Hemispheric lateralization in reasoning
Benjamin O. Turner
Nicole Marinsek
Emily Ryhal
Michael B. Miller
October 2, 2015
Abstract
A growing body of evidence suggests that reasoning in humans relies on a number of related processes whose
neural loci are largely lateralized to one hemisphere or the other. A recent review of this evidence 1 concluded
that the patterns of lateralization observed are organized according to two complementary tendencies. The
left hemisphere attempts to reduce uncertainty by drawing inferences or creating explanations, even at
the cost of ignoring conflicting evidence or generating implausible explanations. Conversely, the right
hemisphere aims to reduce conflict, by rejecting or refining explanations which are no longer tenable in
the face of new evidence. In healthy adults, the hemispheres work together to achieve a balance between
certainty and consistency, and a wealth of neuropsychological research supports the notion that upsetting
this balance results in various failures in reasoning, including delusions. However, the support for this
model from the neuroimaging literature is mixed. Here we examine the evidence for this framework from
across multiple research domains, including an activation likelihood estimation analysis of fMRI studies of
reasoning. Our results suggest a need to either revise this model as it applies to healthy adults, or else to
develop better tools for assessing lateralization in these individuals.
Keywords: Split-brain, unilateral brain damage, meta-analysis
Introduction
even more obvious asymmetry. The great majority
of humans are right hand dominant, and it is clear
that this dominance is due to a cerebral asymmetry
in the control of the hands rather than to mechanical or structural differences in the hands 4 . Even
though both of these processes are lateralized to the
left hemisphere and often assumed to be related, the
cerebral dominance for language is only weakly correlated with cerebral dominance for handedness 5 . And,
there are well-documented specializations in the right
hemisphere as well, including spatial processing and
face perception 6–8 .
Hemispheric specialization is a widely accepted principle of cortical organization in the human brain.
However, this specialization is best understood in
the context of the overwhelming symmetry evident
in the brains and bodies of most organisms, including humans 2,3 . For example, the nervous system of
most organisms evolved symmetrical bilateral control
of both sides of the body in order to move in a linear
fashion through a natural environment (unless the organism prefers to move in circles). Likewise, nervous
systems evolved symmetrical bilateral sensitivity to
predators approaching from either side of the visual
field. Across all species, this principle of symmetrical
bilateral control and sensitivity is ubiquitous across
brain functions and development.
So, the evolution of hemispheric specializations
swam against the tide of this overwhelming symmetry. The two clearest examples of this in the human
brain are language processing and handedness. The
lateralization of language processing has been known
since the pioneering work of Marc Dax, Paul Broca,
John Hughlings Jackson, Carl Wernicke, and others
on aphasic patients in the 1800s, showing that brain
damage restricted to the left hemisphere in most individuals was associated with deficits in language
production and comprehension. Handedness is an
The reasons for hemispheric lateralization and the
extent to which it is unique to humans is not well
understood. Anatomical studies of the asymmetry
in microcolumns and macrocolumns in cortical regions that support language processing suggest that
lateralization may be fixed (or at least, strongly predisposed) in the human brain—that is, early ontogenetic or genetically determined events might have led
to asymmetries in cytoarchitecture that in turn led
to hemispheric differences in function 9–11 —although
this relationship is not perfect 12 . It has also been suggested that sophisticated functions like skilled manual operations (praxis) and language involve heavy
processing demands. Such processing may require
localized neural circuitry that is efficiently encap1
sulated in one region and does not depend on the
transfer of information across the corpus callosum 3,8 .
Indeed, several functional specializations may have
evolved in this way within the human brain, including face recognition and theory of mind. And precursors of some of these specializations may be evident in other species as well. Corballis 5 suggests
that language may have slowly evolved through a
dual-stream system that underlies praxis and communication, leading to the lateralized gestural system that is evident in several primate species. The
final step in vocalizing these gestures within hominins
may have accelerated this lateralization and opened
up the hands for sophisticated tool use, and the voice
for more expressive communication.
Aside from these specialized brain regions that may
be lateralized to one hemisphere or the other, there
may also be pairs of identically homologous brain regions (across the two hemispheres) that nevertheless
carry out different functions, with one hemisphere
biased toward a different processing strategy than
the other hemisphere. Indeed, fMRI studies of intrinsic brain activity suggest that laterality can vary
independently across at least four different brain networks 13 . This duality of function may be adaptive in that it increases the neural capacity of the
brain 8,14,15 . Indeed, this has often been the argument
used to explain language processing in the left hemisphere versus spatial processing in the right 15 . An
even more striking example of hemispheric dual function is the unihemispheric slow-wave sleep that has
been observed in a number of terrestial and aquatic
animals, a phenomenon in which half the brain engages in deep slow-wave sleep while the other half is
awake and alert 16,17 .
There are also a number of examples across species
of hemispheric biases in functions. For example, reliable asymmetries have been observed in human brain
activity associated with approach and avoidance (or
conversely, distinct affective outcomes have been observed in patients with damage to one or the other
hemisphere) 18 . These findings accord with evidence
from the animal literature, including research on nonhuman primates and in rodents 19 . Similarly, “handedness” has been observed across multiple species,
including some species of birds and ungulates, along
with other behavioral asymmetries such as the direction of a dog’s tail-wag depending on whether it
is being approached by an unfamiliar dog or by its
owner 3 or the direction in which a toad attacks 14 .
Asymmetries are also present at the neural level in
other species—for example, in the population coding properties of honeybees’ left versus right primary
olfactory centers 20 .
We suggest this may be the case for reasoning:
that each hemisphere operates according to different
cost functions, which has the effect of predisposing
each hemisphere to certain types of processes. That
is, rather than asserting that any given process is
localized exclusively to a particular hemisphere, we
merely claim that the operational principles guiding each hemisphere renders it more suitable for certain processes. In much the same way that the right
hemisphere has come to be understood to have some
limited linguistic capacity 5 , each hemisphere may be
able to perform some of the functions we ascribe to
the contralateral hemisphere, but will tend to do so
in ways that are inefficient or strange.
The particular distinction at the center of our theory of reasoning is between uncertainty and conflict.
We propose that the left hemisphere is primarily concerned with reducing uncertainty, which it does by
generating explanations, filling in gaps in information, and drawing inferences, which are all manifestations of the left hemisphere’s guiding principle of
eliminating uncertainty. On the other hand, the right
hemisphere is sensitive primarily to conflict, which it
relieves by abandoning untenable hypotheses, initiating shifts to new hypotheses, and operating in a generally inhibitory or cautious fashion. In the healthy
brain, the hemispheres should synergistically cooperate in order to balance the need for fast but flexible hypothesis-making. However, when this balance
is upset, for example by damage to only one hemisphere, the behavioral effects will reflect the unhindered operation of the spared hemisphere, and should
have a systematic form depending on which hemisphere is damaged.
Ours is hardly the first theory to propose that each
hemisphere operates according to different principles,
nor to point out the effect these differences might
have on reasoning. For example, a number of theories
have characterized the hemispheres in similar (or at
least compatible) ways. Ramachandran 21 suggested
that the left hemisphere is willing to accept minor
discrepancies in order to maintain a status quo, and
that a “devil’s advocate” in the right hemisphere generates a paradigm shift in the case that the discrepancy exceeds some threshold. Likewise, Corballis 2
proposed that the left hemisphere is responsible for
“generativity,” while the right hemisphere operates
according to an analogue mode of representation. On
the basis of a comparative psychological approach,
Rogers 22 concluded that the left hemisphere is used
to consider multiple alternatives and choose strategies sequentially, and also in the manipulation of objects, while the right hemisphere operates in a parallel fashion and specializes in rapid, “species-typical”
responses. Similarly, Bradshaw and Nettleton 23 asserted that the “analytic” left hemisphere is adept at
2
operating in the sequential or temporal domain (including, e.g., perceiving duration or rhythm), while
the “holistic” right hemisphere is characterized by
spatial processing. And Beeman and Bowden 24 proposed that strength of the left hemisphere is defined
by relatively fine semantic coding, while the right
hemisphere’s strength draws on coarse semantic coding.
The role of laterality in emotion has also spawned
a number of general theories of hemispheric specialization. For instance, the left hemisphere has repeatedly been associated with positive emotional valence
and approach behaviors, and the right hemisphere
with negative emotional valence and avoidance behaviors 25–27 . Several theories have proposed that
these tendencies are in fact driven by a more basic
division, for instance, associating the left hemisphere
with “energy enrichment” and the right hemisphere
with “energy expenditure” 25 ; or that the left hemisphere acts as a deliberate, conscious approach system (including goal-oriented and exploratory behavior), while the right hemisphere operates in a nondeliberate, non-conscious fashion, and that it is especially attentive to novelty (or threat), which leaves it
better positioned to drive avoidant behavior” 26 .
It is clear that most of these theories ascribe similar roles to the two hemispheres, which are broadly
consistent with the distinctions we make in our own
theory. The left hemisphere is consistently identified with local processing (in time as well as possibly
in psychological space) and a generally more “acceptant” attitude (for instance, in ignoring conflicting
evidence or being approach-centric). The right hemisphere is somewhat less clearly characterized, but is
nonetheless associated with more global processing,
which it may carry out in parallel, and that it is more
“skeptical” than the left hemisphere. These commonalities align well with our proposal that the left hemisphere operates in a space defined by certainty rather
than conflict—perhaps because its scope is too narrow to detect such conflict—and vice versa for the
right hemisphere. These theories exist at different
levels of explanation, both in terms of the ramifications of the distinctions they identify as well as the
proposed origins or mechanisms underlying those distinctions, but they all seem to be recognizing a common truth.
Researchers must be careful, however, to not be
too fast and loose with claims about hemispheric
asymmetries. Previous claims of asymmetries have
been found to be more apparent than real. For example, one of the most prominent theories of memory of the last three decades—the Hemispheric Encoding/Retrieval Asymmetry (HERA) model 28,29 —
has been repeatedly shown to fail to account for
the patterns of results from studies of split-brain patients 30 , as well as neuroimaging data from healthy
adults 31,32 . It has been proposed that hemispheric
effects in memory experiments instead reflect other
distinctions, for instance, the nature of the studied
material 30 . Similarly, theories proposing strict lateralization of functions related to face processing have
been refuted; while there may be differences in how
faces are processed by each hemisphere 33 , there is
clearly some face-processing capacity in each hemisphere 34 .
In light of the possibility of misconstruing the nature of hemispheric asymmetry in reasoning, clear
tests of laterality are needed to verify any such claims
of asymmetry. A common theme in the abovediscussed cases of “apparent” laterality has been the
diverging conclusions reached when using different research methodologies, so in this article, we review the
various lines of evidence regarding the hemispheric
laterality of reasoning. In addition to discussing the
evidence per se, we also point out potential strengths
and weaknesses of each line of evidence. We conclude
by applying a test of laterality to the neuroimaging evidence using an activation likelihood estimation
(ALE) meta-analysis of the existing fMRI literature,
the results of which confirm that neuroimaging represents an approach for which especial care is needed
when investigating laterality.
Studies of split-brain patients
There is probably no condition more appropriate for
asking questions about laterality than the split-brain.
Split-brain patients are those patients who have undergone corpus callosotomy, a surgical procedure in
which the corpus callosum is severed, as a treatment
for intractable epilepsy. In these patients, the two
hemispheres are almost completely unable to communicate (although the extent of the resection varies
from patient to patient, and the anterior and posterior commissures are usually left intact). To the
degree that there is competition between the hemispheres in the healthy brain, the “split-brain” allows
for investigation of the operation of each hemisphere
when freed completely of such competition. Corpus
callosotomies also carry lower concern for traumainduced processes (apoptosis, cortical reorganization,
etc.) than the various types of damage present in
patients with lesions. The experimental procedures
for addressing questions of laterality are also fairly
straightforward; the primary concerns are ensuring
information reaches the correct hemisphere, and finding a suitable means for querying the (languagedeficient) right hemisphere.
