ArticlePDF AvailableLiterature Review

The Cognitive Neuroscience of Insight

Abstract

Insight occurs when a person suddenly reinterprets a stimulus, situation, or event to produce a nonobvious, nondominant interpretation. This can take the form of a solution to a problem (an "aha moment"), comprehension of a joke or metaphor, or recognition of an ambiguous percept. Insight research began a century ago, but neuroimaging and electrophysiological techniques have been applied to its study only during the past decade. Recent work has revealed insight-related coarse semantic coding in the right hemisphere and internally focused attention preceding and during problem solving. Individual differences in the tendency to solve problems insightfully rather than in a deliberate, analytic fashion are associated with different patterns of resting-state brain activity. Recent studies have begun to apply direct brain stimulation to facilitate insight. In sum, the cognitive neuroscience of insight is an exciting new area of research with connections to fundamental neurocognitive processes.
PS65CH13-Kounios ARI 30 August 2013 16:0
R
E
V
I
E
W
S
I
N
A
D
V
A
N
C
E
The Cognitive Neuroscience
of Insight
John Kounios1and Mark Beeman2
1Department of Psychology, Drexel University, Philadelphia, Pennsylvania 19102;
email: john.kounios@gmail.com
2Department of Psychology, Northwestern University, Evanston, Illinois 60208;
email: mjungbee@northwestern.edu
Annu. Rev. Psychol. 2014. 65:13.1–13.23
The Annual Review of Psychology is online at
http://psych.annualreviews.org
This article’s doi:
10.1146/annurev-psych-010213-115154
Copyright c
2014 by Annual Reviews.
All rights reserved
Keywords
attention, cognitive enhancement, creativity, hemispheric asymmetry,
problem solving
Abstract
Insight occurs when a person suddenly reinterprets a stimulus, situation, or
event to produce a nonobvious, nondominant interpretation. This can take
the form of a solution to a problem (an “aha moment”), comprehension of a
joke or metaphor, or recognition of an ambiguous percept. Insight research
began a century ago, but neuroimaging and electrophysiological techniques
have been applied to its study only during the past decade. Recent work
has revealed insight-related coarse semantic coding in the right hemisphere
and internally focused attention preceding and during problem solving. In-
dividual differences in the tendency to solve problems insightfully rather
than in a deliberate, analytic fashion are associated with different patterns of
resting-state brain activity. Recent studies have begun to apply direct brain
stimulation to facilitate insight. In sum, the cognitive neuroscience of in-
sight is an exciting new area of research with connections to fundamental
neurocognitive processes.
13.1
PS65CH13-Kounios ARI 30 August 2013 16:0
Contents
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.2
WHAT IS INSIGHT? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.3
SCOPE OF THE REVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.4
COGNITIVE PSYCHOLOGY OF INSIGHT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.5
NEURAL BASIS OF INSIGHT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.7
Hemispheric Asymmetry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.7
Neural Correlates of Insight Solving . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.8
Preparation for Insight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13.10
Resting-State Brain Activity and Individual Differences. . . . . . . . . . . . . . . . . . . . . . . . . . .13.11
ATTENDINGIN, OUT,ANDAROUND........................................13.12
FACTORS THAT INFLUENCE THE LIKELIHOOD OF INSIGHT . . . . . . . . . . . .13.13
Mood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13.13
Mood and Attention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13.14
Models of Attention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13.15
Cognitive Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13.15
STIMULATINGINSIGHT ......................................................13.16
FUTURE DIRECTIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13.17
INTRODUCTION
In an article in the Annual Review of Astronomy and Astrophysics, William Wilson Morgan (1988)
summarized several of the groundbreaking scientific contributions he made over his long career.
One of these was the discovery of the structure of the Milky Way galaxy. What isn’t obvious from
his article is how he came to make this discovery (Sheehan 2008).
In 1951, Morgan had been calculating the distances of OB associations, which are groups of hot,
bright stars. OB associations are considered “star nurseries” because these stars are young. One
evening, he finished his work for the night and started to walk home from the Yerkes Observatory.
He glanced up at the sky to observe the stars that he had been studying and had what he called a
“flash inspiration ...a creative intuitional burst”: These stars are organized in a three-dimensional,
strand-like structure. Galaxies come in a variety of forms, but he knew that in spiral galaxies OB
associations reside in the galactic arms. Morgan understood that the strand-like form was a galactic
arm and that he had directly apprehended the spiral structure of the Milky Way, a realization that
he substantiated with data that he presented at a conference a few months later.
Morgan’s breakthrough realization was an insight, colloquially known as an “aha moment”—a
sudden, conscious change in a person’s representation of a stimulus, situation, event, or problem
(Kaplan & Simon 1990). Awareness of this kind of representational change, though abrupt, takes
place after a period of unconscious processing (van Steenburgh et al. 2012). Because insights are
largely a product of unconscious processing, when they emerge, they seem to be disconnected
from the ongoing stream of conscious thought. In contrast, analytic thought is deliberate and
conscious and is characterized by incremental awareness of a solution (Smith & Kounios 1996).
Although Morgan’s insight was literally on a cosmic scale, the phenomenon of insight, in a
more modest guise, is a common experience that occurs in perception, language comprehension,
problem solving, and other domains of cognition (van Steenburgh et al. 2012). It is therefore of
interest to ask what happened in Morgan’s brain and in the brains of many other people when they
have had an insight. This article reviews relevant cognitive neuroscience research and an emerging
13.2 Kounios ·Beeman
PS65CH13-Kounios ARI 30 August 2013 16:0
theoretical framework that is progressing toward an answer to this question. Before describing
this work, we circumscribe the insight phenomenon to specify the domain of this review.
WHAT IS INSIGHT?
Insight is often defined as a sudden change in or the formation of a concept or other type of
knowledge representation, often leading to the solution of a problem. These changes are thought
to have certain attributes. For example, insights are frequently accompanied by a burst of emotion,
including a highly positive surprise at either the content or manner of the realization. In contrast,
analytic solutions are not typically accompanied by an emotional response except perhaps for a
sense of satisfaction resulting from completing the task. However, though not an unusual con-
comitant, a conscious emotional response is not a necessary feature of insight. Participants in many
studies have solved dozens of verbal puzzles with insight (e.g., Jung-Beeman et al. 2004, Smith &
Kounios 1996) without reports of multiple bursts of emotion.
Another feature is that insights often break an impasse or mental block produced because a
solver initially fixated on an incorrect solution strategy or strong but ultimately unhelpful as-
sociations of a problem. The breaking of an impasse is accompanied by the reinterpretation or
restructuring of a problem to reveal a new, often simple, solution or solution strategy. Some re-
searchers implicitly consider problem restructuring and the breaking of an impasse to be defining
features of insight (e.g., Cranford & Moss 2012). However, this view excludes prominent types of
insights, such as those that occur (a) when the solution suddenly intrudes on a person’s awareness
when he or she is not focusing on any solution strategy, (b) when an insight pointing to a solution
occurs while a person is actively engaged in analytic processing but has not yet reached an impasse,
and (c) when a person has a spontaneous realization that does not relate to any explicitly posed
problem. We therefore do not consider the breaking of an impasse to be a precondition for insight.
Thus, there are a number of potential definitions of insight, depending on which combination
of features one selects. Very narrowly defined, insight could be thought of as a sudden solution to a
problem preceded by an impasse and problem restructuring and followed by a positive emotional
response. In contrast, the broadest definition of insight is the common nonscientific one in which an
insight is any deep realization, whether sudden or not. Within cognitive psychology and cognitive
neuroscience, inconsistency exists concerning what we consider to be a basic criterion for insight,
namely, suddenness. For example, a number of purported insight studies do not specifically isolate
and focus on solutions that occurred suddenly (e.g., Luo & Niki 2003, Wagner et al. 2004).
Another broad use of the term insight can be found in clinical psychology, in which insight
refers to self-awareness, often of one’s own symptoms, functional deficits, or other kind of predica-
ment. The clinical and nonscientific uses of the term do not require suddenness of realization or
any accompanying emotional response. Indeed, in clinical psychology, the lack of an emotional
response could itself be considered a symptom signifying a lack of insight.
The issue of defining insight is not an exercise in pedantry. When insight is defined too broadly,
it includes so many diverse, loosely related phenomena that it becomes virtually impossible for
researchers to draw general conclusions. For example, one recent review of cognitive neuroscience
research on creativity and insight lumps together widely diverse studies characterized by a variety of
definitions, assumptions, experimental paradigms, empirical phenomena, analytical methods, and
stages of the solving process (and inconsistent experimental rigor). Unsurprisingly, because of such
indiscriminate agglomeration, that review failed to find much consistency across studies, leading
the authors to pronounce a negative verdict on the field (Dietrich & Kanso 2010). In contrast,
going to the other extreme by adopting an overly narrow definition of insight can lead one to miss
important large-scale generalizations that cut across particular experimental paradigms.
www.annualreviews.org The Cognitive Neuroscience of Insight 13.3
PS65CH13-Kounios ARI 30 August 2013 16:0
Thus, progress in studying insight can be facilitated or enhanced by “carving nature at its joints”
and adopting a middle-path definition of insight to guide the selection of empirical phenomena
and the development of experimental paradigms for its study. Specifically, we define insight as
any sudden comprehension, realization, or problem solution that involves a reorganization of
the elements of a person’s mental representation of a stimulus, situation, or event to yield a
nonobvious or nondominant interpretation. Insights are not confined to any particular domain of
understanding, but we do not include all sudden realizations within this definition. For example,
the reading of an isolated word starts with unconscious processing, which is followed by a sudden
conscious realization of the word’s meaning. But this is not an insight, because it doesn’t involve
reorganizing a mental representation to arrive at a nonobvious or nondominant interpretation.
Insights may be especially salient when they follow an impasse, but impasse is not a necessary
precondition for insight; otherwise, spontaneous sudden realizations would be excluded because
they are not associated with an explicit problem whose solution is blocked by another idea. Insights
are often accompanied by surprise and a positive burst of conscious emotion, but we do not consider
these to be defining features because individual insights in a sequence of insights, as occur in many
experimental studies, don’t all elicit such conscious affective responses. (Of course, this doesn’t
exclude the possibility that all insights may be accompanied by unconscious affective responses;
cf. Topolinski & Reber 2010.) Phenomena such as impasse and emotion play important roles
in problem solving and are worthy of study. However, isolating the core processes of insight
is a prerequisite for investigating it. To accomplish this, we adopt a “Goldilocks” approach—
neither too much nor too little—and argue that this strategy can, and has, enabled progress in
understanding insight’s neurocognitive substrates.
It is also critical to recognize that insight involves several component processes working to-
gether and unfolding over time. Experimental paradigms that emphasize one process over another
will reveal different parts of this network. Such results may appear complex but actually paint a
richer picture of insight, just as studying encoding and retrieval, or implicit and explicit learning,
paints a more complete picture of how the brain supports memory.
SCOPE OF THE REVIEW
This review discusses the current state of cognitive neuroscience research on insight. Though
we also discuss selected behavioral cognitive studies that inform the neuroscientific framework
we describe, we do not provide an overall review of the relevant cognitive literature here. Re-
cent reviews of the cognitive literature are available elsewhere (e.g., van Steenburgh et al. 2012).
Moreover, our discussion of neuroscientific studies is not exhaustive. We focus on those that meet
several methodological criteria.
The first desideratum is that a study must demonstrably isolate the insight phenomenon. Some
studies present problems to participants and simply assume that the solutions are the result of
insight rather than analytical thought. However, as described below, many types of problems can
be solved by either insight or analysis (Bowden et al. 2005). Therefore, with some exceptions,
we do not discuss studies that do not demonstrate that participants’ solutions were, in fact, a
product of insight. One exception to this criterion is studies that use classic insight problems, such
as the Nine-Dot Problem, that have been used by researchers for many decades and for which
a consensus has been tacitly reached—though perhaps not yet with sufficient justification—that
solutions to these problems are usually achieved by insight (e.g., Chi & Snyder 2012).
