Content uploaded by Ayelet Fishbach
Author content
All content in this area was uploaded by Ayelet Fishbach on Dec 27, 2017
Content may be subject to copyright.
ABSTRACT: The authors tested the hypothesis that a
broad or narrow scope of perceptual attention engen-
ders an analogously broad or narrow focus of concep-
tual attention, which in turn bolsters or undermines
creative generation. In the first two experiments, par-
ticipants completed visual tasks that forced them to fo-
cus perceptual attention on a comparatively broad or
narrow visual area. As predicted, broad, compared to
narrow initial focusing of perceptual attention subse-
quently led to generation of more original uses for a
brick (Experiment 1) and generation of more unusual
category exemplars (Experiment 2). In Experiment 3,
participants were merely asked to contract their
frontalis versus corrugator muscles, producing rudi-
mentary peripheral feedback associated with broad
versus narrow perceptual focus. As predicted, frontalis
contraction, relative to corrugator contraction, led to
the production of more original uses for a pair of scis-
sors. Together, these three experiments provided con-
verging initial support for our attentional priming
hypothesis, suggesting that situationally induced vari-
ations in the scope of perceptual attention (and simple
cues associated with such variations) may corre-
spondingly expand or constrict the focus of conceptual
attention within the semantic network, thereby improv-
ing or diminishing creativity.
In his widely influential theory of individual differences
in creativity, Mednick (1962) argued that the ability to
produce creative solutions is substantially influenced by
ideographic variation in the strength distributions of as-
sociative responses. For individuals with relatively steep
associative response gradients, the most conventional
responses to a given stimulus are overwhelmingly high
in associative strength, ensuring that these responses
will almost always be produced. For instance, within the
context of a word association task, for any given word
presented (e.g., “table”), individuals with more steeply
graded associative distributions will typically offer only
the most dominant or stereotyped responses (e.g.,
“chair,” “cloth”). In contrast, for individuals with rela-
tively flat response gradients, the associative strengths
of responses with respect to a given stimulus are more
equally distributed, increasing the likelihood that
remotely associated responses (e.g., “leg,” “fable”) will
be produced. Inasmuch as it is “among...more remote
responses that the requisite elements and mediating
terms for a creative solution will be lurking” (Mednick,
1962, p. 223), Mednick argued that individuals with
flatter associative gradients—those more likely to gen-
erate unconventional responses to a given stimulus—
should demonstrate greater creativity, on average, than
those with steeper associative gradients. This prediction
Creativity Research Journal Copyright 2003 by
2003, Vol. 15, Nos. 2 & 3, 277–286 Lawrence Erlbaum Associates, Inc.
Attentional Priming Effects on Creativity
Ronald S. Friedman
University of Maryland–College Park
Ayelet Fishbach
University of Chicago
Jens Förster
International University Bremen
Lioba Werth
University of Wurzburg
Correspondence and requests for reprints should be sent to Ronald
S. Friedman, Department of Psychological Sciences, University of
Missouri-Columbia, Columbia, MO 65211. E-mail: friedmann@
missouri.edu
Creativity Research Journal 277
was empirically supported by Mednick’s (1962) own
studies revealing positive correlations between per-
formance on the Remote Associates Test (higher scores
posited to indicate flatter associative gradients) and var-
ious indices of creativity (e.g., originality of independ-
ent research).
Although couched in the language of behaviorism,
according to Martindale (1981, 1995), Mednick’s the-
ory may be readily reconstrued in the language of cog-
nitive science. Essentially, Martindale suggested that
the individual differences in the slope of associative
gradients, as described by Mednick (1962), may be
understood in cognitive terms as individual differences
in the chronic scope (i.e., breadth or narrowness) of
conceptual attention, or attentional selection of inter-
nal, conceptual representations (as opposed to external
percepts). According to this view, individuals de-
scribed by Mednick as having steep associative gradi-
ents may be seen as tending to focus conceptual atten-
tion narrowly, leading to the strong activation of only a
few proximal nodes (e.g., “chair,” “cloth”), when
primed with a given input (e.g., “table”). Correspond-
ingly, individuals described by Mednick as having rel-
atively flat associative gradients may be seen as in-
clined to focus conceptual attention broadly (or to
defocus conceptual attention), engendering the weaker
activation of more, and thereby more remote (e.g.,
“leg,” “fable”) nodes in memory. In Martindale’s
reconceptualization, it is the enhanced activation of
these more remotely associated concepts that accounts
for the latter individuals’ superior propensity to access
and creatively recombine disparate ideas.
Nomothetic Implications: An Attentional
Priming Hypothesis
Although Martindale’s (1995) cognitive reconceptu-
alization of Mednick’s theory was formulated to account
for individual differences in creativity, his analysis has
clear nomothetic implications. Specifically, if ideo-
graphic variations in the scope of conceptual attention in-
fluence the creative generation of ideas, then situational
factors that transiently impact the breadth or narrowness
of conceptual attention (and thereby the extensiveness of
spreading activation) should analogously influence cre-
ativity. We posit that one such factor which may situa-
tionally influence the scope of conceptual attention, and
thereby creativity, is the scope of perceptual attention.