3
Some of the earliest evidence regarding the diverging tendencies of the two hemispheres with respect to
reasoning comes from studies of split-brain patients
conducted by Michael Gazzaniga. For example, Gazzaniga and Smylie 35 used a task that required participants to infer the relationship between objects (for
instance, a wooden log and a match would lead to the
conclusion of a bonfire). The left hemisphere was able
to perform the task while the right hemisphere was
unable to do so (see also 36 ). Moreover, the left hemisphere draws connections even between unrelated objects. In an experiment where each hemisphere independently selected an object based on a cue presented only to that hemisphere, the left hemisphere
was asked to explain the choices of both hemispheres.
One subject whose left hemisphere had selected a
chicken (in response to a picture of a chicken claw)
and whose right hemisphere had selected a shovel (in
response to a picture of snow) explained that, “The
chicken claw goes with the chicken and you need a
shovel to clean out the chicken shed,” despite the
fact that there was no extrinsic connection between
these objects 8 . The tendency of the left hemisphere
to create explanations led Gazzaniga to formulate the
idea of the Left Hemisphere Interpreter, which has a
propensity to explain everything it encounters.
Although the left hemisphere is able to draw inferences and generate explanations, it is less capable of
revising those explanations. When instructed specifically to adopt a new strategy, the left hemisphere
seems to be able to do so 37 , but there are also cases
wherein the left hemisphere perseverates with an explanation even in the presence of new information
(but without explicit instructions to update) 38 . The
left hemisphere is also prone to errors of commission.
For example, in recognition memory experiments, the
left hemisphere falsely identifies related lures as having been studied before, while the right hemisphere
correctly rejects these lures 39–41 . This is again consistent with the idea that the left hemisphere bridges
gaps between studied items and related, non-studied
items. The left hemisphere also strives to establish
patterns where there are none, even if doing so is
suboptimal: in a probability-matching experiment
wherein participants were asked to guess which of two
stimuli would appear on the following trial, the right
hemisphere adopted the optimal strategy (nearly always choosing the more frequent stimulus), while the
left hemisphere matched the baserates 42 . However,
note that this behavior is abolished if the stimuli differ in a way to which the left hemisphere is insensitive (e.g., using face stimuli), in which case the right
hemisphere actually exhibits matching behavior 43 .
The available evidence from split-brain patients
tells a fairly coherent story regarding the role of each
hemisphere in reasoning. The left hemisphere is able
to make inferences and to draw connections between
distinct pieces of information, but is unable to abandon the explanations it comes up with, either ignoring conflicting information or expanding the explanation. The right hemisphere is more veridical, and
although it did not demonstrate any other specific
strengths in regards to reasoning, the deficits demonstrated by the left hemisphere in these tasks suggest
that in healthy brains, the right hemisphere is responsible for halting perseveration of incorrect hypotheses
and reigning in the left hemisphere’s propensity for
filling in gaps. As mentioned above, this line of evidence is particularly compelling because of the purity
of hemispheric isolation. It is also worth restating
that to the degree that the hemispheres act in opposition in the intact brain, the functioning of each
hemisphere in the split brain reflects the operation
of that hemisphere released from the influence of the
other hemisphere (a point to which we will return
later) 44 .
Studies of unilaterally brain-damaged
(non-delusional) patients
Whereas the surgery that results in the split-brain
condition is relatively rare, brain damage or surgical
intervention limited to a single cerebral hemisphere
is far more common, although correspondingly less
homogeneous. The most common causes of unilateral brain damage include stroke and other related
ischemic events, various traumatic events such as motor vehicle accidents, and surgical intervention (for
instance, to treat epilepsy). The deficits associated
with unilateral brain damage are therefore extremely
varied, and depend on factors including the extent
of the damage, the locus (e.g., involvement of white
matter), the age at which damage occurred, and the
environment in which it occurred (e.g., in a surgical
setting with relatively less extensive damage outside
the focally damaged region compared to a traumatic
event in which nearly all of cortex may undergo postinjury responses). Despite this heterogeneity, there
are a number of phenomena relating to reasoning
which have been reliably observed in brain-damaged
patients.
Consistent with the results of studies in split-brain
patients, patients with right hemisphere brain damage are still able to draw inferences 45–50 , and in fact
are often subject to making the same sorts of hyperinferential errors as the left hemisphere in split-brain
patients 42,49 . Similarly, patients with right hemisphere damage can be “captured” by initial hypotheses and be unable to update or abandon those hy4
potheses 51–54 . Conversely, patients with left hemisphere damage have difficulty drawing inferences 50 ,
but remain sensitive to conflict, for example between
real-world knowledge and logic 45 or semantic inconsistencies 47 .
Despite the general consistency of this evidence,
several studies from this domain do report findings
that are inconsistent 53,55,56 . In addition to the possibility of cortical reorganization or other compensatory or injury-induced neural changes with unilateral brain damage, we should expect a less consistent effect in terms of how thoroughly the undamaged
hemisphere is released from the influence of the damaged hemisphere. That is, because damage to any
particular brain region may be incomplete, it may
retain the ability to partially influence the contralateral hemisphere, and thereby mitigate the measurable effects of unchecked operation of the contralateral hemisphere. Notwithstanding these concerns,
the results from studies of brain-damaged patients
are broadly consistent with our theory.
An influential theory of delusions 68 serves to explain both why delusions arise in these patients, and
also why split-brain (and right-hemisphere-damaged)
patients are surprisingly free of delusional behavior.
This theory hypothesizes that delusions require two
factors, of which only one is present in split-brain
patients: a dysfunctional belief evaluation system,
and an impairment that demands explanation. Delusional patients satisfy both factors, insofar as they
have hypoactivity in the right hemisphere, and the
left hemisphere is confronted with information for
which it has no schema (for example, the absence of a
limb, or degraded input from another damaged neural system) 68 . Delusions in these patients represent
the left hemisphere’s best attempt to explain such
phenomena. On the other hand, split brain patients
(that is, their disconnected speaking left hemisphere)
satisfy the first factor, but in the absence of any such
“unexplained” phenomena, we do not expect splitbrain patients’ left hemispheres to develop delusions.
By linking delusions simultaneously to a leftward shift in the left-/right-hemisphere balance, and
to a pattern of deficits characterized by excessive
inference-making and imperviousness to conflicting
information, it is possible to use evidence from studies of delusional patients to address questions of laterality in reasoning. The general conclusions we
reach are the same, although the form these deficits
take is more extreme than in most non-delusional
cases. This may be because these patients are confronted with information for which they have no established model, and for which they therefore generate bizarre explanations 72,73 . However, as with
brain damage that does not lead to delusions, there
may well be other confounded changes (besides hemispheric asymmetries) that accompany delusional disorders and lead to some of the effects we observe.
Moreover, because of their deficits, delusional patients present particular challenges from an experimental standpoint.
Studies of patients with delusional
disorders
Delusional disorders are associated with “fixed beliefs that are not amenable to change in light of conflicting evidence” 57 and have been consistently linked
with a shift in the balance of hemispheric control toward the left hemisphere. Generally, this is the result of damage to the right hemisphere 21,44,58–60 , but
there are also instances of left hemisphere hyperactivity in the absence of right hemisphere damage 44,61,62 .
As with unilateral brain damage that does not lead
to delusions, delusional patients are a very heterogeneous group, though they are definitionally united by
a common set of diagnostic symptoms.
Once again, the pattern of deficits observed across
a variety of classes of delusional disorder are in line
with what we predict on the basis of the left hemisphere hyperactivity (or right hemisphere hypoactivity) that seems to underlie most delusions. Delusional patients draw conclusions prematurely 63–67
and fail to move away from those conclusions appropriately 59,68–70 . Neuroimaging studies with delusional patients have also demonstrated that the right
hemispheres of delusional individuals with intact
right hemispheres have lost sensitivity to conflicts between evidence and beliefs 71 . These deficits combine
to produce strikingly abnormal behavior. For example, patients with anosognosia will deny that a limb
is paralyzed, even immediately after having been unable to perform a task with the affected limb 59 .
Behavioral studies in healthy adults
In healthy adults, the options for assessing laterality
are limited. Up until the past two decades, virtually the only option was carefully designed behavioral studies. In the intact brain, information travels
freely between the hemispheres, so the outcome measures are generally limited to subtle differences in, for
instance, processing speed as a function of presentation lateralization. Moreover, to whatever degree
there is tonic (rather than phasic, or event-related)
influence of one hemisphere on the other, lateralized
effects should be yet more muted. However, there
5
is less cause for concern that results may be due to
compensatory or other outside factors.
As it relates to lateralization in reasoning, there
are several distinct lines of evidence. The first comes
from priming studies of inference-making in language
tasks 24,74 . Although limited, this evidence is in line
with our predictions, insofar as the left hemisphere
was shown to be more involved in inference-making
than the right hemisphere. A second line of evidence comes from a study using rapid visual halffield stimulus presentation 75 , which demonstrated a
left hemisphere advantage in a visuospatial transitive inference task 76 . The last two lines of evidence
come from the persuasion literature, in which it has
been demonstrated that participants are more persuasible when they covertly direct their attention
to the left (which causes increased activation in the
right hemisphere) 77,78 , and are also more persuasible if they are inconsistently handed (which has been
shown to correlate with increased right hemisphere
activity and increased interhemispheric connectivity 79 ) 80,81 . Both of these are likewise in line with
our predictions: presuming participants begin with
some default belief which they must abandon—i.e.,
from which they must be persuaded—greater right
hemisphere activity (or better communication with
the left hemisphere) should be helpful.
tions of laterality. These methods differ in terms
of their spatial and temporal resolution, but nearly
all commonly used methods are spatially resolved
enough to distinguish between hemispheres. Like
neurostimulation, neuroimaging carries the benefit
of revealing the workings of the intact brain, although they do not allow for direct experimental
control in the same way. However, as with behavioral studies of healthy adults, neuroimaging studies reveal the activity of the hemispheres when they
are communicating normally. In fact, the behavioral
paradigms used in neuroimaging generally are not designed with lateralization in mind—in that they do
not attempt to strictly isolate the processes we believe to be lateralized—so the ability to separate the
hemispheres depends entirely on the common logic
of neuroimaging that any region active during a task
is associated (usually, causally) with performance or
processing of that task. Moreover, questions of laterality require direct contrasts between the hemispheres in order to avoid committing the “neuroimager’s fallacy” 209 —interpreting a difference in significance as implying a significant difference—which very
few studies do. We discuss these issues, in addition
to several others, in more detail below.
Unsurprisingly, the evidence for our theory from
fMRI is somewhat mixed. In line with our predictions, multiple studies have identified left-lateralized
activity (based on inspection of the reported peaks)
associated with reasoning about past events 108 , making inferences about related sentences 47,210–212 , attempting to generate explanations 213 , and inferring
a rule 178 . Conversely, right-lateralized activity has
been observed when logic conflicts with prior beliefs 49,125,214–216 , when participants are preparing for
a set shift 217,218 or receiving feedback necessitating
such a shift 219 , and generally in context monitoring and inhibition 56,220,221 , including inhibiting outlandish hypotheses 222 . However, most of this evidence is fairly qualitative and lacks direct contrasts
between the hemispheres; there are also studies whose
results seem to at least partially contradict our predictions 218,223 .
Whereas most of the research from the previous
lines was directly addressing questions of laterality,
and is fairly straightforward to combine, the results
from fMRI generally lack a focus on lateralization,
and because they are presented as a series of peaks
of activation spanning the entire brain, it is more difficult to qualitatively assess correspondence between
studies. In order to address these limitations, we undertook an activation likelihood estimation (ALE)
meta-analysis of fMRI studies of reasoning. This
analysis technique allows us to assemble a quantitative understanding of the cumulative patterns of
Cognitive neuroscience studies in
healthy adults
By far the most widely used approach to investigating the localization of cognitive processes in healthy
adults is cognitive neuroscience, in particular functional magnetic resonance imaging, but also including electroencephalography (which we will not discuss
here) and transcranial magnetic stimulation (TMS).
Neurostimulation methods offer a means to emulate
certain types of brain damage, and also lend themselves to causal investigations into the role of particular brain regions. However, suppression or enhancement of any given region is not complete; certain areas are beyond the reach of existing stimulation technologies; and the exact mechanism of some
of these technologies is still not fully understood.