A number of studies examine brain activity when people recognize rather than generate so-
lutions (e.g., Ludmer et al. 2011, Luo et al. 2011, Metuki et al. 2012). People may feel a sense
of insight upon recognizing solutions, but these postsolution recognition processes differ from
13.4 Kounios ·Beeman
PS65CH13-Kounios ARI 30 August 2013 16:0
the processes responsible for generating the solutions. Once people see a solution word, they can
perform a directed semantic memory search to connect the solution to the problem rather than an
open-ended search for associations that might lead to the solution. Although solution recognition
is itself interesting, it differs from pure insight.
A second criterion is that a candidate study must use an appropriate control or comparison
condition. For example, in studies that use remote associates problems or anagrams, insight so-
lutions can be directly compared to analytic solutions for the same type of problem because this
comparison controls for all factors except for the cognitive solving strategy—insight versus ana-
lytic processing—that is the factor of primary interest. We therefore do not focus on studies that
directly compare neural activity for sets of problems that differ in complexity, solving duration
(e.g., Aziz-Zadeh et al. 2013), visual content, working-memory load, and so forth (e.g., Sheth et al.
2008) because differences in cognitive strategy are confounded with these ancillary factors. We
wish to highlight how insight solving differs from analytic solving when other factors are held
relatively constant.
Other studies are not discussed here due to methodological issues that cannot be addressed
on a study-by-study basis in an article of this scope, such as problematic baselining of neural
activity (e.g., Sandk ¨
uhler & Bhattacharya 2008). Another type of methodological issue involves
the integration over time of functional magnetic resonance imaging (fMRI) signal. One study
attempted to use both subjective (self-report) and objective measures to distinguish insight from
analytic solving (Aziz-Zadeh et al. 2009). Unfortunately, the objective measure was speed of
solution: It was assumed that fast solutions were achieved with insight and slow solutions were
achieved analytically. Not only is this assumption questionable, it also completely confounds the
experimental contrast with the duration of solving effort. Because fMRI signal is integrated over
time—the longer an area is active, the more the measured signal will increase—it is very sensitive
to such confounds. Thus, it is impossible to know which effects were real and which were confound
related.
It is important to note that the studies that are not discussed here due to methodological issues
are not entirely uninformative. However, careful consideration must be given to each of these
issues in the context of interpreting the results.
COGNITIVE PSYCHOLOGY OF INSIGHT
Much of the cognitive psychology research on insight done over the past three decades aimed
to clarify the relationship between insightful and analytic thought (Sternberg & Davidson 1995).
Early gestalt studies distinguished insight and analysis almost solely on the basis of the informal
conscious experience of a problem solution emerging suddenly versus gradually. To extend this
research, cognitive psychologists attempted to uncover more formal evidence to distinguish these
two types of processing. A prominent example is a pioneering series of studies done by Janet
Metcalfe during the 1980s. For example, Metcalfe & Wiebe (1987) focused on metacognitive
characteristics of insight such as participants’ feelings of “warmth” (i.e., closeness to solution) while
working on insight and analytic problems. Participants reported a gradual increase in feelings of
warmth leading up to analytic solutions, but little or no warmth preceding insights until shortly
before they solved the problem. Moreover, insight problems that were accompanied by feelings
of warmth usually elicited incorrect solutions.
Metcalfe & Wiebe’s (1987) study was groundbreaking in showing a behavioral difference be-
tween insight and analytic solving beyond factors that differentially affected solution rates for
these two types of problems. However, because Metcalfe’s study sampled participants’ feelings
only once every 15 seconds, it did not directly address one of the central characteristics thought
www.annualreviews.org The Cognitive Neuroscience of Insight 13.5
PS65CH13-Kounios ARI 30 August 2013 16:0
to distinguish insight and analytic solving, namely, the suddenness of solution. Rather, her pro-
cedure was designed to examine changes over time in participants’ feelings about their closeness
to solution.
However, it is possible to measure the accrual of solution information with higher temporal
resolution using the speed-accuracy decomposition procedure (Kounios et al. 1987, Meyer et al.
1988). This technique revealed no discernable partial response information preceding the solution
when people solve insight-like anagrams (Smith & Kounios 1996). Thus, insight solving occurs
in a discrete transition from a state of no conscious information about the solution to the final
complete solution, with no intermediate states. In contrast, for similar speed-accuracy decompo-
sition studies of other (noninsight) tasks, such as lexical decision, semantic verification, short-term
recognition memory, and long-term recognition memory, people show evidence of substantial
partial information (Smith & Kounios 1996). This finding objectively validated the conscious
experience of the abruptness of insight.
The conscious experience of insight directly relates to unconscious processing that precedes it.
When people solve problems (anagrams), they solve better and experience their solutions as more
insight-like when, prior to solution, solution-related words are presented to them subliminally
(Bowden 1997). The fact that a subliminal prime can spark a later insight supports the hypothesis
that insight solutions are preceded by substantial unconscious processing rather than sponta-
neously generated. Similarly, when people respond to solution words before solving a problem,
the amount of semantic priming for solution words—an index of related unconscious processing—
is directly related to how they experience the recognition of the solution. Specifically, people show
more solution priming when they recognize solution words with a feeling of sudden insight than
when they recognize the words without an insight experience (Bowden & Jung-Beeman 2003b).
These studies are also notable for a methodological innovation. Much of the insight literature
compares performance on so-called insight problems with performance on analytic problems, a
distinction based largely on researchers’ intuitions or introspections about sudden versus grad-
ual solution. This phenomenological difference had rarely been measured and quantified in a
rigorous way. Furthermore, applying the monikers “insight” and “analytic” to specific problems
assumes that all participants will always solve insight problems insightfully and analytic problems
analytically—hardly a safe assumption. To put insight research on a firmer empirical founda-
tion, Bowden (1997) developed a procedure for soliciting participants’ trial-by-trial judgments of
whether a solution had been derived by insight or analysis. This technique has been validated by
subsequent studies that have shown that the number of insight solutions and analytic solutions
to a series of problems varies independently as a function of factors such as mood (Subramaniam
et al. 2009) and meaningfully with respect to cognitive strategies (Kounios et al. 2008) and brain
activations ( Jung-Beeman et al. 2004, Kounios et al. 2006, Subramaniam et al. 2009). The insight
judgment procedure has thus provided a foundation for subsequent neuroimaging studies of in-
sight because it allows researchers to isolate the insight phenomenon by controlling for ancillary
differences between problems that were solved insightfully and analytically (Bowden et al. 2005,
Kounios & Beeman 2009).
The development of short problems solvable by insight (Bowden & Jung-Beeman 2003a) has
also proved useful in later neuroscience studies. Early studies of insight typically posed a small
number of complex problems to participants. Most participants take many minutes to solve such
problems, when they are able to solve them. However, neuroimaging and electrophysiological
methods require many trials to accurately record brain activity. An alternative approach uses a
relatively large number of structurally identical verbal problems, called remote associates prob-
lems, modeled after one type of problem developed by Mednick for his remote associates test of
creativity (Mednick 1962). Bowden & Jung-Beeman (2003a) developed a set of compound remote
13.6 Kounios ·Beeman
PS65CH13-Kounios ARI 30 August 2013 16:0
associates problems that consist of three words (e.g., pine, crab, sauce). The participant’s task is
to think of a single solution word (apple) that will form a compound or familiar phrase with each
of the three problem words (pineapple, crabapple, applesauce).
Remote associates problems are well suited to neuroimaging and electrophysiological studies.
Large numbers of these problems have been developed, allowing for neuroimaging and elec-
trophysiological studies with a sufficient number of trials per condition. Other types of short
problems can serve this function as well. For example, anagrams have also been used with the
insight judgment procedure (Bowden 1997, Kounios et al. 2008).
NEURAL BASIS OF INSIGHT
Hemispheric Asymmetry
Much of the research on the neural basis of insight has been framed by hemispheric differences,
namely, that the right hemisphere contributes relatively more to insight solving than to analytic
solving, whereas the left hemisphere contributes more to analytic solving than to insight solving.
This hypothesis particularly influenced the experimental methods and predictions of early cog-
nitive neuroscience studies of insight. For instance, several studies used visually lateralized probe
words to detect and compare semantic processing in the hemispheres while participants worked
on remote associates problems. On trials for which participants failed to solve problems within a
time limit, they still showed semantic priming for the solution words by responding to solution
word probes more quickly than to unrelated word probes. Importantly, this solution priming was
especially pronounced when the solution word probes were presented to the left visual field, thus
being directed initially to the right hemisphere (Beeman & Bowden 2000, Bowden & Beeman
1998). Furthermore, enhanced priming in the right hemisphere occurred only when participants
reported that they recognized a solution word probe with a feeling of insight (Bowden & Jung-
Beeman 2003b).
This rightward asymmetry of insight processing was predicted (Bowden & Beeman 1998) on the
basis of prior evidence of right hemisphere involvement in integrating distant semantic relations
in language input (e.g., St George et al. 1999) as well as a theoretical framework that describes
the right hemisphere as engaging in relatively coarser semantic coding than the left hemisphere
( Jung-Beeman 2005). This framework incorporates neuropsychological and neurological evidence
of subtle comprehension deficits following right hemisphere brain damage with neuroanatomical
findings of asymmetric neuronal wiring.
According to the coarse semantic coding framework, when readers or listeners encounter a word
or concept, they activate a semantic field related to the word: a subset of features, properties, and
associations of that word. Evidence suggests that the left hemisphere strongly activates a relatively
smaller semantic field of features, those most closely related to the dominant interpretation or the
current context; in contrast, the right hemisphere weakly activates a relatively broader semantic
field, including features that are distantly related to the word or context (Chiarello 1988, Chiarello
et al. 1990). Despite some obvious limitations, coarser semantic coding in the right hemisphere
has one big advantage: The less sharply each word’s meaning is specified, the more likely it is
to connect to other words and concepts. This is a key ingredient for drawing inferences (Virtue
et al. 2006, 2008), extracting the gist (St George et al. 1999), comprehending figurative language
(Mashal et al. 2008), and for insight.
The coarse semantic coding notion is more than a metaphor. Rather, it potentially links asym-
metric semantic processing to asymmetric brain wiring. Aside from some size asymmetries in par-
ticular regions of cortex (such as Broca’s area and Wernicke’s area), lateralized cytoarchitectonic
www.annualreviews.org The Cognitive Neuroscience of Insight 13.7
PS65CH13-Kounios ARI 30 August 2013 16:0
differences also influence how neurons integrate inputs (for a review, see Hutsler & Galuske 2003).
In brief, pyramidal neurons collect inputs through their dendrites. Differences in synaptic distri-
butions along dendrites influence the type of inputs that cause these pyramidal neurons to fire. The
range of cortical area over which neurons collect inputs could be termed their input fields. In as-
sociation cortices in or near language-critical areas, such as Wernicke’s area, Broca’s area, and the
anterior temporal cortex, right hemisphere neurons have larger input fields than do left hemisphere
neurons (e.g., Jacob et al. 1993, Scheibel et al. 1985, Seldon 1981). Specifically, right hemisphere
pyramidal neurons have more synapses overall and especially more synapses far from the cell body.
This indicates that they have larger input fields than corresponding left hemisphere pyramidal
neurons. Because cortical connections are spatially organized, the right hemisphere’s larger in-
put fields collect more differentiated inputs, perhaps requiring a variety of inputs to fire. The left
hemisphere’s smaller input fields collect highly similar inputs, likely causing the neuron to respond
best to somewhat redundant inputs. Outputs from neurons appear to show similar asymmetry;
for example, axons in superior temporal cortex are longer in the right hemisphere than in the left
hemisphere, favoring more integrative processing in the right hemisphere (Tardif & Clarke 2001).
