Divergent Thinking as an Indicator of Creative Potential

What I read

In this article, the authors provide a thorough history of theoretical and experimental approaches to measuring divergent thinking.

First, the authors describe that divergent thinking is a “good metaphor of the kind of cognition that should lead to original ideas.” They indicate quickly—in the second paragraph—that “Divergent thinking is not the same as creative thinking. Divergent thinking often leads to originality, and originality is the central feature of creativity, but someone can do well on a test of divergent thinking and never actually perform in a creative fashion.”

Next, the authors work through an extremely detailed survey of academic work that has attempted to measure creativity through various ways. They again indicate that “tests of divergent thinking have dominated the field of creativity assessment for several decades. This has created one problem, namely that occasionally they are regarded as tests of creativity. As noted, that is not a tenable view.” They reference Guilford as the first to tie divergent thinking to creative potential, and indicate that most tests look only at fluency, or production of ideas, but fluency should not be the only indicator explored; they cite themselves as indicating that originality and flexibility are more tied to creativity.

Next, the authors describe that reliability is the stability of a test, and that fluency can easily be identified by counting ideas, originality is “determined by compiling ideas within a sample and identifying which are given infrequently”, and judges, who “can be objective and give reliability ratings,” can also be used. Validity is next described, as the opposite of bias. A complicated part of assessing validity of divergent thinking is that divergence is tied to convergent thinking, as with ideation, variation development, and evaluation.

The authors then describe, in great detail, the various tests that have been used historically to measure divergent thinking; Guilford’s tests “are still employed, and some have the advantage precisely because they have been used so many times.” Dozens of other tests are described, showing slight differences between each one. These include tests by Runco, Dow, and Smith; Runco and Acar; Abedi; Cheung; Wallach and Kogan; Torrance; Cropley; Harrington, Block and Block; Runco, Illies, and Eisenman; Runco and Charles; Vernon; Hocevar and Michae; Vernon; Runco, Okuda and Thurnston; Manske and Davis; Milgram, Milgram, Rosenbloom, and Rabkin; Moran Milgram, Sawyers and Fu; Mouchiroud and Lubar; Seddon and Snider; Mitchell, Bossomaier and Pallier; Seddon; Snyder et al; Runco and Mraz; Charles and Runco; Runco and Albert; Clark and Mirels; Michael and Wright; Zarnegar, Hocevar and Michael; Mednick; Mumford, Marks, Connelly, Zaccaro and Johnson; and Vincent, Decker and Mumford.

The authors then spend an equal amount of time discussing how other skills related to creativity, such as judgment, evaluation, and convergence, have been explored. Some evaluation research has reframed creativity away from novelty, and towards creative problem solving approaches. For example, “Basadur, Wakabayashi, and Graen (1990) defined evaluation as a kind of thinking style. They then examined style as a moderator of the impact of training. Four styles of creative problem solving were compared: generator (experience and ideation), conceptualizer (thinking and ideation), implementor (experience and evaluation), and optimizer (ideation and evaluation).”

The authors conclude by noting the back-and-forth of creativity research. It began by focusing on divergent thinking as a primary indicator of creativity. Then, tests of divergent thinking were rejected. And now, “The bedrock of that reasonable theory is that DT tests are not tests of creativity. They are estimates of the potential for creative problem solving. DT is not synonymous with creativity.”

What I learned and what I think

The paper’s internal mechanics confuse me, and its relationship to the authors outside perspectives confuse me.

By mechanics, I mean how the internal structure intends to support a conclusion. The conclusion is that “the word is out, DT is not synonymous with creativity but DT tests provide useful estimates of meaningful potential.” The paper begins with a similar assertion. But the whole article in-between is a discussion about how the creativity research community has framed divergent thinking as synonymous with creativity, including the author himself. Maybe he intended nuance between “synonymous with” and “relationship with” and I missed it, but it sure seems like one is being equated with the other.

This paper was published in February of 2012. As I understand it, Runco published “The standard definition of creativity” a month earlier, in January of 2012. In that article, he indicates that “if something is not unusual, novel, or unique, it is commonplace, mundane, orconventional. It is not original, and therefore not creative.” But in this article, he indicates that there is much more to creativity; for example, he describes that “It may be effectiveness or aesthetic appeal, but in some way creative things are all appropriate, as well as original.” Appropriateness is almost by definition commonplace, mundane, or conventional; it’s also a quality of good design. Again, maybe I’m missing a nuance here, but I don’t think so.

This type of work is also cementing my distrust and disappointment with the academic reliance on quantitative scoring. I realize this is a fundamental part of cognitive psychology, and the researchers are working with a same goal in mind as I am: to understand how creativity works. But I’m just increasingly aware of the gap between these studies and what happens in real life; there’s almost no applicability or overlap. If that’s true, what’s the point? If we try to observe “purity” of a behavior, but that purity never actually manifests, then it’s meaningless.

Also, a learning for me; I didn’t realize it was good form to heavily lean on your own research in producing new ideas for the academic community, but this paper includes 20 papers by Runco himself. This seems weird to me? I haven’t seen this in other survey-type papers before.