As with behavioral studies, there is scant evidence
from this domain regarding the lateralization of reasoning, but what there is goes directly against our
predictions: disruption of right hemisphere increased
(rather than decreased) conflict sensitivity as measured by the belief-bias effect 82,83 .
Neuroimaging methods are currently far more
prevalent than neurostimulation methods, although
they too are used only infrequently to address ques6
Search term
Abstract reasoning
Analogical reasoning
Categorical reasoning
Conceptual reasoning
Deduction tasks
Deductive reasoning
Delis-Kaplan
Everyday reasoning
Inductive reasoning
Inferential reasoning
Raven’s progressive matrices
Remote associate’s test
Syllogism
Syllogistic
Twenty questions test
Wason selection
Wisconsin card sorting task
Search results
Allowed
Not relevant
32
25
7
20
7
31
5
15
29
10
21
23
11
8
5
9
70
7
12
2
6
3
13
2
5
13
4
13
11
7
3
2
5
17
4 84–87
3 91–93
1 103
105–108
4
0
1 114
2 127,128
4 129–132
2 134,135
3 147–149
2 151,152
8 164–171
0
0
2 185,186
2 187,188
6 192–197
Excluded
Failed
Used
0
0
0
0
0
0
3 88–90
7 96–102
1 104
1 110
111–113
3
9 118–126
0
1 133
6 141–146
0
6 158–163
0
6 176–181
3 182–184
0
3 189–191
8 201–208
2
94,95
0
1 109
0
2 115,116
0
0
3 136–138
1 150
153–156
4
3 172–174
0
0
0
0
3 198–200
1 117
0
0
2 139,140
0
1 157
0
1 175
0
0
0
0
Table 1: Summary counts of ALE search results. “Allowed” are papers that passed our initial exclusion
screening (see Appendix); “Not relevant” are those papers for which no contrast was deemed relevant;
“Excluded” are those papers that failed our secondary exclusion screening; “Failed” are those papers for
which a technical issue prevented inclusion; and “Used” are those papers which contributed at least one
relevant contrast (see Table 5).
ALE meta-analysis of fMRI studies of
reasoning
activation across multiple studies (a summary of the
topics covered is given in Table 1). Briefly, ALE
treats each study or contrast as providing a Gaussian cluster of activity centered on the reported peak
location, with width and height dictated by the sample size of the study; information is aggregated across
studies by considering how much activation is present
in each voxel 224 .
Our primary hypotheses concerned the patterns of
activity associated with labels related to either hypothesis formation or hypothesis evaluation. Our
terms “building a model” and “rule finding” were
taken to load on the former, and “statement verification” and “rule checking” on the latter. In addition to the analyses using these core terms, we ran
ALE analyses on two other terms—“set-shifting” and
“conflict”—which we posited may load more heavily
on the right hemisphere than “statement verification”
or “rule checking” because they correspond better
with our prediction that the right hemisphere rejects
(rather than merely checks) incongruent hypotheses.
Figure 1 presents the thresholded results for the ALE
analyses using each of these labels, while Figure 2
(Appendix) gives unthresholded results. The coordinates for each cluster in Figure 1 are given in Tables 3–4.
We searched the literature for fMRI studies involving reasoning using the search terms listed in Table 1
in conjunction with the term “fMRI.” Then, two labelers (the first two authors) read redacted versions of
each paper (in which the results were removed) and
ascribed a label to each fMRI contrast. The labels
attempted to describe the cognitive process or processes that the contrast was designed to identify (for
example, “negative feedback processing” or “building a model”). Once all contrasts were labeled, the
labelers reconciled their labels to create one set of
descriptive labels for each contrast (Appendix, Table 5). ALE analysis were performed on the labels
we deemed most relevant. Additional details are provided in the Appendix; below, we present the results
of this analysis.
Although visual inspection of these figures suggests
clear patterns of lateralization (or a lack thereof), we
conducted a series of t-tests to quantify the degree
of lateralization in each figure. The results of those
tests are presented in Table 2, which were conducted
to test the hypothesis of greater activity in the left
hemisphere than the right hemisphere. These tests
7
confirm the general pattern of results evident in the
figures: there is strong left lateralization for both
“building a model” as well as “statement verification”
frontally, along with moderate right lateralization for
“conflict” frontally, striking bilaterality for “rule finding” and “rule checking,” and a lack of any consistent
pattern for “set-shifting.”
According to our framework, we predicted left lateralization for the first two labels in Table 2 and right
lateralization for the last four. What we observed
bears little resemblance to those predictions. In particular, although “building a model” demonstrated
the expected left lateralization, “statement verification” exhibited a very similar (rather than opposite)
left-lateralized pattern. Similarly, “rule finding” and
“rule checking,” which should have opposite patterns
of lateralization, were both fairly bilateral. Finally,
“set-shifting” and “conflict” were both expected to
be strongly right-lateralized, but the former showed
no consistent pattern, and although the latter was
right-lateralized frontally, this was not significant (after corrections), and the rest of the brain was mixed.
However, there are a number of complicating factors
which might account for this disagreement.
First and most obviously, the majority of the tasks
we included were not designed to separate out the
various processes in which we are interested. Although we attempted to “enforce” such separation
through our labeling scheme, and by using exclusive
ALE analyses, this process was only an approximation. Also, we did not attempt to exclude other processes which may have been present in some of the
contrasts. For instance, we noted processes associated with decision-making, abstract thinking, and
cognitive effort, each of which may be bilateral (or
lateralized opposite to the primary label) and may
therefore give the appearance of bilaterality for the
associated label, but did not exclude contrasts with
these labels, as doing so would have left us with too
few contrasts.
Second, the way in which we applied our labels
inadvertently ended up mapping more or less onto
a distinction between tasks that relied on linguistic
processes and those that relied on spatial or visual
processes. For example, most tasks that employed
written syllogisms were given the labels “building a
model” to describe the process of creating a cognitive model based on the premises of the syllogism
and “statement verification” to describe the process
of evaluating the validity of the conclusion. Similarly, tasks such as the Wisconsin Card Sorting Task
(WCST) were given the labels “rule finding” to describe the process of searching for an unknown rule
“rule checking” to describe implementing a rule and
using feedback to verify its accuracy. This distinction
was unintentional, but may explain the stark divide
between these two sets of labels, with the former being strongly left-lateralized (as expected for linguistic
processes) and the latter more bilateral.
In addition to this linguistic/non-linguistic distinction, our labels differed in another way: namely,
it is theoretically plausible that during the “statement verification” phase of tasks to which that label
was applied (e.g., syllogism tasks), there was still a
considerable amount of hypothesis formation ongoing (as the participant attempted to incorporate the
syllogism’s conclusion into their model, for example),
while it seems somewhat less likely that participants
in these sorts of tasks were engaging in “statement
verification” during the “building a model” phases of
these taks. Likewise, during the “rule finding” phase
of tasks given that label (e.g., Raven’s Progressive
Matrices or the WCST), it may be that participants
were engaging in a more rapid process of alternating between “finding” and “checking” their rules, but
that the “rule checking” phase truly contained relatively less “rule finding.”
Third, the interpretation of BOLD activity as reflecting causal processing is speculative. Particularly
when it comes to regions which are posited to operate in roughly analogous ways and to act on approximately the same input, the assumption that the
more active region is the causally responsible region
requires careful thought. It may be that greater activity reflects greater (ineffective) effort, for example. Moreover, in the intact brain, even if there existed a situation that loaded purely on the processes
of only one hemisphere, we are agnostic to how the
hemispheres would interact—perhaps the other hemisphere would adopt a supportive role, rather than either continuing in its normal mode or “turning off.”
Finally, outside the limitations of the tasks we considered, our specific label-dependent meta-analysis
approach, or the ambiguities surrounding the interpretation of BOLD activity differences in these situations, it may be that fMRI is fundamentally ill-suited
to addressing questions of lateralization in the intact
brain. Given the small effect sizes associated with
most higher-order processes in fMRI, it may be that
the biases in hemispheric dominance during performance of a reasoning task are simply too subtle to be
detected using fMRI, even with the power afforded by
an ALE approach. Or it may be that the dual effects
of hemispheric isolation present in patient studies—
pure activation of one hemisphere and release from
contribution by the other hemisphere—are responsible for the effects observed in those studies, and that
homeostatic processes are powerful enough in the intact brain to prevent any asymmetries detectable by
fMRI.
8
Figure 1: ALE results for each label. Translucent clusters correspond to Type II-matched thresholding; solid
clusters correspond to Type I-matched thresholding. Values are ALE scores.
9
Label
Frontal
Parietal
Occipital
Temporal
Cerebellar
Subcortical
Building a model
Rule finding
Statement verification
Rule checking
Set-shifting
Conflict
4.722
-1.077
3.782
-0.350
0.574
-2.469
1.849
-0.310
2.698
1.247
1.310
2.772
3.520
0.887
0.479
-0.829
2.663
2.987
0.482
0.505
1.740
0.896
0.649
-1.272
0.910
0.329
-0.351
0.263
-2.692
0.892
2.084
1.151
-1.776
0.287
0.445
-0.301
Table 2: Results of tests of laterality for Figure 2. t-values have been converted to z-values for comparability
across regions. Test is of left hemisphere greater than right hemisphere (so positive values denote left > right,
negative values right > left). Values in italics are significant at p < 0.05, two-tailed, Bonferroni-corrected
for multiple comparisons.
For researchers interested in using fMRI to address
questions of lateralization, this is not to say there is
no hope. Indeed, we see several possible ways forward. The most obvious approach is to endorse the
assumption that BOLD activity reflects effective processing, and to design a task that separates the processes ascribed to each hemisphere as fully as possible. For example, a hypothetical task that included
some events or epochs that loaded entirely on hypothesis formation and others that loaded entirely
on hypothesis evaluation would offer a strong test of
lateralization in reasoning in the intact brain.
challenges—for instance, sifting out the various activation associated with any particular task to identify
the activity associated with the putative common underlying process—it might be more straightforward
than attempting to parse processes in tasks in which
those processes are inextricably linked.
The most obvious remedy for these problems in
cognitive neuroscience lies in neuromodulatory techniques, including TMS and transcranial direct current stimulation. Because these methods can (reversibly) emulate brain damage, they allow a direct
link between the patient literature and studies of laterality in healthy adults. In fact, these methods allow direct testing of causal questions in a way that
is impossible in patients (for instance, by targeting
each hemisphere in turn in the same individual 45 ).
As discussed above, these methods are not without
their limitations, but they may nonetheless represent
the best path forward for investigating questions of
lateralization in healthy adults.
Unfortunately, as stated previously, our framework
presumes that both hemispheres will attempt to use
all available information. Thus, although they will
perform different computations on that information,
the net amount of activity in each hemisphere may
be similar, even during times when one or the other is
ultimately driving behavior. However, even given the
limitations of interpreting activity in any such task, it
may still be possible to answer questions of laterality.
For example, differences in asymmetry during differ- Conclusion
ent events (or from other measures entirely) could be
related with, for example, different types of reasoning The combined evidence from across multiple methoderrors across subjects.
ologies suggests that successful reasoning is the reAnother approach would be to conduct a meta- sult of the interaction of two lateralized sets of proanalysis using a broader survey of topics. Indeed, rea- cesses. The left hemisphere has consistently been assoning studies may ironically be a suboptimal domain sociated with explanation, and the right hemisphere
in which to study our proposed primary function of with monitoring and inhibition. There is little questhe right hemisphere—that is, to overhaul implausi- tion regarding these general tendencies, particularly
ble or inappropriate hypotheses. In most reasoning when the balance between the two hemispheres is upstudies, participants dont need to overhaul their be- set. However, it is less clear how individuated the
liefs. Instead, they perform simple evaluations like hemispheres are in the intact brain. Although we do
checking a rule or verifying if a statement matches not believe that the tendencies we have described are
an inference, which may load onto the right hemi- epiphenomenal outcomes of brain damage, it may be
sphere according to our theory, but which is likely do that these tendencies are exaggerated in such cases.
so relatively weakly. One of the strengths of MarinThe other possible explanation for the results we
sek et al. 1 was that it related evidence from a mul- see in fMRI is methodological. Some isolated fMRI
titude of tasks to a common set of processes thought studies do lend support for our framework, so the
to be involved in reasoning. Although this presents ambiguous results from our ALE analysis may reflect
10
limitations of the ALE analysis itself—for example,
using tasks that failed to sufficiently separate the processes, or settling on a labeling scheme that happened
to pit language- and spatial-processing-related tasks.