These neuroanatomical asymmetries could contribute to the right hemisphere’s bias to en-
gage in coarser semantic coding and the left hemisphere’s bias to engage in finer (i.e., less coarse)
semantic coding. As previously noted ( Jung-Beeman 2005), there is a huge gap between descrip-
tions of dendritic branching and modes of language processing or problem solving. However, the
asymmetries that exist in neuronal wiring almost certainly influence information processing, and
the asymmetries that indisputably exist in language processing must have some neuroanatomical
basis. The coarser semantic coding framework attempts to bridge that gap. In so doing, it also
provides an avenue for future research on the relationship between neural microcircuitry and
higher cognitive functions.
Neural Correlates of Insight Solving
Further specification of the neural bases of insight can be achieved through neuroimaging studies.
These studies have identified a number of distinct components of insight and have generally
supported the idea that the right hemisphere contributes relatively more to insight than to analytic
solving.
One early neuroimaging study of insight isolated neural correlates of the insight experience
with both fMRI and high-density EEG in separate experiments matched as closely as possible for
procedure ( Jung-Beeman et al. 2004). EEG has excellent temporal resolution but limited spatial
resolution. It is therefore good at circumscribing a neural process in time. fMRI has excellent spatial
resolution but limited temporal resolution and is therefore best suited to localize a neural event in
space. Together these techniques were able to isolate insight’s neural correlates in both space and
time. This combination of methods was crucial, because fMRI’s power to localize insight-related
neural activity would have been less informative without knowing whether these neural correlates
occurred before, after, or at the moment of solution. A neural correlate of the insight experience
itself would have to occur at, or immediately prior to, the moment of conscious awareness of a
solution.
At the moment when people solve problems by insight, relative to solving identical problems
by analytic processing, EEG shows a burst of high-frequency (gamma-band) EEG activity over
the right temporal lobe, and fMRI shows a corresponding change in blood flow in the medial
aspect of the right anterior superior temporal gyrus ( Jung-Beeman et al. 2004) (Figure 1). In
the initial fMRI experiment, this right temporal area was the only area exceeding strict statistical
thresholds, but weak activity was detected in other areas, including bilateral hippocampus and
13.8 Kounios ·Beeman
PS65CH13-Kounios ARI 30 August 2013 16:0
a
c
fMRI
EEG gamma
b
Time to solution
Insight
Analytic
EEG power
Figure 1
Neural correlates of insight. (a) Insight-related blood oxygen–level dependent (BOLD) activity in the right
anterior superior temporal gyrus recorded by functional magnetic resonance imaging (fMRI). (b) Insight-
related gamma-band oscillatory activity recorded by electroencephalogram (EEG) over the anterior right
temporal lobe. (c) Time course of insight- and analysis-related gamma-band EEG power recorded at a right
anterior electrode. The vertical gray line marks the point in time at which participants made a bimanual
button press to indicate that they had solved a problem. EEG power leading up to insight and analytic
solutions diverges at approximately 300 ms before the bimanual button press. Taking into consideration that
a button press requires about 300 milliseconds to initiate and execute (Smith & Kounios 1996), the
insight-related burst of gamma activity occurred at approximately the time at which the solution to the
problem became available to participants. Adapted from Jung-Beeman et al. (2004), with permission.
parahippocampal gyri and anterior and posterior cingulate cortex. In a later replication with
more participants and stronger imaging methods (Subramaniam et al. 2009), the same network of
areas all far exceeded critical statistical threshold, with the right anterior temporal region again
being the strongest. The close spatial and temporal correspondence of the fMRI and EEG results
obtained by Jung-Beeman et al. suggested that they were produced by the same underlying brain
activation. This right temporal brain response was identified as the main neural correlate of the
insight experience because (a) it occurred at about the moment when participants realized the
solution to each of these problems, (b) the same region is involved in other tasks demanding
semantic integration (St George et al. 1999); and (c) gamma-band activity has been proposed to be
a mechanism for binding information as it emerges into consciousness (Tallon-Baudry & Bertrand
1999). Alternative interpretations of this finding were rejected based on considerations of timing,
functional neuroanatomy, etc.
The burst of gamma-band EEG activity in the right temporal lobe was not unexpected, given
earlier visual half-field studies (Beeman & Bowden 2000, Bowden & Beeman 1998, Bowden &
Jung-Beeman 2003b). However, the EEG results revealed another, totally unexpected, finding.
The insight-related gamma-band activity was immediately preceded by a burst of alpha-band
activity (10 Hz) measured over right occipital cortex ( Jung-Beeman et al. 2004) (see Figure 2).
www.annualreviews.org The Cognitive Neuroscience of Insight 13.9
PS65CH13-Kounios ARI 30 August 2013 16:0
1.0 × 10–10
6.0 × 10–11
2.0 × 10–11
–1.0 × 10–10
–6.0 × 10–11
–2.0 × 10–11
00
5.0 × 10–12
1.5 × 10–11
2.5 × 10–11
–5.0 × 10–12
–1.5 × 10–11
–2.5 × 10–11
Time (sec)
Alpha insight eect
Gamma insight eect
Alpha
insight eect
Gamma
insight eect
–0.5 –1.0 –1.5 –2.0 R
Figure 2
The time course of the insight effect. Alpha power (9.8 Hz at right parietal-occipital electrode PO8) and
gamma power (39 Hz at right temporal electrode T8) for the insight effect (i.e., correct insight solutions
minus correct noninsight solutions, in v2). The left y-axis shows the magnitude of the alpha insight effect
(purple line); the right y-axis applies to the gamma insight effect ( green line). The x-axis represents time (in
seconds). The gray arrow and R(at 0.0 sec) signify the time of the button-press response. Note the transient
enhancement of alpha on insight trials (relative to noninsight trials) prior to the gamma burst.
Alpha-band oscillations reflect neural inhibition; occipital alpha reflects inhibition of visual inputs,
that is, sensory gating ( Jensen & Mazaheri 2010). It appears likely that the preinsight alpha burst
reflects transient sensory gating that reduces noise from distracting inputs to facilitate retrieval
of the weakly and unconsciously activated solution represented in the right temporal lobe ( Jung-
Beeman et al. 2004; cf. Wu et al. 2009). This idea is analogous to the common behavior of closing or
averting one’s eyes to avoid distractions that would otherwise interfere with intense mental effort.
The discovery of transient sensory gating immediately preceding the insight-related burst
of gamma-band activity suggested a promising research strategy. Previous behavioral research
had demonstrated the discrete, all-or-nothing nature of insight solutions (Smith & Kounios
1996). On the other hand, visual hemifield solution-priming studies showed that insight, though
consciously abrupt, is preceded by unconscious processing, primarily in the right hemisphere
(e.g., Bowden & Beeman 1998). So, an insightful solution is a discrete phenomenon in terms
of its availability to awareness, but it is preceded by unconscious neural precursors. It should
therefore be possible to trace these neural precursors backward in time from the gamma burst at
the moment of insight to reveal the brain mechanisms that unfold to produce an insight.
Preparation for Insight
Before people even start to tackle a problem, their state of mind—and their brain activity—
predisposes them to solve either by insight or analytic processing. Participants’ neural activity,
assessed with both EEG and fMRI during a task-free preparation phase prior to each remote
associates problem, shows that such predispositions do occur: Distinct patterns of neural activity
precede problems that people eventually solve by insight versus those that they solve by analysis
13.10 Kounios ·Beeman
PS65CH13-Kounios ARI 30 August 2013 16:0
(Kounios et al. 2006). EEG showed that preparation for analytic solving involves increased neural
activity (i.e., decreased alpha-band activity) measured over visual cortex, hypothesized to reflect
outward focus of attention directed to the computer monitor on which the next problem in the
sequence was to be displayed. Both EEG and fMRI revealed that preparation for insight solving
involves activation of the anterior cingulate and bilateral temporal cortices. The temporal lobe
activation suggests that cortical regions involved in lexical and semantic processing are prepared to
respond. Previous research implicates anterior cingulate cortex in monitoring other brain regions
for conflicting action tendencies (Botvinick et al. 2004). Kounios et al. (2006) expanded on this
notion of conflicting action tendencies to propose that the anterior cingulate’s role in problem
solving is to detect the activation of conflicting solution possibilities or strategies. If the anterior
cingulate is sufficiently activated at the time a problem is presented, then it can detect the weak
activation of nondominant solution possibilities, enabling attention to switch to one of these weakly
activated ideas. Switching attention to a nonobvious solution brings the idea to awareness as an
insight. However, when the anterior cingulate is relatively deactivated prior to the presentation
of a problem, attention is dominated by the more obvious associations and solution possibilities
afforded by the problem.
Thus, the transient state of one’s attentional focus, varying from trial to trial, helps to de-
termine the range of potential solutions that a person is prepared to consider when a problem
is presented: Outwardly directed attention coupled with low anterior cingulate activity focuses
processing on the dominant features or possibilities of a situation; inwardly directed attention and
high anterior cingulate activity heightens sensitivity to weakly activated remote associations and
long-shot solution ideas.
One important, but unresolved, issue is to what extent such preparatory activity is under vol-
untary control and to what extent spontaneous shifts of attention may be involved. Kounios et al.
(2006) found no evidence of any trial-to-trial sequential clusters of insight or analytic solutions
that would suggest slow spontaneous shifts of attention. However, this kind of analysis would not
be capable of showing attention shifts on a timescale shorter than the duration of a single trial (i.e.,
approximately 10 sec.). Identifying any neural correlates of possible strategic or spontaneous atten-
tion shifts is therefore an important focus for future investigations. Progress in elucidating the role
of attention shifts would suggest practical techniques for controlling or enhancing cognitive style.
Resting-State Brain Activity and Individual Differences
Given that transient shifts in attention influence whether people solve by analysis or by insight,
do any longer-lasting states or traits influence this preparatory activity and the corresponding
predisposition to solve by insight or analysis? One approach is to examine whether individual
differences in resting-state brain activity while people have no task to perform or any particular
expectation about what will follow may influence their subsequent problem-solving style. In one
study, we recorded participants’ resting-state EEGs before tasking them with solving a series of
anagrams (Kounios et al. 2008). Participants were classified as high insight or low insight based on
the proportion of problems they solved with insight. These groups exhibited different patterns of
resting-state EEGs, suggesting that the insightful and analytic cognitive styles have their origins in,
or are at least related to, distinct patterns of resting-state neural activity. The differences between
these patterns highlighted two general phenomena.
Insightful individuals show greater right hemisphere activity at rest, relative to analytic indi-
viduals, consistent with the idea of insight-related right hemisphere bias described above. The
fact that an insight-related hemispheric difference can be found in the resting state suggests that
the functional hemispheric asymmetry occurring during problem solution ( Jung-Beeman et al.
www.annualreviews.org The Cognitive Neuroscience of Insight 13.11
PS65CH13-Kounios ARI 30 August 2013 16:0
2004) may have its origin in structural hemispheric differences among people, such as the cytoar-
chitectonic differences described above (Jung-Beeman 2005) or in asymmetries of structural or
functional connectivity.
Insightful individuals also showed greater diffuse activation of visual cortex compared to an-
alytical individuals, even when resting-state EEGs were measured while participants’ eyes were
closed (Kounios et al. 2008). This finding mirrors earlier behavioral research showing that highly
creative individuals tend to have diffuse attention when at rest or when cognitive resources are not
dominated by a task (e.g., Ansburg & Hill 2003, Carson et al. 2003).
Resting-state EEG can be perturbed by stimuli but otherwise is relatively stable; in fact, be-
havioral genetics studies show that individual differences in resting-state EEG have a substantial
genetic loading largely attributable to individual differences in gray matter and white matter
volume (Smit et al. 2012). It is not yet known whether insight-related individual differences in
resting-state EEG are a subset of the genetically loaded individual differences in EEG or brain
volume, but if so, this would be a promising avenue of investigation into the stability and origins
of cognitive style.