Why and how might the breadth or narrowness of
perceptual attention—attention directed to external
percepts—affect the scope of conceptual attention?
According to a large and burgeoning number of
cognitive theorists (Anderson & Spellman, 1995;
Neill & Westberry, 1987; Neumann & DeSchepper,
1992; Posner, 1987), the attentional selection mecha-
nism utilized on a perceptual level (e.g., to visually
focus upon percept X, while excluding percepts Y
and Z) may be identical, or at least highly correlated,
with the attentional mechanism utilized to select con-
ceptual nodes within the semantic network (i.e., to
regulate the extensiveness of spreading activation).
That is, a narrowing or broadening of attention to el-
ements of external perceptual input may be achieved
by means of the same mechanism used to narrow or
broaden attention to internal, conceptual representa-
tions. For example, when viewing a vase full of
flowers, attention may be directed narrowly (e.g., to
the vase alone, or the petals alone) or broadly (e.g.,
to the vase, stems, and petals together). Likewise,
when the concept of “flowers” is semantically
primed, conceptual attention may be directed nar-
rowly (e.g., allocating activation to the concepts of
“vase” alone or “petals” alone), or broadly (e.g., col-
lectively allocating activation to the concepts “vase,”
“stems,” and “petals”). According to the “identity”
hypothesis, these two varieties of attentional narrow-
ing and broadening share the same underlying mech-
anism: They differ in content (perceptual vs. concep-
tual), not process.
The Attentional Priming Hypothesis
Assuming that perceptual and conceptual selection
are indeed regulated by the same basic mechanism, it
is possible that broadening or narrowing the scope of
perceptual attention, by directing its focus to a wider
or more limited visual area, may activate or prime the
attentional mechanism, so that, for at least a brief pe-
riod of time, conceptual attention is correspondingly
broadened or narrowed. In other words, broadening or
narrowing the scope of perceptual attention may sub-
sequently engender more or less extensive (broader or
narrower) spreading activation, with respect to con-
structs entered into working memory. More concretely
speaking, we propose that if individuals are first
directed to focus their visual attention broadly or nar-
rowly on some initial stimulus, then (on an ostensibly
R. S. Friedman, A. Fishbach, J. Förster, and L. Werth
278 Creativity Research Journal
unrelated second task) are presented with a given
semantic construct to process, spreading activation
from this construct (e.g., “bird,” or “brick,” or “scis-
sors”) will be more extensive when individuals ini-
tially have to focus their perceptual attention broadly,
and spreading activation will be less extensive when
individuals initially have to focus their perceptual at-
tention narrowly. Concisely stated, broad or narrow
perceptual attention primes broad or narrow concep-
tual attention.
In terms of creativity, we predict that broad situa-
tionally induced perceptual attention should momen-
tarily widen the scope of conceptual attention, enhanc-
ing spreading activation to remote associates, and
thereby transiently bolstering the generation of cre-
ative ideas. In contrast, narrow situationally induced
perceptual attention should temporarily constrict the
scope of conceptual attention, reducing spreading acti-
vation to remote associates, and thereby diminishing
creative generation. In this study, we tested these pre-
dictions in three experiments, two manipulating the
scope of perceptual attention prior to gauging the abil-
ity to produce innovative alternatives; and one manip-
ulating facial muscular cues associated with broad
versus narrow perceptual attention prior to assessing
originality of generated alternatives.
Preliminary Experiment
Before attempting the aforementioned creativity
experiments, we first needed to empirically address a
critical, preliminary issue: If a situational manipula-
tion of the scope of perceptual attention is to prime,
that is, to have a lingering effect on the scope of con-
ceptual attention, it should be at least potent enough to
have lingering effects on another subsequent measure
of perceptual attentional scope. Or, as restated, if ma-
nipulations of broad or narrow perceptual scope can-
not carry over to influence perceptual attention on a
subsequent task, there is no reason to believe that such
manipulations can carry over to influence the breadth
or narrowness of conceptual scope (and thereby cre-
ativity) on a subsequent task. To ensure that situational
manipulations of perceptual attentional scope can have
a significantly lingering influence, we conducted a
preliminary experiment in which we manipulated the
breadth or narrowness of visual attention, then gauged
residual effects on the scope of perceptual attention
using an ostensibly unrelated, subsequent measure.
Manipulating Scope
To experimentally vary perceptual attentional
scope, we devised a visual search-based manipulation.