Alternatively, current fMRI methods may simply be
inadequate to detect lateralization, at least without
careful attention to issues specific to tests of laterality. Either way, future work is clearly needed before
we can fully understand the role hemispheric lateralization plays in reasoning in the intact brain.
Acknowledgments
This work has been supported by the Institute
for Collaborative Biotechnologies through grant
W911NF-09-0001 from the U.S. Army Research Office. The content of the information does not necessarily reflect the position or the policy of the Government, and no official endorsement should be inferred.
11
Appendix
We conducted an ALE meta-analysis in conjunction
with manual labeling of every contrast from each paper included in the meta-analysis. Each of these steps
is described below.
Papers included in meta-analysis
To begin our meta-analysis, we conducted a series
of searches using Thomson Reuters Web of Science,
using the search terms given in Table 1. These
searches were conducted between August–December
2014. Every paper returned by each search was included initially (although duplicates were arbitrarily
assigned to a single search term). An initial exclusion screening removed papers that met any of the
following descriptions: posters, conference abstracts,
patient-only studies, book chapters, reports without
foci in MNI or Talaraich space, other meta-analyses,
review papers, prefaces, and TMS studies. The peak
coordinates for each contrast in every remaining paper was recorded, along with the sample size for the
contrast.
Generating labels
Because our interest is in particular subprocesses of
reasoning, many of which were not the explicit focus of the included papers, we manually labeled each
contrast. This process had several steps: first, in order to avoid bias on the part of the labelers, each
paper was redacted independently by a research assistant, which involved blacking out every reference
to laterality (including the terms “right” or “left,”
along with all figures depicting SPMs and all references to any specific coordinates); the list of contrasts
and coordinates was likewise redacted to remove the
coordinates associated with each peak.
Next, two labelers (the first and second author)
read each paper independently and assigned a label
to every included contrast. The order in which the
labelers proceeded through the papers was partially
randomized: the first 17 papers spanned all 17 search
terms, and then the remaining papers were fully random. Each labeler proceeded through the first group
of 17 papers, and each subsequent group of 10 papers, in a different order (to reduce order effects).
After the initial group of 17 papers, and periodically
thereafter, the two labelers compared their respective lists of labels, but not to which contrasts those
labels were applied, in order to develop a common
lexicon while maintaining an independent use of that
lexicon inasmuch as possible. Finally, after both labelers completed labeling the entire set of included
papers, they met and reconciled the labels applied
to every contrast. In cases of disagreement, the labelers debated until agreement was reached as to the
appropriate label or labels.
Any papers with no relevant labels were excluded
from subsequent consideration; the number of papers
excluded for each search term is given in Table 1. Papers were additionally excluded if they included only
patients or patients outside the age range of 18–65,
if they used only a modality other than fMRI, or if
the analyses were other than standard GLM-based
analyses (e.g., functional connectivity analyses). An
additional five papers were excluded due to technical issues associated with converting coordinates to
Talairach space, or difficulty retrieving an electronic
copy of the article. Table 5 displays the labels assigned to the contrasts from every paper with at least
one contrast with a relevant label.
ALE meta-analyses
All meta-analyses were run with the labels generated as described above, using GingerALE v. 2.3.2.
All defaults were left unchanged, except for the
thresholding. Thresholding in GingerALE uses a
simulation-based approach to bootstrap the null distribution based on the actual observed number of
foci, number of groups, and sample sizes (but with
random locations), and to determine a threshold cluster size. We thresholded our maps in two ways:
first, in order to focus on the largest effects, we ran
our analyses using cluster-based thresholding with a
voxel-level threshold of p < 0.0001 and cluster-level
threshold of p < 0.01. Next, because our hypotheses are as much about an absence of activity in one
hemisphere as the presence of activity in the other
(although we actually do not believe that the segregation of processes to hemispheres is absolute, but is
instead characterized by more graded imbalances in
hemispheric control), we used cluster-based thresholding with a voxel- and cluster-level thresholds set
to approximately equate power across the maps—in
particular, we estimated effect sizes from the map
with most power (the “building a model” map), and
applied the thresholds needed in each other map in
order to achieve the same power, assuming the same
effect size. In effect, the results of the first analysis
are meant to identify the dominant regions associated
with each label, while the results of the second analysis are meant to confirm (or disconfirm) absences
of activity. The conservative threshold used in the
primary analysis was chosen a priori, agnostic to the
results.
We ran ALE analyses for several of our labels. Our
primary focus was on the distinction between pro-
12
Figure 2: Unthresholded ALE results for each label. Maps show max-projection of ALE scores along the z
dimension.
cesses associated with forming hypotheses and those
associated with evaluating hypotheses. Contrasts
aligned with either of these general processes were included only if they included the associated label and
not contrasting labels. Note that this is not the same
as running a direct contrast between the “included”
and “excluded” labels, but is just meant to achieve
greater specificity for the “included” label. In every
case, we visualize the unthresholded results of these
ALE analyses using a “glass brain” approach, maxprojecting along the z dimension in order to highlight patterns of hemispheric asymmetry (Figure 2).
We also present more standard 3D renderings of the
thresholded results, created using caret v. 5.65 after
transforming the thresholded maps from Talairach to
MNI space (Figure 1). Tables 3–4 present the coordinates for every cluster in Figure 1.
Testing hemispheric dominance
In order to get coarse metrics of the degree of hemispheric dominance for each ALE map, we computed
paired-samples t-tests across hemispheres for each of
the major lobes plus subcortex. In particular, in any
given map and for each lobe, we took the difference
between each voxel in one hemisphere and the corresponding contralateral voxel (i.e., the voxel with the
same y and z coordinates, and the sign-flipped x coordinate). Then we computed a t-test across all of
these differences, correcting the degrees of freedom
by estimating the number of resels in each lobe in order to account for the spatial smoothness inherent in
the ALE maps (in particular, we took the volume of
a resel to be the cube of the median FWHM used to
construct any given map). This approach naturally
balances the tradeoffs between amplitude and spatial extent of activation, because either a single high
(but narrow) peak, or a shallow (but broad) plateau,
in only one hemisphere will increase the value of the
13
resulting t statistic. However, caution is advised in
interpreting the results, which may in some cases be
based on relatively few independent studies.
14
Label
Region
x
y
z
mm3Type I
mm3Type II
Peak score
Building a model
L IFG
L MFG
L M1
L precuneus
R precuneus
L caudate
R SFG
−48
−6
−46
0
14
−15
23
11
9
−1
−61
−72
8
53
23
50
44
34
48
7
14
1384
1024
1000
720
520
488
376
n/a
n/a
n/a
n/a
n/a
n/a
n/a
0.042
0.028
0.033
0.028
0.025
0.018
0.022
Rule finding
L M1
L IPL
L fusiform
R MFG
R MFG
L M1
−44
−37
−41
45
24
−40
23
−52
−57
17
−8
−1
35
44
−10
35
55
40
1128
416
408
312
216
216
5232
3320
1424
4272
2528
2288
0.030
0.017
0.021
0.017
0.017
0.018
Statement verification
L MFG
−41
7
41
880
4864
0.025
Rule checking
L IFG
−44
38
3
544
1424
0.023
Set-shifting
L MFG
−31
44
15
632
4752
0.022
Conflict
R IFG
R IFG
52
26
13
29
29
−7
1976
712
6672
3688
0.027
0.014
Table 3: Cluster coordinates for “conservative” clusters from Figure 1. All coordinates are in Talairach
space and represent “center” of cluster. “Region” and “peak score” correspond to highest peak per cluster.
Because each cluster here is also present in a “liberal” version (except for “building a model,” for which the
“liberal” and “conservative” analyses were one and the same), the volume of each cluster is given according
to both thresholds, and these clusters are excluded from Table 4.
15
Label
Region
x
y
z
mm3
Peak score
Rule finding
R IPL
R insula
L insula
L MFG
R precuneus
L putamen
R IFG
40
32
−28
2
4
−29
54
−45
23
24
23
−71
2
17
43
3
5
45
50
5
10
2880
1336
1312
1208
1184
536
520
0.015
0.015
0.014
0.011
0.015
0.011
0.010
Statement verification
L IPL
L precuneus
L precuneus
L MFG
R SPL
R MFG
R IPL
L MFG
R occipital
−42
−14
−23
−6
24
31
39
−44
27
51
−72
−71
5
−62
−5
−52
42
−84
42
50
32
49
51
58
52
5
−6
3672
1256
1200
1144
1024
920
840
760
648
0.018
0.015
0.013
0.013
0.017
0.014
0.015
0.014
0.010
Rule checking
L STG
R occipital
R SFG
R IPL
L PCC
L ITG
L precuneus
R putamen
L insula
R caudate
R IFG
−53
5
3
40
−9
−55
−8
24
−32
17
49
−29
−86
18
−53
−38
−28
−54
0
0
−12
17
13
21
50
37
11
−14
60
−9
15
21
6
1152
1048
1008
720
696
648
632
616
608
600
576
0.017
0.015
0.014
0.010
0.015
0.013
0.014
0.013
0.012
0.013
0.009
Set-shifting
R claustrum
L M1
L caudate
L MFG
R cerebellum
R AG
L precuneus
L precuneus
34
−41
−10
−5
33
34
−32
−5
11
−13
11
18
−53
−55
−61
−73
5
30
9
43
−40
34
43
41
6984
4352
4272
4000
2936
2936
2864
2512
0.011
0.010
0.010
0.017
0.009
0.009
0.008
0.015
Conflict
L precuneus
−36
−74
39
2240
0.014
Table 4: Cluster coordinates for “liberal” clusters from Figure 1. All coordinates are in Talairach space and
represent “center” of cluster. “Region” and “peak score” correspond to highest peak per cluster.
16
Table 5: Labels associated with every contrast from each “used” paper (see Table 1).
17
Table 5 continued.
18
Table 5 continued.
19
Table 5 continued.
20
Table 5 continued.
21
Bibliography
[1]
[2]
[12]
Steven A Chance. The cortical microstructural
basis of lateralized cognition: a review. Frontiers in psychology, 5, 2014.
Nicole Marinsek, Benjamin O Turner, Michael
Gazzaniga, and Michael B Miller. Divergent
hemispheric reasoning strategies: reducing un- [13] Hesheng Liu, Steven M Stufflebeam, Jorge
Sepulcre, Trey Hedden, and Randy L Buckner.
certainty versus resolving inconsistency. FronEvidence from intrinsic activity that asymmetiers in human neuroscience, 8, 2014.
try of the human brain is controlled by multiple
factors. Proceedings of the National Academy of
Michael C Corballis. Laterality and human evoSciences, 106(48):20499–20503, 2009.
lution. Psychological Review, 96(3):492, 1989.
[3]
Michael C Corballis. The evolution and genet- [14] Giorgio Vallortigara. The evolutionary psychology of left and right: costs and benefits of laterics of cerebral asymmetry. Philosophical Transalization. Developmental psychobiology, 48(6):
actions of the Royal Society B: Biological Sci418–427, 2006.
ences, 364(1519):867–879, 2009.
[4]
James Steele and Natalie Uomini. Humans, [15] William D Hopkins and Claudio Cantalupo.
Theoretical speculations on the evolutionary
tools and handedness. Stone knapping: the necorigins of hemispheric specialization. Current
essary conditions for a uniquely hominin beDirections in Psychological Science, 17(3):233–
haviour (eds V. Roux & B. Bril), pages 217–
237, 2008.
239, 2005.
[5]
Michael Corballis. Whats left in language? be- [16] Niels C Rattenborg, Steven L Lima, and
Charles J Amlaner. Half-awake to the risk of
yond the classical model. Annals of the New
predation. Nature, 397(6718):397–398, 1999.
