Emerging Trends in The Social and Behavioral Sciences

Insight

Item

Title

Insight

Author

Erickson, Brian

Kounios, John

Research Area

Development

Topic

Skills and Talent Development

Abstract

Insight, also known as the Aha phenomenon, is the sudden awareness of the solution to a problem. In contrast, analysis is problem solving by consciously and deliberately manipulating the elements of a problem. The Gestalt psychologists began studying insight about a century ago. On the basis of their research with complex “insight problems,” they characterized insight as a reinterpretation or restructuring of one's representation of a stimulus or situation after a period of unconscious processing. The emergence of cognitive psychology later during the twentieth century led to another period of advancement in insight research during the 1980s and 1990s. This work further characterized the unconscious nature of the processing leading up to an insight. More recently, the development of techniques for measuring and manipulating brain function has sparked a new renaissance in insight research. Cognitive neuroscience research has highlighted the key role of the right hemisphere and has discovered a number of neural precursors to insight, including its origins in patterns of resting‐state brain activity and in neural preparatory activity immediately before a problem is presented. The latest trend is work aimed at developing techniques to enhance insight, including recent research showing that direct stimulation of the right hemisphere can facilitate the solving of insight problems. Cognitive neuroscience approaches should continue to fuel rapid advances and may lead to the development of practical technologies for insight enhancement.

Related Essays

Models of Revealed Preference (Economics), Abi Adams and Ian Crawford

To Flop Is Human: Inventing Better Scientific Approaches to Anticipating Failure (Methods), Robert Boruch and Alan Ruby

Misinformation and How to Correct It (Psychology), John Cook et al.

Four Psychological Perspectives on Creativity (Psychology), Rodica Ioana Damian and Dean Keith Simonton

Youth Entrepreneurship (Psychology), William Damon et al.

Resilience (Psychology), Erica D. Diminich and George A. Bonanno

Empathy Gaps between Helpers and Help‐Seekers: Implications for Cooperation (Psychology), Vanessa K. Bohns and Francis J. Flynn

Controlling the Influence of Stereotypes on One's Thoughts (Psychology), Patrick S. Forscher and Patricia G. Devine

Setting One's Mind on Action: Planning Out Goal Striving in Advance (Psychology), Peter M. Gollwitzer

Regulatory Focus Theory (Psychology), E. Tory Higgins

Social Aspects of Memory (Psychology), William Hirst and Charles B. Stone

Emotion and Decision Making (Psychology), Jeff R. Huntsinger and Cara Ray

Implicit Memory (Psychology), Dawn M. McBride

Embodied Knowledge (Psychology), Diane Pecher and René Zeelenberg

Creativity in Teams (Psychology), Leigh L. Thompson and Elizabeth Ruth Wilson

Theory of Mind (Psychology), Henry Wellman

Identifier

etrds0183

extracted text

Insight
BRIAN ERICKSON and JOHN KOUNIOS

Abstract
Insight, also known as the Aha phenomenon, is the sudden awareness of the solution
to a problem. In contrast, analysis is problem solving by consciously and deliberately
manipulating the elements of a problem. The Gestalt psychologists began studying
insight about a century ago. On the basis of their research with complex “insight
problems,” they characterized insight as a reinterpretation or restructuring of one’s
representation of a stimulus or situation after a period of unconscious processing.
The emergence of cognitive psychology later during the twentieth century led to
another period of advancement in insight research during the 1980s and 1990s. This
work further characterized the unconscious nature of the processing leading up to
an insight. More recently, the development of techniques for measuring and manipulating brain function has sparked a new renaissance in insight research. Cognitive
neuroscience research has highlighted the key role of the right hemisphere and has
discovered a number of neural precursors to insight, including its origins in patterns of resting-state brain activity and in neural preparatory activity immediately
before a problem is presented. The latest trend is work aimed at developing techniques to enhance insight, including recent research showing that direct stimulation
of the right hemisphere can facilitate the solving of insight problems. Cognitive neuroscience approaches should continue to fuel rapid advances and may lead to the
development of practical technologies for insight enhancement.

When a person solves a problem, he or she takes a situation’s initial state
and transforms it into a goal state with whatever tools or “operators” are
available. There are two general strategies for accomplishing this transformation. Analytic thought involves deliberately, methodically, and consciously
applying these operators to effect the transformation. Insight is the sudden
awareness of the solution after a period of unconscious processing (i.e., the
“Aha” phenomenon).
HISTORICAL PERSPECTIVES
Insight has been a topic of interest since antiquity, but became a topic of scientific study only in the early twentieth century. Behaviorists such as Edward
Emerging Trends in the Social and Behavioral Sciences. Edited by Robert Scott and Stephen Kosslyn.
© 2015 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.

1

2

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Thorndike viewed learning and problem solving as a trial-and-error process by which response tendencies are gradually strengthened by rewards or
weakened by punishments. In contrast, the Gestalt psychologists recognized
that the behaviorist view is insufficient, as shown by studies that demonstrated discrete, all-or-none transformations in visual perception. Furthermore, Wolfgang K˝ohler’s studies of problem solving in apes demonstrated
what appeared to be sudden insights in problem solving. His most famous
study involved placing a bunch of bananas behind a fence out of the reach of a
chimpanzee. Two bamboo sticks were available to the chimp, but neither was
long enough to reach the bananas and pull them within reach. After a period
of frustration and inaction, the chimp spontaneously arrived at the solution,
namely, jamming one stick into the other—bamboo shafts are hollow—to
make a longer rod that could be used to reach the bananas. Importantly, the
chimpanzee had not been rewarded for intermediate steps that would have
brought him incrementally closer to achieving this solution, such as holding
both sticks at the same time or knocking them together. The use and construction of the tool apparently came to the chimpanzee suddenly and all at
once—an insight.
Gestalt psychologists went on to demonstrate similar examples of insightful problem solving in humans. Their methodology was based around a corpus of “insight problems” that typically elicit an “Aha!” experience when
they are solved. This work helped characterize insight but did not lead to
great progress in understanding its underlying mechanisms. One of the main
limitations with the Gestalt psychologists’ research on insight was that it
relied on an informal consensus regarding which problems were to be considered insight problems and which would be considered analytic problems.
Thus, they did not isolate objective behavioral or experimental correlates of
insight.
RESTRUCTURING AS INSIGHT SOLVING
Many classic insight problems are difficult to solve because they encourage
the assumption of erroneous constraints on the solution. For instance, the
very difficult nine-dot problem consists of a 3 × 3 matrix of dots (Figure 1a).
Solvers are instructed to draw four straight lines that pass through all nine
dots without lifting the pen or retracing any lines. The solution to this
problem is simple, but in many laboratory studies fewer than 5% of subjects
successfully solve it within the allotted time. Subjects fail to solve the
problem because they assume that the solution must stay inside the square
figure implied by the dots and that pivots can only be made on dots—neither
stipulation being part of the explicit instructions. When one discards these
constraints by restructuring one’s representation of the problem so that it

Insight

(a)

3

(b)

Figure 1 The nine-dot problem. (a) Subjects’ task is to draw four straight lines
connecting all 9 dots without lifting the pen from the paper and without retracing
any lines. (b) A solution to the nine-dot problem. Source: Adapted from
http://en.wikipedia.org/wiki/Thinking_outside_the_box

includes the blank background on which the nine-dot figure is presented,
the solution becomes fairly obvious (Figure 1b).
INSIGHT AND FUNCTIONAL FIXATIONS
The Gestalt psychologists’ main work was on visual perception. They
noted that any visual object or scene is ambiguous in the sense that it can
be interpreted in more than one way. One example of this is the necker
cube (Figure 2a), which can be interpreted in either of two orientations
(Figure 2b). When the viewer focuses attention on the lower square, this
square seems to be the front face of the cube; when the viewer focuses on the
upper square, that appears to be the front face. Moreover, the shift between
these two perceptual representations is sudden and discrete, as in an insight.
The Gestalt psychologists generalized from such perceptual phenomena to
problem solving. They viewed the main difficultly in solving a problem to
be that the would-be solver started with the wrong initial representation of
the problem. After restructuring, the new representation would immediately
suggest a solution that would be experienced as a sudden insight.
One type of misrepresentation, called functional fixation, is the tendency to
use objects only for their traditional functions. For example, in the classic candle problem, participants are directed to support a candle on a wall, given
only a box of tacks and a match. The fact that the box is holding the tacks
tends to discourage solvers from seeing it as a critical part of the solution: a

4

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

(a)

(b)

Figure 2 A Necker cube. (a) The Necker cube is an ambiguous figure that can be
viewed as existing in different orientations. (b) Switching between perceived
orientations is instantaneous. Source: Adapted from http://en.wikipedia.org/
wiki/Necker_cube

shelf for the candle, to be tacked to the wall. Solution rates improve, however,
when the box is presented to solvers independently of the tacks. Thus, participants tend to see the box only as a container when it is initially presented
as a container. Functional fixedness is reduced when the box is not presented
in its typical guise. This facilitates the solver’s restructuring of their idea of
the box.
THE MODERN STUDY OF INSIGHT
The emergence of modern cognitive psychology brought greater methodological and theoretical sophistication to the study of insight. For example,
Metcalfe and Wiebe were among the first to identify an objective difference
between insight and analytic problem solving. Their participants worked on
a series of insight and analytical problems while periodically rating how close
they felt to solving each problem (“warmth”). While working on analytic
problems, participants indicated that they felt gradually increasing warmth
before solving each problem. In contrast, while working on insight problems,
participants reported low warmth until just before solving the problem. Furthermore, for those few insight problems during which participants reported
gradually increasing warmth, their reported solutions were often incorrect.
Metcalfe and Wiebe’s study was groundbreaking in showing a clear
behavioral distinction between insight and analytic solving. However,
their study did not actually show that insight solutions were derived in a
discrete all-or-none manner, which is one of the defining features of insight.
Following up on their work, Smith and Kounios used a new experimental
procedure to force participants to make their best guess about a solution

Insight

5

immediately before they normally would have solved the problem. Mathematical analyses enabled them to determine whether problem solving
involves the gradual accumulation of solution information over time or
whether solution information becomes available in a single all-or-none
burst, as in insight. They used insight-like anagram problems and discovered that participants had accrued no measurable solution information
before achieving the solution, showing that solution information became
available to subjects in an all-or-none manner. This study confirmed the
sudden, discrete nature of insight, and differentiated it from prior findings
of gradual information accrual in noninsight tasks.
IMPASSE AND INCUBATION
When solvers reach a state at which they have failed to solve a problem and
cannot generate new ideas or strategies, they are said to be at impasse. Verbal
protocols in which the solver explains his or her thinking while working on
a problem have identified features of impasse before solving classic insight
problems. Results obtained with the verbal protocol technique have been corroborated by eye tracking studies. Before an insight solution, the duration
of gaze fixation (how long the solver looks at a single part of the problem)
increases. This increasing fixation means that the solver is running out of
ideas to try to solve the problem, signaling impasse. However, gaze fixation
tends to increase most on features of the problem that are relevant to the solution, indicating that solvers may unconsciously be collecting the information
necessary for restructuring.
Why is impasse frequently linked to insight solving? One theory is that
impasse often leads solvers to take a break from the problem, which allows
incubation. Incubation breaks have been shown to increase the solution rate
for classic insight problems. This can happen because time away from a problem can allow selective forgetting of incorrect strategies and assumptions,
making retrieval of the correct solution more likely. Incubation by selective
forgetting is supported by evidence that for incubation to impart beneficial
effects, problems must be completely removed from view. Furthermore,
incubation is helpful only when the original problem encourages spurious
assumptions and is most successful when the solver becomes engaged in
another task.
Another potential mechanism for incubation during a break is that solvers
might continue to unconsciously work on the problem; however, incubation
of essentially any length appears equally effective in boosting solution, which
would not be expected if complex unconscious processing were the cause.
Nevertheless, some form of unconscious processing may facilitate insight in
some situations.