Studies of insight-related resting-state brain activity may also provide a link to recent so-
cial psychological research on construal level. According to construal level theory, psychological
“distance”—thinking about things that are far away in space or time, or about people that are
different from oneself—engages abstract thinking (Trope & Liberman 2010). Based on this idea,
F¨
orster et al. (2004) predicted that priming people to think about the distant future would bias them
to think abstractly, which in turn would induce a person to think more insightfully and creatively;
conversely, priming people to think about the near future would bias them to think concretely and
therefore analytically. Their studies supported these predictions: Participants primed by asking
them to think about the distant future subsequently did better on insight and creativity tasks; those
asked to think about the near future did better on analytic tasks. These results are particularly
interesting because construal-level priming may influence task-related cognition by transiently al-
tering the resting-state brain activity from which task-related brain activity emerges. A potentially
fruitful line of research would be to examine how various types of priming might influence the ten-
dency to solve problems insightfully or analytically by imposing transient changes in resting-state
brain activity.
Thus, distinct patterns of neural activity are associated with insight versus analytic solving at
the moment of insight, in the last two seconds leading up to that moment, in the preparation
phase prior to presentation of a problem, and even in resting-state brain activity of individuals
who tend to solve by insight contrasted with those who tend to solve analytically. These findings
objectively substantiate an abundance of behavioral evidence that indicates insight solving differs
from analytic solving and that these solving styles result from different tunings of the network of
brain areas involved in problem solving. Moreover, the specific areas associated at each distinct
stage of solving (or preparation) help inform theories of how insight is different from analysis.
Compared to analytic solving, insight requires greater input from and integration of relatively
coarser semantic processing of right hemisphere temporal areas, greater sensitivity to competing
responses in cognitive control mechanisms supported by the anterior cingulate cortex, and a
relative emphasis on internal processing and de-emphasis on external stimuli.
ATTENDING IN, OUT, AND AROUND
A prominent theme in insight and creativity research is the role that attention deployment plays
in cognitive style. Though a number of behavioral studies have observed that highly creative
individuals have diffuse attention, taken as a whole, neuroimaging and electrophysiological studies
13.12 Kounios ·Beeman
PS65CH13-Kounios ARI 30 August 2013 16:0
of insight suggest that attention plays a more nuanced role. Neural activity during the preparatory
phase suggests that attention can also be focused outwardly on external objects or inwardly on
internal knowledge in memory (see Chun et al. 2011), which influences the likelihood that people
will solve with insight (Kounios et al. 2006, Wegbreit et al. 2012).
Occipital alpha-band EEG activity reflects visual cortex inhibition that protects fragile internal
processing from potentially interfering or distracting perceptual inputs (Ray & Cole 1985). Levels
of occipital alpha during the different phases leading up to the solution of a problem suggest
that (a) during the resting state, insightful individuals have more externally oriented attention
than do analytical individuals (Kounios et al. 2008); (b) during the preparation phase prior to the
presentation of a problem that will be solved by insight, there is greater internal focus of attention
(Kounios et al. 2006); and (c) just before the emergence of an insight, there is another brief burst
of inward focus ( Jung-Beeman et al. 2004).
Thus, the notion that insight is associated with diffuse attention appears to be an oversimpli-
fication. Insightful individuals may generally have more diffuse and outwardly directed attention,
but successful insight solving involves transiently redirecting attention inwardly during the prepa-
ration for and solving of a problem. It therefore appears that the tendency to solve problems
insightfully is associated with broad perceptual intake as the default mode of resting-state atten-
tion deployment, coupled with the tendency to focus inwardly in preparation for, and during,
solving. In contrast, analytical people’s resting-state attention is less outwardly focused during the
resting state and less inwardly directed during preparation and solving.
FACTORS THAT INFLUENCE THE LIKELIHOOD OF INSIGHT
Understanding the factors that can increase or reduce the likelihood of experiencing insight is
important for both theoretical and practical reasons. Besides contributing toward the develop-
ment of a theoretical model of insight, understanding these factors also suggests strategies for its
enhancement outside the laboratory. Here we focus on the interrelated factors of mood, attention,
and cognitive control.
Mood
Positive affect enhances insight and other forms of creativity, both when the mood occurs naturally
and when it is induced in the laboratory (e.g., Ashby et al. 1999, Isen et al. 1987). Though some
aspects of creative production may be impeded by positive mood or aided by other moods such as
depression (Verhaeghen et al. 2005), insight and related processes seem to benefit from greater
positive mood (or reduced anxiety) either when participants enter the lab in a relatively positive
mood or after they watch funny film clips (H. Mirous & M. Beeman, manuscript submitted;
Subramaniam et al. 2009). Facilitation of insight by positive mood has also been demonstrated in
the workplace, as documented by diaries and self-reports (Amabile et al. 2005).
Mood influences other cognitive abilities that are related to insight and creativity. For example,
positive mood facilitates intuition, the ability to make decisions or judgments about stimuli without
conscious access to the information or processes influencing their behavior. When people are
working on remote associates problems, they show better-than-chance judgment about whether
individual problems are solvable before they are able to state the solution; such intuitive judgments
are improved after people recall happy autobiographical events and are impeded after recalling
sad autobiographical events (Bolte et al. 2003).
Finding or intuiting the presence of a solution to a remote associates problem requires a person
to access weak associations of the problem words because the solution is typically not a strong
www.annualreviews.org The Cognitive Neuroscience of Insight 13.13
PS65CH13-Kounios ARI 30 August 2013 16:0
associate of all three problem words. Thus, the presentation of the problem will evoke only weak
activation of the solution’s representation. Positive mood seems to broaden the scope of semantic
processing to make such weak associations more accessible (Isen & Daubman 1984, Isen et al.
1985). For instance, the amplitude of the N400 component of the event-related potential (ERP)
is inversely proportional to the relatedness of a word to its semantic context (Kounios 1996), and
N400 semantic relatedness effects are modulated by mood. When a positive mood is induced,
target words that loosely fit their semantic context elicit a smaller N400 than when the participant
is in a neutral mood (Federmeier et al. 2001). This indicates that a positive mood makes a word
seem less incongruent with its semantic context. Additionally, when people listen to stories that
imply specific causal events without stating them explicitly, they show sensitivity to semantic
information that is related to the implicit inference. Specifically, they read or respond to probe
words that are related to an implicit inference more quickly than they read or respond to unrelated
probe words. Such inference-related priming normally occurs earlier and more strongly for words
presented to the right hemisphere (left visual field) than for words presented to the left hemisphere
(right visual field) (Beeman et al. 2000). But people show stronger inference-related priming to
inference-related words presented in the middle of the visual field while listening to stories after
watching funny film clips than after watching emotionally neutral films; they show no inference-
related priming after watching scary film clips that induce anxiety (H. Mirous & M. Beeman,
manuscript submitted).
Recently it has also been argued that the relation between broad associations and positive
mood is bidirectional, making it possible to induce a positive mood by instructing or otherwise
inducing participants to process remote associations (Bar 2009, Bruny´
e et al. 2013). According
to this idea, a positive mood both facilitates insight and is enhanced by it. This shows deep
integration of cognitive and affective processes and relates to many everyday behaviors, such as
peoples’ enjoyment of verbal puzzles, especially those with surprising solutions.
The fact that people engage in coarser semantic coding when they are in a positive mood
compared to when they are in neutral or anxious mood raises the possibility that positive mood
may selectively activate right hemisphere semantic processing. However, as of yet, there is no
evidence to support this hypothesis. For example, the neuroimaging study of insight and mood by
Subramaniam et al. (2009) found no evidence of lateralized differences in brain activity that were
attributable to mood.
Mood and Attention
A more likely hypothesis is that mood influences the likelihood of insight by modulating attention
or cognitive control, which in turn modulates semantic processing. For example, anxiety narrows
the scope of attention by eliciting excessive focus on the center of one’s field of vision—which
usually includes the source of the threat—to the exclusion of peripheral information (Easterbrook
1959). From the evolutionary standpoint, this makes great sense: Early humans spotting a lion
on the African savannah would not want to be distracted by less important stimuli. In contrast,
positive mood appears to broaden attention. It increases the perception and utilization of global
and peripheral perceptual features at the expense of the local details of complex stimuli by spatially
broadening the “spotlight” of attention (Gasper & Clore 2002). For example, in a task that requires
participants to respond to a centrally located target stimulus and disregard other stimuli that flank
it, a positive mood increases both facilitation and interference of target processing attributable
to the flanking stimuli (Rowe et al. 2007). Beyond visual processing, positive mood seems to
broaden the processing of novel and varied stimuli, stimulating exploratory behavior (Fredrickson
& Branigan 2005).
13.14 Kounios ·Beeman
PS65CH13-Kounios ARI 30 August 2013 16:0
Models of Attention
One possibility is that mood affects the balance between two brain systems: an anterior attention
network that maintains top-down control over perceptual processing in the service of goals and
a posterior network involved in bottom-up attentional capture by salient stimuli. According to
attentional control theory, anxiety shifts the balance away from the top-down system toward
the bottom-up system (Derakshan & Eysenck 2009), leading to enhanced distractibility by task-
irrelevant stimuli, especially threatening stimuli (Bar-Haim et al. 2007). This view suggests that
anxiety shifts attention toward external stimuli and away from internal representations, states, and
goals; positive affect may have the opposite effect.
In practice, attentional control theory makes predictions that are similar to those of the spotlight
model of attention. Moreover, it links positive mood with internally focused attention in a way that
is consistent with research implicating both of these factors in preparation for insight (Kounios
et al. 2006, Subramaniam et al. 2009). Attentional control theory is therefore a promising direction
for future insight research.
One important question is why changes in the breadth of perceptual attention due to mood
and other factors should be related to changes in the breadth or narrowness of thought to include
or exclude remote associations, what Rowe et al. (2007) called conceptual attention. Rowe et al.
demonstrated that a positive mood both increases the breadth of visual attention to include stimuli
that flank a target and enhances performance on remote associates problems that, as noted, require
access to distant associations of the problem words. Rowe et al. (2007) argued that perceptual and
conceptual attention are closely linked.
Current attention theory can be expanded to include both phenomena. The biased competi-
tion model of attention characterizes the neural computations subserving vision as a process of
competition between the representations of the stimuli in the visual field (Desimone 1998). This
competition can be biased to favor a particular stimulus by a variety of factors, including externally
driven (bottom-up) and internally driven (top-down) processes. Such biasing is manifested as an
increase in the neural activity subserving one of the representations.
A similar mechanism may underlie the conceptual processing that occurs during problem solv-
ing. According to this idea, the presentation of a problem activates a number of associations in
a person’s memory. Dominant associations are strongly activated; nondominant ones are weakly
activated. All of these associations compete for processing resources, though under normal cir-
cumstances, the dominant associations win the competition for further processing. However,
top-down mechanisms can bias this competition toward weak associations by actively selecting
nondominant representations; by expanding the scope of attention to boost activation of non-
dominant representations; or, simply, by not suppressing the less dominant associations when the
dominant ones capture the spotlight. Dominant associations are already as activated as they can
be, so expanding the scope of attention would benefit weak associations more than it would do
for strong ones, giving weak associations a greater opportunity to capture attention and spark
an insight. Bottom-up processes can also bias the competition between ideas. Stimuli in the sur-
rounding environment that are related to the solution can intervene to bias processing toward a
weak association by acting as a hint that triggers an insight (Bowden 1997, Seifert et al. 1995).
Thus, the biased competition model may be generalized to explain the role of conceptual attention
in problem solving by insight.
Cognitive Control
Other research has focused on the role of cognitive control, especially the ability to maintain
or switch between different thoughts, actions, and goals. People often focus on a task or goal
www.annualreviews.org The Cognitive Neuroscience of Insight 13.15
PS65CH13-Kounios ARI 30 August 2013 16:0
to shield it from distraction. In other cases, especially in creative tasks, people need to flexibly
switch between different processes, associations, or goals. These two functions, task shielding
and task switching, appear to be in direct competition: The need for flexible switching demands
relaxation of task shielding and leaves processing open to distraction (Dreisbach 2012). Pos-
itive affect enhances task switching but yields increased distractibility (Dreisbach & Goschke
2004).