York Academy of Sciences, 1316, 2015.
[6]
Roger Sperry. Some effects of disconnecting the
cerebral hemispheres. Bioscience reports, 2(5):
265–276, 1982.
[7]
Michael C Corballis. The lopsided ape: Evolution of the generative mind. Oxford University
Press, 1991.
[8]
[17]
Jennifer L Lapierre, Peter O Kosenko, Oleg I
Lyamin, Tohru Kodama, Lev M Mukhametov,
and Jerome M Siegel. Cortical acetylcholine
release is lateralized during asymmetrical slowwave sleep in northern fur seals. The Journal
of Neuroscience, 27(44):11999–12006, 2007.
[18]
Richard J Davidson. Cerebral asymmetry, emotion, and affective style. The MIT Press, 1995.
M S Gazzaniga. Cerebral specialization and interhemispheric communication: does the cor- [19] Richard J Davidson, Daren C Jackson, and
pus callosum enable the human condition?
Ned H Kalin. Emotion, plasticity, context,
Brain : a journal of neurology, 123 ( Pt 7:1293–
and regulation: perspectives from affective neu1326, 2000.
roscience. Psychological bulletin, 126(6):890,
2000.
[9] Jeffrey Hutsler and Ralf AW Galuske. Hemispheric asymmetries in cerebral cortical net- [20] Elisa Rigosi, Albrecht Haase, Lisa Rath, Gianworks. Trends in neurosciences, 26(8):429–435,
franco Anfora, Giorgio Vallortigara, and Paul
2003.
Szyszka. Asymmetric neural coding revealed by
in vivo calcium imaging in the honey bee brain.
[10] Pierre-Yves Herve, Laure Zago, Laurent PeProceedings of the Royal Society of London B:
tit, Bernard Mazoyer, and Nathalie TzourioBiological Sciences, 282(1803):20142571, 2015.
Mazoyer. Revisiting human hemispheric specialization with neuroimaging. Trends in cog- [21] V. S. Ramachandran. The evolutionary biolnitive sciences, 17(2):69–80, 2013.
ogy of self-deception, laughter, dreaming and
depression: Some clues from anosognosia. Med[11] Gregor Kasprian, Georg Langs, Peter C Brugical Hypotheses, 47(5):347–362, 1996.
ger, Mario Bittner, Michael Weber, Mavilde
Arantes, and Daniela Prayer. The prenatal ori- [22] Lesley J Rogers. Lateralised brain function in
gin of hemispheric asymmetry: an in utero neuanurans: Comparison to lateralisation in other
roimaging study. Cerebral Cortex, page bhq179,
vertebrates. Laterality: Asymmetries of Body,
2010.
Brain and Cognition, 7(3):219–239, 2002.
22
[23]
John L Bradshaw and Norman C Nettleton.
The nature of hemispheric specialization in
man. Behavioral and Brain Sciences, 4(01):51–
63, 1981.
evidence from word impairments in prosopagnosia and face impairments in pure alexia.
Cerebral Cortex, page bhs390, 2012.
[35]
[24]
Michael S. Gazzaniga and Charlotte S.
Smylie. DISSOCIATION OF LANGUAGE
AND COGNITION. Brain, 107:145–153, 1984.
M J Beeman and E M Bowden. The right hemisphere maintains solution-related activation for
yet-to-be-solved problems. Memory & cogni- [36] Matthew E Roser, Jonathan a Fugelsang,
tion, 28(7):1231–41, 2000. ISSN 0090-502X.
Kevin N Dunbar, Paul M Corballis, and
Michael S Gazzaniga. Dissociating processes
[25] AD Craig. Forebrain emotional asymmetry: a
supporting causal perception and causal inferneuroanatomical basis? Trends in cognitive science in the brain. Neuropsychology, 19(5):591–
ences, 9(12):566–571, 2005.
602, 2005. ISSN 0894-4105.
[26] Claude M J Braun. Evolution of hemispheric [37] J E Ledoux, G L Risse, S P Springer, D H Wilspecialisation of antagonistic systems of manson, and M S Gazzaniga. Cognition and comagement of the body’s energy resources. Latermissurotomy. Brain : a journal of neurology,
ality, 12(5):397–427, 2007. ISSN 1357-650X.
100 Pt 1:87–104, 1977.
[27]
Guido Gainotti. Emotional behavior and hemi- [38] Michael B Miller, Walter Sinnott-Armstrong,
spheric side of the lesion. Cortex, 8(1):41–55,
Liane Young, Danielle King, Aldo Paggi, Mara
1972.
Fabri, Gabriele Polonara, and Michael S Gazzaniga. Abnormal moral reasoning in complete
[28] Endel Tulving, Shitu Kapur, FI Craik, Morris
and partial callosotomy patients. NeuropsyMoscovitch, and Sylvain Houle. Hemispheric
chologia, 48(7):2215–20, 2010. ISSN 1873-3514.
encoding/retrieval asymmetry in episodic
memory: positron emission tomography find- [39] Elizabeth A Phelps and Michael S Gazzaniga.
NOTE HEMISPHERIC DIFFERings. Proceedings of the National Academy of
ENCES
IN
MNEMONIC PROCESSING : OF
Sciences, 91(6):2016–2020, 1994.
LEFT HEMISPHERE INTERPRETATION.
[29] Reza Habib, Lars Nyberg, and Endel Tulving.
Neuropsychologia, 30:293–297, 1992.
Hemispheric asymmetries of memory: the hera
model revisited. Trends in cognitive sciences, 7 [40] Janet Metcalfe, Margaret Funnell, and
Michael S Gazzaniga.
Right-hemisphere
(6):241–245, 2003.
memory superiority: Studies of a split-brain
patient. Psychological Science, pages 157–164,
[30] Michael B Miller, Alan Kingstone, and
1995.
Michael S Gazzaniga. Hemispheric encoding
asymmetry is more apparent than real. Journal [41] Claude M J Braun, Josée Delisle, Anik Guiof cognitive neuroscience, 14(5):702–708, 2002.
mond, and Rafaël Daigneault. Post unilat[31]
[32]
Andy CH Lee, Trevor W Robbins, and
Adrian M Owen. Episodic memory meets working memory in the frontal lobe: functional neuroimaging studies of encoding and retrieval.
Critical Reviews in Neurobiology, 14(3-4), 2000. [42]
PC Fletcher and R N A Henson. Frontal lobes
and human memory insights from functional
neuroimaging. Brain, 124(5):849–881, 2001.
eral lesion response biases modulate memory:
crossed double dissociation of hemispheric specialisations. Laterality, 14(2):122–64, 2009.
ISSN 1464-0678.
George Wolford, Michael B Miller, and Michael
Gazzaniga. The left hemisphere’s role in hypothesis formation. The Journal of Neuroscience, 2000.
[43] Michael B Miller and Monica ValsangkarMing Meng, Tharian Cherian, Gaurav Singal,
Smyth. Probability matching in the right hemiand Pawan Sinha. Lateralization of face prosphere. Brain and cognition, 57(2):165–167,
cessing in the human brain. Proceedings of
2005.
the Royal Society B: Biological Sciences, 279
[44] Claude M J Braun and Sabrina Suffren. A gen(1735):2052–2061, 2012.
eral neuropsychological model of delusion. Cog[34] Marlene Behrmann and David C Plaut. Bilatnitive neuropsychiatry, 16(1):1–39, 2011. ISSN
eral hemispheric processing of words and faces:
1464-0619.
[33]
23
[45]
V L Deglin and M Kinsbourne. Divergent
thinking styles of the hemispheres: how syllogisms are solved during transitory hemisphere
suppression. Brain and cognition, 31(3):285–
307, 1996.
[46]
R Caplan and M Dapretto. Making sense dur- [55] Mark Beeman. Semantic processing in the right
ing conversation: an fMRI study. Neuroreport,
hemisphere may contribute to drawing infer12(16):3625–32, 2001. ISSN 0959-4965.
ences from discourse. Brain and language, 44
(1):80–120, 1993.
Evelyn C Ferstl, Thomas Guthke, and D Yves
von Cramon. Text comprehension after brain [56] Donald T Stuss and Michael P Alexander. Is
injury: left prefrontal lesions affect inference
there a dysexecutive syndrome? Philosophiprocesses.
Neuropsychology, 16(3):292–308,
cal transactions of the Royal Society of Lon2002. ISSN 0894-4105.
don. Series B, Biological sciences, 362(1481):
901–15, 2007. ISSN 0962-8436.
Connie A Tompkins, Wiltrud Fassbinder, Margaret Lehman Blake, Annette Baumgaertner, [57] American Psychiatric Association et al. Diagand Nandini Jayaram. Inference generation
nostic and statistical manual of mental disorduring text comprehension by adults with right
R
ders, (DSM-5).
American Psychiatric Pub,
hemisphere brain damageactivation failure ver2013.
sus multiple activation. Journal of Speech, Language, and Hearing Research, 47(6):1380–1395, [58] S. P. Stone, P. W. Halligan, and R. J. Green2004.
wood. The Incidence of Neglect Phenomena
and Related Disorders in Patients with an
Vinod Goel. Anatomy of deductive reasonAcute Right or Left-Hemisphere Stroke. Age
ing. Trends in cognitive sciences, 11(10):435–
and Ageing, 22:46–52, 1993.
41, 2007. ISSN 1364-6613.
[59] Anthony J Marcel, Richard Tegnér, and Ian
Carlo Reverberi, Tim Shallice, Serena
Nimmo-Smith. Anosognosia for plegia: speciD’Agostini, Miran Skrap, and Luca L.
ficity, extension, partiality and disunity of bodBonatti. Cortical bases of elementary deily unawareness. Cortex; a journal devoted to
ductive reasoning: Inference, memory, and
the study of the nervous system and behavior,
metadeduction.
Neuropsychologia, 47(4):
40(1):19–40, 2004.
1107–1116, 2009. ISSN 0028-3932.
[47]
[48]
[49]
[50]
Ildebrando Appollonio, Silvia Drei, and Paolo
Nichelli. Lesions to right prefrontal cortex
impair real-world planning through premature
commitments. Neuropsychologia, 51(4):713–24,
2013. ISSN 1873-3514.
H Brownell. Deficits in Right Brain-Damaged. [60] Orrin Devinsky. Delusional misidentifications
and duplications: right brain lesions, left brain
Brain and Language, 27:310–321, 1986.
delusions. Neurology, 72(1):80–87, 2009.
[52] W J Lombardi, P J Andreason, K Y Sirocco,
D E Rio, R E Gross, J C Umhau, and D W [61] Fritz Leutmezer, Ivo Podreka, Susanne Asenbaum, Uwe Pietrzyk, Helmut Lucht, Claude
Hommer. Wisconsin Card Sorting Test perBack, Norbert Benda, and Christoph Baumformance following head injury: dorsolateral
gartner. Postictal psychosis in temporal lobe
fronto-striatal circuit activity predicts perseepilepsy. Epilepsia, 44(4):582–90, 2003. ISSN
veration. Journal of clinical and experimental
0013-9580.
neuropsychology, 21(1):2–16, 1999. ISSN 13803395.
[62] Armida Mucci, Silvana Galderisi, Paola Bucci,
E Tresca, a Forte, Thomas Koenig, and Mario
[53] D T Stuss, B Levine, M P Alexander, J Hong,
Maj. Hemispheric lateralization patterns and
C Palumbo, and L Hamer. Wisconsin Card
psychotic experiences in healthy subjects. PsySorting Test performance in patients with fochiatry research, 139(2):141–54, 2005. ISSN
cal frontal and posterior brain damage : e ects
0165-1781.
of lesion location and test structure on separable cognitive processes. Neuropsychologia, 38:
[63] SF Huq, PA Garety, and DR Hemsley. Proba388–402, 2000.
bilistic judgements in deluded and non-deluded
[54] Vinod Goel, Oshin Vartanian, Angela Bartolo,
subjects. The Quarterly Journal of ExperimenLila Hakim, Anna Maria Ferraro, Valeria Isella,
tal Psychology, 40(4):801–812, 1988.