6

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

NEW METHODOLOGIES FOR STUDYING INSIGHT
The body of research on the classic nine-dot, candle, and related problems
has a distinct limitation: Such “insight problems” were assumed to require
insight for their solution, but lacked any empirical support for this assumption. Furthermore, work comparing the solving of such “insight” and “analytic” problems makes a related assumption, namely, that these two classes
of problems differ only in the processing strategy that subjects use to solve
them. While the former assumption may eventually be shown to be generally
valid for classic insight problems, the latter is clearly not. Insight and analytic
problems typically differ from each other in a number of ways, such as their
complexity, familiarity, visual versus verbal content, and so forth. Therefore,
observed differences in how people solve these problems cannot be entirely
attributed to the use of an insight or an analytic strategy and may, in fact, be
due to these ancillary factors.
To circumvent these limitations, Edward Bowden and colleagues developed a new methodology for studying insight. They adapted one type of
problem originally developed for a test of creativity called the Remote Associates Test. Each of these compound remote associates (CRA) problems consists
of three words (e.g., pine, sauce, crab). The solver must generate a fourth word
that can be combined with each of the problem words to form a compound
or familiar phrase (apple: pineapple, applesauce, crabapple).
Solving a CRA problem requires a solver to access weak, remote associations of the problem words. For example, the word pine strongly evokes
the association tree, sauce evokes tomato, and crab, seafood. To find the solution word, the solver must retrieve the word apple, which is weakly associated with each of the problem words. This is thought to involve a type of
restructuring similar to what solvers must accomplish while tackling a classic
insight problem.
An interesting feature of CRAs is that they are neither insight nor analytic
problems. After participants solve a CRA, they are asked to report whether
they solved it in a deliberate, methodical (i.e., analytical) manner, or whether
they had solved it with insight (i.e., an “Aha”), a distinction with which participants are familiar. Virtually all participants report at least a few insight
solutions and a few analytic solutions. On average, about half of the solutions
are of each type, although participants exhibit substantial variation about
this mean.
This approach yields important benefits for insight research. Most importantly, insight and analytic solutions can be directly compared because the
problems that evoked these two types of solutions do not differ in complexity,
length, or any of the other factors that ordinarily distinguish classic insight
problems from analytic ones. Second, because these problems are short and

Insight

7

yield solutions within a few seconds, participants can solve many of them
within a single session. This opens up the possibility of using these problems
for neuroimaging studies of insight, because all neuroimaging techniques
require multiple trials to yield the necessary signal-to-noise ratio.
THE HEMISPHERIC BASIS OF INSIGHT
Research has associated creative cognition with the brain’s right cerebral
hemisphere. In particular, the solution of CRA problems with insight is
supported by visual hemifield studies. In these studies, a stimulus (such
as a word) is directly presented to either the right or left hemisphere by
presenting a stimulus to the left or right of a fixation point, respectively. In
several experiments by Beeman and Bowden (2000) CRAs were presented
with a deadline short enough that participants often failed to solve them.
Immediately upon reaching the deadline, either the solution word or an
irrelevant word was presented to the left visual field (which projects to
the right hemisphere) or the right visual field (which projects to the left
hemisphere). Participants were instructed to vocalize this probe word as
quickly as possible. They pronounced solution words more quickly than
irrelevant words. They also pronounced solution words more quickly when
they were presented to the right hemisphere. Furthermore, solution words
that were presented to the right hemisphere which also elicited an “Aha”
feeling of recognition were pronounced more quickly still. Such results
support the notion that insights are preceded by unconscious processing in
which the solution word is weakly represented in the right hemisphere.
On the basis of visual hemifield findings and a variety of other types of
research, Beeman proposed that the brain’s hemispheres process semantic
information differently. The left hemisphere engages in fine semantic coding
in which each word strongly activates a small number of closely related
associates, while the right hemisphere engages in coarse semantic coding
in which a word weakly activates a relatively large number of remote
associates. Insight and related forms of creative cognition primarily rely on
right hemisphere coarse semantic coding that allows a person to access the
nondominant meanings necessary for problem restructuring.
THE NEUROIMAGING OF INSIGHT
The first neuroimaging study to isolate insight was done by Beeman,
Kounios, and colleagues. This study used functional magnetic resonance
imaging (fMRI) and electroencephalography (EEG) to record participants’
neural activity while they solved CRA problems. fMRI provides excellent
spatial resolution but only modest temporal resolution, so it is the tool of

8

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

choice for localizing brain areas that are active during a cognitive task,
although it gives less precise information about the timing of this activity. In
contrast, EEG affords excellent temporal resolution, but only modest spatial
resolution, so it is ideally suited to delineating the timing of neural events.
Together, fMRI and EEG provide powerful localization of neural activity in
both space and time.
This work found that the moment of solving a CRA problem by insight
(but not analytic thought) was accompanied by a burst of high-frequency
(“gamma-band”) EEG activity measured over the right temporal lobe. The
timing of this activity coincided with participants’ awareness of the solution.
fMRI localized this activity to a brain structure called the right anterior superior
temporal gyrus (aSTG). Prior research supports the involvement of the right
aSTG region in integration of remotely associated pieces of semantic information, as occurs in jokes, metaphors, and other types of figurative language
processing.
This study revealed an additional phenomenon associated with insight
solutions: Immediately before the burst of high-frequency EEG activity
coincident with the sudden awareness of the solution, there was a burst of
lower frequency “alpha-band” activity measured over the right posterior
cortex. EEG alpha waves are generally associated with inhibition of neural
activity and, when measured over the visual cortex, are understood to reflect
inhibition of visual inputs to the brain. The researchers’ interpretation of this
finding was that the brain reduces sensory inputs briefly in order to increase
the signal-to-noise ratio of the weakly activated solution represented in the
right temporal cortex. This enables retrieval of the solution into awareness,
which is accomplished as a sudden insight.
The discovery of a neural correlate of insight and its immediate precursor suggested a research strategy of finding additional precursors. This is
accomplished by tracing neural activity back in time starting at the moment
of insight. This is important for two reasons. First, the discovery that insight
and analytic solving involve different cognitive strategies and patterns of
neural activity raises the question of what factors determine which of these
strategies are applied toward a given problem. Second, if there is a sequence
of neural precursors to insight and analytic solving, this raises the possibility that each of these precursors can potentially be influenced to change the
cognitive strategy that a person uses to solve a problem.
THE PREPARED MIND
Louis Pasteur once said, “Chance favors only the prepared mind.” To look
for a neural basis for Pasteur’s claim, Kounios and Beeman examined one
type of neural precursor of insight—brain activity immediately before the

Insight

9

presentation of each CRA problem. They hypothesized that a person prepares for or anticipates solving an upcoming problem by adopting a pattern
of neural activity that will promote either an insightful or analytic solution.
They found that preparation for solving an upcoming problem with insight
involves (relative to preparation for analytic solving) greater activity in the
anterior cingulate and in the right and left temporal lobes. By contrast, preparation for analytic solving involves increased activity in the visual cortex. The
general interpretation of these results was that preparation for analytic solving involves outward focus of attention on the screen on which the problem is
about to be displayed, while preparation for insight involves inward focus of
attention and priming of brain areas involved in processing words and concepts. The isolation of different brain states corresponding to insightful and
analytic modes of thought further supports the idea that cognitive solving
strategies can be systematically primed.
One way to prime insightful thought is by manipulating mood. A number
of studies have shown that a positive mood facilitates creative thought, while
a negative mood facilitates analytic thought. The effect of positive mood has
been localized in the brain by fMRI studies of CRA problems. In the brief
preparatory phase just preceding a CRA solution by insight, a number of
brain areas are activated. An increasingly positive mood is associated with
activation of a different set of areas. These two sets of brain activations have
an area in common: the anterior cingulate, a region involved in cognitive
control and executive processes.
Prior research has shown that the anterior cingulate is involved in the
detection of cognitive conflict in the brain, that is, the detection of competing
response tendencies. Ordinarily, when conflicting tendencies are activated,
such as pressing a button on the right and pressing a button on the left
with the same hand, the anterior cingulate suppresses the weaker tendency
to let the stronger one dominate. In insightful problem solving, a weaker
tendency, that is, an obscure solution possibility or a remote association,
must be allowed to dominate. According to this idea, activation of the
anterior cingulate permits detection of weak solution ideas that might
otherwise be suppressed. Thus, positive mood sensitizes a person to remote
associations and “long-shot” ideas during the preparation phase.
THE INSIGHTFUL BRAIN AT REST
If the adoption of an insight mode of processing on a problem is influenced
by one’s pattern of brain activity immediately before problem presentation,
what determines whether one adopts an insightful or analytic pattern of
activity in this preparatory period? It has been shown that resting-state brain
activity—the pattern of neural activity that occurs when a person relaxes