Here, participants were provided with a series of 18
computerized displays, each containing nine digits
(0–9), with each digit subtending a visual angle of ap-
proximately 1 degree. Half of these displays contained
the digit “3” and half did not. Each display was pre-
sented on screen for 1 sec, after which participants
were asked to respond as quickly as possible whether
or not a 3 was present, using the “y” or “n” keys,
respectively. There was a 2-sec intertrial interval be-
tween displays. In the broad attentional scope condi-
tion, the nine digits were randomly scattered about the
periphery of a 9 ×13-in. CRT display (excluding the
area within a 2-in. radius from the center), rendering it
necessary to visually search a broad perceptual area to
make the requisite presence/absence decision. Corre-
spondingly, in the narrow attentional scope condition,
the nine digits were randomly scattered about within
relatively close proximity to the center of the CRT dis-
play (within a 2-in. radius), rendering it necessary to
restrict search to a narrow visual area to make each
presence–absence decision. We posited that the re-
peated requirement to broaden or narrow the scope of
perceptual attention across 18 trials would prime the
attentional system, so that a relative broadening or nar-
rowing of perceptual attention could be detected on a
subsequent, unrelated task.
Measuring Attentional Priming
To gauge the extent of attentional priming, we uti-
lized a paper-and-pencil drawing task, in which partic-
ipants were given 1 min to complete a cartoon picture
(involving a doctor bicycling past a camel) by con-
necting numbered dots in ascending order. Accurate
completion of this task requires narrowed perceptual
attention to ensure the pencil mark actually connects
with each dot. Broader attention (e.g., to wider seg-
ments of the picture or to the scene as a whole) de-
tracts from focus on the component dots themselves
and hampers performance.
Method
Participants were 28 undergraduates of the Univer-
sity of Maryland who were informed that they would
be completing a study involving “a few separate tasks”
Attentional Priming and Creativity
Creativity Research Journal 279
and were paid $5 for their participation. Upon arrival,
participants were randomly assigned to complete ei-
ther the broad or narrow versions of the attentional
scope manipulation (the 3 task). After completing this
task, participants were administered the connect-the-
dots task, which, again, was meant to assess priming
of perceptual attentional scope.
Results
Performance scores on the connect-the-dots task
were computed by taking the difference between the
order of the final dot each participant marked (out of
96) and the number of dots which they missed (i.e.,
which their pencil mark did not intersect). To reiterate,
it was predicted that the performance of participants in
the narrow perceptual attention condition should be su-
perior to that of those in the broad condition, inasmuch
as fine-grained, narrowly directed attention should en-
hance focus on the low-level components essential to
the task (e.g., the dots) rather than on more global,
albeit task-irrelevant, aspects of the picture. These pre-
dictions were supported: Mnarrow = 75.6, Mbroad = 68.7,
t(26) = 2.30, p= .03. Theoretically speaking, this sug-
gests that manipulating the breadth or narrowness of
perceptual attention may at least briefly prime the at-
tentional system, leading to broader or narrower per-
ceptual attention on a physically and psychologically
separate task (at least on one presented after merely a
brief delay). These preliminary findings convinced us
that attentional priming manipulations may indeed be
potent enough to influence subsequent attentional pro-
cessing. The question then remained, whether this sub-
sequent processing includes the conceptual attentional
selection involved in creative generation? Does percep-
tual attentional priming influence the production of
innovative alternatives?
Experiment 1
Overview
As an initial test of the attentional priming hypothe-
sis, we used the same experimental manipulation of per-
ceptual attentional scope utilized in the preliminary ex-
periment (the 3 presence–absence task). As dependent
measures of originality (a critical component of creativ-
ity), we administered two brief tasks, the first requiring
participants to generate a creative alternative use for a
brick, the second asking participants to devise a creative
title for a photograph. It was predicted that participants
who had to search a broad visual area for the appearance
of a 3 (broad perceptual attention condition) would gen-
erate more creative responses on these tasks than those
who had to search a narrow visual area for the appear-
ance of a 3 (narrow perceptual attention condition). In
theoretical terms, this would suggest that compared to a
narrow perceptual focus, a broad focus of perceptual
attention gives rise to a broader focus of conceptual at-
tention, increasing the accessibility of remote, atypical
associations, and thereby enhancing innovation.
Method
Participants. Participants were 47 undergradu-
ates of the University of Maryland who were recruited
to complete a study involving “a few separate tasks.”
Participants were run individually and were paid $5
for their participation.
Procedure. Upon arrival, participants were ran-
domly assigned to complete either the broad or narrow
versions of the attentional scope manipulation (the 3
task, described in the previous section). After complet-
ing this task, participants were administered the two
generation measures. In the first measure, participants
were given 1 min to write down the most creative use
for a brick that they could generate. They were asked
to produce a use for a brick that was “neither typical
nor virtually impossible” (Friedman & Förster, 2001).
After time had elapsed, participants were then pro-
vided with a photograph of a Rottweiler stretched out
on a plush bedspread holding a bagel in its mouth. Par-
ticipants were given 1 min to generate a creative title
for this photograph. Following completion of these
tasks, participants were debriefed, paid, and released.