As noted above, preparation for insight involves increased activity in the anterior cingulate
during the preparatory period preceding a problem (Kounios et al. 2006). Anterior cingulate acti-
vation is hypothesized to be a sign that problem solvers are sensitized to competing, nondominant
associations that they can switch to, resulting in an insight. When a person is in a positive mood,
the preparation period shows stronger anterior cingulate activation than occurs for people not in
a positive mood. In fact, the anterior cingulate was found to be the only brain area whose activa-
tion varies with mood, preparation for insight versus analytic processing, and later insight versus
analytic solving (Subramaniam et al. 2009).
The anterior cingulate has long been recognized as a critical component of the cognitive control
network. One hypothesis, backed up by substantial evidence, is that this brain region monitors
other regions for competing action tendencies or stimulus representations (Kerns et al. 2004,
Weissman et al. 2005). It has been proposed to be an interface between emotion and cognition
(Allman et al. 2001, Bush et al. 2000, Lane et al. 1998) in part because some (ventral) regions of
the cingulate are important for emotional processing (Mayberg et al. 1999).
Another brain area implicated in insight-related cognitive control is prefrontal cortex. Consid-
erable evidence supports the idea that prefrontal cortex exerts control over other brain regions in
response to input from the anterior cingulate signaling the presence of cognitive conflict (Miller
& Cohen 2001). According to this idea, modulation of insight solving due to changes in an-
terior cingulate activity should be mediated, at least in part, by control signals originating in
prefrontal cortex that limit the range of possibilities that a person considers when working on
a problem. This limiting function is ordinarily helpful because it focuses the solver on a small
number of the most viable solution paths to avoid computational overload. However, it can be
a hindrance when a person tries to solve a problem whose solution lies on a nonobvious solu-
tion path. In support of this idea, patients with damage to lateral prefrontal cortex were better
able to solve matchstick insight problems than were healthy control participants (Reverberi et al.
2005).
STIMULATING INSIGHT
One limitation of neuroimaging and electrophysiological studies is that they are inherently
correlational—they don’t directly show that the recorded patterns of brain activity cause the
measured changes in behavior or experience. But the advent of brain stimulation techniques now
affords the opportunity to treat brain activity as an independent variable rather than a dependent
one.
Recent efforts have applied one brain-stimulation technique, transcranial direct current
stimulation (tDCS), to attempt to enhance insight solving. Two recent studies have yielded
promising results (Chi & Snyder 2011, 2012). Researchers tested the hemispheric hypothesis of
insight by applying facilitatory (anodal) stimulation to right frontal-temporal cortex and inhibitory
(cathodal) stimulation to left frontal-temporal cortex. This pattern of stimulation, but not the re-
verse hemispheric pattern, significantly enhanced solution rates for the nine-dot problem and for
an insight matchstick problem. This stimulation protocol yielded especially dramatic enhancement
for the classic nine-dot problem, increasing the solution rate from 0% to 40%. Additional studies
13.16 Kounios ·Beeman
PS65CH13-Kounios ARI 30 August 2013 16:0
have found that stimulation interfering with left dorsolateral prefrontal cortex facilitates solving
compound remote associates problems (Metuki et al. 2012) or flexibly generating unusual uses for
objects (Chrysikou et al. 2013). These studies suggest that such tasks benefit from the release of
cognitive control that would otherwise maintain focus on fine semantic coding in the left
hemisphere; however, neither of these studies specifically contrasted insight versus analytic
performance.
Such stimulation studies are encouraging initial efforts that provide striking support for the
claim that insight depends relatively more on right than on left temporal lobe processes. They also
raise the alluring possibility that someday brain stimulation techniques will be refined to the point
at which individuals grappling with difficult problems may have the option of donning a “thinking
cap” that will increase their ability to find solutions. However, it should be kept in mind that these
early studies—similar to other groundbreaking studies—raise as many questions as they answer.
For example, this stimulation protocol simultaneously stimulated right frontal-temporal cortex
and inhibited left frontal-temporal cortex (Chi & Snyder 2012). It is not yet known whether it was
the right frontal-temporal stimulation, the left frontal-temporal inhibition, or the combination
of the two that facilitated solving these insight problems. Moreover, it is not yet known whether
this tDCS protocol actually increased the probability that participants solved these problems
insightfully or whether it increased the probability that they solved the problems analytically. As
we have noted, most problems can be solved by either strategy. The fact that researchers have
considered the nine-dot and matchstick-type problems to be insight problems doesn’t preclude
the possibility that they can be solved analytically. These studies did not assess the strategies
that their participants employed, so all that is known is that the solution rates for these problems
increased rather than how or why they increased. Thus, much foundational work must be done
before tDCS can be considered a realistic possibility for adaptively modifying people’s cognitive
strategies.
Pharmacological intervention is another route to insight enhancement. To date, we are aware
of no studies that use drugs to attempt selective facilitation of insight solving. However, the
recent demonstration that alcohol can enhance insight, but not analytic, solving of remote as-
sociates problems shows that pharmacological promotion of insight is achievable ( Jarosz et al.
2012).
FUTURE DIRECTIONS
Neuroimaging and electrophysiological techniques have begun to reveal neural substrates of in-
sight that were invisible to behavioral research. This has led to progress in understanding how
insight emerges from more basic cognitive mechanisms. Technologies for stimulating insightful
thought are becoming available, including intervention by direct brain stimulation.
Nevertheless, from our current vantage point, it is important to keep in mind that the surface has
barely been scratched. Research has shown that multiple component processes and corresponding
neural substrates are involved, and some of these are susceptible to subtle shifts in attention,
mood, and other factors. Refined methods will expand on the research we have described and
contribute new findings from connectivity and network analyses. And one can only guess what
will be uncovered by future studies of insight-related individual differences in neuroanatomy,
cytoarchitectonics, and genetics. The psychopharmacology of insight and creativity, currently
virtually unexplored, holds out the promise of contributing both to our scientific understanding
of insight and to methods for enhancing it. Further research will reveal the limits and applicability
of brain stimulation, neurofeedback, and cognitive training techniques for enhancing insight and,
more generally, influencing and optimizing cognitive styles to suit different circumstances. The
www.annualreviews.org The Cognitive Neuroscience of Insight 13.17
PS65CH13-Kounios ARI 30 August 2013 16:0
study of insight began in the early twentieth century, but a century from now, researchers may
look back at the early twenty-first century as the beginning of a golden age of insight research.
SUMMARY POINTS
1. Insight is any sudden comprehension, realization, or problem solution that involves a
reorganization of the elements of a person’s mental representation of a stimulus, situation,
or event to yield a nonobvious or nondominant interpretation. Insight is sudden, but it
is preceded by substantial unconscious processing.
2. Some critical components of insight are preferentially associated with the right cerebral
hemisphere. Insight culminates with a sharp increase in neural activity in the right anterior
temporal lobe at the moment of insight.
3. Insights are immediately preceded by a transient reduction of visual inputs that apparently
reduce distractions and boost the signal-to-noise ratio of the solution.
4. Neural activity immediately before the presentation of an expected problem predicts
whether that problem will be solved by insight or analytically. Such preparation for
insight involves inwardly directed attention; preparation for analysis involves outwardly
directed attention.
5. Resting-state neural activity biases later processing to favor insight or analytical problem
solving.
6. Positive mood facilitates insight by increasing attentional scope to include weakly acti-
vated solution possibilities.
7. Direct stimulation of right frontal-temporal cortex coupled with inhibition of left frontal-
temporal cortex enhances solving of insight problems.
8. Cognitive neuroscience methods have contributed exciting new results and theories of
insight; nevertheless, insight research is still in a very early stage. Continuing applications
of new methods, paradigms, and models hold much promise for additional substantial
progress.
FUTURE ISSUES
1. How and when can insight be facilitated? People often ask how they can foster more
insightful thinking, even though in many situations and for many problems, analytic
processing would be more effective. At the least, problems must be deeply analyzed [what
Graham Wallas (1926) termed immersion] before an insight solution can be achieved. So
the question becomes when insight should be facilitated. We suggest that the right time
to facilitate insight is when semantic integration processes activate the representation
of a potential solution to a level just below the threshold for consciousness, setting the
stage for an aha moment. Thus, honing intuition to sense the presence of a subthreshold
solution may be the first step toward facilitating insight. At this time, inducing a positive
mood and broadening attention through various means, and not directly focusing on the
problem, is likely to increase the chance of achieving insight.
13.18 Kounios ·Beeman
PS65CH13-Kounios ARI 30 August 2013 16:0
2. What individual differences in attention or cognitive control are most conducive to
insight? Recent work shows that while in the resting state and not engaged in any task,
individuals who tend to solve by insight deploy attention differently from individuals who
are more analytic. What is the best way to characterize this pattern of differences? Do
high-insight individuals deploy attention externally at rest and internally during solving?
Or do they just deploy their attention in a less-focused manner? Do these differences
primarily occur in bottom-up attention or primarily in top-down attention and cognitive
control? How tightly coupled are insight-related individual differences in visual cortex
and those in the anterior cingulate? Do individual differences in insightfulness have a
genetic basis?
3. How do individual differences in processes that support insight interact? In general, pos-
itive mood facilitates insight by encouraging broader or less selective attention. How-
ever, a greater tendency or ability to solve problems with insight may also be associated
with individual differences in other factors that cause broad associative thinking, such as
schizotypy (Folley & Park 2005). It is possible that for individuals who typically think
more broadly, anxiety would facilitate solving by focusing their attention to harness their
broader associative processes toward a useful solution.
4. How does insight problem solving relate to other creative behavior? Insight is considered
a critical facet of creative cognition. However, creativity is a highly complex behavior.
Although aspects of creativity may be entirely unrelated, others are likely linked, perhaps
sharing similar patterns of attention. How do insight, intuitive decision making, divergent
thinking, and creative achievement relate to each other, and to attention and cognitive
control?
DISCLOSURE STATEMENT
The authors are not aware of any affiliations, memberships, funding, or financial holdings that
might be perceived as affecting the objectivity of this review.
ACKNOWLEDGMENTS
Preparation of this article and some of the research described therein was supported by National
Science Foundation grant 1144976 to J.K. and John Templeton Foundation grant number 24467.
LITERATURE CITED
Allman J, Hakeem A, Erwin JM, Nimchinsky E, Hof P. 2001. The anterior cingulate cortex: the evolution of
an interface between emotion and cognition. Ann. N. Y. Acad. Sci. 935:107–17
Amabile TM, Barsade SG, Mueller JS, Staw BM. 2005. Affect and creativity at work. Adm. Sci. Q. 50:367–403
Ansburg PI, Hill K. 2003. Creative and analytic thinkers differ in their use of attentional resources. Personal.
Individ. Differ. 34:1141–52
Ashby FG, Isen AM, Turken U. 1999. A neuropsychological theory of positive affect and its influence on
cognition. Psychol. Rev. 106:529–50
Aziz-Zadeh L, Kaplan JT, Iacoboni M. 2009. “Aha!”: the neural correlates of verbal insight solutions. Hum.
Brain Mapp. 30:908–16
Aziz-Zadeh L, Liew S-L, Dandekar F. 2013. Exploring the neural correlates of visual creativity. Soc. Cogn.
Affect. Neurosci. 8: 475–80
www.annualreviews.org The Cognitive Neuroscience of Insight 13.19
PS65CH13-Kounios ARI 30 August 2013 16:0
Bar M. 2009. A cognitive neuroscience hypothesis of mood and depression. Trends Cogn. Sci. 13:456–63
Bar-Haim Y, Lamy D, Pergamin L, Bakermans-Kranenburg MJ, van IJzendoorn MH. 2007. Threat-related
attentional bias in anxious and nonanxious individuals: a meta-analytic study. Psychol. Bull. 133:1–24
Beeman MJ, Bowden EM. 2000. The right hemisphere maintains solution-related activation for yet-to-be-
solved problems. Mem. Cogn. 28:1231–41
Beeman MJ, Bowden EM, Gernsbacher MA. 2000. Right and left hemisphere cooperation for drawing pre-
dictive and coherence inferences during normal story comprehension. Brain Lang. 71:310–36
Bolte A, Goschke T, Kuhl J. 2003. Emotion and intuition: effects of positive and negative mood on implicit
judgments of semantic coherence. Psychol. Sci. 14:416–21
Botvinick MM, Cohen JD, Carter CS. 2004. Conflict monitoring and anterior cingulate cortex: an update.