10

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

with no task to perform—is influential to an individual’s tendency to prepare
to process an expected problem either insightfully or analytically. In a study
by Subramanian, participants had their resting-state activity recorded before
solving anagrams. They were then divided into a high-insight group and
high-analytic group based on the number of insightful versus analytic solutions they had produced. When the resting-state data is compared between
these two groups, the patterns of neural activity differ in several ways. Most
notably, high-insight solvers tend to have more right-hemisphere activity, as
well as greater activity in the anterior cingulate.
INSIGHT ENHANCEMENT
The discovery of neural precursors to insight suggests that it should be possible to develop a practical technology for enhancing or promoting insightful
thought. Although such work is in a very early phase, clear directions are
emerging. A straightforward example is the manipulation of mood to facilitate insight or analysis (whichever is more appropriate to the situation at
hand). Another approach has emerged from social psychology research on
mindset priming. A mindset is a theorized set of cognitive processes that can
be evoked by a situation or stimulus. Once evoked, these processes bias a person to process information in a particular way. For example, temporal construal
priming has been used to evoke a mindset conducive to insightful thought.
Thinking about an event that will take place in the distant future tends to
promote abstract thought, while thinking about near future events promotes
more concrete thinking. Research has shown that abstract thought generated through this method facilitates insight solving, while concrete thinking
facilitates analysis. Other work has shown the effectiveness of counterfactual
mindset priming, in which a person must consider a series of hypothetical
statements constructed by changing one or more elements of a scenario, for
example, “If I had brought my umbrella today, then I would not have gotten wet.” Thinking about such counterfactual statements has been shown to
prime insight.
A more dramatic approach to insight enhancement is by direct stimulation
of the brain. Two recent studies by Chi and Snyder used transcranial direct
current stimulation (tDCS) while participants attempted to solve insight
problems. tDCS is a very weak DC current that travels through the scalp and
skull and across the brain. When this mild electrical stimulation was applied
over right frontal-temporal cortex (slightly anterior to the aSTG), it increased
solution rates to these problems. In a particularly striking demonstration,
this pattern of stimulation increased solution rates for the classic and very
difficult nine-dot problem from 0% to 40%. These results are consistent with

Insight

11

visual half-field and neuroimaging findings pointing to a key role for the
right hemisphere in insight.
Although promising, research on the effects of neurostimulation on insight
is still in a very early phase. Significant questions remain. For example,
while the tDCS studies by Chi and Snyder showed increased solution
rates for insight problems with right-hemisphere stimulation, they did
not verify that their subjects actually solved the problems with insight.
Of course, for practical purposes the important thing is that a problem is
solved, however this occurs. Nevertheless, further development of this
approach will depend on an understanding of the cognitive mechanisms
involved. It is also unclear what was altered by tDCS at the level of neural
circuitry and functional neuroanatomy. Chi and Snyder’s neurostimulation protocol did not permit them to infer whether the obtained insight
enhancement was due to right-hemisphere stimulation or left-hemisphere
inhibition (because of their placement of the reference electrode). Additional
studies will undoubtedly clarify these effects, especially by combining
tDCS with EEG to ascertain the effects of the neurostimulation protocol
on brain activity. If insight enhancement through neurostimulation is
eventually proved to be practical and reliable, then there will be potential
for widespread real-world applications in education, business, psychotherapy, scientific research, or any other field in which problems must be
solved.
Other neurotechnologies for insight enhancement have yet to be systematically investigated. For example, to date, pharmacological approaches have
received virtually no attention. Neurofeedback, also known as EEG biofeedback, may help to modulate the neural processes subserving insightful and
analytic thought.
CONCLUSIONS
After a spurt of cognitive psychology research on insight during the 1980s,
there was a period of little progress. During the past decade, cognitive
neuroscience approaches have reinvigorated insight research by revealing
the outlines of its functional neuroanatomy and by elucidating the involvement of hitherto unsuspected cognitive processes, such as sensory gating
immediately before insight. This recent progress has stimulated promising
work aimed at facilitating insight. This work will ultimately spur additional
research that attempts to identify entire networks of brain regions whose
cooperative efforts make insight possible.
Another problem that will undoubtedly receive great attention in the coming years is the issue of how to define the relationship between insight and
the larger domain of creativity. This relationship has not yet been clearly

12

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

delineated, largely because these two concepts have themselves not yet been
clearly defined. However, recent progress in insight research bodes well for
creativity research. As neuroscience methods help to clarify the nature of creativity, as they have begun to do for insight, the relationship between insight
and creativity will become clearer.
In sum, there is reason for great optimism concerning the future of insight
research and for the future of insight itself. Insight is a powerful human
ability. Understanding and harnessing it has the potential to be a disruptive
development in human history that will contribute to our understanding of
what it is to be human.
FURTHER READING
Beeman, M. J., & Bowden, E. M. (2000). The right hemisphere maintains solutionrelated activation for yet-to-be-solved problems. Memory & Cognition, 28, 1231–
1241. doi:10.3758/BF03211823
Bowden, E. M., & Beeman, M. J. (1998). Getting the right idea: Semantic activation
in the right hemisphere may help solve insight problems. Psychological Science, 9,
435–440. doi:10.1111/1467-9280.00082
Chi, R. P., & Snyder, A. W. (2012). Brain stimulation enables the solution of an
inherently difficult problem. Neuroscience Letters, 515, 121. doi:10.1016/j.neulet.
2012.03.012
Holyoak, K. J. & Morrison, R. G. (Eds.) (2012). The Oxford handbook of thinking and
reasoning. New York, NY: Oxford University Press.
Kounios, J., & Beeman, M. (2009). The Aha! moment: The cognitive neuroscience
of insight. Current Directions in Psychological Science, 18, 210–216. doi:10.1111/
j.1467-8721.2009.01638.x
Luo, J., & Knoblich, G. (2007). Studying insight problem solving with neuroscientific
methods. Methods, 42, 77–86. doi:10.1016/j.ymeth.2006.12.005
Metcalfe, J., & Wiebe, D. (1987). Intuition in insight and noninsight problem solving.
Memory & Cognition, 15, 238–246. doi:10.3758/BF03197722
Sawyer, K. (2011). The cognitive neuroscience of creativity: A critical review. Creativity Research Journal, 23, 137–154. doi:10.1080/10400419.2011.571191
Smith, R. W., & Kounios, J. (1996). Sudden insight: All-or-none processing revealed
by speed accuracy decomposition. Journal of Experimental Psychology: Learning,
Memory, and Cognition, 22, 1443–1462. doi:10.1037/0278-7393.22.6.1443
Sternberg, R. J. & Davidson, J. E. (Eds.) (1995). The nature of insight. Cambridge, MA:
MIT Press.
Subramaniam, K., Kounios, J., Parrish, T. B., & Jung-Beeman, M. (2009). A brain
mechanism for facilitation of insight by positive affect. Journal of Cognitive Neuroscience, 21, 415–432. doi:10.1162/jocn.2009.21057

Insight

13

BRIAN ERICKSON SHORT BIOGRAPHY
Brian Erickson received his BS in mechanical engineering and MA in neural engineering from Drexel University. He is currently a doctoral student in
Drexel’s Applied Brain and Cognitive Sciences (ACBS) program studying the
basis of creativity and insight in the brain. Erickson’s current research focus
is the identification of stable neural profiles that distinguish insightful versus
analytic solvers by examining differences in resting-state electroencephalograms. He is also investigating the use of transcranial direct current brain
stimulation (tDCS) to facilitate insight.
Personal Site: https://sites.google.com/site/ericksongrad/
JOHN KOUNIOS SHORT BIOGRAPHY
John Kounios is a Professor of Psychology at Drexel University. He received
his BA from Haverford College and PhD in experimental psychology from
the University of Michigan. Kounios has held research and faculty positions
at Princeton University, Tufts University, the Boston Veterans Affairs Medical
Center, and the University of Pennsylvania. He has published on a variety of
topics in cognitive psychology and cognitive neuroscience. Kounios’ current
research focus is the neural basis of insight and creativity. His research has
been funded by grants from the National Science Foundation, the National
Institute of Mental Health, the National Institute of Deafness and Other Communication Disorders, and the National Institute of Aging. His research has
been reported by The New Yorker, The New York Times, The Wall Street Journal,
National Public Radio, and other print and electronic media.
Personal site: https://sites.google.com/site/johnkounios/
RELATED ESSAYS
Models of Revealed Preference (Economics), Abi Adams and Ian Crawford
To Flop Is Human: Inventing Better Scientific Approaches to Anticipating
Failure (Methods), Robert Boruch and Alan Ruby
Misinformation and How to Correct It (Psychology), John Cook et al.
Four Psychological Perspectives on Creativity (Psychology), Rodica Ioana
Damian and Dean Keith Simonton
Youth Entrepreneurship (Psychology), William Damon et al.
Resilience (Psychology), Erica D. Diminich and George A. Bonanno
Empathy Gaps between Helpers and Help-Seekers: Implications for Cooperation (Psychology), Vanessa K. Bohns and Francis J. Flynn
Controlling the Influence of Stereotypes on One’s Thoughts (Psychology),
Patrick S. Forscher and Patricia G. Devine

14

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Setting One’s Mind on Action: Planning Out Goal Striving in Advance
(Psychology), Peter M. Gollwitzer
Regulatory Focus Theory (Psychology), E. Tory Higgins
Social Aspects of Memory (Psychology), William Hirst and Charles B. Stone
Emotion and Decision Making (Psychology), Jeff R. Huntsinger and Cara Ray
Implicit Memory (Psychology), Dawn M. McBride
Embodied Knowledge (Psychology), Diane Pecher and René Zeelenberg
Creativity in Teams (Psychology), Leigh L. Thompson and Elizabeth Ruth
Wilson
Theory of Mind (Psychology), Henry Wellman



Insight
BRIAN ERICKSON and JOHN KOUNIOS

Abstract
Insight, also known as the Aha phenomenon, is the sudden awareness of the solution
to a problem. In contrast, analysis is problem solving by consciously and deliberately
manipulating the elements of a problem. The Gestalt psychologists began studying
insight about a century ago. On the basis of their research with complex “insight
problems,” they characterized insight as a reinterpretation or restructuring of one’s
representation of a stimulus or situation after a period of unconscious processing.
The emergence of cognitive psychology later during the twentieth century led to
another period of advancement in insight research during the 1980s and 1990s. This
work further characterized the unconscious nature of the processing leading up to
an insight. More recently, the development of techniques for measuring and manipulating brain function has sparked a new renaissance in insight research. Cognitive
neuroscience research has highlighted the key role of the right hemisphere and has
discovered a number of neural precursors to insight, including its origins in patterns of resting-state brain activity and in neural preparatory activity immediately
before a problem is presented. The latest trend is work aimed at developing techniques to enhance insight, including recent research showing that direct stimulation
of the right hemisphere can facilitate the solving of insight problems. Cognitive neuroscience approaches should continue to fuel rapid advances and may lead to the
development of practical technologies for insight enhancement.

When a person solves a problem, he or she takes a situation’s initial state
and transforms it into a goal state with whatever tools or “operators” are
available. There are two general strategies for accomplishing this transformation. Analytic thought involves deliberately, methodically, and consciously
applying these operators to effect the transformation. Insight is the sudden
awareness of the solution after a period of unconscious processing (i.e., the
“Aha” phenomenon).
HISTORICAL PERSPECTIVES
Insight has been a topic of interest since antiquity, but became a topic of scientific study only in the early twentieth century. Behaviorists such as Edward
Emerging Trends in the Social and Behavioral Sciences. Edited by Robert Scott and Stephen Kosslyn.
© 2015 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.