No participants voiced any suspicions regarding the con-
nection between the 3 task and the creativity measures.
Data coding. For the purposes of analysis, our de-
pendent variable was a composite of the creativity of the
brick uses and the photograph titles generated by partic-
ipants. To assess the creativity of the responses tendered,
nine independent scorers (all members of the psychol-
ogy department at the University of Maryland) were
asked to separately rate the creativity of the brick uses
and the photograph titles that participants generated on a
R. S. Friedman, A. Fishbach, J. Förster, and L. Werth
280 Creativity Research Journal
Likert scale (“How creative is this response?”) anchored
at 1 (not at all creative) and 9 (very creative). These rat-
ings (interrater reliability: α= .76) were used to com-
pute originality scores for each participant on each task,
which were then averaged to form the composite score.
An example of a relatively innovative solution on the
brick task was “to grind [it] up and use [it] as makeup”;
an example of a relatively mundane solution was “to
throw [it] through a window.” On the photograph task, an
example of a relatively innovative title was “Betty the
Beagle Beds a Bagel”; an example of a relatively mun-
dane title was “Dog Who Breaks Rules.”
Results
To assess the attentional priming hypothesis that
broadly focused perceptual attention would enhance in-
novation compared to narrowly focused perceptual at-
tention, a ttest was conducted comparing mean creativ-
ity scores within the two attentional scope conditions
(broad vs. narrow). As predicted, participants who com-
pleted the broadly focusing visual search (3) task, prior
to the generation tasks, demonstrated more originality
(M= 4.19) than those who completed the narrowly
focusing visual search task, (M= 3.57), t(45) = 2.31,
p= .02. This finding comprises the first evidence con-
sistent with the notion that situational variations in the
scope of perceptual attention may analogously influ-
ence the scope of conceptual attention, thereby affecting
the generation of innovative alternatives. Theoretically
speaking, these results suggest that, relative to a narrow
perceptual focus, a broad focus of perceptual attention
widens the focus of conceptual attention, rendering re-
mote, unconventional associations more accessible and
consequently bolstering innovation. In terms of the
present tasks, broadening of perceptual attention may
have expanded the scope of conceptual attention, so that
more unusual features of a brick (i.e., features irrelevant
to construction) or more remote associations to details
of the Rottweiler photograph (e.g., the use of a bed for
sex rather than sleep), became available for use in gen-
erating original responses.
Experiment 2
Overview
One question that remains from Experiment 1 con-
cerns the nature of the manipulation of perceptual
attentional scope. Here, participants were explicitly
asked to perform a visual search for a piece of infor-
mation (3) within either a broad or narrow perceptual
space. However, the provision of these explicit search
instructions raises the issue of whether active search of
a broad or narrow visual area is required to produce at-
tentional priming effects on creativity. Theoretically,
we posit no such requirement: The scope of conceptual
attention, with its concomitant effects on creativity,
should be influenced by the initial scope of the area of
perceptual focus, regardless of whether an active
search was underway for an object within that percep-
tual area. To demonstrate this empirically, in this exper-
iment, we devised a new manipulation of perceptual at-
tentional scope, in which participants were merely
asked to repeatedly focus their attention on broad or
narrow segments of a set of visual scenes (state maps).
Another limitation of the first experiment involved
the dependent measures. Although the brick-use and
title-generation tasks are face-valid indexes of creativ-
ity, these measures do not clearly reveal how concep-
tual broadening or narrowing influence the originality
of responses. Again, we have argued that broad or nar-
row perceptual focus should prime broader or nar-
rower conceptual focus, facilitating or impairing the
accessibility of remote, unconventional associates,
and correspondingly bolstering or diminishing innova-
tion. To provide more direct evidence for the mediat-
ing role of conceptual attentional scope, in Experiment
2, we devised a new originality measure in which par-
ticipants were to generate unusual exemplars for a se-
ries of categories. Presumably, broader conceptual at-
tention (i.e., more extensive spreading activation)
should enable generation of more atypical, creative ex-
emplars for any given category. Notably, this task is
closely akin to the sort of word association task
described by Mednick (1962) in conceptualizing the
underlying, associative basis of creativity.
Method
Participants. Participants were 62 undergradu-
ates of the University of Maryland, who were re-
cruited to complete a study involving “a few separate
tasks.” Participants were run individually and were
paid $5 for their participation.
Procedure. As discussed earlier, in this experi-
ment, we devised a novel manipulation of perceptual
Attentional Priming and Creativity
Creativity Research Journal 281
scope. Here, participants were presented with a series of
seven Rand McNally U.S. state maps. One map was pre-
sented on the computer screen for 5 sec before each ex-
perimental trial (see later). Each digitized map image
subtended nearly the entire area of the computer screen.
