Effects of subliminal hints on insight problem solving
Masasi Hattori &Steven A. Sloman &Ryo Orita
Published online: 8 February 2013
#Psychonomic Society, Inc. 2013
Abstract Two experiments tested a total of 509 participants
on insight problems (the radiation problem and the nine-dot
problem). Half of the participants were first exposed to a 1-
min movie that included a subliminal hint. The hint raised the
solution rate of people who did not recognize it. In addition,
the way they solved the problem was affected by the hint. In
Experiment 3, a novel technique was introduced to address
some methodological concerns raised by Experiments 1and 2.
A total of 80 participants solved the 10-coin problem, and half
of them were exposed to a subliminal hint. The hint facilitated
solving the problem, and it shortened the solution time. Some
implications of subliminal priming for research on and theo-
rizing about insight problem solving are discussed.
Keywords Subliminal priming .Implicit cognition .
Convergence problem .Nine-dot problem .Ten-coin problem
A striking aspect of insight problem solving occurs when
people reach an impasse despite having all the knowledge
required for a solution. To explain how people resolve
impasses, some theorists emphasize changes in heuristic
search of the problem space (e.g., Chronicle, MacGregor, &
Ormerod 2004;Kaplan&Simon,1990), while others empha-
size changes in the distribution of activation in memory (e.g.,
Ohlsson, 2011; Seifert, Meyer, Davidson, Patalano, & Yaniv,
1995). The former view puts the theoretical workload on
controlled, attention-demanding, and conscious processes,
the latter view on automatic, effortless, and unconscious pro-
cesses such as the spread of activation.
Gick and Holyoak (1980) long ago demonstrated the
importance of awareness in accessing relevant information.
They used Duncker’s(1945)radiation problem,which
requires devising a method to destroy an inoperable tumor
in a patient’s stomach by a kind of ray without causing any
serious damage to surrounding healthy tissue. Presenting a
story that was structurally analogous to the solution (i.e., to
converge multiple low-intensity rays) did not enhance par-
ticipants’success unless it was explicitly presented as a hint.
Lockhart, Lamon, and Gick (1988) claimed that conceptual
processing of the prime triggers an awareness of the rele-
vance of the information and is required for transfer. On the
other hand, Maier (1931) observed that people can make use
of a hint incidentally given to them to solve the two-string
(pendulum) problem even if they were not aware of the hint.
Schunn and Dunbar (1996) also showed that people can
transfer their knowledge analogically to help solve even a
complex problem without awareness that they are doing so.
We examine the contribution of hints that are primed
subliminally to solving insight problems. Like Maier’s
(1931), most problem-solving studies have been concerned
with transfer from analogical tasks, or supraliminal priming,
in which an individual is fully aware of the stimulus, al-
though he or she may not be aware that it is actually a hint.
This method, however, may be susceptible to both memory
failures (e.g., participants may merely have forgotten being
aware) and demand characteristics (Bowden, 1997).
Subliminal priming has neither disadvantage and, thus,
offers an ideal method for examining the role of awareness
in the resolution of impasses. To our knowledge, however,
Electronic supplementary material The online version of this article
(doi:10.3758/s13423-013-0389-0) contains supplementary material,
which is available to authorized users.
M. Hattori (*)
Department of Psychology, Ritsumeikan University,
56-1 Toji-in Kitamachi, Kita-ku,
Kyoto 603-8577, Japan
S. A. Sloman
Department of Cognitive, Linguistic, and Psychological Sciences,
Brown University, Providence, RI, USA
Department of Psychology, Ritsumeikan University, Kyoto, Japan
Psychon Bull Rev (2013) 20:790–797
this technique has not been applied to insight problem
solving, except by Nishimura and Suzuki (2006), who re-
port, without describing their methods in detail, subliminal
priming of the solution time for the T Puzzle.