1

2

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Thorndike viewed learning and problem solving as a trial-and-error process by which response tendencies are gradually strengthened by rewards or
weakened by punishments. In contrast, the Gestalt psychologists recognized
that the behaviorist view is insufficient, as shown by studies that demonstrated discrete, all-or-none transformations in visual perception. Furthermore, Wolfgang K˝ohler’s studies of problem solving in apes demonstrated
what appeared to be sudden insights in problem solving. His most famous
study involved placing a bunch of bananas behind a fence out of the reach of a
chimpanzee. Two bamboo sticks were available to the chimp, but neither was
long enough to reach the bananas and pull them within reach. After a period
of frustration and inaction, the chimp spontaneously arrived at the solution,
namely, jamming one stick into the other—bamboo shafts are hollow—to
make a longer rod that could be used to reach the bananas. Importantly, the
chimpanzee had not been rewarded for intermediate steps that would have
brought him incrementally closer to achieving this solution, such as holding
both sticks at the same time or knocking them together. The use and construction of the tool apparently came to the chimpanzee suddenly and all at
once—an insight.
Gestalt psychologists went on to demonstrate similar examples of insightful problem solving in humans. Their methodology was based around a corpus of “insight problems” that typically elicit an “Aha!” experience when
they are solved. This work helped characterize insight but did not lead to
great progress in understanding its underlying mechanisms. One of the main
limitations with the Gestalt psychologists’ research on insight was that it
relied on an informal consensus regarding which problems were to be considered insight problems and which would be considered analytic problems.
Thus, they did not isolate objective behavioral or experimental correlates of
insight.
RESTRUCTURING AS INSIGHT SOLVING
Many classic insight problems are difficult to solve because they encourage
the assumption of erroneous constraints on the solution. For instance, the
very difficult nine-dot problem consists of a 3 × 3 matrix of dots (Figure 1a).
Solvers are instructed to draw four straight lines that pass through all nine
dots without lifting the pen or retracing any lines. The solution to this
problem is simple, but in many laboratory studies fewer than 5% of subjects
successfully solve it within the allotted time. Subjects fail to solve the
problem because they assume that the solution must stay inside the square
figure implied by the dots and that pivots can only be made on dots—neither
stipulation being part of the explicit instructions. When one discards these
constraints by restructuring one’s representation of the problem so that it

Insight

(a)

3

(b)

Figure 1 The nine-dot problem. (a) Subjects’ task is to draw four straight lines
connecting all 9 dots without lifting the pen from the paper and without retracing
any lines. (b) A solution to the nine-dot problem. Source: Adapted from
http://en.wikipedia.org/wiki/Thinking_outside_the_box

includes the blank background on which the nine-dot figure is presented,
the solution becomes fairly obvious (Figure 1b).
INSIGHT AND FUNCTIONAL FIXATIONS
The Gestalt psychologists’ main work was on visual perception. They
noted that any visual object or scene is ambiguous in the sense that it can
be interpreted in more than one way. One example of this is the necker
cube (Figure 2a), which can be interpreted in either of two orientations
(Figure 2b). When the viewer focuses attention on the lower square, this
square seems to be the front face of the cube; when the viewer focuses on the
upper square, that appears to be the front face. Moreover, the shift between
these two perceptual representations is sudden and discrete, as in an insight.
The Gestalt psychologists generalized from such perceptual phenomena to
problem solving. They viewed the main difficultly in solving a problem to
be that the would-be solver started with the wrong initial representation of
the problem. After restructuring, the new representation would immediately
suggest a solution that would be experienced as a sudden insight.
One type of misrepresentation, called functional fixation, is the tendency to
use objects only for their traditional functions. For example, in the classic candle problem, participants are directed to support a candle on a wall, given
only a box of tacks and a match. The fact that the box is holding the tacks
tends to discourage solvers from seeing it as a critical part of the solution: a

4

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

(a)

(b)

Figure 2 A Necker cube. (a) The Necker cube is an ambiguous figure that can be
viewed as existing in different orientations. (b) Switching between perceived
orientations is instantaneous. Source: Adapted from http://en.wikipedia.org/
wiki/Necker_cube

shelf for the candle, to be tacked to the wall. Solution rates improve, however,
when the box is presented to solvers independently of the tacks. Thus, participants tend to see the box only as a container when it is initially presented
as a container. Functional fixedness is reduced when the box is not presented
in its typical guise. This facilitates the solver’s restructuring of their idea of
the box.
THE MODERN STUDY OF INSIGHT
The emergence of modern cognitive psychology brought greater methodological and theoretical sophistication to the study of insight. For example,
Metcalfe and Wiebe were among the first to identify an objective difference
between insight and analytic problem solving. Their participants worked on
a series of insight and analytical problems while periodically rating how close
they felt to solving each problem (“warmth”). While working on analytic
problems, participants indicated that they felt gradually increasing warmth
before solving each problem. In contrast, while working on insight problems,
participants reported low warmth until just before solving the problem. Furthermore, for those few insight problems during which participants reported
gradually increasing warmth, their reported solutions were often incorrect.
Metcalfe and Wiebe’s study was groundbreaking in showing a clear
behavioral distinction between insight and analytic solving. However,
their study did not actually show that insight solutions were derived in a
discrete all-or-none manner, which is one of the defining features of insight.
Following up on their work, Smith and Kounios used a new experimental
procedure to force participants to make their best guess about a solution

Insight

5

immediately before they normally would have solved the problem. Mathematical analyses enabled them to determine whether problem solving
involves the gradual accumulation of solution information over time or
whether solution information becomes available in a single all-or-none
burst, as in insight. They used insight-like anagram problems and discovered that participants had accrued no measurable solution information
before achieving the solution, showing that solution information became
available to subjects in an all-or-none manner. This study confirmed the
sudden, discrete nature of insight, and differentiated it from prior findings
of gradual information accrual in noninsight tasks.
IMPASSE AND INCUBATION
When solvers reach a state at which they have failed to solve a problem and
cannot generate new ideas or strategies, they are said to be at impasse. Verbal
protocols in which the solver explains his or her thinking while working on
a problem have identified features of impasse before solving classic insight
problems. Results obtained with the verbal protocol technique have been corroborated by eye tracking studies. Before an insight solution, the duration
of gaze fixation (how long the solver looks at a single part of the problem)
increases. This increasing fixation means that the solver is running out of
ideas to try to solve the problem, signaling impasse. However, gaze fixation
tends to increase most on features of the problem that are relevant to the solution, indicating that solvers may unconsciously be collecting the information
necessary for restructuring.
Why is impasse frequently linked to insight solving? One theory is that
impasse often leads solvers to take a break from the problem, which allows
incubation. Incubation breaks have been shown to increase the solution rate
for classic insight problems. This can happen because time away from a problem can allow selective forgetting of incorrect strategies and assumptions,
making retrieval of the correct solution more likely. Incubation by selective
forgetting is supported by evidence that for incubation to impart beneficial
effects, problems must be completely removed from view. Furthermore,
incubation is helpful only when the original problem encourages spurious
assumptions and is most successful when the solver becomes engaged in
another task.
Another potential mechanism for incubation during a break is that solvers
might continue to unconsciously work on the problem; however, incubation
of essentially any length appears equally effective in boosting solution, which
would not be expected if complex unconscious processing were the cause.
Nevertheless, some form of unconscious processing may facilitate insight in
some situations.

6

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

NEW METHODOLOGIES FOR STUDYING INSIGHT
The body of research on the classic nine-dot, candle, and related problems
has a distinct limitation: Such “insight problems” were assumed to require
insight for their solution, but lacked any empirical support for this assumption. Furthermore, work comparing the solving of such “insight” and “analytic” problems makes a related assumption, namely, that these two classes
of problems differ only in the processing strategy that subjects use to solve
them. While the former assumption may eventually be shown to be generally
valid for classic insight problems, the latter is clearly not. Insight and analytic
problems typically differ from each other in a number of ways, such as their
complexity, familiarity, visual versus verbal content, and so forth. Therefore,
observed differences in how people solve these problems cannot be entirely
attributed to the use of an insight or an analytic strategy and may, in fact, be
due to these ancillary factors.
To circumvent these limitations, Edward Bowden and colleagues developed a new methodology for studying insight. They adapted one type of
problem originally developed for a test of creativity called the Remote Associates Test. Each of these compound remote associates (CRA) problems consists
of three words (e.g., pine, sauce, crab). The solver must generate a fourth word
that can be combined with each of the problem words to form a compound
or familiar phrase (apple: pineapple, applesauce, crabapple).
Solving a CRA problem requires a solver to access weak, remote associations of the problem words. For example, the word pine strongly evokes
the association tree, sauce evokes tomato, and crab, seafood. To find the solution word, the solver must retrieve the word apple, which is weakly associated with each of the problem words. This is thought to involve a type of
restructuring similar to what solvers must accomplish while tackling a classic
insight problem.
An interesting feature of CRAs is that they are neither insight nor analytic
problems. After participants solve a CRA, they are asked to report whether
they solved it in a deliberate, methodical (i.e., analytical) manner, or whether
they had solved it with insight (i.e., an “Aha”), a distinction with which participants are familiar. Virtually all participants report at least a few insight
solutions and a few analytic solutions. On average, about half of the solutions
are of each type, although participants exhibit substantial variation about
this mean.
This approach yields important benefits for insight research. Most importantly, insight and analytic solutions can be directly compared because the
problems that evoked these two types of solutions do not differ in complexity,
length, or any of the other factors that ordinarily distinguish classic insight
problems from analytic ones. Second, because these problems are short and

Insight

7

yield solutions within a few seconds, participants can solve many of them
within a single session. This opens up the possibility of using these problems
for neuroimaging studies of insight, because all neuroimaging techniques
require multiple trials to yield the necessary signal-to-noise ratio.
THE HEMISPHERIC BASIS OF INSIGHT
Research has associated creative cognition with the brain’s right cerebral
hemisphere. In particular, the solution of CRA problems with insight is
supported by visual hemifield studies. In these studies, a stimulus (such
as a word) is directly presented to either the right or left hemisphere by
presenting a stimulus to the left or right of a fixation point, respectively. In
several experiments by Beeman and Bowden (2000) CRAs were presented
with a deadline short enough that participants often failed to solve them.
Immediately upon reaching the deadline, either the solution word or an
irrelevant word was presented to the left visual field (which projects to
the right hemisphere) or the right visual field (which projects to the left
hemisphere). Participants were instructed to vocalize this probe word as
quickly as possible. They pronounced solution words more quickly than
irrelevant words. They also pronounced solution words more quickly when
they were presented to the right hemisphere. Furthermore, solution words
that were presented to the right hemisphere which also elicited an “Aha”
feeling of recognition were pronounced more quickly still. Such results
support the notion that insights are preceded by unconscious processing in
which the solution word is weakly represented in the right hemisphere.
On the basis of visual hemifield findings and a variety of other types of
research, Beeman proposed that the brain’s hemispheres process semantic
information differently. The left hemisphere engages in fine semantic coding
in which each word strongly activates a small number of closely related
associates, while the right hemisphere engages in coarse semantic coding
in which a word weakly activates a relatively large number of remote
associates. Insight and related forms of creative cognition primarily rely on
right hemisphere coarse semantic coding that allows a person to access the
nondominant meanings necessary for problem restructuring.
THE NEUROIMAGING OF INSIGHT
The first neuroimaging study to isolate insight was done by Beeman,
Kounios, and colleagues. This study used functional magnetic resonance
imaging (fMRI) and electroencephalography (EEG) to record participants’
neural activity while they solved CRA problems. fMRI provides excellent
spatial resolution but only modest temporal resolution, so it is the tool of