Trends Cogn. Sci. 8:539–46
Bowden EM. 1997. The effect of reportable and unreportable hints on anagram solution and the aha! experi-
ence. Conscious. Cogn. 6:545–73
Bowden EM, Beeman MJ. 1998. Getting the right idea: Semantic activation in the right hemisphere may help
solve insight problems. Psychol. Sci. 9(6):435–40
Bowden EM, Jung-Beeman M. 2003a. Normative data for 144 compound remote associate problems. Behav.
Res. Methods 35(4):634–39
Bowden EM, Jung-Beeman M. 2003b. Aha! Insight experience correlates with solution activation in the right
hemisphere. Psychon. Bull. Rev. 10:730–37
Bowden EM, Jung-Beeman M, Fleck J, Kounios J. 2005. New approaches to demystifying insight. Trends
Cogn. Sci. 9(7):322–28
Bruny´
e TT, Gagnon SA, Paczynski M, Shenhav A, Mahoney CR, Taylor HA. 2013. Happiness by association:
Breadth of free association influences affective states. Cognition 127:93–98
Bush G, Luu P, Posner MI. 2000. Cognitive and emotional influence in anterior cingulate cortex. Trends Cogn.
Sci. 4:215–22
Carson SH, Peterson JB, Higgins DM. 2003. Decreased latent inhibition is associated with increased creative
achievement in high-functioning individuals. J. Personal. Soc. Psychol. 85:499–506
Chi RP, Snyder AW. 2011. Facilitate insight by non-invasive brain stimulation. PLoS ONE 6(2):e16655
Chi RP, Snyder AW. 2012. Brain stimulation enables the solution of an inherently difficult problem. Neurosci.
Lett. 515:121–24
Chiarello C. 1988. Lateralization of lexical processes in the normal brain: a review of visual half-field research.
In Contemporary Reviews in Neuropsychology, ed. HA Whitaker, pp. 36–76. New York: Springer-Verlag
Chiarello C, Burgess C, Richards L, Pollock A. 1990. Semantic and associative priming in the cerebral hemi-
spheres: some words do, some words don’t ...sometimes, some places. Brain Lang. 38:75–104
Chrysikou EG, Hamilton RH, Coslett HB, Datta A, Bikson M, Thompson-Schill SL. 2013. Noninvasive
transcranial direct current stimulation over the left prefrontal cortex facilitates cognitive flexibility in tool
use. Cogn. Neurosci. 4:81–89
Chun MM, Golomb JD, Turk-Browne NB. 2011. A taxonomy of external and internal attention. Annu. Rev.
Psychol. 62:73–101
Cranford EA, Moss J. 2012. Is insight always the same? A protocol analysis of insight in compound remote
associate problems. J. Probl. Solving 4(2):128–53
Derakshan N, Eysenck MW. 2009. Anxiety, processing efficiency, and cognitive performance: new develop-
ments from attentional control theory. Eur. Psychol. 14:168–76
Desimone R. 1998. Visual attention mediated by biased competition in extrastriate visual cortex. Philos. Trans.
R. Soc. Lond. B 353:1245–55
Dietrich A, Kanso R. 2010. A review of EEG, ERP, and neuroimaging studies of creativity and insight. Psychol.
Bull. 136:822–48
Dreisbach G. 2012. Mechanisms of cognitive control: the functional role of task rules. Curr. Dir. Psychol. Sci.
21:227–31
Dreisbach G, Goschke T. 2004. How PA modulates cognitive control: reduced perseveration at the cost of
increased distractibility. J. Exp. Psychol.: Learn. Mem. Cogn. 30(2):343–53
Easterbrook JA. 1959. The effect of emotion on cue utilization and the organization of behavior. Psychol. Rev.
66:183–201
13.20 Kounios ·Beeman
PS65CH13-Kounios ARI 30 August 2013 16:0
Federmeier KD, Kirson DA, Moreno EM, Kutas M. 2001. Effects of transient, mild mood states on semantic
memory organization and use: an event-related potential investigation in humans. Neurosci. Lett. 305:149–
52
Demonstrates that
schizotypes show more
creativity than normal
control participants in a
divergent thinking task.
Folley BS, Park S. 2005. Verbal creativity and schizotypal personality in relation to prefrontal hemi-
spheric laterality: a behavioral and near-infrared optical imaging study. Schizophr. Res. 80:271–82
F¨
orster J, Friedman RS, Liberman N. 2004. Temporal construal effects on abstract and concrete thinking:
consequences for insight and creative cognition. J. Personal. Soc. Psychol. 87:177–89
Fredrickson B, Branigan C. 2005. Positive emotions broaden the scope of attention and thought-action reper-
toires. Cogn. Emot. 19:313–32
Gasper K, Clore GL. 2002. Attending to the big picture: mood and global versus local processing of visual
information. Psychol. Sci. 13:34–40
Hutsler J, Galuske RA. 2003. Hemispheric asymmetries in cerebral cortical networks. Trends Neurosci. 26:429–
35
Isen AM, Daubman KA. 1984. The influence of affect on categorization. J. Personal. Soc. Psychol. 47:1206–17
Isen AM, Daubman KA, Nowicki GP. 1987. Positive affect facilitates creative problem solving. J. Personal.
Soc. Psychol. 52:1122–31
Isen AM, Johnson MM, Mertz E, Robinson GF. 1985. The influence of positive affect on the unusualness of
word associations. J. Personal. Soc. Psychol. 48:1413–26
Jacob R, Schall M, Scheibel AB. 1993. A quantitative dendritic analysis of Wernicke’s area in humans. II.
Gender, hemispheric, and environmental factors. J. Comp. Neurol. 327:97–111
Jarosz AF, Colflesh GJ, Wiley J. 2012. Uncorking the muse: Alcohol intoxication facilitates creative problem
solving. Conscious. Cogn. 21:487–93
Jensen O, Mazaheri A. 2010. Shaping functional architecture by oscillatory alpha activity: gating by inhibition.
Front. Hum. Neurosci. 4:186
Jung-Beeman M. 2005. Bilateral brain processes for comprehending natural language. Trends Cogn. Sci. 9:512–
18
Jung-Beeman M, Bowden EM, Haberman J, Frymiare JL, Arambel-Liu S, et al. 2004. Neural activity when
people solve verbal problems with insight. PLoS Biol. 2(4):e97
Kaplan CA, Simon HA. 1990. In search of insight. Cogn. Psychol. 22:374–419
Kerns JG, Cohen JD, MacDonald AW III, Cho RY, Stenger VA, Carter CS. 2004. Anterior cingulate, conflict
monitoring and adjustments in control. Science 303:1023–26
Kounios J. 1996. On the continuity of thought and the representation of knowledge: Electrophysiological
and behavioral time-course measures reveal levels of structure in semantic memory. Psychon. Bull. Rev.
3:265–86
Kounios J, Beeman M. 2009. The Aha! moment. The cognitive neuroscience of insight. Curr. Dir. Psychol.
Sci. 18(4):210–16
Kounios J, Fleck JI, Green DL, Payne L, Stevenson JL, et al. 2008. The origins of insight in resting-state
brain activity. Neuropsychologia 46(1):281–91
Kounios J, Frymiare JL, Bowden EM, Fleck JI, Subramaniam K, et al. 2006. The prepared mind: Neural activity
prior to problem presentation predicts subsequent solution by sudden insight. Psychol. Sci. 17(10):882–90
Kounios J, Osman AM, Meyer DE. 1987. Structure and process in semantic memory: new evidence based on
speed–accuracy decomposition. J. Exp. Psychol.: Gen. 116:3–25
Lane RD, Reiman EM, Axelrod B, Yun L, Holmes A. 1998. Neural correlates of emotional awareness: evidence
of an interaction between emotion and attention in the anterior cingulate cortex. J. Cogn. Neurosci. 10:525–
35
Ludmer R, Dudai Y, Rubin N. 2011. Uncovering camouflage: Amygdala activation predicts long-term memory
of induced perceptual insight. Neuron 69:1002–14
Luo J, Li W, Fink A, Jia L, Xiao X, et al. 2011. The time course of breaking mental sets and forming novel
associations in insight-like problem solving: an ERP investigation. Exp. Brain Res. 212:583–91
Luo J, Niki K. 2003. Function of hippocampus in “insight” of problem solving. Hippocampus 13:316–23
Mashal N, Faust M, Hendler T, Jung-Beeman M. 2008. Hemispheric differences in processing the literal
interpretation of idioms: converging evidence from behavioral and fMRI studies. Cortex 44:848–60
www.annualreviews.org The Cognitive Neuroscience of Insight 13.21
PS65CH13-Kounios ARI 30 August 2013 16:0
Mayberg HS, Liotti M, Brannan SK, McGinnis S, Mahurin RK, et al. 1999. Reciprocal limbic-cortical function
and negative mood: converging PET findings in depression and normal sadness. Am. J. Psychiatry 156:675–
82
Mednick S. 1962. The associative basis of the creative process. Psychol. Rev. 69(3):220–32
Metcalfe J, Wiebe D. 1987. Intuition in insight and noninsight problem solving. Mem. Cogn. 15(3):238–46
Metuki N, Sela T, Lavidor M. 2012. Enhancing cognitive control components of insight problem solving by
anodal tDCS of the left dorsolateral prefrontal cortex. Brain Stimul. 5:110–15
Meyer DE, Irwin DE, Osman AM, Kounios J. 1988. The dynamics of cognition and action: mental processes
inferred from speed-accuracy decomposition. Psychol. Rev. 95:183–237
Miller EK, Cohen JD. 2001. An integrative theory of prefrontal cortex function. Annu. Rev. Neurosci. 24:167–
202
Morgan WW. 1988. A morphological life. Annu. Rev. Astron. Astrophys. 26:1–10
Ray WJ, Cole HW. 1985. EEG alpha activity reflects attentional demands, and beta activity reflects emotional
and cognitive processes. Science 228:750–52
Reverberi C, Toraldo A, D’Agostini S, Skrap M. 2005. Better without (lateral) frontal cortex? Insight problems
solved by frontal patients. Brain 128:2882–90
Rowe G, Hirsch JB, Anderson AK. 2007. Positive affect increases the breadth of attentional selection. Proc.
Natl. Acad. Sci. USA 101:383–88
Sandk ¨
uhler S, Bhattacharya J. 2008. Deconstructing insight: EEG correlates of insightful roblem solving.
PLoS ONE 3(1):e1459
Scheibel AB, Fried I, Paul L, Forsythe A, Tomiyasu U, et al. 1985. Differentiating characteristics of the human
speech cortex: a quantitative Golgi study. In The Dual Brain: Hemispheric Specialization in Humans,ed.DF
Benson, E Zaidel, pp. 65–74. New York: Guilford
Seifert CM, Meyer DE, Davidson N, Patalano AL, Yaniv I. 1995. Demystification of cognitive insight: op-
portunistic assimilation and the prepared-mind hypothesis. In TheNatureofInsight, ed. R Sternberg, JE
Davidson, pp. 65–124. Cambridge, MA: MIT Press
Seldon HL. 1981. Structure of human auditory cortex. II. Cytoarchitectonics and dendritic distributions. Brain
Res. 229:277–94
Sheehan W. 2008. W. W. Morgan and the discovery of the spiral arm structure of our galaxy. J. Astron. Hist.