Evidence does suggest that subliminal priming affects
people’s preferences (Kunst-Wilson & Zajonc, 1980), inter-
personal judgments (Bargh & Pietromonaco, 1982), brand
choice (Karremans, Stroebe, & Claus, 2006), motivation in
classrooms (Radel, Sarrazin, Legrain, & Gobancé, 2009),
and even higher-order goals (Légal, Chappé, Coiffard, &
Villard-Forest, 2012). There is, however, not yet agreement
on the definition of a perceptual threshold (Snodgrass,
Bernat, & Shevrin, 2004), and some researchers still dismiss
subliminal perception (see Holender, 1986). Most studies
define subliminal perception in terms of subjective report. In
this study, we use a more conservative method to test
awareness, forced choice recognition between the true prime
and similar distractors.
In Experiments 1and 2, the effect of subliminal priming on
insight problem solving was examined using a short movie
that included a hint that participants were not aware they had
seen. In Experiment 3, a new method was used to address
concerns about the methodology of Experiments 1and 2.
In the first experiment, we examined whether subliminal
priming can affect human insight problem solving using
the radiation problem. Half of participants were given an
unrecognized hint during the attempt to solve the problem.
Participants and design
A total of 206 undergraduate students (89 female and 117
male;age:19–24 years, M= 20.1, SD =1.1)from
Ritsumeikan University and Ryukoku University were test-
ed. They were randomly assigned to either a hint or a no-
Materials and apparatus
The hint stimulus was presented as part of a 56-s movie,
composed from one hint image (exposed 33 ms × 60 times),
two mask images, three filler images, and one fixation image
(Fig. 1;see the online materials for more detail). The movie
was projected on a screen at the front of the room by a liquid
crystal display projector. The radiation problem was then
presented in a booklet along with diagrams (available online).
In the recognition task, participants were presented four fig-
ures (see Fig. 2), and their task was to choose the one that they
thought had been shown with a confidence rating—“sure,”
The experiment was administered in Japanese to groups in
two different classrooms. Two minutes after participants
started to solve the problem, they were asked to engage in
an “irrelevant”task for 1 min. While participants in the hint
condition watched the hint movie (described as “an irrele-
vant short movie”) on the screen, participants in the no-hint
condition tackled the dummy calculation task (e.g., 23+18 = )
to try to solve as many problems as possible while not looking
at the screen. They were given a total of 9 min to complete the
problem, including the initial 2-min trial period, the 1-min
exposure or calculation period, and the second 6-min trial
period. Finally, they were asked whether they have seen the
radiation problem before and chose an answer among the
options “yes,”“no,”and “I am not sure, but I might have seen
Results and discussion
Of the 206 participants, 10 reported that they had seen the
radiation problem before, 9 in the hint condition correctly
identified the hint image with certainty, and 2 in the no-hint
condition reported that they mistakenly had a glance at the
hint movie. Data from all these participants were excluded
from analysis. Of the remaining 185 participants, 56 %
(49/88) solved the problem in the hint condition, whereas
only 37 % (36/97) succeeded in the no-hint condition, χ
N= 185) = 6.41, p= .01, ϕ= .19.
In the recognition task, 55 % (48/88) selected the
correct hint image (2 out of 88 did not answer) signif-
icantly more than chance (i.e., 25 %), p<10
, but the
majority (69 % = 33/48) of them reported that it was a
“guess,”while the others reported “half-sure.”This
means either that some of participants saw the hint
with some degree of awareness or that they were
biased to select the hint image. We did verify in pilot
experiments that most people failed to recognize the
hint. But if they saw it, correct responders in the
recognition task should show a higher solution rate
than incorrect responders. They did not. Solution rates
for the two subgroups of the hint group were very close,
58 % (28/48) versus 55 % (21/38), χ
(1, N= 86) =
0.082, p=.78,ϕ= .03. This suggests that correct iden-
tifiers did not have the hint image available to awareness
Bowden (1997) revealed the effect of subliminal priming on ana-
grams. Anagrams, however, are not insight problems according to
Weisberg’s(1995) taxonomy, because solving an anagram does not
require changes in the problem representation.
Psychon Bull Rev (2013) 20:790–797 791
but, instead, were biased to choose it for some other
reason. Perhaps it simply looked more plausible than
the other images.