8

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

choice for localizing brain areas that are active during a cognitive task,
although it gives less precise information about the timing of this activity. In
contrast, EEG affords excellent temporal resolution, but only modest spatial
resolution, so it is ideally suited to delineating the timing of neural events.
Together, fMRI and EEG provide powerful localization of neural activity in
both space and time.
This work found that the moment of solving a CRA problem by insight
(but not analytic thought) was accompanied by a burst of high-frequency
(“gamma-band”) EEG activity measured over the right temporal lobe. The
timing of this activity coincided with participants’ awareness of the solution.
fMRI localized this activity to a brain structure called the right anterior superior
temporal gyrus (aSTG). Prior research supports the involvement of the right
aSTG region in integration of remotely associated pieces of semantic information, as occurs in jokes, metaphors, and other types of figurative language
processing.
This study revealed an additional phenomenon associated with insight
solutions: Immediately before the burst of high-frequency EEG activity
coincident with the sudden awareness of the solution, there was a burst of
lower frequency “alpha-band” activity measured over the right posterior
cortex. EEG alpha waves are generally associated with inhibition of neural
activity and, when measured over the visual cortex, are understood to reflect
inhibition of visual inputs to the brain. The researchers’ interpretation of this
finding was that the brain reduces sensory inputs briefly in order to increase
the signal-to-noise ratio of the weakly activated solution represented in the
right temporal cortex. This enables retrieval of the solution into awareness,
which is accomplished as a sudden insight.
The discovery of a neural correlate of insight and its immediate precursor suggested a research strategy of finding additional precursors. This is
accomplished by tracing neural activity back in time starting at the moment
of insight. This is important for two reasons. First, the discovery that insight
and analytic solving involve different cognitive strategies and patterns of
neural activity raises the question of what factors determine which of these
strategies are applied toward a given problem. Second, if there is a sequence
of neural precursors to insight and analytic solving, this raises the possibility that each of these precursors can potentially be influenced to change the
cognitive strategy that a person uses to solve a problem.
THE PREPARED MIND
Louis Pasteur once said, “Chance favors only the prepared mind.” To look
for a neural basis for Pasteur’s claim, Kounios and Beeman examined one
type of neural precursor of insight—brain activity immediately before the

Insight

9

presentation of each CRA problem. They hypothesized that a person prepares for or anticipates solving an upcoming problem by adopting a pattern
of neural activity that will promote either an insightful or analytic solution.
They found that preparation for solving an upcoming problem with insight
involves (relative to preparation for analytic solving) greater activity in the
anterior cingulate and in the right and left temporal lobes. By contrast, preparation for analytic solving involves increased activity in the visual cortex. The
general interpretation of these results was that preparation for analytic solving involves outward focus of attention on the screen on which the problem is
about to be displayed, while preparation for insight involves inward focus of
attention and priming of brain areas involved in processing words and concepts. The isolation of different brain states corresponding to insightful and
analytic modes of thought further supports the idea that cognitive solving
strategies can be systematically primed.
One way to prime insightful thought is by manipulating mood. A number
of studies have shown that a positive mood facilitates creative thought, while
a negative mood facilitates analytic thought. The effect of positive mood has
been localized in the brain by fMRI studies of CRA problems. In the brief
preparatory phase just preceding a CRA solution by insight, a number of
brain areas are activated. An increasingly positive mood is associated with
activation of a different set of areas. These two sets of brain activations have
an area in common: the anterior cingulate, a region involved in cognitive
control and executive processes.
Prior research has shown that the anterior cingulate is involved in the
detection of cognitive conflict in the brain, that is, the detection of competing
response tendencies. Ordinarily, when conflicting tendencies are activated,
such as pressing a button on the right and pressing a button on the left
with the same hand, the anterior cingulate suppresses the weaker tendency
to let the stronger one dominate. In insightful problem solving, a weaker
tendency, that is, an obscure solution possibility or a remote association,
must be allowed to dominate. According to this idea, activation of the
anterior cingulate permits detection of weak solution ideas that might
otherwise be suppressed. Thus, positive mood sensitizes a person to remote
associations and “long-shot” ideas during the preparation phase.
THE INSIGHTFUL BRAIN AT REST
If the adoption of an insight mode of processing on a problem is influenced
by one’s pattern of brain activity immediately before problem presentation,
what determines whether one adopts an insightful or analytic pattern of
activity in this preparatory period? It has been shown that resting-state brain
activity—the pattern of neural activity that occurs when a person relaxes

10

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

with no task to perform—is influential to an individual’s tendency to prepare
to process an expected problem either insightfully or analytically. In a study
by Subramanian, participants had their resting-state activity recorded before
solving anagrams. They were then divided into a high-insight group and
high-analytic group based on the number of insightful versus analytic solutions they had produced. When the resting-state data is compared between
these two groups, the patterns of neural activity differ in several ways. Most
notably, high-insight solvers tend to have more right-hemisphere activity, as
well as greater activity in the anterior cingulate.
INSIGHT ENHANCEMENT
The discovery of neural precursors to insight suggests that it should be possible to develop a practical technology for enhancing or promoting insightful
thought. Although such work is in a very early phase, clear directions are
emerging. A straightforward example is the manipulation of mood to facilitate insight or analysis (whichever is more appropriate to the situation at
hand). Another approach has emerged from social psychology research on
mindset priming. A mindset is a theorized set of cognitive processes that can
be evoked by a situation or stimulus. Once evoked, these processes bias a person to process information in a particular way. For example, temporal construal
priming has been used to evoke a mindset conducive to insightful thought.
Thinking about an event that will take place in the distant future tends to
promote abstract thought, while thinking about near future events promotes
more concrete thinking. Research has shown that abstract thought generated through this method facilitates insight solving, while concrete thinking
facilitates analysis. Other work has shown the effectiveness of counterfactual
mindset priming, in which a person must consider a series of hypothetical
statements constructed by changing one or more elements of a scenario, for
example, “If I had brought my umbrella today, then I would not have gotten wet.” Thinking about such counterfactual statements has been shown to
prime insight.
A more dramatic approach to insight enhancement is by direct stimulation
of the brain. Two recent studies by Chi and Snyder used transcranial direct
current stimulation (tDCS) while participants attempted to solve insight
problems. tDCS is a very weak DC current that travels through the scalp and
skull and across the brain. When this mild electrical stimulation was applied
over right frontal-temporal cortex (slightly anterior to the aSTG), it increased
solution rates to these problems. In a particularly striking demonstration,
this pattern of stimulation increased solution rates for the classic and very
difficult nine-dot problem from 0% to 40%. These results are consistent with

Insight

11

visual half-field and neuroimaging findings pointing to a key role for the
right hemisphere in insight.
Although promising, research on the effects of neurostimulation on insight
is still in a very early phase. Significant questions remain. For example,
while the tDCS studies by Chi and Snyder showed increased solution
rates for insight problems with right-hemisphere stimulation, they did
not verify that their subjects actually solved the problems with insight.
Of course, for practical purposes the important thing is that a problem is
solved, however this occurs. Nevertheless, further development of this
approach will depend on an understanding of the cognitive mechanisms
involved. It is also unclear what was altered by tDCS at the level of neural
circuitry and functional neuroanatomy. Chi and Snyder’s neurostimulation protocol did not permit them to infer whether the obtained insight
enhancement was due to right-hemisphere stimulation or left-hemisphere
inhibition (because of their placement of the reference electrode). Additional
studies will undoubtedly clarify these effects, especially by combining
tDCS with EEG to ascertain the effects of the neurostimulation protocol
on brain activity. If insight enhancement through neurostimulation is
eventually proved to be practical and reliable, then there will be potential
for widespread real-world applications in education, business, psychotherapy, scientific research, or any other field in which problems must be
solved.
Other neurotechnologies for insight enhancement have yet to be systematically investigated. For example, to date, pharmacological approaches have
received virtually no attention. Neurofeedback, also known as EEG biofeedback, may help to modulate the neural processes subserving insightful and
analytic thought.
CONCLUSIONS
After a spurt of cognitive psychology research on insight during the 1980s,
there was a period of little progress. During the past decade, cognitive
neuroscience approaches have reinvigorated insight research by revealing
the outlines of its functional neuroanatomy and by elucidating the involvement of hitherto unsuspected cognitive processes, such as sensory gating
immediately before insight. This recent progress has stimulated promising
work aimed at facilitating insight. This work will ultimately spur additional
research that attempts to identify entire networks of brain regions whose
cooperative efforts make insight possible.
Another problem that will undoubtedly receive great attention in the coming years is the issue of how to define the relationship between insight and
the larger domain of creativity. This relationship has not yet been clearly