Herit. 11:3–21
Sheth BR, Sandkuhler S, Bhattacharya J. 2008. Posterior beta and anterior gamma oscillations predict cognitive
insight. J. Cogn. Neurosci. 21:1269–79
Smit DJ, Boomsma DI, Schnack HG, Hulshoff Pol HE, de Geus EJ. 2012. Individual differences in EEG
spectral power reflect genetic variance in gray and white matter volumes. Twin Res. Hum. Genet. 15:384–92
Smith RW, Kounios J. 1996. Sudden insight: all-or-none processing revealed by speed–accuracy decomposi-
tion. J. Exp. Psychol.: Learn. Mem. Cogn. 22(6):1443–62
St George M, Kutas M, Martinez A, Sereno MI. 1999. Semantic integration in reading: engagement of the
right hemisphere during discourse processing. Brain 122:1317–25
Sternberg RJ, Davidson JE. 1995. The Nature of Insight. Cambridge, MA: MIT Press
Subramaniam K, Kounios J, Parrish TB, Jung-Beeman M. 2009. A brain mechanism for facilitation of insight
by positive affect. J. Cogn. Neurosci. 21:415–32
Tallon-Baudry C, Bertrand O. 1999. Oscillatory gamma activity in humans and its role in object representation.
Trends Cogn. Sci. 3:151–62
Tardif E, Clarke S. 2001. Intrinsic connectivity of human auditory areas: a tracing study with Dil. Eur. J.
Neurosci. 13:1045–50
Topolinski S, Reber R. 2010. Gaining insight into the “aha” experience. Curr. Dir. Psychol. Sci. 19:402–5
Trope Y, Liberman N. 2010. Construal-level theory of psychological distance. Psychol. Rev. 117:440–63
van Steenburgh J, Fleck JI, Beeman M, Kounios J. 2012. Insight. In The Oxford Handbook of Thinking and
Reasoning, ed. K Holyoak, R Morrison, pp. 475–91. New York: Oxford Univ. Press
Verhaeghen P, Joormann J, Khan R. 2005. Why we sing the blues: the relation between self-reflective rumi-
nation, mood, and creativity. Emotion 5:226–32
Virtue S, Haberman J, Clancy Z, Parrish T, Jung-Beeman M. 2006. Neural activity of inferences during story
comprehension. Brain Res. 1084:104–14
13.22 Kounios ·Beeman
PS65CH13-Kounios ARI 30 August 2013 16:0
Virtue S, Parrish T, Beeman M. 2008. Inferences during story comprehension: cortical recruitment affected
by predictability of events and working memory capacity. J. Cogn. Neurosci. 20:2274–84
Describes a theory of
stages leading to insight
or “illumination”;
immersion in the facts
of a problem is the first
stage.
Wallas G. 1926. The Art of Thought. New York: Harcourt Brace
Wagner U, Gais S, Haider H, Verleger R, Born J. 2004. Sleep inspires insight. Nature 427(6972):352–55
Wegbreit E, Suzuki S, Grabowecky M, Kounios J, Beeman M. 2012. Visual attention modulates insight versus
analytic solving of verbal problems. J. Probl. Solving 4(2):artic. 5
Weissman DH, Gopalakrishnan A, Hazlett CJ, Woldorff MG. 2005. Dorsal anterior cingulate cortex resolves
conflict from distracting stimuli by boosting attention toward relevant events. Cereb. Cortex 15:229–37
Wu L, Knoblich G, Wei G, Luo J. 2009. How perceptual processes help to generate new meaning: an EEG
study of chunk decomposition in Chinese characters. Brain Res. 1296:104–12
www.annualreviews.org The Cognitive Neuroscience of Insight 13.23
... Since Guilford (1956) [112] first proposed divergent thinking (DT) and convergent thinking (CT) processes as two core processes of creative thinking, a variety of paradigms and materials have been employed to explore the cognitive and neural mechanisms of these processes. As a measure of CT, a compound remote associates measure (CRA) was developed patterned after the remote associates test (RAT) devised by Mednick [15,16]. These problems consist of three words (e.g. ...
... The CRA is well-suited for EEG and neuroimaging research because it involves brief timing epochs and the availability of a large number of normed task items supporting wellpowered designs, which has contributed to wide use of CRA for cognitive and brain-based studies of the CT component of CPS [17,18]. Notably, although the CRA presents puzzles to be solved, it is not a measure of problem solving in a traditional sense because (unlike classical CPS paradigms such as the candle problem, or the nine-dot problem) [16,17]), it emphasizes word associations rather than solution generation to meet the situational demands of a presented problem. ...
... In addition to regions associated with DT, several regions linked to CT, such as the right STG and the left MTG are also recruited in the prototype heuristic paradigm [72,74]. Right STG is associated with finding convergent solution words that satisfy multiple associative constraints in the CRA [5,16]. Like the CRA, the prototype heuristic paradigm requires convergent constraint satisfaction, as the problem solver first represents the required function of the problem, and then uses this required function to search their memory for alignable features in learned prototypes. ...
Article
Creative problem solving (CPS) in real-world contexts often relies on reorganization of existing knowledge to serve new, problem-relevant functions. However, classic creativity paradigms that minimize knowledge content are generally used to investigate creativity, including CPS. We argue that CPS research should expand consideration of knowledge-rich problem contexts, both in novices and experts within specific domains. In particular, paradigms focusing on creative analogical transfer of knowledge may reflect CPS skills that are applicable to real-world problem solving. Such paradigms have begun to provide process-level insights into cognitive and neural characteristics of knowledge-rich CPS and point to multiple avenues for fruitfully expanding inquiry into the role of crystalized knowledge in creativity.
... In addition to the ability for divergent thinking, lower WM has also been associated with enhanced insight-based reasoning (DeCaro et al., 2016). Insight refers to the sudden reinterpretation of a stimulus, situation, or event to produce a non-obvious interpretation, seemingly disconnected from the stream of conscious thought, that finds a solution to a problem or the comprehension of a joke or metaphor (Kounios and Beeman, 2014). Insight-based reasoning contrasts with deliberate, analytic, incremental problem solving, which is also associated with different patterns of brain activity (Kounios and Beeman, 2014). ...
... Insight refers to the sudden reinterpretation of a stimulus, situation, or event to produce a non-obvious interpretation, seemingly disconnected from the stream of conscious thought, that finds a solution to a problem or the comprehension of a joke or metaphor (Kounios and Beeman, 2014). Insight-based reasoning contrasts with deliberate, analytic, incremental problem solving, which is also associated with different patterns of brain activity (Kounios and Beeman, 2014). Given these characteristics, insight-based reasoning may be considered as a more explorative process and analytical reasoning of a more exploitative one. ...
Article
Full-text available
We raise the new possibility that people diagnosed with developmental dyslexia (DD) are specialized in explorative cognitive search, and rather than having a neurocognitive disorder, play an essential role in human adaptation. Most DD research has studied educational difficulties, with theories framing differences in neurocognitive processes as deficits. However, people with DD are also often proposed to have certain strengths – particularly in realms like discovery, invention, and creativity – that deficit-centered theories cannot explain. We investigate whether these strengths reflect an underlying explorative specialization. We re-examine experimental studies in psychology and neuroscience using the framework of cognitive search, whereby many psychological processes involve a trade-off between exploration and exploitation. We report evidence of an explorative bias in DD-associated cognitive strategies. High DD prevalence and an attendant explorative bias across multiple areas of cognition suggest the existence of explorative specialization. An evolutionary perspective explains the combination of findings and challenges the view that individuals with DD have a disorder. In cooperating groups, individual specialization is favored when features that confer fitness benefits are functionally incompatible. Evidence for search specialization suggests that, as with some other social organisms, humans mediate the exploration–exploitation trade-off by specializing in complementary strategies. The existence of a system of collective cognitive search that emerges through collaboration would help to explain our species’ exceptional adaptiveness. It also aligns with evidence for substantial variability during our evolutionary history and the notion that humans are adapted not to a particular habitat but to variability itself. Specialization creates interdependence and necessitates balancing complementary strategies. Reframing DD therefore underscores the urgency of changing certain cultural practices to ensure we do not inhibit adaptation. Key improvements would remove cultural barriers to exploration and nurture explorative learning in education, academia, and the workplace, as well as emphasize collaboration over competition. Specialization in complementary search abilities represents a meta-adaptation; through collaboration, this likely enables human groups (as a species and as cultural systems) to successfully adapt. Cultural change to support this system of collaborative search may therefore be essential in confronting the challenges humanity now faces.
... However, in some cases, the players can also fail the task several times, and try again until they suddenly have an Aha! moment and find a solution which is later further optimized. The Aha! moment is identified as a solution that appears suddenly thanks to intuition (Chronicle et al., 2004;Kounios and Beeman, 2014) and transforms the processing of the problem: when a solution arises, the information that was difficult to process can be processed more fluidly (Topolinski and Reber, 2010). This experience is also called "insight problem-solving" or "insight problem" (Smith, 1995) and can be conceptualized as a two-phase process (Chronicle et al., 2004): (i) the fruitless search/blocking phase in the face of a situation or questioning, (ii) the sudden emergence of a solution after a mental pause or re-evaluation of the problem. ...
Article
Full-text available
During the learning process, a child develops a mental representation of the task he or she is learning. A Machine Learning algorithm develops also a latent representation of the task it learns. We investigate the development of the knowledge construction of an artificial agent through the analysis of its behavior, i.e., its sequences of moves while learning to perform the Tower of Hanoï(TOH) task. The TOH is a well-known task in experimental contexts to study the problem-solving processes and one of the fundamental processes of children’s knowledge construction about their world. We position ourselves in the field of explainable reinforcement learning for developmental robotics, at the crossroads of cognitive modeling and explainable AI. Our main contribution proposes a 3-step methodology named Implicit Knowledge Extraction with eXplainable Artificial Intelligence (IKE-XAI) to extract the implicit knowledge, in form of an automaton, encoded by an artificial agent during its learning. We showcase this technique to solve and explain the TOH task when researchers have only access to moves that represent observational behavior as in human-machine interaction. Therefore, to extract the agent acquired knowledge at different stages of its training, our approach combines: first, a Q-learning agent that learns to perform the TOH task; second, a trained recurrent neural network that encodes an implicit representation of the TOH task; and third, an XAI process using a post-hoc implicit rule extraction algorithm to extract finite state automata. We propose using graph representations as visual and explicit explanations of the behavior of the Q-learning agent. Our experiments show that the IKE-XAI approach helps understanding the development of the Q-learning agent behavior by providing a global explanation of its knowledge evolution during learning. IKE-XAI also allows researchers to identify the agent’s Aha! moment by determining from what moment the knowledge representation stabilizes and the agent no longer learns.
... Being accompanied by positive emotional feedback, insight is also interpreted as the Aha/Eureka moment (Jung-Beeman et al., 2004;Tik et al., 2018). A sure and concrete response usually comes after such effortful thinking (Jung-Beeman et al., 2004;Kounios & Beeman, 2014). Recently, the dual-process theory of thought between associative process (also called System 1, intuition, and fast thinking) and hypothetical process (also called System 2, deliberation, and slow thinking) has been exploited to better understand human thought. ...
Article
Full-text available
The dual-process theory that two different systems of thought coexist in creative thinking has attracted considerable attention. In the field of creative thinking, divergent thinking (DT) is the ability to produce multiple solutions to open-ended problems in a short time. It is mainly considered an associative and fast process. Meanwhile, insight, the new and unexpected comprehension of close-ended problems, is frequently marked as a deliberate and time-consuming thinking process requiring concentrated effort. Previous research has been dedicated to revealing their separate neural mechanisms, while few studies have compared their differences and similarities at the brain level. Therefore, the current study applied Activation Likelihood Estimation to decipher common and distinctive neural pathways that potentially underlie DT and insight. We selected 27 DT studies and 30 insight studies for retrospective meta-analyses. Initially, two single analyses with follow-up contrast and conjunction analyses were performed. The single analyses showed that DT mainly involved the inferior parietal lobe (IPL), cuneus, and middle frontal gyrus (MFG), while the precentral gyrus, inferior frontal gyrus (IFG), parahippocampal gyrus (PG), amygdala (AMG), and superior parietal lobe were engaged in insight. Compared to insight, DT mainly led to greater activation in the IPL, the crucial part of the default mode network. However, insight caused more significant activation in regions related to executive control functions and emotional responses, such as the IFG, MFG, PG, and AMG. Notably, the conjunction analysis detected no overlapped areas between DT and insight. These neural findings implicate that various neurocognitive circuits may support DT and insight.