In the next experiment, designed to
examine the facilitation effect in a different problem,
we examine this issue directly.
In Experiment 1, an unrecognized hint raised the solution
rate of the radiation problem by 50 %, suggesting that
subliminal priming helped to solve an insight problem. In
this experiment, we tried to replicate the results using a
different problem that was less text based and more purely
spatial, the nine-dot problem.
Participants and design
A total of 133 undergraduate students from Brown
University participated in the experiment. They were
randomly assigned to either a hint or a no-hint condition. Of
these, 4 did not follow the instructions correctly, and their data
were excluded. The remaining 129 participants consisted of
59 female, 65 male, and 5 unknown (age range: 17–53 years,
Materials and apparatus
A hint movie for the nine-dot problem and an answer booklet
were used as in Experiment 1(Fig. 1;see the online materials
for details). The nine-dot problem was followed by a recog-
nition inquiry (only for the hint condition), where participants
were asked whether they saw the hint and they chose an
answer from “yes,”“no,”and “I thought I saw something,
but I didn’t recognize it clearly.”In the recognition task, they
were given a forced choice among four figures—the true hint
image and the identical image rotated by 90°, 180°, and 270°.
The experiment was administered to two groups in different
classrooms, and the procedure was similar to Experiment 1.
One minute after the experiment started, participants were
shown the hint movie or given the dummy calculation task.
They were given a total of 5 min to solve the problem.
Of the 129 participants, 55 reported that they had seen the nine-
dot problem before, and 3 finished the task successfully within
1 min (i.e., before the exposure period). All their data were
excluded from further analysis, unless otherwise stated. Of the
remaining 71 participants, 2 out of the 28 participants tested in
the hint condition reported having recognized the hint, but they
all failed to correctly identify the true hint image. Solution rates
in the hint and no-hint conditions were 29 % (8/28) and 9 %
(4/43), respectively, p=.05(Fisher’s exact test), ϕ=.25(see
Fig. 3). This result can be compared with an effect of a similar
Unfortunately, we did not collect data on the recognition task from
the no-hint group. Such data would have revealed whether there are
biases in favor of one of the response options.
18 sec × 3 rep
133 ms 767 ms Hint
33 ms Mask
133 ms Filler
701 ms 200 ms Hint
33 ms Mask
200–734 ms Filler
Fig. 1 Schematic description
of a sequence of frames
including implicit hint stimulus
presented in Experiments 1 (I)
and 2 (II). These sequences
(18 s long each) are looped
three times with a 1-s blank
between each loop and formed
a total of a 56-s long movie that
included 60 hint cuts
792 Psychon Bull Rev (2013) 20:790–797
supraliminal hint reported by Chronicle, Ormerod, and
MacGregor (2001), who used a shading pattern overlaid on
the array of nine dots as a hint without mentioning its relevance.
Their solution rate was 16 % (5/31), a similar effect size, ϕ=
If the hint facilitates solving the problem, solutions should be
similar to the one suggested by the hint. Figure 3also shows the
proportion of solutions that had the same structure as the hint, as
opposed to one of the other three solutions (see Fig. 4). Of the
successful solvers in the hint condition, 88 % (7/8) matched the
hint, whereas only 25 % (1/4) of successful solvers did so in the
no-hint condition, exactly what would be expected by chance.
The difference between the two conditions was marginally
significant, p=.07(Fisher’s exact test), although the effect size
was large ϕ= .63.
Results regarding participants’awareness were dissociated
from the effect of the hint. Only 5 out of 57 participants
(including the 27 who knew the problem and the 2 who finished
within 1 min) in the hint condition reported that they had
recognized the hint, but 3 of the 5 failed to correctly identify
the target in the recognition task. The distribution of their
choices in the recognition task also indicated that they did not
explicitly recognize the hint. The frequency of choosing each
solution (Fig. 4) was 17, 13, 14, and 9, respectively (4 chose
nothing), no different than a uniform distribution, χ
=2.47,p= .48. Recognition confidence also failed to predict
the ability to solve the problem. The proportions correct were
40 % (2/5), 22 % (2/9), and 30 % (13/43) for participants who
reported “yes,”“unsure,”and “no”to the recognition inquiry,
respectively, p=.79(Fisher’s exact test), Cramer’sV=.09(a
very small effect size).