12

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

delineated, largely because these two concepts have themselves not yet been
clearly defined. However, recent progress in insight research bodes well for
creativity research. As neuroscience methods help to clarify the nature of creativity, as they have begun to do for insight, the relationship between insight
and creativity will become clearer.
In sum, there is reason for great optimism concerning the future of insight
research and for the future of insight itself. Insight is a powerful human
ability. Understanding and harnessing it has the potential to be a disruptive
development in human history that will contribute to our understanding of
what it is to be human.
FURTHER READING
Beeman, M. J., & Bowden, E. M. (2000). The right hemisphere maintains solutionrelated activation for yet-to-be-solved problems. Memory & Cognition, 28, 1231–
1241. doi:10.3758/BF03211823
Bowden, E. M., & Beeman, M. J. (1998). Getting the right idea: Semantic activation
in the right hemisphere may help solve insight problems. Psychological Science, 9,
435–440. doi:10.1111/1467-9280.00082
Chi, R. P., & Snyder, A. W. (2012). Brain stimulation enables the solution of an
inherently difficult problem. Neuroscience Letters, 515, 121. doi:10.1016/j.neulet.
2012.03.012
Holyoak, K. J. & Morrison, R. G. (Eds.) (2012). The Oxford handbook of thinking and
reasoning. New York, NY: Oxford University Press.
Kounios, J., & Beeman, M. (2009). The Aha! moment: The cognitive neuroscience
of insight. Current Directions in Psychological Science, 18, 210–216. doi:10.1111/
j.1467-8721.2009.01638.x
Luo, J., & Knoblich, G. (2007). Studying insight problem solving with neuroscientific
methods. Methods, 42, 77–86. doi:10.1016/j.ymeth.2006.12.005
Metcalfe, J., & Wiebe, D. (1987). Intuition in insight and noninsight problem solving.
Memory & Cognition, 15, 238–246. doi:10.3758/BF03197722
Sawyer, K. (2011). The cognitive neuroscience of creativity: A critical review. Creativity Research Journal, 23, 137–154. doi:10.1080/10400419.2011.571191
Smith, R. W., & Kounios, J. (1996). Sudden insight: All-or-none processing revealed
by speed accuracy decomposition. Journal of Experimental Psychology: Learning,
Memory, and Cognition, 22, 1443–1462. doi:10.1037/0278-7393.22.6.1443
Sternberg, R. J. & Davidson, J. E. (Eds.) (1995). The nature of insight. Cambridge, MA:
MIT Press.
Subramaniam, K., Kounios, J., Parrish, T. B., & Jung-Beeman, M. (2009). A brain
mechanism for facilitation of insight by positive affect. Journal of Cognitive Neuroscience, 21, 415–432. doi:10.1162/jocn.2009.21057

Insight

13

BRIAN ERICKSON SHORT BIOGRAPHY
Brian Erickson received his BS in mechanical engineering and MA in neural engineering from Drexel University. He is currently a doctoral student in
Drexel’s Applied Brain and Cognitive Sciences (ACBS) program studying the
basis of creativity and insight in the brain. Erickson’s current research focus
is the identification of stable neural profiles that distinguish insightful versus
analytic solvers by examining differences in resting-state electroencephalograms. He is also investigating the use of transcranial direct current brain
stimulation (tDCS) to facilitate insight.
Personal Site: https://sites.google.com/site/ericksongrad/
JOHN KOUNIOS SHORT BIOGRAPHY
John Kounios is a Professor of Psychology at Drexel University. He received
his BA from Haverford College and PhD in experimental psychology from
the University of Michigan. Kounios has held research and faculty positions
at Princeton University, Tufts University, the Boston Veterans Affairs Medical
Center, and the University of Pennsylvania. He has published on a variety of
topics in cognitive psychology and cognitive neuroscience. Kounios’ current
research focus is the neural basis of insight and creativity. His research has
been funded by grants from the National Science Foundation, the National
Institute of Mental Health, the National Institute of Deafness and Other Communication Disorders, and the National Institute of Aging. His research has
been reported by The New Yorker, The New York Times, The Wall Street Journal,
National Public Radio, and other print and electronic media.
Personal site: https://sites.google.com/site/johnkounios/
RELATED ESSAYS
Models of Revealed Preference (Economics), Abi Adams and Ian Crawford
To Flop Is Human: Inventing Better Scientific Approaches to Anticipating
Failure (Methods), Robert Boruch and Alan Ruby
Misinformation and How to Correct It (Psychology), John Cook et al.
Four Psychological Perspectives on Creativity (Psychology), Rodica Ioana
Damian and Dean Keith Simonton
Youth Entrepreneurship (Psychology), William Damon et al.
Resilience (Psychology), Erica D. Diminich and George A. Bonanno
Empathy Gaps between Helpers and Help-Seekers: Implications for Cooperation (Psychology), Vanessa K. Bohns and Francis J. Flynn
Controlling the Influence of Stereotypes on One’s Thoughts (Psychology),
Patrick S. Forscher and Patricia G. Devine

14

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Setting One’s Mind on Action: Planning Out Goal Striving in Advance
(Psychology), Peter M. Gollwitzer
Regulatory Focus Theory (Psychology), E. Tory Higgins
Social Aspects of Memory (Psychology), William Hirst and Charles B. Stone
Emotion and Decision Making (Psychology), Jeff R. Huntsinger and Cara Ray
Implicit Memory (Psychology), Dawn M. McBride
Embodied Knowledge (Psychology), Diane Pecher and René Zeelenberg
Creativity in Teams (Psychology), Leigh L. Thompson and Elizabeth Ruth
Wilson
Theory of Mind (Psychology), Henry Wellman


Insight
BRIAN ERICKSON and JOHN KOUNIOS

Abstract
Insight, also known as the Aha phenomenon, is the sudden awareness of the solution
to a problem. In contrast, analysis is problem solving by consciously and deliberately
manipulating the elements of a problem. The Gestalt psychologists began studying
insight about a century ago. On the basis of their research with complex “insight
problems,” they characterized insight as a reinterpretation or restructuring of one’s
representation of a stimulus or situation after a period of unconscious processing.
The emergence of cognitive psychology later during the twentieth century led to
another period of advancement in insight research during the 1980s and 1990s. This
work further characterized the unconscious nature of the processing leading up to
an insight. More recently, the development of techniques for measuring and manipulating brain function has sparked a new renaissance in insight research. Cognitive
neuroscience research has highlighted the key role of the right hemisphere and has
discovered a number of neural precursors to insight, including its origins in patterns of resting-state brain activity and in neural preparatory activity immediately
before a problem is presented. The latest trend is work aimed at developing techniques to enhance insight, including recent research showing that direct stimulation
of the right hemisphere can facilitate the solving of insight problems. Cognitive neuroscience approaches should continue to fuel rapid advances and may lead to the
development of practical technologies for insight enhancement.

When a person solves a problem, he or she takes a situation’s initial state
and transforms it into a goal state with whatever tools or “operators” are
available. There are two general strategies for accomplishing this transformation. Analytic thought involves deliberately, methodically, and consciously
applying these operators to effect the transformation. Insight is the sudden
awareness of the solution after a period of unconscious processing (i.e., the
“Aha” phenomenon).
HISTORICAL PERSPECTIVES
Insight has been a topic of interest since antiquity, but became a topic of scientific study only in the early twentieth century. Behaviorists such as Edward
Emerging Trends in the Social and Behavioral Sciences. Edited by Robert Scott and Stephen Kosslyn.
© 2015 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.

1

2

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Thorndike viewed learning and problem solving as a trial-and-error process by which response tendencies are gradually strengthened by rewards or
weakened by punishments. In contrast, the Gestalt psychologists recognized
that the behaviorist view is insufficient, as shown by studies that demonstrated discrete, all-or-none transformations in visual perception. Furthermore, Wolfgang Kőhler’s studies of problem solving in apes demonstrated
what appeared to be sudden insights in problem solving. His most famous
study involved placing a bunch of bananas behind a fence out of the reach of a
chimpanzee. Two bamboo sticks were available to the chimp, but neither was
long enough to reach the bananas and pull them within reach. After a period
of frustration and inaction, the chimp spontaneously arrived at the solution,
namely, jamming one stick into the other—bamboo shafts are hollow—to
make a longer rod that could be used to reach the bananas. Importantly, the
chimpanzee had not been rewarded for intermediate steps that would have
brought him incrementally closer to achieving this solution, such as holding
both sticks at the same time or knocking them together. The use and construction of the tool apparently came to the chimpanzee suddenly and all at
once—an insight.
Gestalt psychologists went on to demonstrate similar examples of insightful problem solving in humans. Their methodology was based around a corpus of “insight problems” that typically elicit an “Aha!” experience when
they are solved. This work helped characterize insight but did not lead to
great progress in understanding its underlying mechanisms. One of the main
limitations with the Gestalt psychologists’ research on insight was that it
relied on an informal consensus regarding which problems were to be considered insight problems and which would be considered analytic problems.
Thus, they did not isolate objective behavioral or experimental correlates of
insight.
RESTRUCTURING AS INSIGHT SOLVING
Many classic insight problems are difficult to solve because they encourage
the assumption of erroneous constraints on the solution. For instance, the
very difficult nine-dot problem consists of a 3 × 3 matrix of dots (Figure 1a).
Solvers are instructed to draw four straight lines that pass through all nine
dots without lifting the pen or retracing any lines. The solution to this
problem is simple, but in many laboratory studies fewer than 5% of subjects
successfully solve it within the allotted time. Subjects fail to solve the
problem because they assume that the solution must stay inside the square
figure implied by the dots and that pivots can only be made on dots—neither
stipulation being part of the explicit instructions. When one discards these
constraints by restructuring one’s representation of the problem so that it

Insight

(a)

3

(b)

Figure 1 The nine-dot problem. (a) Subjects’ task is to draw four straight lines
connecting all 9 dots without lifting the pen from the paper and without retracing
any lines. (b) A solution to the nine-dot problem. Source: Adapted from
http://en.wikipedia.org/wiki/Thinking_outside_the_box

includes the blank background on which the nine-dot figure is presented,
the solution becomes fairly obvious (Figure 1b).
INSIGHT AND FUNCTIONAL FIXATIONS
The Gestalt psychologists’ main work was on visual perception. They
noted that any visual object or scene is ambiguous in the sense that it can
be interpreted in more than one way. One example of this is the necker
cube (Figure 2a), which can be interpreted in either of two orientations
(Figure 2b). When the viewer focuses attention on the lower square, this
square seems to be the front face of the cube; when the viewer focuses on the
upper square, that appears to be the front face. Moreover, the shift between
these two perceptual representations is sudden and discrete, as in an insight.
The Gestalt psychologists generalized from such perceptual phenomena to
problem solving. They viewed the main difficultly in solving a problem to
be that the would-be solver started with the wrong initial representation of
the problem. After restructuring, the new representation would immediately
suggest a solution that would be experienced as a sudden insight.
One type of misrepresentation, called functional fixation, is the tendency to
use objects only for their traditional functions. For example, in the classic candle problem, participants are directed to support a candle on a wall, given
only a box of tacks and a match. The fact that the box is holding the tacks
tends to discourage solvers from seeing it as a critical part of the solution: a

4

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

(a)

(b)