For each map, participants randomly assigned to the
broad perceptual scope condition were asked to attend
to the entire state, rather than any particular city or re-
gion, for the entire time that the map appeared on
screen. Correspondingly, participants randomly as-
signed to the narrow perceptual scope condition were
instructed to focus their attention on a red star (marking
a city within the center of each map), rather than any
other area of the map, for the entire time that the map
appeared on screen. The seven state maps used, along
with the central cities attended to in the narrow scope
condition, are as follows: Arkansas (Little Rock); Iowa
(Ames); Massachusetts (Worcester); Pennsylvania
(Selinsgrove); South Carolina (Columbia); Tennessee
(McMinnville); and Wisconsin (Wisconsin Rapids).
Following the presentation of each state map, par-
ticipants were provided on the computer with one of
seven category names (“birds,” “colors,” “fruits,” “fur-
niture,” “sports,” “vegetables,” and “vehicles”). For
each category, participants were asked to type as
quickly as possible the name of the most unusual ex-
emplar of that category that they could think of. The
computer recorded their entries and response times for
use in analysis. Following completion of the computer
program, participants were debriefed, paid, and
released. No participants voiced any hypothesis-
consistent suspicions.
Data coding. To assess the originality of the re-
sponses generated, 13 independent scorers (all mem-
bers of the psychology department at the University of
Maryland) were asked to rate how unusual each exem-
plar was for its category, on a Likert scale anchored at
1 (very typical) and 9 (very atypical ). Ratings (inter-
rater reliability: α= .93) were averaged across cate-
gories to form a composite originality score for each
participant.
Results
To assess the main experimental hypothesis that
broadly focused perceptual attention would enhance
generation of innovative exemplars, compared to
narrowly focused perceptual attention, a ttest was
conducted, comparing mean atypicality scores within
the broad- and narrow-scope conditions. Consistent
with predictions, participants who focused broadly on
the state maps, prior to response generation, produced
more unusual exemplars (M= 3.41) than those who
focused narrowly on the state maps, (M= 3.04),
t(60) = 1.87, p= .06. Analyses of log-transformed re-
sponse times revealed no reliable difference between
experimental conditions whatsoever (t> .19), suggest-
ing that participants in the broad perceptual scope
group were not more innovative than those in the nar-
row group merely because they took longer to generate
their responses.
Analyses of the zero-order correlations, between
originality (atypicality of exemplars) and generation
time, revealed substantial differences between the broad
and narrow conditions. Specifically, within the broad
perceptual attention condition, the correlation between
atypicality of generated exemplars and response time
was strong and positive, r(29) = .46, p= .01, suggesting
that increased response time perpetuated spreading acti-
vation, enabling access to more remote, unusual exem-
plars. However, within the narrow perceptual attention
condition, there was no correlation between response
time and the originality of responses, r(33) = .08, ns.
However, on average, participants in the narrow group
took just as much time as those in the broad group to
generate their responses, and their originality was not
appreciably enhanced by increased rumination. This
may suggest that narrow perceptual attention con-
stricted the scope of conceptual attention, so that
spreading activation to innovative exemplars was at
least momentarily inhibited.
The results of Experiment 2 conceptually replicate
those of the first experiment, providing further evi-
dence that situationally induced variations in the scope
of perceptual attention may influence creativity. These
findings also suggest that active visual search (as in
the procedure of Experiment 1) is unnecessary for at-
tentional priming effects—prolonged attention to a
broad or narrow visual area is alone sufficient to bol-
ster or impair creative generation. Furthermore, the re-
sults of Experiment 2 more clearly suggest that alter-
ing the scope of perceptual attention analogously
alters the scope of conceptual attention. Presumably,
the generation of atypical exemplars of a given cate-
gory requires broadening conceptual focus within the
semantic network to render fringe exemplars more ac-
cessible. As such, the fact that participants in the broad
R. S. Friedman, A. Fishbach, J. Förster, and L. Werth
282 Creativity Research Journal
perceptual scope condition generated more atypical ex-
emplars than those in the narrow condition implies that
their scope of conceptual attention was probably wider
than that of the narrowly focused group. This assump-
tion is further supported by the finding that atypicality
of generated exemplars was uncorrelated with response
time in the narrow condition, suggesting that spreading
activation within this group was constrained by nar-
rowed attentional selection on a conceptual level.
Experiment 3
Overview
Experiments 1 and 2 provided evidence that the
scope of perceptual attention influences that of con-
ceptual attention and thereby affects originality—a
fundamental component of creativity. In an ex-
ploratory vein, our final experiment tested the possi-
bility that simple bodily cues associated with, but not
necessarily evocative of, perceptual broadening or nar-
rowing, may, by dint of this association, come to inde-
pendently prime conceptual attention and thereby af-
fect creative generation.