The hint tripled the solution rate. Together with the results of
Experiment 1, the results in the hint condition (i.e., a high
solution rate, a high likelihood of solving the problem in a
way consistent with the hint, and a low recognition rate) suggest
an effect of subliminal stimulation. The method of Experiments
1and 2may raise some concerns, however. First, the effects
might be caused by the mask or filler image instead of the hint.
Second, insight might have been hampered by the calculation
task in the control conditions. These interpretations actually are
not consistent with the finding that solutions matched the hint in
the hint condition. Additionally, the latter interpretation is ren-
dered suspect by the higher solution rates in the no-hint condi-
tions (i.e., 37 % and 9 %), as compared with the typical solution
rates in the literature (9 % and 4 %, respectively).
do not have direct evidence that the explicit images and the
calculation task were irrelevant to the facilitation effect. Third,
it might be regarded as a problem that all the participants did not
receive exactly the same hint stimulus. That is, because the
experiment was administered to participants as a group, the
distance and angle to the hint varied with their seating positions.
All of these concerns were addressed in the next experiment.
In this experiment, the subliminal priming effect was examined
by a more strictly controlled method than in the previous experi-
ments. In order to generalize the results, we used a different
insight problem than for previous experiments, the 10-coin
problem. The problem is to turn a triangle composed of 10
coins upside down by moving no more than 3 coins (see Fig. 5).
Participants and design
A total of 80 adults (39 female and 41 male; age: 18–25 years
including 1 unknown, M=21.2,SD = 1.6) were tested. They
were randomly assigned to either a hint or a no-hint condition.
Since they actually did not include a control condition, we used a
solution rate of the control group, 0% (0/27), reported in MacGregor,
Ormerod, and Chronicle (2001).
The radiation problem is based on data from a total of 250 partic-
ipants from 10 control conditions reported in Gick and Holyoak
(1980), Spencer and Weisberg (1986), Holyoak and Koh (1987), and
Thomas and Lleras (2007,2009). The nine-dot problem is based on
data from a total of 284 participants from seven control groups (or
eliminated data) reported in Burnham and Davis (1969), Weisberg and
Alba (1981), MacGregor et al. (2001), and Kershaw and Ohlsson
Participants actually were assigned to one of four conditions: 2 (hint
vs. no hint) × 2 (instruction vs. no instruction). In the instruction
conditions, they were encouraged to generate novel ideas. A directive
like “Think unconventionally”was displayed on the center of the
screen for 1 s immediately after each prime (Fig. 5). No message was
displayed in the no-instruction condition. There was no main effect of
the instruction, χ
(1) = 0.15, p= .71, nor an interaction with the hint,
(1) = 0.15, p= .71, by a two-way ANOVA based on a chi-square
distribution. We therefore refrain from further discussion of this
Fig. 2 Four alternatives
prepared for the recognition
task in Experiment 1.
Participants forcedly chose one
that they thought had been
Psychon Bull Rev (2013) 20:790–797 793
Materials and apparatus
Participants solved the 10-coin problem on a 10.1-in. tablet
computer (NEC PC-LT550FS) using a touch pen. All the
operations, including receiving the hint and answering post-
questions, were carried out on the device.
The experiment was administered in Japanese individual-
ly. Participants were able to move any coins on the
screen or to return to the initial state whenever they
desired. In the hint condition, the hint was periodically
displayed on the screen as the problem was solved
(Fig. 5). In the no-hint condition, the hint was replaced
by a blank screen. Before they started, participants read
general instructions on the screen. They were notified in
advance that they would see irregular polygons (i.e., a
pattern mask) every 10 s and were falsely instructed that
the experiment aimed to examine the effect of “irrele-
vant”visual stimuli during problem solving. The task
ended when they successfully solved the problem. After
they solved the problem or 4 min had passed, they were
given several questions identical to those in Experiment 2
(see the online materials).