[51]
24
[64]
R E Dudley, C H John, a W Young, and D E [74] Chivon Powers, Rachel Bencic, William S HorOver. Normal and abnormal reasoning in peoton, and Mark Beeman. Hemispheric inference
ple with delusions. The British journal of clinpriming during comprehension of conversations
ical psychology / the British Psychological Soand narratives Author ’ s personal copy. Neuciety, 36 ( Pt 2):243–258, 1997.
ropsychologia, 50:2577–2583, 2012.
[65]
Charles R. Conway, Anna M. Bollini, Bre- [75] Michael S Gazzaniga, Joseph E Bogen, and
vick G. Graham, Richard S E Keefe, SuRoger W Sperry. Some functional effects of secsan S. Schiffman, and Joseph P. McEvoy. Sentioning the cerebral commissures in man. Prosory acuity and reasoning in delusional disorceedings of the National Academy of Sciences,
der. Comprehensive Psychiatry, 43(3):175–178,
48(10):1765–1769, 1962.
2002. ISSN 0010440X.
[76] Michael S Vendetti, Elizabeth L Johnson, ConSteffen Moritz and Todd S Woodward. Jumpnor J Lemos, and Silvia A Bunge. Hemispheric
ing to conclusions in delusional and nondifferences in relational reasoning: novel indelusional schizophrenic patients. The British
sights based on an old technique. Frontiers in
journal of clinical psychology / the British
human neuroscience, 9, 2015.
Psychological Society, 44(Pt 2):193–207, 2005.
[77] Roger A Drake. Processing persuasive arguISSN 0144-6657.
ments: Recall and recognition as a function of
Debbie M. Warman, Paul H. Lysaker, Joel M.
agreement and manipulated activation asymMartin, Louanne Davis, and Samantha L. Haumetry. Brain and Cognition, 15(1):83–94, 1991.
denschield. Jumping to Conclusions and the
Continum of Delusional Beliefs. Behavior Re- [78] Roger A Drake. Processing persuasive arguments: 2. discounting of truth and relevance as
search and Therapy, 45(6):1255–1269, 2007.
a function of agreement and manipulated actiMax Coltheart, Peter Menzies, and John Sutvation asymmetry. Journal of Research in Perton. Abductive inference and delusional besonality, 27(2):184–196, 1993.
lief. Cognitive neuropsychiatry, 15(1):261–287,
[79] Eric Prichard, Ruth E Propper, and Stephen D
2010. ISSN 1464-0619.
Christman. Degree of handedness, but not diMax Coltheart. The neuropsychology of delurection, is a systematic predictor of cognitive
sions. Annals of the New York Academy of Sciperformance. Frontiers in psychology, 4, 2013.
ences, 1191:16–26, 2010. ISSN 1749-6632.
[80] Stephen D Christman, Bradley R Henning, AnMax Coltheart, Robyn Langdon, and Ryan
drew L Geers, Ruth E Propper, and ChristoMcKay. Delusional belief. Annual review of
pher L Niebauer. Mixed-handed persons are
psychology, 62:271–98, 2011. ISSN 1545-2085.
more easily persuaded and are more gullible:
Interhemispheric interaction and belief updatP R Corlett, G K Murray, G D Honey, M R F
ing. Laterality, 13(5):403–426, 2008.
Aitken, D R Shanks, T W Robbins, E T Bullmore, a Dickinson, and P C Fletcher. Disrupted [81] John D Jasper, J Scott Kunzler, Eric C
prediction-error signal in psychosis: evidence
Prichard, and Stephen D Christman. Individfor an associative account of delusions. Brain
ual differences in information order effects: The
: a journal of neurology, 130(Pt 9):2387–400,
importance of right-hemisphere access in be2007. ISSN 1460-2156.
lief updating. Acta psychologica, 148:115–122,
2014.
Jeffrey W Cooney and Michael S Gazzaniga.
Neurological disorders and the structure of hu- [82] Takeo Tsujii, Sayako Masuda, Takekazu
man consciousness. Trends in Cognitive SciAkiyama, and Shigeru Watanabe. The role of
ences, 7(4):161–165, 2003.
inferior frontal cortex in belief-bias reasoning:
An rTMS study. Neuropsychologia, 48(7):2005–
Ryan McKay, Robyn Langdon, and Max Colt2008, 2010. ISSN 1873-3514.
heart. Models of misbelief: Integrating motivational and deficit theories of delusions. Con- [83] Takeo Tsujii, Kaoru Sakatani, Sayako Masuda,
sciousness and Cognition, 16(4):932–941, 2007.
Takekazu Akiyama, and Shigeru Watanabe.
ISSN 1090-2376.
Evaluating the roles of the inferior frontal gyrus
[66]
[67]
[68]
[69]
[70]
[71]
[72]
[73]
25
and superior parietal lobule in deductive reasoning: an rTMS study. NeuroImage, 58(2):
640–6, 2011. ISSN 1095-9572.
[84]
[93]
Patrick CM Wong, Marc Ettlinger, and Jing
Zheng. Linguistic grammar learning and drd2taq-ia polymorphism. PloS one, 8(5):e64983,
2013.
Joshua D Greene, Leigh E Nystrom, Andrew D
Engell, John M Darley, and Jonathan D Cohen. [94] Samantha B Wright, Bryan J Matlen, Carol L
The neural bases of cognitive conflict and conBaym, Emilio Ferrer, and Silvia A Bunge. Neutrol in moral judgment. Neuron, 44(2):389–400,
ral correlates of fluid reasoning in children and
2004.
adults. Frontiers in Human Neuroscience, 1,
2007.
[85] Jennifer S Beer, Michael V Lombardo, and
Jamil Palacios Bhanji. Roles of medial prefrontal cortex and orbitofrontal cortex in self- [95] Daniel C Krawczyk, M Michelle McClelland,
Colin M Donovan, Gail D Tillman, and
evaluation. Journal of Cognitive Neuroscience,
Mandy J Maguire. An fmri investigation of
22(9):2108–2119, 2010.
cognitive stages in reasoning by analogy. Brain
[86] Daniel R Hawes, Alexander Vostroknutov, and
research, 1342:63–73, 2010.
Aldo Rustichini. Experience and abstract reasoning in learning backward induction. Fron- [96] Silvia A Bunge, Carter Wendelken, David
Badre, and Anthony D Wagner. Analogical
tiers in neuroscience, 6, 2012.
reasoning and prefrontal cortex: evidence for
[87] Nicole Malfait, Pierre Fonlupt, Laurie Censeparable retrieval and integration mechanisms.
telles, Bruno Nazarian, Liana E Brown, and
Cerebral cortex, 15(3):239–249, 2005.
Anne Caclin. Different neural networks are
involved in audiovisual speech perception de- [97] Soohyun Cho, Teena D Moody, Leonardo Ferpending on the context. Journal of cognitive
nandino, Jeanette A Mumford, Russell A Polneuroscience, 26(7):1572–1586, 2014.
drack, Tyrone D Cannon, Barbara J Knowlton,
and Keith J Holyoak. Common and dissociable prefrontal loci associated with component
mechanisms of analogical reasoning. Cerebral
Cortex, 20(3):524–533, 2010.
[88]
Rebecca J Melrose, Renee M Poulin, and Chantal E Stern. An fmri investigation of the role of
the basal ganglia in reasoning. Brain research,
1142:146–158, 2007.
[89]
Bernardo Perfetti, Aristide Saggino, Antonio [98] Adam E Green, David JM Kraemer,
Ferretti, Massimo Caulo, Gian Luca Romani,
Jonathan A Fugelsang, Jeremy R Gray,
and Marco Onofrj. Differential patterns of corand Kevin N Dunbar.
Connecting long
tical activation as a function of fluid reasoning
distance: semantic distance in analogical reacomplexity. Human brain mapping, 30(2):497–
soning modulates frontopolar cortex activity.
510, 2009.
Cerebral Cortex, 20(1):70–76, 2010.
[90]
Deepika Bagga, Namita Singh, Sadhana Singh, [99] Emmanuelle Volle, Sam J Gilbert, Roland G
Shilpi Modi, Pawan Kumar, D Bhattacharya,
Benoit, and Paul W Burgess. Specialization of
Mohan L Garg, and Subash Khushu. Assessthe rostral prefrontal cortex for distinct analogy
ment of abstract reasoning abilities in alcoholprocesses. Cerebral Cortex, 20(11):2647–2659,
dependent subjects: an fmri study. Neuroradi2010.
ology, 56(1):69–77, 2014.
[91] Daniel C Krawczyk, M Michelle McClelland, [100] Adam Hampshire, Russell Thompson, John
Duncan, and Adrian M Owen. Lateral preand Colin M Donovan. A hierarchy for relafrontal cortex subregions make dissociable contional reasoning in the prefrontal cortex. Cortributions during fluid reasoning. Cerebral Cortex, 47(5):588–597, 2011.
tex, page bhq085, 2010.
[92] Chantel S Prat, Robert A Mason, and Marcel Adam Just. An fmri investigation of ana- [101] Adam E Green, David JM Kraemer,
logical mapping in metaphor comprehension:
Jonathan A Fugelsang, Jeremy R Gray,
the influence of context and individual cogniand Kevin N Dunbar. Neural correlates of
tive capacities on processing demands. Journal
creativity in analogical reasoning. Journal of
of Experimental Psychology: Learning, MemExperimental Psychology: Learning, Memory,
ory, and Cognition, 38(2):282, 2012.
and Cognition, 38(2):264, 2012.
26
[102] Christine E Watson and Anjan Chatterjee. A [111] Vinod Goel and Raymond J Dolan. Differenbilateral frontoparietal network underlies visutial involvement of left prefrontal cortexin inospatial analogical reasoning. Neuroimage, 59
ductive and deductive reasoning. Cognition, 93
(3):2831–2838, 2012.
(3):B109–B121, 2004.
[103] Yuanyuan Li, Fei Li, Ning He, Lanting Guo, [112] Claire Landmann, Stanislas Dehaene, Sabina
Pappata, Antoinette Jobert, Michel BottlaenXiaoqi Huang, Su Lui, and Qiyong Gong.
der, Dimitri Roumenov, and Denis Le Bihan.
Neural hyperactivity related to working memDynamics of prefrontal and cingulate activity
ory in drug-naive boys with attention deficit
during a reward-based logical deduction task.
hyperactivity disorder. Progress in NeuroCerebral Cortex, 17(4):749–759, 2007.
Psychopharmacology and Biological Psychiatry,
53:116–122, 2014.
[113] Martin M Monti, Daniel N Osherson, Michael J
Martinez, and Lawrence M Parsons. Func[104] Adam E Green, Jonathan A Fugelsang,
tional neuroanatomy of deductive inference:
David JM Kraemer, Noah A Shamosh, and
a language-independent distributed network.
Kevin N Dunbar. Frontopolar cortex mediates
Neuroimage, 37(3):1005–1016, 2007.
abstract integration in analogy. Brain research,
1096(1):125–137, 2006.
[114] Charlotte E Hartwright, Ian A Apperly, and
Peter C Hansen. Representation, control, or
[105] Lorraine K Tyler, Emmanuel A Stamatakis,
reasoning? distinct functions for theory of
Peter Bright, K Acres, S Abdallah, JM Rodd,
mind within the medial prefrontal cortex. Jourand HE Moss. Processing objects at different
nal
of cognitive neuroscience, 26(4):683–698,
levels of specificity. Journal of cognitive neuro2014.
science, 16(3):351–362, 2004.
[115] Mathilde Bonnefond, Ira Noveck, Sylvain Baillet, Anne Cheylus, Claude Delpuech, Olivier
Bertrand, Pierre Fourneret, Van der Henst,
et al. What meg can reveal about inference
making: The case of if... then sentences. Human brain mapping, 34(3):684–697, 2013.
[107] Barbara Rutter, Sören Kröger, Rudolf Stark,
Jan Schweckendiek, Sabine Windmann, Chris- [116] Jérôme Prado, Rachna Mutreja, and James R
tiane Hermann, and Anna Abraham. Can
Booth. Fractionating the neural substrates of
clouds dance? neural correlates of passive contransitive reasoning: Task-dependent contriceptual expansion using a metaphor processing
butions of spatial and verbal representations.
task: implications for creative cognition. Brain
Cerebral Cortex, 23(3):499–507, 2013.
and cognition, 78(2):114–122, 2012.
