that, according to honing theory, creativity is the process by
which one hones a worldview, and each idea the creator
comes up with is a different expression of the same
underlying core network of understandings, beliefs, and
attitudes. A worldview has a characteristic structure, and the
creator’s various outputs are reflections of that structure,
and they are related to one another, and potentially pave the
way for one another. Thus honing theory predicts that
creative individuals have a recognizable style.
There is evidence that human creativity is more consistent
with honing theory than with competing theories of
creativity with respect to developmental antecedents of
creativity, personality traits of creative individuals, and
studies of lifespan creativity (Gabora, under revision). This
paper reports on the results of creative style experiments
that provide further support for the theory. The goal of the
first two studies was to find empirical evidence for the
common belief that there really is such a thing as
recognizable style or voice. Although artists have no doubt
this is true, it has not been studied by psychologists, and as
we have seen, most theories of creativity do not predict it.
The goal of the third study was to test a more controversial
prediction of honing theory, the prediction that the structure
of a worldview manifests in a unique and recognizable way,
to varying degrees, through different creative outlets. Thus
for example, you might recognize someone’s art by
knowing how they dress or decorate.
Study 1: Within-domain Recognizability of Artistic
Style
The first study tested the hypothesis that individuals who
are highly familiar with the art of a given artist will
recognize other works by that artist that they have not
encountered before.
Method
Participants The research was conducted with 10 University
of British Columbia undergraduates majoring in art who
were highly familiar with five well-known artists, and with
each other’s art.
Materials and Procedures Prior to the study, participants
were instructed to bring from home a recently completed
painting that they had never discussed with or shown to any
of their classmates. They were asked to hide their signatures
or any other identifying feature of the painting. Before the
study, the paintings were examined to ensure that signatures
and any other identifying features had been covered.
At the beginning of the study, the art students were shown
three well-known paintings by each of five well-known
artists as a refresher. The well-known artists were Picasso,
Monet, Van Gogh, Dahli, and Andy Warhol. These artists
were decided upon because previous discussion with the
class indicated that all students were highly familiar with
them. The students were then shown ten unfamiliar (rare or
newly completed) works that they had not studied in class.
Signatures on all artworks were covered by black tape. The
art students were given a questionnaire and asked to guess
which famous artist did each painting. For each answer they
were also asked to state how certain they were on a 3-point
scale that they had not encountered the work before.
They were also shown the paintings by their fellow
classmates that they had never seen before. The rationale for
showing classmates’ paintings was to control for the
possibility that with the well-known artists, a participant
who, though not recognizing the creative voice, might guess
above chance levels to which era or country the artist
belonged. The only sufficiently large number of artists from
the same era and locale that the students were familiar with
were their own classmates. As with the famous artists, they
were asked to guess which classmate did each painting, and
to state how certain they were on a 3-point scale that they
had not encountered the work before.
The participants were debriefed, and the results were
analyzed. If a participant had encountered a work before, or
was uncertain about having encountered it before, the score
for this question was not included in the analysis. Less than
5% of scores were not included in the analysis.
Analysis The data were analyzed to determine if the
participants correctly identified the artists at above-chance
levels. First, a proportion correct score for each participant
was computed. For example, if a participant correctly
identified seven out of 10 possible artists, the proportion
correct score for that person was .70. Then, the proportion
correct score that would have been obtained on the basis on
random guesses for each question was computed. For
example, for the well-known artists, since there were 5 of
them, the proportion correct based on random guesses was
.20. One-sample t-tests were then computed comparing the
average proportion correct scores to the proportion correct
values that would have been obtained had participants been
randomly guessing. A one-sample randomization test
(Manly, 2007) was used to compute the p-levels for these t-
test values, given the small sample sizes, and .05 was used
as the criterion for statistical significance.
Results
The results are divided into two sections: recognition of
famous artists, and recognition of classmates’ art.
Recognition of Famous Artists For the task in which art
students were asked which famous artist painted each
painting, the mean proportion correct was .78 (SD = .12).
The proportion correct that would have been obtained on the
basis of random guesses was .20. This difference is
statistically significant, t(9) = 15.3, p < .0001, r (effect size)
= .98. Thus art students were able to distinguish above
chance which famous artists created pieces of art they had
not seen before.
Recognition of Classmates’ Art A similar result was
obtained with works of art by the students themselves. The
mean proportion correct was .74 (SD = .29). The proportion