Figure 2 A Necker cube. (a) The Necker cube is an ambiguous figure that can be
viewed as existing in different orientations. (b) Switching between perceived
orientations is instantaneous. Source: Adapted from http://en.wikipedia.org/
wiki/Necker_cube

shelf for the candle, to be tacked to the wall. Solution rates improve, however,
when the box is presented to solvers independently of the tacks. Thus, participants tend to see the box only as a container when it is initially presented
as a container. Functional fixedness is reduced when the box is not presented
in its typical guise. This facilitates the solver’s restructuring of their idea of
the box.
THE MODERN STUDY OF INSIGHT
The emergence of modern cognitive psychology brought greater methodological and theoretical sophistication to the study of insight. For example,
Metcalfe and Wiebe were among the first to identify an objective difference
between insight and analytic problem solving. Their participants worked on
a series of insight and analytical problems while periodically rating how close
they felt to solving each problem (“warmth”). While working on analytic
problems, participants indicated that they felt gradually increasing warmth
before solving each problem. In contrast, while working on insight problems,
participants reported low warmth until just before solving the problem. Furthermore, for those few insight problems during which participants reported
gradually increasing warmth, their reported solutions were often incorrect.
Metcalfe and Wiebe’s study was groundbreaking in showing a clear
behavioral distinction between insight and analytic solving. However,
their study did not actually show that insight solutions were derived in a
discrete all-or-none manner, which is one of the defining features of insight.
Following up on their work, Smith and Kounios used a new experimental
procedure to force participants to make their best guess about a solution

Insight

5

immediately before they normally would have solved the problem. Mathematical analyses enabled them to determine whether problem solving
involves the gradual accumulation of solution information over time or
whether solution information becomes available in a single all-or-none
burst, as in insight. They used insight-like anagram problems and discovered that participants had accrued no measurable solution information
before achieving the solution, showing that solution information became
available to subjects in an all-or-none manner. This study confirmed the
sudden, discrete nature of insight, and differentiated it from prior findings
of gradual information accrual in noninsight tasks.
IMPASSE AND INCUBATION
When solvers reach a state at which they have failed to solve a problem and
cannot generate new ideas or strategies, they are said to be at impasse. Verbal
protocols in which the solver explains his or her thinking while working on
a problem have identified features of impasse before solving classic insight
problems. Results obtained with the verbal protocol technique have been corroborated by eye tracking studies. Before an insight solution, the duration
of gaze fixation (how long the solver looks at a single part of the problem)
increases. This increasing fixation means that the solver is running out of
ideas to try to solve the problem, signaling impasse. However, gaze fixation
tends to increase most on features of the problem that are relevant to the solution, indicating that solvers may unconsciously be collecting the information
necessary for restructuring.
Why is impasse frequently linked to insight solving? One theory is that
impasse often leads solvers to take a break from the problem, which allows
incubation. Incubation breaks have been shown to increase the solution rate
for classic insight problems. This can happen because time away from a problem can allow selective forgetting of incorrect strategies and assumptions,
making retrieval of the correct solution more likely. Incubation by selective
forgetting is supported by evidence that for incubation to impart beneficial
effects, problems must be completely removed from view. Furthermore,
incubation is helpful only when the original problem encourages spurious
assumptions and is most successful when the solver becomes engaged in
another task.
Another potential mechanism for incubation during a break is that solvers
might continue to unconsciously work on the problem; however, incubation
of essentially any length appears equally effective in boosting solution, which
would not be expected if complex unconscious processing were the cause.
Nevertheless, some form of unconscious processing may facilitate insight in
some situations.

6

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

NEW METHODOLOGIES FOR STUDYING INSIGHT
The body of research on the classic nine-dot, candle, and related problems
has a distinct limitation: Such “insight problems” were assumed to require
insight for their solution, but lacked any empirical support for this assumption. Furthermore, work comparing the solving of such “insight” and “analytic” problems makes a related assumption, namely, that these two classes
of problems differ only in the processing strategy that subjects use to solve
them. While the former assumption may eventually be shown to be generally
valid for classic insight problems, the latter is clearly not. Insight and analytic
problems typically differ from each other in a number of ways, such as their
complexity, familiarity, visual versus verbal content, and so forth. Therefore,
observed differences in how people solve these problems cannot be entirely
attributed to the use of an insight or an analytic strategy and may, in fact, be
due to these ancillary factors.
To circumvent these limitations, Edward Bowden and colleagues developed a new methodology for studying insight. They adapted one type of
problem originally developed for a test of creativity called the Remote Associates Test. Each of these compound remote associates (CRA) problems consists
of three words (e.g., pine, sauce, crab). The solver must generate a fourth word
that can be combined with each of the problem words to form a compound
or familiar phrase (apple: pineapple, applesauce, crabapple).
Solving a CRA problem requires a solver to access weak, remote associations of the problem words. For example, the word pine strongly evokes
the association tree, sauce evokes tomato, and crab, seafood. To find the solution word, the solver must retrieve the word apple, which is weakly associated with each of the problem words. This is thought to involve a type of
restructuring similar to what solvers must accomplish while tackling a classic
insight problem.
An interesting feature of CRAs is that they are neither insight nor analytic
problems. After participants solve a CRA, they are asked to report whether
they solved it in a deliberate, methodical (i.e., analytical) manner, or whether
they had solved it with insight (i.e., an “Aha”), a distinction with which participants are familiar. Virtually all participants report at least a few insight
solutions and a few analytic solutions. On average, about half of the solutions
are of each type, although participants exhibit substantial variation about
this mean.
This approach yields important benefits for insight research. Most importantly, insight and analytic solutions can be directly compared because the
problems that evoked these two types of solutions do not differ in complexity,
length, or any of the other factors that ordinarily distinguish classic insight
problems from analytic ones. Second, because these problems are short and

Insight

7

yield solutions within a few seconds, participants can solve many of them
within a single session. This opens up the possibility of using these problems
for neuroimaging studies of insight, because all neuroimaging techniques
require multiple trials to yield the necessary signal-to-noise ratio.
THE HEMISPHERIC BASIS OF INSIGHT
Research has associated creative cognition with the brain’s right cerebral
hemisphere. In particular, the solution of CRA problems with insight is
supported by visual hemifield studies. In these studies, a stimulus (such
as a word) is directly presented to either the right or left hemisphere by
presenting a stimulus to the left or right of a fixation point, respectively. In
several experiments by Beeman and Bowden (2000) CRAs were presented
with a deadline short enough that participants often failed to solve them.
Immediately upon reaching the deadline, either the solution word or an
irrelevant word was presented to the left visual field (which projects to
the right hemisphere) or the right visual field (which projects to the left
hemisphere). Participants were instructed to vocalize this probe word as
quickly as possible. They pronounced solution words more quickly than
irrelevant words. They also pronounced solution words more quickly when
they were presented to the right hemisphere. Furthermore, solution words
that were presented to the right hemisphere which also elicited an “Aha”
feeling of recognition were pronounced more quickly still. Such results
support the notion that insights are preceded by unconscious processing in
which the solution word is weakly represented in the right hemisphere.
On the basis of visual hemifield findings and a variety of other types of
research, Beeman proposed that the brain’s hemispheres process semantic
information differently. The left hemisphere engages in fine semantic coding
in which each word strongly activates a small number of closely related
associates, while the right hemisphere engages in coarse semantic coding
in which a word weakly activates a relatively large number of remote
associates. Insight and related forms of creative cognition primarily rely on
right hemisphere coarse semantic coding that allows a person to access the
nondominant meanings necessary for problem restructuring.
THE NEUROIMAGING OF INSIGHT
The first neuroimaging study to isolate insight was done by Beeman,
Kounios, and colleagues. This study used functional magnetic resonance
imaging (fMRI) and electroencephalography (EEG) to record participants’
neural activity while they solved CRA problems. fMRI provides excellent
spatial resolution but only modest temporal resolution, so it is the tool of

8

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

choice for localizing brain areas that are active during a cognitive task,
although it gives less precise information about the timing of this activity. In
contrast, EEG affords excellent temporal resolution, but only modest spatial
resolution, so it is ideally suited to delineating the timing of neural events.
Together, fMRI and EEG provide powerful localization of neural activity in
both space and time.
This work found that the moment of solving a CRA problem by insight
(but not analytic thought) was accompanied by a burst of high-frequency
(“gamma-band”) EEG activity measured over the right temporal lobe. The
timing of this activity coincided with participants’ awareness of the solution.
fMRI localized this activity to a brain structure called the right anterior superior
temporal gyrus (aSTG). Prior research supports the involvement of the right
aSTG region in integration of remotely associated pieces of semantic information, as occurs in jokes, metaphors, and other types of figurative language
processing.
This study revealed an additional phenomenon associated with insight
solutions: Immediately before the burst of high-frequency EEG activity
coincident with the sudden awareness of the solution, there was a burst of
lower frequency “alpha-band” activity measured over the right posterior
cortex. EEG alpha waves are generally associated with inhibition of neural
activity and, when measured over the visual cortex, are understood to reflect
inhibition of visual inputs to the brain. The researchers’ interpretation of this
finding was that the brain reduces sensory inputs briefly in order to increase
the signal-to-noise ratio of the weakly activated solution represented in the
right temporal cortex. This enables retrieval of the solution into awareness,
which is accomplished as a sudden insight.
The discovery of a neural correlate of insight and its immediate precursor suggested a research strategy of finding additional precursors. This is
accomplished by tracing neural activity back in time starting at the moment
of insight. This is important for two reasons. First, the discovery that insight
and analytic solving involve different cognitive strategies and patterns of
neural activity raises the question of what factors determine which of these
strategies are applied toward a given problem. Second, if there is a sequence
of neural precursors to insight and analytic solving, this raises the possibility that each of these precursors can potentially be influenced to change the
cognitive strategy that a person uses to solve a problem.
THE PREPARED MIND
Louis Pasteur once said, “Chance favors only the prepared mind.” To look
for a neural basis for Pasteur’s claim, Kounios and Beeman examined one
type of neural precursor of insight—brain activity immediately before the

Insight

9

presentation of each CRA problem. They hypothesized that a person prepares for or anticipates solving an upcoming problem by adopting a pattern
of neural activity that will promote either an insightful or analytic solution.
They found that preparation for solving an upcoming problem with insight
involves (relative to preparation for analytic solving) greater activity in the
anterior cingulate and in the right and left temporal lobes. By contrast, preparation for analytic solving involves increased activity in the visual cortex. The
general interpretation of these results was that preparation for analytic solving involves outward focus of attention on the screen on which the problem is
about to be displayed, while preparation for insight involves inward focus of
attention and priming of brain areas involved in processing words and concepts. The isolation of different brain states corresponding to insightful and
analytic modes of thought further supports the idea that cognitive solving
strategies can be systematically primed.
One way to prime insightful thought is by manipulating mood. A number
of studies have shown that a positive mood facilitates creative thought, while
a negative mood facilitates analytic thought. The effect of positive mood has
been localized in the brain by fMRI studies of CRA problems. In the brief
preparatory phase just preceding a CRA solution by insight, a number of
brain areas are activated. An increasingly positive mood is associated with
activation of a different set of areas. These two sets of brain activations have
an area in common: the anterior cingulate, a region involved in cognitive
control and executive processes.
Prior research has shown that the anterior cingulate is involved in the
detection of cognitive conflict in the brain, that is, the detection of competing
response tendencies. Ordinarily, when conflicting tendencies are activated,
such as pressing a button on the right and pressing a button on the left
with the same hand, the anterior cingulate suppresses the weaker tendency
to let the stronger one dominate. In insightful problem solving, a weaker
tendency, that is, an obscure solution possibility or a remote association,
must be allowed to dominate. According to this idea, activation of the
anterior cingulate permits detection of weak solution ideas that might
otherwise be suppressed. Thus, positive mood sensitizes a person to remote
associations and “long-shot” ideas during the preparation phase.
THE INSIGHTFUL BRAIN AT REST
If the adoption of an insight mode of processing on a problem is influenced
by one’s pattern of brain activity immediately before problem presentation,
what determines whether one adopts an insightful or analytic pattern of
activity in this preparatory period? It has been shown that resting-state brain
activity—the pattern of neural activity that occurs when a person relaxes