Results and discussion
Of the 80 participants, 3 reported that they had seen the 10-
coin problem before, and 1 in the hint condition reported
having recognized the hint. All their data were excluded
from further analysis.
The solution rate was 26 % (10/38) in the hint condition,
but only 5 % (2/38) in the no-hint condition, χ
(1, N= 76) =
6.33, p= .012, ϕ= .29. Figure 6shows the cumulative
distribution of successful solvers as time elapsed in each
group. A log-rank test for equality of rise curves (i.e.,
survivor functions) showed a significant difference of solu-
tion times between the two groups, χ
(1, N= 76) = 6.3,
p=.01,ϕ= .29. In the recognition task (25 % chance level),
only 26 % (10/38) selected the correct answer, p=.85
In sum, the effect of Experiments 1and 2was replicated.
The hint quintupled the solution rate and shortened the
solution time. The results suggest that the main cause of
facilitation in the previous experiments was not explicit
images given as masks or inexperience with the calculation
task but the hint itself.
In three experiments, we observed facilitation from sub-
liminal priming on insight problem solving. Hints in-
creased solution rates in all three experiments, despite
participants being both subjectively unaware of primes
and also unable to confidently discriminate the target
The results shed light on how impasses are resolved. The
subliminal priming effect is more consistent with theories
based on activation in memory, including the redistribution
theory of insight (Ohlsson, 2011), than with those based on
awareness. Exposure to a hint can activate insightful ideas
without awareness, increasing the probability of producing a
corresponding strategy and of deactivating inappropriate
ones to escape from the impasse.
One condition likely critical to the effect of subliminal
priming is preparedness. Moss, Kotovsky, and Cagan
(2007) obtained evidence that open goals set in a task (i.e.,
Fig. 4 Four solutions of the
nine-dot problem. Only
solution 1 matches to the
Solution Rate (%)
Fig. 3 Percentage of participants who solved the nine-dot problem in
794 Psychon Bull Rev (2013) 20:790–797
unsolved problems) promoted acquisition of hints implicitly
presented in another task. The opportunistic assimilation
hypothesis (Seifert et al., 1995) supposes that reaching im-
passe sets up “failure indices”in memory and relevant
information later introduced in the environment may lead
to retrieval of these indices, which may result in an insight.
So the procedure we used in Experiments 1and 2(i.e.,
presenting the hint a few minutes after participants started
tackling the problem) might have made participants recep-
tive to priming. Waiting a few minutes allowed them to set
their goals. Indeed, the effectiveness of incubation periods
in creative thinking could involve such goal setting.
Psychologists have argued about whether insights are
initially unconscious or not. Some theorists have claimed
that “insights are not always conscious from the start”
(Siegler, 2000, p. 82) or that first, “unconscious thought
‘boosts’the associative search for creative solutions”and
then “solutions are transferred to consciousness”(Zhong,
Dijksterhuis, & Galinsky, 2008, p. 916). Such claims are
consistent with our present data, but such processes may
require enabling conditions and the absence of disablers.
Conscious verbal processes can interfere with unconscious
processes in insight problem solving (Schooler, Ohlsson, &
Brooks, 1993). Moreover, although conscious control alone
is known to facilitate creativity (e.g., Nickerson, 1999),
intentional activities can hamper unconscious processes.
Mindful students in a classroom were insensitive to sublim-
inal priming (Radel et al., 2009), and similarly, conscious
attention eliminated priming effects on social perception
(Dijksterhuis & Van Knippenberg, 2000). How priming
Time (se c)
Proportion of Participants Successful
0 60 120 180 240
Fig. 6 Proportion of participants who successfully solved the 10-coin
problem, with time in each condition
20.8 sec0 sec 240 sec
Move only three coins and turn the
triangle point downwards
33±7 ms 240±13 ms Hint
33±7 ms 240±13 ms Hint
33±7 ms 240±13 ms
Correct SolutionHint Ima
Fig. 5 Schematic depiction of
the temporal task structure in
Experiment 3. Actual duration
time of each image was
measured by counting the
number of frames (600 fps)
captured by a high-speed
camera (Casio EX-F1). The
means and standard deviations
of values obtained by 10
measurements are shown (M±
SD). In the no-hint condition,
each hint image was replaced
by a blank screen (23±6 ms
Psychon Bull Rev (2013) 20:790–797 795
effects or unconscious processes interact with more inten-
tional and controlled activities is an important open issue.