Article
We compared problem‐solving in four sets of classical insight and analytic problems in the verbal and spatial domains, and examined the impact of externalization (verbalization or sketching). In a within‐participants factorial design, we presented 24 classical insight and analytic problems, half verbal and half spatial. Participants solved these problems either while thinking aloud, while sketching, or in a baseline condition where neither was allowed. Higher solving accuracy was found in the analytic problems than in the insight ones as well as in the verbal problems than in the spatial ones. The verbal problems were also found to be solved faster than were spatial ones; in particular, verbal‐insight problems were solved faster than spatial‐insight and verbal‐analytic ones. Therefore, spatial‐insight problems stand out as particularly hard among the ones typically found in the literature. Surprisingly, no effects of externalization were found on problem‐solving speed or accuracy. We discuss the implications of our results for the theses that insight problems involve special processes.
Article
Full-text available
A popcorn popping is almost magical. And yet, the science of popcorn is safe and clear about the steps until the pop: the components, processes, and results of making popcorn. Nature has its own way to produce surprise in the form of “pops” (i.e., emergence, qualitative shifts). Emergent features spread throughout the life of taxa and individuals. A pop can be sudden and chaotic. And so is creativity. There is no incompatibility between creativity and naturalistic endeavors in science. Creativity is no god given gift blown inside humans. When creativity is defined by originality and spontaneity, it describes a feature with no past or present. I briefly summarize how one can see non-random innovation, no free occurring spontaneity, and non-heuristic effectiveness as features of behaviors that are not necessarily considered creative. Those three features reveal how traditional views of creativity undermine its real determiners and how it can be objectively defined and observed.
Chapter
We will consider the differentialDifferential geometry of the cerebral cortex from the point of view of heterogenesisHeterogenesis. Rather than considering the cortex as a network, as is frequently done, it will be regarded as a neuromagmaNeuromagma of differentialDifferential constraints defined on an assemblageAssemblage of manifoldsManifold. Every point of the manifoldManifold will eventually be joined to other points with integral curves always tangent to admissible distributions of planes generated by vector fields. The dynamics emerging from this virtuality will be closer to a heterogeneous version of fieldField theory than to the connective dynamics of either artificial neural networks, digital networks, or actor networks. In the following, we will outline the evolution of neurogeometrical models of the cortex, starting from the first models, based on group theory, which are homogeneous in space and time, progressing through models showing differentialDifferential heterogeneity in space alone, and ultimately arriving at models of embodied plasticityPlasticity in which a full deployment of spatial and temporal heterogeneity is put in place, opening onto the creation of new geometries and new possibility spaces. We are especially interested in considering perceptionPerception and imaginationImagination as processes of the passive and active compositionComposition of assemblagesAssemblage, in such a manner as to include the possibility of adding, subtracting, and modifying differentialDifferential fields that constitute the immanent plane of compositionComposition of cortical structures.
Article
Predominant ordinary ideas are insufficient for solving insight-like problems; they interfere with subordinate original ideas and can produce a mental impasse. However, how people monitor and control invalid predominant ideas remains largely unknown. In the current study, participants were asked to solve a sequence of several similar practice problems that had the same solution to strengthen a predominant idea; the participants were then presented with an insight-like test problem that could not be solved by the predominant idea. The results showed that if the test problem was similar to the practice problems in which the predominant idea could typically be applied, it elicited greater late sustained potential (LSP) over the whole brain but no conflict-related N2 or N400 components, which suggests that the participants did not experience cognitive conflict and continued to verify the predominant but currently invalid idea. When the test problem differed from the practice problems, the items that participants reported trying to solve elicited greater N2-N400 and LSP over the whole brain, which suggests that the participants experienced cognitive conflict and exerted more reactive control over the invalid predominant idea; in contrast, the items that participants reported thinking about how to solve did not evoke greater conflict-related N2-N400 components and evoked even lower LSP, which likely indicates an ineffective state. These findings demonstrate three kinds of cognitive control toward invalid predominant ideas in situations where they are typically and not typically applied and provide empirical evidence of a mental impasse in insight-like problem-solving behaviors.
Article
Full-text available
The literature on insight lists four main characteristics of this experience: (a) suddenness (the experience is surprising and immediate), ease (the solution is processed without difficulty), positive affect (insights are gratifying), and the feeling of being right (after an insight, problem solvers judge the solution as being true and have confidence in this judgment). Although this phenomenology is well known, no theory has explained why insight feels the way it does. We propose a fluency account of insight: Positive affect and perceived truth and confidence in one's own judgment are triggered by the sudden appearance of the solution for a problem and the concomitant surprising fluency gain in processing. We relate earlier evidence on insight concerning the impact of sudden fluency variations on positive affect and perceived truth and confidence.
Article
Full-text available
Behavioral and neuroimaging findings indicate that distinct cognitive and neural processes underlie solving problems with sudden insight. Moreover, people with less focused attention sometimes perform better on tests of insight and creative problem solving. However, it remains unclear whether different states of attention, within individuals, influence the likelihood of solving problems with insight or with analysis. In this experiment, participants (N = 40) performed a baseline block of verbal problems, then performed one of two visual tasks, each emphasizing a distinct aspect of visual attention, followed by a second block of verbal problems to assess change in performance. After participants engaged in a center-focused flanker task requiring relatively focused visual attention, they reported solving more verbal problems with analytic processing. In contrast, after participants engaged in a rapid object identification task requiring attention to broad space and weak associations, they reported solving more verbal problems with insight. These results suggest that general attention mechanisms influence both visual attention task performance and verbal problem solving.
Article
Full-text available
The literature on insight lists four main characteristics of this experience: (a) suddenness (the experience is surprising and immediate), ease (the solution is processed without difficulty), positive affect (insights are gratifying), and the feeling of being right (after an insight, problem solvers judge the solution as being true and have confidence in this judgment). Although this phenomenology is well known, no theory has explained why insight feels the way it does. We propose a fluency account of insight: Positive affect and perceived truth and confidence in one’s own judgment are triggered by the sudden appearance of the solution for a problem and the concomitant surprising fluency gain in processing. We relate earlier evidence on insight concerning the impact of sudden fluency variations on positive affect and perceived truth and confidence.
Article
Full-text available
Recent neuroscience evidence suggests that some higher-order tasks might benefit from a reduction in sensory filtering associated with low levels of cognitive control. Guided by neuroimaging findings, we hypothesized that cathodal (inhibitory) transcranial direct current stimulation (tDCS) will facilitate performance in a flexible use generation task. Participants saw pictures of artifacts and generated aloud either the object's common use or an uncommon use for it, while receiving cathodal tDCS (1.5 mA) either over left or right PFC, or sham stimulation. A forward digit span task served as a negative control for potential general effects of stimulation. Analysis of voice-onset reaction times and number of responses generated showed significant facilitative effects of left PFC stimulation for the uncommon, but not the common use generation task and no effects of stimulation on the control task. The results support the hypothesis that certain tasks may benefit from a state of diminished cognitive control.
Article
Anterior cingulate cortex (ACC) is a part of the brain's limbic system. Classically, this region has been related to affect, on the basis of lesion studies in humans and in animals. In the late 1980s, neuroimaging research indicated that ACC was active in many studies of cognition. The findings from EEG studies of a focal area of negativity in scalp electrodes following an error response led to the idea that ACC might be the brain's error detection and correction device. In this article, these various findings are reviewed in relation to the idea that ACC is a part of a circuit involved in a form of attention that serves to regulate both cognitive and emotional processing. Neuroimaging studies showing that separate areas of ACC are involved in cognition and emotion are discussed and related to results showing that the error negativity is influenced by affect and motivation. In addition, the development of the emotional and cognitive roles of ACC are discussed, and how the success of this regulation in controlling responses might be correlated with cingulate size. Finally, some theories are considered about how the different subdivisions of ACC might interact with other cortical structures as a part of the circuits involved in the regulation of mental and emotional activity.
Article
Time-course studies of semantic verification are reviewed, discussed, and reinterpreted with the aim of drawing general theoretical conclusions about semantic memory structure. These reaction time, speed-accuracy tradeoff, speed-accuracy decomposition, and event-related (brain) potential (ERP) studies suggest that semantic memory is structured on at least three levels. In particular, specific models of the intermediate (macrostructural) level are discussed and compared. ERP investigations of this level suggest that context-independent and context-dependent types of semantic information are potentially isolable and analyzable.
Article
Cognitive control enables humans to flexibly switch between different thoughts and actions. An important prerequisite for this cognitive flexibility is the human ability to form and apply general task rules. In this article, I review research investigating the functional role of task rules, with an emphasis on two main findings. First, the shielding function of task rules helps guide attention toward task-related information, thereby reducing possible distraction by irrelevant information. Second, this task shielding has to be relaxed when a task rule changes, thereby making the cognitive system more vulnerable to the intrusion of distracting information. Implications for developmental psychology and higher-level cognition are discussed.
Article
Two experiments examined hemispheric differences in information processing that may contribute to solving insight problems. We propose that right-hemisphere (RH) coarse semantic coding is more likely than left-hemisphere (LH) fine semantic coding to activate distantly related information or unusual interpretations of words, and thus more likely to activate solution-relevant information for insight problems. In Experiment 1, after trying to solve insight problems, participants read aloud solution or unrelated target words presented to the left visual field (lvf) or right visual field (rvf). Participants showed greater lvf-RH than rvf-LH priming for solutions for solved problems and priming only in the lvf-RH for unsolved problems. In Experiment 2, participants showed an lvf-RH advantage for recognizing solutions to unsolved problems. These results demonstrate that in a problem-solving context, there was greater activation of solution-relevant information in the RH than in the LH. This activation is useful for recognizing, and perhaps producing, solutions to insight problems.
Article
There have been many attempts to account theoretically for the effects of anxiety on cognitive performance. This article focuses on two theories based on insights from cognitive psychology. The more recent is the attentional control theory (Eysenck, Derakshan, Santos, & Calvo, 2007), which developed from the earlier processing efficiency theory (Eysenck & Calvo, 1992). Both theories assume there is a fundamental distinction between performance effectiveness (quality of performance) and processing efficiency (the relationship between performance effectiveness and use of processing resources), and that anxiety impairs processing efficiency more than performance effectiveness. Both theories also assume that anxiety impairs the efficiency of the central executive component of the working memory system. In addition, attentional control theory assumes that anxiety impairs the efficiency of two types of attentional control: (1) negative attentional control (involved in inhibiting attention to task-irrelevant stimuli); and (2) positive attentional control (involved in flexibly switching attention between and within tasks to maximize performance). Recent (including unpublished) research relevant to theoretical predictions from attentional control theory is discussed. In addition, future directions for theory and research in the area of anxiety and performance are presented.
Article
The extent to which the right hemisphere can subserve language functions is one of the most hotly debated issues in neuropsychology today, as witnessed by the lively interchanges which have appeared in several recent journals (Gazzaniga, 1983a, 1983b; Levy, 1983; Zaidel, 1983b; Myers, 1984; Patterson & Besner, 1984a, 1984b; Rabinowicz & Moscovitch, 1984; Zaidel & Schweiger, 1984). A focal issue in this controversy concerns right hemisphere involvement in the recognition of visually presented words, and much of the debate has turned on differing interpretations of evidence from commissurotomy and acquired dyslexia. Some have argued for extensive participation of the right hemisphere (Coltheart, 1980a; Zaidel, 1983b), while others find little reason to presume any significant role for the right hemisphere (Patterson & Besner, 1984a, 1984b; Gazzaniga, 1983a, 1983b). In the context of such discussions it is somewhat surprising that evidence from lateralization studies with neurologically normal individuals has been given only piecemeal consideration. For more than a decade researchers have been exploring the consequences of selectively stimulating the cerebral hemispheres by using tachistoscopic methods, with words being the most frequently employed stimuli. Any comprehensive model of hemisphere differences in lexical processing must encompass the results of such studies in addition to the evidence from neurologically compromised subjects.