The mechanism most training ignores

Working memory has a limited processing capacity. This is not a metaphor or an approximation — it is a hard architectural constraint of human cognition. George Miller’s foundational research established that working memory holds approximately seven items, and subsequent research by Nelson Cowan narrowed the active processing limit to around two to four chunks. People at every ability level have a ceiling on the mental effort they can exert while learning. When training tasks exceed that ceiling, cognitive overload occurs and learning stops — not because the learner is disengaged or underprepared, but because the cognitive architecture has run out of capacity.

John Sweller’s Cognitive Load Theory, first published in 1988 and one of the most cited frameworks in educational psychology, provides the most practically useful available account of why this matters for training design. The three-load model it proposes gives training designers a diagnostic framework for identifying what is consuming working memory capacity and what can be done about it.

The three-load model

Sweller distinguishes three types of cognitive load that add together to produce total load. Intrinsic load is the inherent complexity of the material — set by the subject matter and the learner’s prior knowledge, and not directly manipulable by the trainer. Extraneous load is generated by the design of the training materials themselves — confusing layouts, redundant information, jargon that adds no conceptual value, and decorative elements that consume working memory without contributing to learning. Germane load is the cognitive effort that actually produces learning — the mental work of building schemas that organise new information into usable knowledge structures.

The commercially important implication follows directly: intrinsic load is fixed; germane load is the goal; extraneous load is what the training designer controls. Minimising extraneous load is the primary lever available to anyone designing training that actually produces learning. The typical corporate training module that crams twelve learning objectives into forty-five minutes, narrated over slides that mirror the narration word-for-word, is maximising extraneous load while minimising the germane processing that produces durable skill.

The redundancy effect: the most common and most fixable design error

Presenting the same information simultaneously in two formats — narrated audio plus on-screen text that mirrors the audio — produces worse learning than presenting the information in one format only. The learner’s working memory is processing two channels carrying identical information, consuming capacity that would otherwise be available for schema formation. This is the redundancy effect, and it is violated in the majority of corporate eLearning modules produced today.

The fix is trivially simple. Either narrate text without displaying it, or display text without narrating it. Either approach eliminates the redundancy-induced extraneous load. The assumption that dual-modality presentation reinforces learning by hitting two channels simultaneously is directly contradicted by the research — it imposes additional processing cost without additional informational value.

The worked example effect: show before asking

One of the most robustly replicated findings in CLT research is the worked example effect. Novice learners who study fully solved examples outperform those who solve equivalent problems independently, because the independent problem-solving approach imposes enormous extraneous load through search — the cognitive cost of trying to figure out what to do next — while the worked example eliminates that search load and redirects the freed capacity toward schema formation.

The practical training design principle is direct. Before asking trainees to practise, show them expertly completed examples and explicitly model the reasoning that produced them. Onboarding that shows new employees high-quality completed examples of the work they will be required to produce is more effective than onboarding that gives them tasks and lets them discover solutions through trial and error. Discovery learning imposes the search load that worked examples eliminate, producing slower skill acquisition with higher error rates — precisely the inverse of what most training is designed to achieve.

The expertise reversal effect: one-size training fails everyone

What constitutes high intrinsic load for a novice constitutes low intrinsic load for an expert, because the expert has already built the schemas that compress multiple cognitive elements into a single chunk. This produces the expertise reversal effect: training materials designed appropriately for novices impose extraneous load on more experienced learners, because the explanatory scaffolding and redundant guidance that novices need actively interferes with the automated schema retrieval that experts use instead.

The commercial implication is that a single training module delivered to a mixed-expertise cohort is suboptimal for both groups simultaneously. The two-year veteran returning from a role change and the first-day hire should not receive the same onboarding programme. The veteran’s training should be shorter, less scaffolded, and more focused on the delta between what they already know and what is genuinely new. The novice’s training should be more guided, more example-rich, and more carefully paced to match the schema-formation capacity that genuine novices have when starting from scratch.

Spacing: the design principle most training completely ignores

The spacing effect is one of the most robust findings in all of learning science and one of the most consistently violated principles in corporate training design. A single intensive training session imposes high total cognitive load while allowing limited time for the consolidation that transfers schemas from working memory to long-term memory. Distributed practice — the same total instruction time spread across multiple sessions with intervening rest periods — produces substantially better retention because each session’s working memory processing has time to consolidate before the next session adds to it.

The standard day-long or multi-day intensive training programme maximises immediate knowledge exposure while minimising the spacing that produces durable retention. The employee who returns from a three-day training intensive having absorbed a fraction of the material, with most forgotten within a week, is the predictable output of a design that treats the brain as a storage device rather than a working memory system with a specific consolidation cycle.

If the cognitive demands of work — the sustained concentration, the information overload, the inability to keep up with what the role requires — are significantly affecting your wellbeing, that deserves attention beyond training design. UK: Samaritans (116 123, free, 24/7). Mind (0300 123 3393). International: iasp.info/resources/Crisis_Centres.

A book worth reading alongside this

Make It Stick: The Science of Successful Learning by Peter Brown, Henry Roediger and Mark McDaniel is the most accessible synthesis of CLT, the spacing effect, the testing effect, and interleaving available to a non-specialist reader. The book translates decades of learning science research into specific, counter-intuitive design principles — the most important being that the training methods that feel most productive to learners and trainers are often the least effective, while the methods that feel most difficult and uncomfortable tend to produce the most durable learning. For any entrepreneur making decisions about how to train their team, it provides the most direct available evidence that the training investment they are currently making is probably designed in exactly the way that minimises its return.

Have questions about this article?

If any part of this article raised questions you want to explore further, courbot.co is built for exactly that. It is courben.co’s AI assistant, designed around the psychology of entrepreneurship. Ask it anything from this article.

If the demands described in this article are significantly affecting your wellbeing, speaking with a psychologist is the right next step. UK: Samaritans (116 123, free, 24/7). Mind (0300 123 3393). BACP therapist finder: bacp.co.uk/search/Therapists. International: iasp.info/resources/Crisis_Centres. Crisis Text Line — text HOME to 741741.

This article is for educational and informational purposes only. Sources: Sweller, J. (1988), Cognitive Science, 12(2), 257–285. Paas, F. & van Merriënboer, J.J.G. (2020), Current Directions in Psychological Science, 29(4), 394–400. Sweller, J., van Merrienboer, J.J.G. & Paas, F. (1998), Educational Psychology Review, 10(3), 251–296.