The rationale for this corollary hypothesis flows
from the principles of learning theory. Over the
course of a lifetime, certain bodily or environmental
cues may consistently precede or co-occur with the
narrowing or broadening of perceptual attention. If
so, then over time, by means of conditioning, these
cues may come to independently narrow or broaden
the scope of conceptual attention, even when they no
longer precede or co-occur with changes in perceptual
attentional scope.
To provide an initial assessment of this hypothesis,
we tested whether facial muscular cues associated
with, but not directly evocative of, narrowed or broad-
ened perceptual scope would themselves influence the
scope of conceptual attention and thereby influence
the generation of innovative alternatives. More specif-
ically, we tested whether contraction of the eyebrow-
furrowing corrugator muscles, a known concomitant
of focused perceptual attention (Cohen, Davidson,
Senulis, Saron, & Weisman, 1992) compared to con-
traction of the eyebrow-raising frontalis muscle, a con-
comitant of widened perceptual scope (antithetical to
corrugator contraction), may diminish originality on a
creative generation task.
Method
Participants. Participants were 19 undergradu-
ates of the University of Würzburg, who were re-
cruited to complete a battery of psychology studies.
Participants were run in groups up to three (separated
so they could not see one another) and received DM 10
for their participation.
Procedure. Upon arrival, participants were given
a packet of paper-and-pencil surveys and instructed to
complete the different tasks contained therein. As a
cover story for the facial muscle manipulation, partic-
ipants were told that in addition to the study they were
participating in at that time, another related study was
concurrently being run by the psychology department,
one concerning muscle contractions involved in com-
puter work. It was explained to participants that in-
stead of working at a computer, they would instead
serve as a control group, engaging in similar muscle
contractions, yet doing so while completing paper-
and-pencil, rather than computerized, tasks.
At this point, participants randomly assigned to the
broad attentional cue condition were asked to contract
their frontalis muscle by raising their brows; those ran-
domly assigned to the narrow attentional cue condition
were asked to contract their corrugator muscles by fur-
rowing their brows. The experimenter did not person-
ally demonstrate these muscle actions, but merely de-
scribed them verbally. All participants were instructed
to maintain these muscle contractions throughout the
course of the subsequent task. In this task, which
served as our dependent measure of originality, partic-
ipants were instructed to generate and write down, on
a preprepared blank sheet of paper, as many creative
uses for a pair of scissors as they could think of. They
were asked to refrain from listing typical uses or from
listing uses that were virtually impossible.
Participants were interrupted after 2 min, told to
stop generating uses, and to fill out a f inal question-
naire. This survey gauged participants’ retrospective
liking of the generation task (“How much did you like
the task?”), anchored at 1 (not at all) and 9 (very
much); their motivation to perform the task (“How
motivated were you to complete the task?”), anchored
at 1 (not at all) and 9 (very much); and the perceived
difficulty of the task (“How difficult was the task?”),
anchored at 1 (not at all difficult) and 9 (very difficult).
These items were included for potential use as statisti-
cal covariates to control the possibility that frontalis
Attentional Priming and Creativity
Creativity Research Journal 283
contraction, relative to corrugator contraction, en-
hances creativity, not by virtue of its association with
broad perceptual attention, but, rather, because it
somehow renders the task at hand more enjoyable
and/or less effortful (cf. Amabile, 1996). Finally, par-
ticipants were probed for suspicions, debriefed, sworn
to secrecy, paid, and released. No suspicions regarding
the connection between the muscle contraction manip-
ulations and the creative generation task were voiced.
Data coding. The main dependent variable was
the originality of the scissor uses generated by partici-
pants. To assess this, 12 independent scorers (all mem-
bers of the psychology department at the University of
Würzburg) were asked to rate the creativity of the dif-
ferent uses participants generated on a Likert scale
(“How creative is this response?”), anchored at 1 (very
uncreative) and 9 (very creative), with an explicit mid-
point of 5 (neither creative nor uncreative). These rat-
ings were used to compute a mean creativity score for
each participant (summed ratings for each response
offered, divided by the total number of responses). In
addition, the total number of responses generated was
recorded for use in analysis. An example of a relatively
innovative response was to use the scissors as a hair dec-
oration; an example of a relatively mundane response
was to use the scissors to cut open a milk container.
Results
We predicted that corrugator contraction, a muscle
action associated with narrowed perceptual attention,
compared to frontalis contraction, a muscle action asso-
ciated with broadened perceptual attention (or at least
antithetical to narrowed attention), would independently
prime narrowed conceptual attention, and thereby di-
minish originality. In terms of the scissor-use generation
task, it was predicted that corrugator contraction would
lead to production of fewer innovative responses than
frontalis contraction. Consistent with this prediction,
participants who engaged in corrugator contraction
(narrow attentional cue) generated significantly fewer
innovative uses for a pair of scissors (M= 4.01) than
those who engaged in frontalis contraction (broad atten-
tional cue), (M= 3.57), t(17) = 2.14, p= .04. There
were no significant differences in the total number of
responses tendered within the two experimental condi-
tions (t< 1), suggesting that participants in the corruga-
tor contraction group were equally able to produce uses
for a pair of scissors, but that the novelty of their uses
was diminished by the relative inaccessibility of re-
motely associated cognitive material. Finally, there were
also no significant differences between conditions (all
ts < 1) in self-reported task liking, motivation, or
difficulty, suggesting that the obtained findings were
not somehow an artifact of these states.
