Myers-Briggs and the technical learner
Early in my career, HR asked me to fill in a Myers-Briggs assessment — 93 questions, submitted, and never mentioned again. No debrief, no design implication, no manager translating the result into how they'd work with me. I am not a psychometrician, and in eighteen years of designing and delivering technical training I have never applied MBTI as an active instrument in a room. What I have done instead is read rooms, with attention rather than instruments — and what that's taught me about learner behaviour is different enough from what MBTI describes that it's worth writing down.
The research case against MBTI is well-documented: a significant proportion of people who retake the assessment within five weeks receive a different type result, and the underlying theory hasn't fared well against the Big Five traits that have considerably stronger empirical support. None of that means personality doesn't influence learning — it means MBTI's particular operationalisation of it is contested, and may still function as a shared vocabulary, which is a narrower claim than its proponents usually make.
But the practical reason it doesn't reach the technical training room is simpler than the research debate: I don't control who attends, getting participants to document their current knowledge level before a programme is already a significant ask most organisations fail to meet, and a pre-training MBTI sits several rungs above that on the ladder of things that won't happen. Even granting a spreadsheet of forty valid results before walking in, I'd do very little with it — because the governing variable in a 40-hour cohort isn't personality type. It's stakes.
When I make the stakes explicit in the first session — forty hours, twenty topics, a capstone assessed by stakeholders, this is your first impression on the people who decide whether you clear probation — that information activates something personality type doesn't touch: the recognition that this week matters. Type might genuinely govern behaviour over a quarter or a year, when someone has the runway to revert to their default state. In forty hours, under assessment conditions, the situation overrides the type. Reading someone's type on day one would tell you who they are in low-stakes conditions. Not who they are here.
What I do instead is watch the first 20 minutes — before content starts, during introductions, the first small ask of the room. Who's pulled their chair back. Who has a question before you've finished your first sentence. Who's already opened their laptop. None of those signals reveal a four-letter type. They reveal something more immediately useful: this person's relationship to this room, right now. That's what you design around.
The caveat that matters: the first 20 minutes doesn't reveal everyone. One participant type — I've come to call them the Sleeper — is actively concealed by the opening of a programme. Quiet all week, unremarkable in every exercise, filed mentally by day four as competent but unexceptional. Then the capstone arrives, and they've built something nobody else in the room could have designed. If you'd profiled them on day one, you'd have filed them correctly and missed them entirely. The profiling isn't a pre-programme exercise. It's the programme itself, if you're paying attention throughout.
Over a thousand-plus participants, a set of behavioural signatures recurs across rooms, cultures, and programmes. Not types in the MBTI sense — not stable, not mutually exclusive, the same person can wear two or three across five days as pressure changes. A field guide, not a diagnostic framework:
| Signature | Signal | Facilitation note |
| The Performer | Hand up before you've finished the sentence. Needs the room to register they're there. | Use them early, then redistribute attention deliberately. |
| The Sleeper | Invisible all week. Then the capstone arrives, and the architecture is the best in the room. | The first 20 minutes won't find them. The 40 hours will, if you leave room for it. |
| The Reactor | Fine until corrected — then a visible recalibration. Correction reads as threat, not information. | Correct privately. When public, frame it as the system's logic, not their error. |
| The Validator | Every new concept met with "yes, that's what I thought." Learns the least, leaves the most satisfied. | Design exercises requiring application, not recognition. |
| The Lone Wolf | Works around the team rather than with it. Excellent solo output; the capstone costs them. | Assign roles in group work. Remove the option to go parallel. |
| The Literalist | Surgical precision on instructions. Falls apart the moment judgment is required beyond them. | Leave deliberate gaps in the brief. The gap is the exercise. |
| The Veteran | Filters every concept through prior experience — sometimes relevant, frequently overapplied. | Acknowledge the experience. Establish early where this system diverges. |
| The Converter | Sceptical through day one. Then a use case lands that maps to their world. Flips completely. | This is why real use cases aren't optional. They need to see their problem before they believe the solution. |
The same person can wear three of these across five days — the Performer on day one, the Pacer on day three, the Sleeper revealed on project day. That is the point. These are not stable identities. They are responses to conditions. Change the pressure, the stakes, the room you leave for someone to arrive at their own pace, and the signature changes with them. A framework that assigns four letters and files you in an HR system cannot account for that — not because the framework is wrong, but because the room is more dynamic than any instrument administered before the programme begins can capture.
The best profiling tool is a well-structured first 20 minutes. Except when it isn't — because the Sleeper only shows up on project day, and only if you left room for them. The real instrument is the forty hours. All of it, if you're paying attention throughout.
Further reading: Myers, I.B., & Briggs, K.C. (1962). Myers-Briggs Type Indicator. Consulting Psychologists Press. · Boyle, G.J. (1995). MBTI: Some psychometric limitations. Australian Psychologist, 30(1). · McCrae, R.R., & Costa, P.T. (1989). Reinterpreting the MBTI from the five-factor model. Journal of Personality, 57(1). · Kolb, D.A. (1984). Experiential Learning. Prentice Hall. · Kahneman, D. (2011). Thinking, Fast and Slow.