different environments, even periodically, is one of the highest-leverage things a coach can do for transfer. A practitioner who has trained only in one gym, however well, has developed skills that are calibrated to that gym. A practitioner who has trained in parks, streets, car parks, sports halls, countryside, and urban spaces has developed movement intelligence that is calibrated to the world.

This is not a logistical argument against indoor training. Indoor environments offer control, safety management, and consistent conditions that are genuinely valuable for early skill acquisition. The argument is for balance: as participants develop, the environment should become a variable rather than a constant. The real world is the ultimate transfer test, and practitioners who never train in it are not fully prepared for it.

What the constraints-led approach has to do with this

The constraints-led approach to coaching, in which environmental, task and individual constraints are manipulated to guide movement development, is partly a transfer strategy. By varying the constraints within which a movement is trained, coaches produce practitioners whose movement solutions are flexible and adaptive rather than rigid and context-dependent.

A practitioner who has learned to land by absorbing into a slightly unstable surface, by landing on surfaces of different textures and dimensions, by arriving at different angles and speeds, has a richer and more transferable landing skill than one who has practised only on a single stable platform at a consistent height. The variability has built something more robust.

This is why the constraints-led approach is not just a technique for making sessions interesting. It is a principled method for developing skills that actually transfer, grounded in what the research on motor learning consistently shows about how durable movement capability is built.

The honest question for coaches

Transfer puts a useful and sometimes uncomfortable question to coaches: are the skills I am developing in my participants actually working in their lives, or only in my sessions?

The coach who has run the same session in the same space with the same obstacles for months may be producing practitioners who are excellent in that space and less capable elsewhere. The coach who varies conditions, changes environments, interleaves skills, and progressively reduces the support and control of the training context is producing practitioners who can actually move.

The test of coaching is not what participants can do in front of you. It is what they can do when you are not there.

For the practical application of these principles to specific movements, see the ADAPT Coaching Database. For the conceptual framework, see the Coaching Glossary. For how these ideas are embedded in the ADAPT certification framework, see the Education Pathway.

Published by ADAPT Qualifications — the world's original parkour coach certification, founded 2008. View upcoming courses.