The results of Experiment 3 supplement those of
the first two experiments, by suggesting that even sim-
ple muscular cues associated with broadened or nar-
rowed perceptual attention may yield attentional prim-
ing effects, surreptitiously impacting the scope of
conceptual attention and thereby influencing the abil-
ity to generate creative alternatives. Moreover, Exper-
iment 3 complements Experiments 1 and 2 by suggest-
ing that attentional priming effects on creativity do not
require explicit, initial instructions to focus perceptual
attention broadly or narrowly. Presumably, the mere
activation of cues associated with a broadened or nar-
rowed focus of perceptual attention may prime the at-
tentional system, such that it momentarily expands or
constricts the scope of conceptual attention and corre-
spondingly enhances or impairs creativity.
General Discussion
These experiments tested the hypothesis that a
broad or narrow scope of perceptual attention engen-
ders an analogously broad or narrow focus of con-
ceptual attention, which in turn bolsters or under-
mines innovation. In Experiment 1, participants
completed a visual search task, which forced them to
focus perceptual attention on a comparatively broad
or narrow visual area. Here, as predicted, broad, com-
pared to narrow focusing of perceptual attention dur-
ing visual search subsequently led to generation of
more original uses for a brick. In Experiment 2, in
place of a visual search-based manipulation, partici-
pants were merely asked to focus their perceptual at-
tention broadly or narrowly on a series of visual
scenes (U.S. state maps). Again, consistent with pre-
dictions, those led to broaden their scope of percep-
tual attention subsequently generated more unusual,
creative, category exemplars than those led to narrow
their focus of perceptual attention. Finally, in Exper-
iment 3, participants were merely asked to contract
their frontalis versus corrugator muscles, producing
rudimentary peripheral feedback associated with
R. S. Friedman, A. Fishbach, J. Förster, and L. Werth
284 Creativity Research Journal
broad versus narrow perceptual focus. As predicted,
frontalis contraction, relative to corrugator contrac-
tion, led to the production of more original uses for a
pair of scissors. Together, these three experiments pro-
vided converging initial support for our attentional
priming hypothesis, suggesting that situationally in-
duced variations in the scope of perceptual attention
(and simple cues associated with such variations) may
correspondingly expand or constrict the focus of con-
ceptual attention within the semantic network, thereby
improving or diminishing creativity.
Future Directions
Attentional priming and memory. If variations
in the scope of perceptual attention significantly affect
creativity, what other fundamental cognitive processes
might they influence? Assuming that changes in the
focus of perceptual attention indeed influence creative
generation by altering the focus of conceptual atten-
tion, this question then becomes: What other funda-
mental cognitive processes might be influenced by
variations in the scope of conceptual attention?
One prominent candidate is memory retrieval.
According to Anderson and his colleagues (Anderson
& Neely, 1996; Anderson & Spellman, 1995), re-
trieval of items from memory may be seen as in-
volving a conceptual focusing of attention. The
perceptual attentional mechanism used to select the
image of a single object within a visual array is con-
ceived of as identical (or closely analogous) to the
conceptual selection mechanism used to retrieve a
given target representation from within its subnet-
work of associated representations. In this view,
narrowing of conceptual attention is seen as critical
for effective retrieval, inasmuch as it serves to shut
out competition from related representations in
memory. Given this reasoning, we might predict
that narrowing perceptual attention may corre-
spondingly constrict the scope of conceptual atten-
tion during retrieval, improving recall of recently
encoded material by inhibiting the activation of
competitor items. In contrast, we might speculate
that broadening perceptual attention may expand
the scope of conceptual attention during recall, di-
minishing interference from highly accessible con-
structs (which are in the focus of attention) and en-
abling retrieval of less recently encoded or more
weakly associated material.
Attentional priming from motivational systems.
Although the present experiments explored the influ-
ence of perceptual attentional scope on conceptual at-
tentional focus, situationally induced variations in the
scope of perceptual attention are most likely not the
only means of indirectly eliciting variations in the fo-
cus of conceptual attention. For instance, according to
Derryberry and Tucker (1994; see also Tucker &
Williamson, 1984), there are two lateralized motiva-
tional/emotional systems in the brain, which function
in part to regulate the breadth of attention on both a
perceptual and conceptual level. The brain’s right
hemisphere is the seat of an appetitive motivational
system, which regulates emotional experience on a
continuum from elation to depression (cf. Higgins,
1997; Watson, Wiese, Vaidya, & Tellegen, 1999) and
serves to expand the focus of attention, again, both
perceptually and conceptually. In corresponding fash-
ion, the left hemisphere is posited as the locus of an
aversive motivational system, which regulates the ex-
perience of anxiety and serves to constrict both per-
ceptual and conceptual attention.