[117] Luca Cocchi, Graeme S Halford, Andrew Za[108] Arnaud D’Argembeau,
Helena Cassol,
lesky, Ian H Harding, Brentyn J Ramm, Tim
Christophe Phillips, Evelyne Balteau, Eric
Cutmore, David HK Shum, and Jason B MatSalmon, and Martial Van der Linden. Brains
tingley. Complexity in relational processing
creating stories of selves: the neural basis of
predicts changes in functional brain network
autobiographical reasoning. Social cognitive
dynamics. Cerebral Cortex, 24(9):2283–2296,
and affective neuroscience, page nst028, 2013.
2014.
[106] Rachel J Garoff-Eaton, Elizabeth A Kensinger,
and Daniel L Schacter. The neural correlates
of conceptual and perceptual false recognition.
Learning & Memory, 14(10):684–692, 2007.
[109] Anna Abraham, Hannes Rakoczy, Markus [118] Nicola Canessa, Alessandra Gorini, Stefano F
Werning, D Yves Von Cramon, and Ricarda I
Cappa, Massimo Piattelli-Palmarini, Massimo
Schubotz. Matching mind to world and vice
Danna, Ferruccio Fazio, and Daniela Perani.
versa: functional dissociations between belief
The effect of social content on deductive reaand desire mental state processing. Social neusoning: an fmri study. Human brain mapping,
roscience, 5(1):1–18, 2010.
26(1):30–43, 2005.
[110] M Layne Kalbfleisch, John W Van Meter, and [119] Thomas Fangmeier, Markus Knauff, ChrisThomas A Zeffiro. The influences of task diftian C Ruff, and Vladimir Sloutsky. fmri evificulty and response correctness on neural sysdence for a three-stage model of deductive reatems supporting fluid reasoning. Cognitive neusoning. Journal of Cognitive Neuroscience, 18
rodynamics, 1(1):71–84, 2007.
(3):320–334, 2006.
27
[120] Carlo Reverberi, Paolo Cherubini, Attilio [130] Monika Sommer, Christoph Rothmayr, Katrin
Rapisarda, Elisa Rigamonti, Carlo Caltagirone,
Döhnel, Jörg Meinhardt, Johannes SchwerdtRichard SJ Frackowiak, Emiliano Macaluso,
ner, Beate Sodian, and Göran Hajak. How
and Eraldo Paulesu. Neural basis of generation
should i decide? the neural correlates of evof conclusions in elementary deduction. Neueryday moral reasoning. Neuropsychologia, 48
roimage, 38(4):752–762, 2007.
(7):2018–2026, 2010.
[121] James K Kroger, Leigh E Nystrom, Jonathan D [131] Gayaneh Szenkovits, Jonathan E Peelle, Dennis Norris, and Matthew H Davis. Individual
Cohen, and Philip N Johnson-Laird. Distinct
differences in premotor and motor recruitment
neural substrates for deductive and mathematduring speech perception. Neuropsychologia, 50
ical processing. Brain research, 1243:86–103,
(7):1380–1392, 2012.
2008.
[122] Thomas Fangmeier and Markus Knauff. Neural [132] Keith J Yoder and Jean Decety. The good,
the bad, and the just: justice sensitivity precorrelates of acoustic reasoning. Brain research,
dicts neural response during moral evaluation
1249:181–190, 2009.
of actions performed by others. The Journal of
Neuroscience, 34(12):4161–4166, 2014.
[123] Martin M Monti, Lawrence M Parsons, and
Daniel N Osherson. The boundaries of language
[133] Markus Ullsperger, Hiroshi Nittono, and
and thought in deductive inference. Proceedings
D Yves Von Cramon. When goals are missed:
of the National Academy of Sciences, 106(30):
dealing with self-generated and externally in12554–12559, 2009.
duced failure. Neuroimage, 35(3):1356–1364,
2007.
[124] Ehsan Shokri-Kojori, Michael A Motes, Bart
Rypma, and Daniel C Krawczyk. The network [134] Nicholas Epley, Benjamin A Converse, Alexa
architecture of cortical processing in visuoDelbosc, George A Monteleone, and John T
spatial reasoning. Scientific reports, 2, 2012.
Cacioppo. Believers’ estimates of god’s be[125] Melanie Stollstorff, Oshin Vartanian, and
Vinod Goel. Levels of conflict in reasoning
modulate right lateral prefrontal cortex. Brain
research, 1428:24–32, 2012.
[126] Camillo Porcaro, Maria Teresa Medaglia,
Ngoc Jade Thai, Stefano Seri, Pia Rotshtein,
and Franca Tecchio. Contradictory reasoning
network: An eeg and fmri study. PloS one, 9
(3):e92835, 2014.
[127] Kristen L Mackiewicz Seghete, Megan M Herting, and Bonnie J Nagel. White matter microstructure correlates of inhibition and taskswitching in adolescents. Brain research, 1527:
15–28, 2013.
liefs are more egocentric than estimates of
other people’s beliefs. Proceedings of the National Academy of Sciences, 106(51):21533–
21538, 2009.
[135] Sjoerd JH Ebisch, Dante Mantini, Roberta Romanelli, Marco Tommasi, Mauro G Perrucci,
Gian Luca Romani, Roberto Colom, and Aristide Saggino. Long-range functional interactions of anterior insula and medial frontal cortex are differently modulated by visuospatial
and inductive reasoning tasks. Neuroimage, 78:
426–438, 2013.
[136] Matthew Weber, Sharon L Thompson-Schill,
Daniel Osherson, James Haxby, and Lawrence
Parsons. Predicting judged similarity of natural categories from their neural representations.
[128] Chad E Forbes, Joshua C Poore, Frank
Neuropsychologia, 47(3):859–868, 2009.
Krueger, Aron K Barbey, Jeffrey Solomon, and
Jordan Grafman. The role of executive function [137] YanHui Yang, PeiPeng Liang, ShengFu Lu,
and the dorsolateral prefrontal cortex in the exKunCheng Li, and Ning Zhong. The role of
pression of neuroticism and conscientiousness.
the dlpfc in inductive reasoning of mci patients
Social neuroscience, 9(2):139–151, 2014.
and normal agings: An fmri study. Science in
China Series C: Life Sciences, 52(8):789–795,
[129] David JM Kraemer, Lauren M Rosenberg,
2009.
and Sharon L Thompson-Schill. The neural correlates of visual and verbal cognitive [138] Peipeng Liang, Yang Mei, Xiuqin Jia, Yanhui
styles. The Journal of Neuroscience, 29(12):
Yang, Shengfu Lu, Ning Zhong, and Kuncheng
3792–3798, 2009.
Li. Brain activation and deactivation in human
28
inductive reasoning: an fmri study. In Brain [147] Roman Liepelt, D Yves Von Cramon, and
Informatics, pages 387–398. Springer, 2010.
Marcel Brass. How do we infer others’ goals
from non-stereotypic actions? the outcome of
[139] Xiuqin Jia, Peipeng Liang, Jie Lu, Yanhui
context-sensitive inferential processing in right
Yang, Ning Zhong, and Kuncheng Li. Cominferior parietal and posterior temporal cortex.
mon and dissociable neural correlates associNeuroimage, 43(4):784–792, 2008.
ated with component processes of inductive
reasoning. NeuroImage, 56(4):2292–2299, 2011. [148] EC Hinton, S Dymond, U Von Hecker, and
CJ Evans. Neural correlates of relational reasoning and the symbolic distance effect: In[140] Xiao Feng, Li Peng, Long Chang-Quan, Lei Yi,
volvement of parietal cortex. Neuroscience, 168
and Li Hong. Relational complexity modulates
(1):138–148, 2010.
activity in the prefrontal cortex during numerical inductive reasoning: An fmri study. Bio[149] Akitoshi Ogawa, Yumiko Yamazaki, Kenichi
logical psychology, 101:61–68, 2014.
Ueno, Kang Cheng, and Atsushi Iriki. Inferential reasoning by exclusion recruits parietal and
[141] Nigel Blackwood, Andrew Simmons, Richard
prefrontal cortices. Neuroimage, 52(4):1603–
Bentall, Robin Murray, Robert Howard, et al.
1610,
2010.
The cerebellum and decision making under uncertainty. Cognitive Brain Research, 20(1):46–
53, 2004.
[142]
[143]
[144]
[145]
[150] Catherine A Hynes, Abigail A Baird, and
Scott T Grafton. Differential role of the orbital frontal lobe in emotional versus cogniPeipeng Liang, Ning Zhong, Shengfu Lu, Jimtive perspective-taking. Neuropsychologia, 44
ing Liu, Yiyu Yao, Kuncheng Li, and Yanhui
(3):374–383, 2006.
Yang. The neural mechanism of human numerical inductive reasoning process: a com- [151] Jeremy R Gray, Gregory C Burgess, Alexanbined erp and fmri study. In Web Intellidre Schaefer, Tal Yarkoni, Randy J Larsen, and
gence Meets Brain Informatics, pages 223–243.
Todd S Braver. Affective personality differences
Springer, 2007.
in neural processing efficiency confirmed using
fmri. Cognitive, Affective, & Behavioral NeuKarsten Specht, Chuh-Hyoun Lie, Nadim Shah,
roscience, 5(2):182–190, 2005.
Gereon R Fink, et al. Disentangling the prefrontal network for rule selection by means of a [152] Moses O Sokunbi, Roger T Staff, Gordon D
Waiter, Trevor S Ahearn, Helen C Fox, Ian J
non-verbal variant of the wisconsin card sorting
Deary, John M Starr, Lawrence J Whalley, and
test. Human brain mapping, 30(5):1734–1743,
Alison D Murray. Inter-individual differences in
2009.
fmri entropy measurements in old age. Biomedical Engineering, IEEE Transactions on, 58
Cristiano Crescentini, Shima Seyed-Allaei,
(11):3206–3214, 2011.
Nicola De Pisapia, Jorge Jovicich, Daniele Amati, and Tim Shallice. Mechanisms of rule ac[153] Meredith A Shafto, Deborah M Burke, Emquisition and rule following in inductive reamanuel A Stamatakis, Phyllis P Tam, and Lorsoning. The Journal of Neuroscience, 31(21):
raine K Tyler. On the tip-of-the-tongue: neu7763–7774, 2011.
ral correlates of increased word-finding failures
in normal aging. Journal of cognitive neuroBeate Dunst, Mathias Benedek, Emanuel Jauk,
science, 19(12):2060–2070, 2007.
Sabine Bergner, Karl Koschutnig, Markus Som-
mer, Anja Ischebeck, Birgit Spinath, Martin [154] Timothy J Silk, Alasdair Vance, Nicole RineArendasy, Markus Bühner, et al. Neural effihart, John L Bradshaw, and Ross Cunnington.
ciency as a function of task demands. IntelliDysfunction in the fronto-parietal network in
gence, 42:22–30, 2014.
attention deficit hyperactivity disorder (adhd):
an fmri study. Brain Imaging and Behavior, 2
[146] Peipeng Liang, Xiuqin Jia, Niels A Taatgen,
(2):123–131, 2008.
Ning Zhong, and Kuncheng Li. Different strategies in solving series completion inductive rea- [155] Paul J Eslinger, Clancy Blair, JianLi Wang,
soning problems: An fmri and computational
Bryn Lipovsky, Jennifer Realmuto, David
study. International Journal of PsychophysiolBaker, Steven Thorne, David Gamson, Erin
ogy, 93(2):253–260, 2014.
Zimmerman, Lisa Rohrer, et al. Developmental
29
shifts in fmri activations during visuospatial relational reasoning. Brain and cognition, 69(1):
1–10, 2009.
of geometric reasoning in adults with highfunctioning autism. PloS one, 7(8):e43220,
2012.