Unconscious processing in insight problem solving cannot
be studied by methods like verbal self-report that require
conscious processing. Thus, we must rely on experimental
methods that tap implicit processes, like the subliminal
Acknowledgements We thank Phil Fernbach, and Hiroaki Suzuki
for their helpful comments on this study. We are also grateful to Yuriko
Shibata for her help in developing experimental materials in a pilot
This research was supported by Grant-in-Aid for Scientific Re-
search 22500247 from Japan Society for the Promotion of Science
and a research grant from the NeuroCreative Lab (NPO) to M.H.
Bargh, J. A., & Pietromonaco, P. (1982). Automatic information pro-
cessing and social perception: The influence of trait information
presented outside of conscious awareness on impression forma-
tion. Journal of Personality and Social Psychology, 43, 437–449.
Bowden, E. M. (1997). The effect of reportable and unreportable hints
on anagram solution and the aha! experience. Consciousness and
Cognition, 6, 545–573. doi:10.1006/ccog.1997.0325
Burnham, C. A., & Davis, K. G. (1969). The nine-dot problem: Beyond
perceptual organisation. Psyochonomic Science, 17, 321–323.
Chronicle, E. P., MacGregor, J. N., & Ormerod, T. C. (2004). What
makes an insight problem? The role of heuristics, goal concep-
tion, and solution recoding in knowledge-lean problems. Journal
of Experimental Psychology: Learning, Memory, and Cognition,
30, 14–27. doi:10.1037/0278-7322.214.171.124
Chronicle, E. P., Ormerod, T. C., & MacGregor, J. N. (2001). When insight
just won't come: The failure of visual cues in the nine-dot problem.
The Quarterly Journal of Experimental Psychology, 54A, 903–919.
Dijksterhuis, A., & Van Knippenberg, A. (2000). Behavioral indeci-
sion: Effects of self-focus on automatic behavior. Social
Cognition, 18, 55–74. doi:10.1521/soco.2000.18.1.55
Duncker, K. (1945). On problem-solving. Psychological Monographs,
58(5), i-113. doi:10.1037/h0093599
Gick, M. L., & Holyoak, K. J. (1980). Analogical problem solving.
Cognitive Psychology, 12, 306–355.
Holender, D. (1986). Semantic activation without conscious identifi-
cation in dichotic listening, parafoveal vision, and visual masking:
A survey and appraisal. Behavioral and Brain Sciences, 9, 1–23.
Holyoak, K., & Koh, K. (1987). Surface and structural similarity in
analogical transfer. Memory & Cognition, 15, 332–340.
Kaplan, C. A., & Simon, H. A. (1990). In search of insight. Cognitive
Psychology, 22, 374–419. doi:10.1016/0010-0285(90)90008-R
Karremans, J. C., Stroebe, W., & Claus, J. (2006). Beyond Vicary’s
fantasies: The impact of subliminal priming and brand choice.
Journal of Experimental Social Psychology, 42, 792–798.
Kershaw, T. C., & Ohlsson, S. (2004). Multiple causes of difficulty in
insight: The case of the nine-dot problem. Journal of
Experimental Psychology: Learning, Memory, and Cognition,
30, 3–13. doi:10.1037/0278-73126.96.36.199
Kunst-Wilson, W. R., & Zajonc, R. B. (1980). Affective discrimination
of stimuli that cannot be recognized. Science, 207, 557–558.
Légal, J.-B., Chappé, J., Coiffard, V., & Villard-Forest, A.
(2012). Don't you know that you want to trust me?
Subliminal goal priming and persuasion. Journal of
Experimental Social Psychology, 48, 358–360. doi:10.1016/
Lockhart, R. S., Lamon, M., & Gick, M. L. (1988). Conceptual transfer
in simple insight problems. Memory & Cognition, 16, 36–44.
