Does spending longer defining the problem actually produce better answers?
Yes, and it was measured in an art studio in 1971. The students who spent longest deciding what the picture would be about produced work judged more original. Problem formulation correlated with rated originality at .54.
Why this is worth money
Almost every audience project starts with a question nobody crafted. A bank assumes its subject is finance. A fashion label assumes its subject is fashion. Everything downstream inherits the assumption.
A wrong lens does not fail loudly. It produces a perfectly competent map and ideas that land on nobody, and by the time anyone notices, the budget is spent and the question is the last thing anyone thinks to check, because it was never treated as a decision.
So the question worth asking is whether the time spent up front is real work or delay. That has an answer.
What the record shows
The art studio study, and the number everyone quotes
Csikszentmihalyi and Getzels (Journal of Personality and Social Psychology, 1971) gave thirty one art students a table of objects and asked them to compose a still life. They recorded what each student did before beginning to draw. How many objects were handled, how unusual the choices were, how much exploring went on, how far the subject changed as the work went along.
Then five well known artists and critics rated the finished drawings, independently. Inter-rater correlations ran from .82 to .98.
The result, in the authors' words: "five of the six behavioral variables were significantly related to the 'originality' of the resulting product", and "The strongest relationship appears between discovery-oriented behavior during the formulation of the problem and the originality of the drawing, as rated by the judges."
Problem formulation against rated originality: r = .54, with thirty one students.
The honest correction, and it belongs on the page
One study, one domain, thirty one people. So the number to trust is not .54.
Abdulla, Paek, Cramond and Runco (Psychology of Aesthetics, Creativity, and the Arts, 2020) pooled forty studies and 6,649 participants across fifty five years. Problem finding and creativity correlate at r = .22, with a confidence interval of .11 to .32 and high heterogeneity.
That is less than half the art studio figure, and it is the number that should be quoted. The relationship is real, significant, and smaller than the famous study suggests.
And the original author went back and found something harder
Csikszentmihalyi followed the same students up eighteen years later. In his own words: "some of the most promising and gifted young fine artists were employed in real estate, were teaching, or remodeling old houses."
Problem finding measured at art school did not carry through to a career. What he concluded from that is not that the finding was wrong, but that individual talent was never the whole system, which is what took him to the systems model of creativity.
We print that because it is the strongest thing anyone could throw at this page, and it came from the man who ran the study.
Where the modern support sits
Reiter-Palmon, Mumford, Boes and Runco (Creativity Research Journal, 1997) gave 195 undergraduates six real life problems and a measure of problem construction ability. Problem construction ability was related to higher quality and more original solutions.
The detail that matters most for us: the advantage was largest when the cues in the problem were inconsistent. In other words, the messier and more ill defined the situation, the more the work of framing it pays.
Which is exactly the situation a client is in when they ask what role they play in somebody's life.
The strongest objection, and the answer to it
The objection is the meta-analysis, and we have already printed it. r = .22 is a modest effect, and the heterogeneity is high, meaning the relationship behaves differently in different domains and at different ages.
So the strong form of the claim, that problem finding is the single determinant of creative quality, does not survive. Anyone asserting the art studio number as a general law is overclaiming.
The form that does survive is narrower and still worth money. Time spent framing the problem is positively related to the originality of what comes out, the effect is largest where the problem is ill defined, and the cost of the extra time is one workshop.
Against a media budget, that is the cheapest insurance available on the whole project.
Where the evidence stops
The 1971 study is thirty one art students composing a still life. It is not a study of commercial briefs, and no figure from it should be presented as one.
The eighteen year follow-up has no published correlation attached to it that we could obtain. It is quoted as the author's own account, which is what it is.
The 1997 problem construction study reports direction, not effect size, in what we could reach. No figure from it appears here.
None of this literature tests lens crafting. It tests the general proposition that framing work predicts output quality. The specific method is ours, and it is not what was measured.
Where the Cone sits
The lens is the decision about what a Cone is a Cone of. Not the product, not the category, the domain the product sits inside.
The test we use is to ask, four times over, what the true role of that thing is in people's lives. If the answer at the end is the category you started with, the lens has not been crafted, it has been assumed.
And it carries a kill question, asked out loud before anything is commissioned. What would tell us this is the wrong question? If nobody can answer that, the framing work has not happened yet.
What to do on Monday
Take the brief on your desk and write down what you think the subject is. Then ask what role that plays in people's lives. Do it four times, following each answer with the same question.
Watch what happens to the subject. If it stays where it started, you are working on a category. If it moves to something a person would recognise as part of their life, you have a lens.
Next is the map, which is what you draw once the lens holds, and the conditions under which your team's intuition about it is worth anything.
Pillar 3. The mapSources
- Csikszentmihalyi, M., & Getzels, J. W. (1971). Discovery-oriented behavior and the originality of creative products: A study with artists. Journal of Personality and Social Psychology, 19(1), 47-52. doi:10.1037/h0031106. Full text reprinted in The Systems Model of Creativity: The Collected Works of Mihaly Csikszentmihalyi (Springer, 2014), chapter 1.
- Getzels, J. W., & Csikszentmihalyi, M. (1976). The Creative Vision: A Longitudinal Study of Problem Finding in Art. New York: Wiley-Interscience.
- Abdulla, A. M., Paek, S. H., Cramond, B., & Runco, M. A. (2020). Problem finding and creativity: A meta-analytic review. Psychology of Aesthetics, Creativity, and the Arts, 14(1), 3-14. doi:10.1037/aca0000194
- Csikszentmihalyi, M. (2014). Introduction to the volume, in The Systems Model of Creativity. Springer. The eighteen year follow-up passage. doi:10.1007/978-94-017-9085-7
- Reiter-Palmon, R., Mumford, M. D., Boes, J. O., & Runco, M. A. (1997). Problem construction and creativity: The role of ability, cue consistency, and active processing. Creativity Research Journal, 10(1), 9-23. doi:10.1207/s15326934crj1001_2