10

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

with no task to perform—is influential to an individual’s tendency to prepare
to process an expected problem either insightfully or analytically. In a study
by Subramanian, participants had their resting-state activity recorded before
solving anagrams. They were then divided into a high-insight group and
high-analytic group based on the number of insightful versus analytic solutions they had produced. When the resting-state data is compared between
these two groups, the patterns of neural activity differ in several ways. Most
notably, high-insight solvers tend to have more right-hemisphere activity, as
well as greater activity in the anterior cingulate.
INSIGHT ENHANCEMENT
The discovery of neural precursors to insight suggests that it should be possible to develop a practical technology for enhancing or promoting insightful
thought. Although such work is in a very early phase, clear directions are
emerging. A straightforward example is the manipulation of mood to facilitate insight or analysis (whichever is more appropriate to the situation at
hand). Another approach has emerged from social psychology research on
mindset priming. A mindset is a theorized set of cognitive processes that can
be evoked by a situation or stimulus. Once evoked, these processes bias a person to process information in a particular way. For example, temporal construal
priming has been used to evoke a mindset conducive to insightful thought.
Thinking about an event that will take place in the distant future tends to
promote abstract thought, while thinking about near future events promotes
more concrete thinking. Research has shown that abstract thought generated through this method facilitates insight solving, while concrete thinking
facilitates analysis. Other work has shown the effectiveness of counterfactual
mindset priming, in which a person must consider a series of hypothetical
statements constructed by changing one or more elements of a scenario, for
example, “If I had brought my umbrella today, then I would not have gotten wet.” Thinking about such counterfactual statements has been shown to
prime insight.
A more dramatic approach to insight enhancement is by direct stimulation
of the brain. Two recent studies by Chi and Snyder used transcranial direct
current stimulation (tDCS) while participants attempted to solve insight
problems. tDCS is a very weak DC current that travels through the scalp and
skull and across the brain. When this mild electrical stimulation was applied
over right frontal-temporal cortex (slightly anterior to the aSTG), it increased
solution rates to these problems. In a particularly striking demonstration,
this pattern of stimulation increased solution rates for the classic and very
difficult nine-dot problem from 0% to 40%. These results are consistent with

Insight

11

visual half-field and neuroimaging findings pointing to a key role for the
right hemisphere in insight.
Although promising, research on the effects of neurostimulation on insight
is still in a very early phase. Significant questions remain. For example,
while the tDCS studies by Chi and Snyder showed increased solution
rates for insight problems with right-hemisphere stimulation, they did
not verify that their subjects actually solved the problems with insight.
Of course, for practical purposes the important thing is that a problem is
solved, however this occurs. Nevertheless, further development of this
approach will depend on an understanding of the cognitive mechanisms
involved. It is also unclear what was altered by tDCS at the level of neural
circuitry and functional neuroanatomy. Chi and Snyder’s neurostimulation protocol did not permit them to infer whether the obtained insight
enhancement was due to right-hemisphere stimulation or left-hemisphere
inhibition (because of their placement of the reference electrode). Additional
studies will undoubtedly clarify these effects, especially by combining
tDCS with EEG to ascertain the effects of the neurostimulation protocol
on brain activity. If insight enhancement through neurostimulation is
eventually proved to be practical and reliable, then there will be potential
for widespread real-world applications in education, business, psychotherapy, scientific research, or any other field in which problems must be
solved.
Other neurotechnologies for insight enhancement have yet to be systematically investigated. For example, to date, pharmacological approaches have
received virtually no attention. Neurofeedback, also known as EEG biofeedback, may help to modulate the neural processes subserving insightful and
analytic thought.
CONCLUSIONS
After a spurt of cognitive psychology research on insight during the 1980s,
there was a period of little progress. During the past decade, cognitive
neuroscience approaches have reinvigorated insight research by revealing
the outlines of its functional neuroanatomy and by elucidating the involvement of hitherto unsuspected cognitive processes, such as sensory gating
immediately before insight. This recent progress has stimulated promising
work aimed at facilitating insight. This work will ultimately spur additional
research that attempts to identify entire networks of brain regions whose
cooperative efforts make insight possible.
Another problem that will undoubtedly receive great attention in the coming years is the issue of how to define the relationship between insight and
the larger domain of creativity. This relationship has not yet been clearly

12

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

delineated, largely because these two concepts have themselves not yet been
clearly defined. However, recent progress in insight research bodes well for
creativity research. As neuroscience methods help to clarify the nature of creativity, as they have begun to do for insight, the relationship between insight
and creativity will become clearer.
In sum, there is reason for great optimism concerning the future of insight
research and for the future of insight itself. Insight is a powerful human
ability. Understanding and harnessing it has the potential to be a disruptive
development in human history that will contribute to our understanding of
what it is to be human.
FURTHER READING
Beeman, M. J., & Bowden, E. M. (2000). The right hemisphere maintains solutionrelated activation for yet-to-be-solved problems. Memory & Cognition, 28, 1231–
1241. doi:10.3758/BF03211823
Bowden, E. M., & Beeman, M. J. (1998). Getting the right idea: Semantic activation
in the right hemisphere may help solve insight problems. Psychological Science, 9,
435–440. doi:10.1111/1467-9280.00082
Chi, R. P., & Snyder, A. W. (2012). Brain stimulation enables the solution of an
inherently difficult problem. Neuroscience Letters, 515, 121. doi:10.1016/j.neulet.
2012.03.012
Holyoak, K. J. & Morrison, R. G. (Eds.) (2012). The Oxford handbook of thinking and
reasoning. New York, NY: Oxford University Press.
Kounios, J., & Beeman, M. (2009). The Aha! moment: The cognitive neuroscience
of insight. Current Directions in Psychological Science, 18, 210–216. doi:10.1111/
j.1467-8721.2009.01638.x
Luo, J., & Knoblich, G. (2007). Studying insight problem solving with neuroscientific
methods. Methods, 42, 77–86. doi:10.1016/j.ymeth.2006.12.005
Metcalfe, J., & Wiebe, D. (1987). Intuition in insight and noninsight problem solving.
Memory & Cognition, 15, 238–246. doi:10.3758/BF03197722
Sawyer, K. (2011). The cognitive neuroscience of creativity: A critical review. Creativity Research Journal, 23, 137–154. doi:10.1080/10400419.2011.571191
Smith, R. W., & Kounios, J. (1996). Sudden insight: All-or-none processing revealed
by speed accuracy decomposition. Journal of Experimental Psychology: Learning,
Memory, and Cognition, 22, 1443–1462. doi:10.1037/0278-7393.22.6.1443
Sternberg, R. J. & Davidson, J. E. (Eds.) (1995). The nature of insight. Cambridge, MA:
MIT Press.
Subramaniam, K., Kounios, J., Parrish, T. B., & Jung-Beeman, M. (2009). A brain
mechanism for facilitation of insight by positive affect. Journal of Cognitive Neuroscience, 21, 415–432. doi:10.1162/jocn.2009.21057

Insight

13

BRIAN ERICKSON SHORT BIOGRAPHY
Brian Erickson received his BS in mechanical engineering and MA in neural engineering from Drexel University. He is currently a doctoral student in
Drexel’s Applied Brain and Cognitive Sciences (ACBS) program studying the
basis of creativity and insight in the brain. Erickson’s current research focus
is the identification of stable neural profiles that distinguish insightful versus
analytic solvers by examining differences in resting-state electroencephalograms. He is also investigating the use of transcranial direct current brain
stimulation (tDCS) to facilitate insight.
Personal Site: https://sites.google.com/site/ericksongrad/
JOHN KOUNIOS SHORT BIOGRAPHY
John Kounios is a Professor of Psychology at Drexel University. He received
his BA from Haverford College and PhD in experimental psychology from
the University of Michigan. Kounios has held research and faculty positions
at Princeton University, Tufts University, the Boston Veterans Affairs Medical
Center, and the University of Pennsylvania. He has published on a variety of
topics in cognitive psychology and cognitive neuroscience. Kounios’ current
research focus is the neural basis of insight and creativity. His research has
been funded by grants from the National Science Foundation, the National
Institute of Mental Health, the National Institute of Deafness and Other Communication Disorders, and the National Institute of Aging. His research has
been reported by The New Yorker, The New York Times, The Wall Street Journal,
National Public Radio, and other print and electronic media.
Personal site: https://sites.google.com/site/johnkounios/
RELATED ESSAYS
Models of Revealed Preference (Economics), Abi Adams and Ian Crawford
To Flop Is Human: Inventing Better Scientific Approaches to Anticipating
Failure (Methods), Robert Boruch and Alan Ruby
Misinformation and How to Correct It (Psychology), John Cook et al.
Four Psychological Perspectives on Creativity (Psychology), Rodica Ioana
Damian and Dean Keith Simonton
Youth Entrepreneurship (Psychology), William Damon et al.
Resilience (Psychology), Erica D. Diminich and George A. Bonanno
Empathy Gaps between Helpers and Help-Seekers: Implications for Cooperation (Psychology), Vanessa K. Bohns and Francis J. Flynn
Controlling the Influence of Stereotypes on One’s Thoughts (Psychology),
Patrick S. Forscher and Patricia G. Devine

14

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Setting One’s Mind on Action: Planning Out Goal Striving in Advance
(Psychology), Peter M. Gollwitzer
Regulatory Focus Theory (Psychology), E. Tory Higgins
Social Aspects of Memory (Psychology), William Hirst and Charles B. Stone
Emotion and Decision Making (Psychology), Jeff R. Huntsinger and Cara Ray
Implicit Memory (Psychology), Dawn M. McBride
Embodied Knowledge (Psychology), Diane Pecher and René Zeelenberg
Creativity in Teams (Psychology), Leigh L. Thompson and Elizabeth Ruth
Wilson
Theory of Mind (Psychology), Henry Wellman



Media

Insight