MacGregor, J. N., Ormerod, T. C., & Chronicle, E. P. (2001).
Information processing and insight: A process model of perfor-
mance on the nine-dot and related problems. Journal of
Experimental Psychology: Learning, Memory, and Cognition,
27, 176–201. doi:10.1037//0278-73188.8.131.52
Maier, N. R. F. (1931). Reasoning in humans: II. The solution of
a problem and its appearance in consciousness. Journal of
Comparative Psychology, 12, 181–194. doi:10.1037/
Moss, J., Kotovsky, K., & Cagan, J. (2007). The influence of
open goals on the acquisition of problem-relevant informa-
tion. Journal of Experimental Psychology: Learning,
Memory, and Cognition, 33, 876–891. doi:10.1037/0278-
Nickerson, R. S. (1999). Enhancing creativity. In R. J. Sternberg (Ed.),
Handbook of creativity (pp. 392–430). Cambridge, UK:
Cambridge University Press.
Nishimura, Y., & Suzuki, H. (2006). Dousatsu mondai kaiketsu-no
seiyaku kanwa-ni okeru senzaiteki jouhou syori [Implicit infor-
mation processing of constraint relaxation in insight problem-
solving]. Cognitive Studies, 13, 136–138.
Ohlsson, S. (2011). Deep learning: How the mind overrides
Radel, R., Sarrazin, P., Legrain, P., & Gobancé, L. (2009). Subliminal
priming of motivational orientation in educational settings: Effect
on academic performance moderated by mindfulness. Journal of
Research in Personality, 43, 695–698. doi:10.1016/
Schooler, J. W., Ohlsson, S., & Brooks, K. (1993). Thoughts beyond
words: When language overshadows insight. Journal of
Experimental Psychology: General, 122, 166–183. doi:10.1037/
Schunn, C. D., & Dunbar, K. (1996). Priming, analogy, and awareness
in complex reasoning. Memory & Cognition, 24, 271–284.
Seifert, C. M., Meyer, D. E., Davidson, N. S., Patalano, A. L., & Yaniv,
I. (1995). Demystification of cognitive insight: Opportunistic
assimilation and the prepared-mind hypothesis. In R. J.
Sternberg & J. E. Davidson (Eds.), The nature of insight (pp.
65–124). Cambridge, MA: MIT Press.
Siegler, R. S. (2000). Unconscious insights. Current Directions in
Psychological Science, 9, 79–83. doi:10.1111/1467-
Snodgrass, M., Bernat, E., & Shevrin, H. (2004). Unconscious percep-
tion: A model-based approach to method and evidence.
Perception & Psychophysics, 66, 846–867. doi:10.3758/
Spencer, R. M., & Weisberg, R. W. (1986). Context-dependent effects
on analogical transfer. Memory & Cognition, 14, 442–449.
Thomas, L. E., & Lleras, A. (2007). Moving eyes and moving thought:
On the spatial compatibility between eye movements and cogni-
tion. Psychonomic Bulletin & Review, 14, 663–668. doi:10.3758/
796 Psychon Bull Rev (2013) 20:790–797
Thomas, L. E., & Lleras, A. (2009). Covert shifts of attention function
as an implicit aid to insight. Cognition, 111, 168–174.
Weisberg, R. W. (1995). Prolegomena to theories of insight in problem
solving: A taxonomy of problems. In R. J. Sternberg & J. E.
Davidson (Eds.), The nature of insight (pp. 157–196).
Cambridge, MA: MIT Press.
Weisberg, R. W., & Alba, J. W. (1981). Gestalt theory, insight, and past
experience: Reply to Dominowski. Journal of Experimental
Psychology: General, 110, 193–198. doi:10.1037/0096-3445.
Zhong, C.-B., Dijksterhuis, A., & Galinsky, A. D. (2008). The merits
of unconscious thought in creativity. Psychological Science, 19,
Psychon Bull Rev (2013) 20:790–797 797