Derryberry and Tucker’s (1994) theory implies that
activation of the appetitive motivational system, pro-
posed to broaden the focus of conceptual attention,
should bolster creativity, and that activation of the
aversive system, posited to narrow conceptual atten-
tional scope, should impair it. We have recently col-
lected a great deal of evidence ostensibly consistent
with this notion (Friedman & Förster, 2001, 2002).
Here, participants were asked to perform arm motor
actions associated with appetitive versus aversive mo-
tivation (i.e., arm flexion vs. arm extension) or were
exposed to external approach versus avoidance-related
cues (i.e., pictures of a mouse searching for cheese vs.
escaping an owl). During, or shortly after the presenta-
tion of these internal or external appetitive and aver-
sive motivational cues, participants were administered
various measures of creative problem solving and
originality (e.g., the brick-use task). Relative to avoid-
ance-related cues, approach-related cues were found
to significantly facilitated creativity across tasks. In
light of Derryberry and Tucker’s (1994) theory, these
results may be seen as demonstrating that activation of
the appetitive or aversive motivational systems, via
presentation of internal and external motivational
cues, broadened or narrowed conceptual attention,
thereby bolstering or impairing creative cognition. We
are still in the process of determining whether the
Attentional Priming and Creativity
Creativity Research Journal 285
aforementioned effects may in fact be understood as
reflecting attentional priming of creativity by differen-
tial motivational systems.
At best, this research can only serve as a point of
departure en route to further developing the seminal
theoretical propositions advanced by Mednick
(1962) and refined by Martindale (1995). Wherever
it leads, we hope this line of work will appreciably
contribute to understanding the intriguing, yet still
mostly mysterious, relationship between attentional
processes and creativity.
References
Amabile, T. M. (1996). Creativity in context. Boulder, CO:
Westview Press.
Anderson, M. C., & Neely, J. H. (1996). Interference and inhibition
in memory retrieval. In E. L. Bjork & R. A. Bjork (Eds.),
Memory (pp. 237–313). New York: Academic Press.
Anderson, M. C., & Spellman, B. A. (1995). On the status of in-
hibitory mechanisms in cognition: Memory retrieval as a
model case. Psychological Review, 102, 68–100.
Cohen, B. H., Davidson, R. J., Senulis, J. A., Saron, C. D., &
Weisman, D. R. (1992). Muscle tension patterns during
auditory attention. Biological Psychology, 33, 133–156.
Derryberry, D., & Tucker, D. M. (1994). Motivating the focus of at-
tention. In P. M. Niedenthal & S. Kitayama (Eds.), The heart’s
eye: Emotional influences in perception and attention
(pp. 167–196). San Diego, CA: Academic Press.
Friedman, R. S., & Förster, J. (2001). The effects of promotion and
prevention cues on creativity. Journal of Personality and Social
Psychology, 81, 1001–1013.
Friedman, R. S., & Förster, J. (2002). The influence of approach and
avoidance motor actions on creative cognition. Journal of
Experimental Social Psychology, 38, 41–55.
Higgins, E. T. (1997). Beyond pleasure and pain. American
Psychologist, 52, 1280–1300.
Martindale, C. (1981). Cognition and consciousness. Homewood,
IL: Dorsey.
Martindale, C. (1995). Creativity and connectionism. In S. M.
Smith, T. B. Ward, & R. A. Finke (Eds.), The creative cognition
approach (pp. 249–268). Cambridge, MA: Bradford
Mednick, S. A. (1962). The associative basis of the creative process.
Psychological Review, 69, 220–232.
Neill, W. T., & Westberry, R. L. (1987). Selective attention and
the supression of cognitive noise. Journal of Experimental
Psychology: Learning, Memory, and Cognition, 13, 327–334.
Neumann, E., & DeSchepper, B. G. (1992). An inhibition-based
fan effect: Evidence for an active suppression mechanism in
selective attention. Canadian Journal of Psychology, 46,
11–50.
Posner, M. I. (1987). Selective attention and cognitive control.
Trends in Neuroscience, 10, 13–17.
Tucker, D. M., & Williamson, P. A. (1984). Asymmetric neural con-
trol systems in human self-regulation. Psychological Review,
91, 185–215.
Watson, D., Wiese, D., Vaidya, J., & Tellegen, A. (1999).
The two general activation systems of affect: Structural
findings, evolutionary considerations, and psychobiological
evidence. Journal of Personality and Social Psychology, 76,
820–838.
R. S. Friedman, A. Fishbach, J. Förster, and L. Werth
286 Creativity Research Journal