[156] Wenjing Yang, Peiduo Liu, Dongtao Wei, [164] F Liegeois, A Connelly, J Helen Cross,
SG Boyd, DG Gadian, F Vargha-Khadem, and
Wenfu Li, Glenn Hitchman, Xueping Li, Jiang
T Baldeweg. Language reorganization in chilQiu, and Qinglin Zhang. Females and males
dren with early-onset lesions of the left hemirely on different cortical regions in ravens masphere: an fmri study. Brain, 127(6):1229–
trices reasoning capacity: Evidence from a
1236, 2004.
voxel-based morphometry study. PloS one, 9
(3):e93104, 2014.
[165] PV Rekkas and R Todd Constable. Evidence
that autobiographic memory retrieval does not
[157] John G Geake and Peter C Hansen. Functional
become independent of the hippocampus: an
neural correlates of fluid and crystallized analofmri study contrasting very recent with remote
gizing. Neuroimage, 49(4):3489–3497, 2010.
events. Journal of Cognitive Neuroscience, 17
(12):1950–1961, 2005.
[158] Hiromi Masunaga, Ryuta Kawashima, John L
Horn, Yuko Sassa, and Atsushi Sekiguchi. Neu[166] Lynn Nadel, Jenna Campbell, and Lee Ryan.
ral substrates of the topology test to measure
Autobiographical memory retrieval and hipfluid reasoning: An fmri study. Intelligence, 36
pocampal activation as a function of repetition
(6):607–615, 2008.
and the passage of time. Neural plasticity, 2007,
2007.
[159] Carter Wendelken, Denis Nakhabenko, Sarah E
Donohue, Cameron S Carter, and Silvia A [167] Paulien M de Vries, Bauke M de Jong, Daryl E
Bunge. brain is to thought as stomach is to??:
Bohning, John A Walker, Mark S George, and
investigating the role of rostrolateral prefrontal
Klaus L Leenders. Changes in cerebral activacortex in relational reasoning. Journal of Cogtions during movement execution and imagery
nitive Neuroscience, 20(4):682–693, 2008.
after parietal cortex tms interleaved with 3t
mri. Brain research, 1285:58–68, 2009.
[160] Isabelle Soulières, Michelle Dawson, Fabienne
Samson, Elise B Barbeau, Cherif P Sahyoun, [168] Nikhil Sharma, Jean-Claude Baron, and
Gary E Strangman, Thomas A Zeffiro, and
James B Rowe. Motor imagery after stroke: reLaurent Mottron. Enhanced visual processing
lating outcome to motor network connectivity.
contributes to matrix reasoning in autism. HuAnnals of neurology, 66(5):604–616, 2009.
man brain mapping, 30(12):4082–4107, 2009.
[169] Christine N Smith and Larry R Squire. Medial temporal lobe activity during retrieval of
[161] Maria Golde, D Yves von Cramon, and Risemantic memory is related to the age of the
carda I Schubotz. Differential role of anterior
memory. The journal of Neuroscience, 29(4):
prefrontal and premotor cortex in the process930–938, 2009.
ing of relational information. NeuroImage, 49
(3):2890–2900, 2010.
[170] Fraser Milton, Christopher R Butler, Abdelmalek Benattayallah, and Adam ZJ Zeman.
[162] Manuel Desco, Francisco J Navas-Sanchez,
The neural basis of autobiographical memory
Javier Sanchez-González, Santiago Reig, Olalla
deficits in transient epileptic amnesia. NeuRobles, Carolina Franco, Juan A Guzmánropsychologia,
50(14):3528–3541, 2012.
De-Villoria, Pedro Garcı́a-Barreno, and Celso
Arango. Mathematically gifted adolescents use [171] Hein JF van Marle, Erno J Hermans,
more extensive and more bilateral areas of the
Shaozheng Qin, Sebastiaan Overeem, and
fronto-parietal network than controls during
Guillén Fernández. The effect of exogenous corexecutive functioning and fluid reasoning tasks.
tisol during sleep on the behavioral and neural
Neuroimage, 57(1):281–292, 2011.
correlates of emotional memory consolidation
in humans. Psychoneuroendocrinology, 38(9):
[163] Takashi Yamada, Haruhisa Ohta, Hiromi
1639–1649, 2013.
Watanabe, Chieko Kanai, Masayuki Tani, Taisei Ohno, Yuko Takayama, Akira Iwanami, [172] Eleanor A Maguire, Dharshan Kumaran,
Nobumasa Kato, and Ryuichiro Hashimoto.
Demis Hassabis, and Michael D Kopelman. AuFunctional alterations in neural substrates
tobiographical memory in semantic dementia:
30
a longitudinal fmri study. Neuropsychologia, 48
(1):123–136, 2010.
GL Romani. Framing deductive reasoning with
emotional content: An fmri study. Brain and
cognition, 87:153–160, 2014.
[173] Pekka S Miettinen, Maija Pihlajamäki, Anne M
Jauhiainen, Eini Niskanen, Tuomo Hänninen, [182] Vinod Goel, Milan Makale, and Jordan Grafman. The hippocampal system mediates logical
Ritva Vanninen, and Hilkka Soininen. Strucreasoning about familiar spatial environments.
ture and function of medial temporal and posJournal of Cognitive Neuroscience, 16(4):654–
teromedial cortices in early alzheimers disease.
664, 2004.
European Journal of Neuroscience, 34(2):320–
330, 2011.
[183] Mark C Eldaief, Thilo Deckersbach, Lindsay E Carlson, Jan C Beucke, and Darin D
[174] Sarah D Broicher, Lars Frings, Hans-Jürgen
Dougherty. Emotional and cognitive stimuli
Huppertz, Thomas Grunwald, Martin Kurthen,
differentially engage the default network durGünter Krämer, and Hennric Jokeit. Altering inductive reasoning. Social cognitive and
ations in functional connectivity of the amygaffective neuroscience, page nsr003, 2011.
dala in unilateral mesial temporal lobe epilepsy.
Journal of neurology, 259(12):2546–2554, 2012.
[184] Carlo Reverberi, Luca L Bonatti, Richard SJ
Frackowiak, Eraldo Paulesu, Paolo Cherubini,
[175] Xiuqin Jia, Shengfu Lu, Ning Zhong, Yiyu Yao,
and Emiliano Macaluso. Large scale brain actiKuncheng Li, and Yanhui Yang. Common and
vations predict reasoning profiles. Neuroimage,
distinct neural substrates of forward-chaining
59(2):1752–1764, 2012.
and backward-chaining syllogistic reasoning.
In Complex Medical Engineering, 2009. CME.
[185] Jennifer J Vannest, Prasanna R Karunanayaka,
ICME International Conference on, pages 1–6.
Mekibib Altaye, Vincent J Schmithorst,
IEEE, 2009.
Elena M Plante, Kenneth J Eaton, Jerod M
Rasmussen, and Scott K Holland. Compar[176] Ira A Noveck, Vinod Goel, and Kathleen W
ison of fmri data from passive listening and
Smith. The neural basis of conditional reasonactive-response story processing tasks in chiling with arbitrary content. Cortex, 40(4):613–
dren. Journal of Magnetic Resonance Imaging,
622, 2004.
29(4):971–976, 2009.
[177] John G Geake and Peter C Hansen. Neural correlates of intelligence as revealed by fmri of fluid [186] Christian Sorg, Andrei Manoliu, Susanne Neufang, Nicholas Myers, Henning Peters, Dirk
analogies. Neuroimage, 26(2):555–564, 2005.
Schwerthöffer, Martin Scherr, Mark Mühlau,
[178] Diana Rodriguez-Moreno and Joy Hirsch. The
Claus Zimmer, Alexander Drzezga, et al. Indynamics of deductive reasoning: An fmri increased intrinsic brain activity in the striatum
vestigation. Neuropsychologia, 47(4):949–961,
reflects symptom dimensions in schizophrenia.
2009.
Schizophrenia bulletin, 39(2):387–395, 2013.
[179] Carlo Reverberi, Paolo Cherubini, Richard SJ [187] Deidre L Reis, Marc A Brackett, Noah A
Frackowiak, Carlo Caltagirone, Eraldo Paulesu,
Shamosh, Kent A Kiehl, Peter Salovey, and
and Emiliano Macaluso. Conditional and sylJeremy R Gray. Emotional intelligence predicts
logistic deductive tasks dissociate functionally
individual differences in social exchange reasonduring premise integration. Human brain maping. NeuroImage, 35(3):1385–1391, 2007.
ping, 31(9):1430–1445, 2010.
[188] Anthony I Jack, Abigail J Dawson, Katelyn L
[180] Aneta Brzezicka, Grzegorz Sedek, Artur
Begany, Regina L Leckie, Kevin P Barry, AnMarchewka, Mateusz Gola, Katarzyna Jedgela H Ciccia, and Abraham Z Snyder. fmri
noróg, Leszek Królicki, and Andrzej Wróbel. A
reveals reciprocal inhibition between social and
role for the right prefrontal and bilateral pariphysical cognitive domains. NeuroImage, 66:
etal cortex in four-term transitive reasoning:
385–401, 2013.
an fmri study with abstract linear syllogism
tasks. Acta neurobiologiae experimentalis, 71 [189] Laurence Fiddick, Maria Vittoria Spampinato,
and Jordan Grafman. Social contracts and
(4):0, 2011.
precautions activate different neurological sys[181] M Brunetti, MG Perrucci, MR Di Naccio,
tems: An fmri investigation of deontic reasonA Ferretti, C Del Gratta, C Casadio, and
ing. NeuroImage, 28(4):778–786, 2005.
31
[190] Elsa Ermer, Scott A Guerin, Leda Cosmides, [198] Marie-Anne Hénaff, Dimitri Bayle, Pierre
John Tooby, and Michael B Miller. Theory of
Krolak-Salmon, and Pierre Fonlupt. Cortical
mind broad and narrow: Reasoning about sodynamics of a self driven choice: A meg study
cial exchange engages tom areas, precautionary
during a card sorting task. Clinical Neurophysreasoning does not. Social Neuroscience, 1(3iology, 121(4):508–515, 2010.
4):196–219, 2006.
[199] Ruben Martins, France Simard, Jean-Sebastien
Provost, and Oury Monchi. Changes in re[191] JiMei Liu, Meng Zhang, Jerwen Jou, Xin Wu,
gional and temporal patterns of activity assoWei Li, and Jiang Qiu. Neural bases of falsificiated with aging during the performance of a
cation in conditional proposition testing: Evilexical set-shifting task. Cerebral Cortex, 22(6):
dence from an fmri study. International Jour1395–1406, 2012.
nal of Psychophysiology, 85(2):249–256, 2012.
[192] Seiki Konishi, Junichi Chikazoe, Koji Jimura, [200] Anya Pedersen, Andreas Wilmsmeier, Karl H
Wiedl, Jochen Bauer, Kerstin Kueppers, Katja
Tomoki Asari, and Yasushi Miyashita. NeuKoelkebeck, Waldemar Kohl, Harald Kugel,
ral mechanism in anterior prefrontal cortex for
Volker Arolt, and Patricia Ohrmann. Anterior
inhibition of prolonged set interference. Procingulate cortex activation is related to learning
ceedings of the National Academy of Sciences of
potential on the wcst in schizophrenia patients.
the United States of America, 102(35):12584–
Brain and cognition, 79(3):245–251, 2012.
12588, 2005.
[193] Xavier Caldú, Pere Vendrell, David Bartrés- [201] Oury Monchi, Michael Petrides, Julien Doyon,
Ronald B Postuma, Keith Worsley, and Alain
Faz, Inmaculada Clemente, Núria Bargalló,
Dagher. Neural bases of set-shifting deficits
Marı́a Ángeles Jurado, Josep Maria Serrain parkinson’s disease. The Journal of neuroGrabulosa, and Carme Junqué. Impact of the
science, 24(3):702–710, 2004.
comt val 108/158 met and dat genotypes on
prefrontal function in healthy subjects. Neu[202] Anna B Smith, Eric Taylor, Mick Brammer,
roimage, 37(4):1437–1444, 2007.
and Katya Rubia. Neural correlates of switching set as measured in fast, event-related func[194] Seiki Konishi, Hiroki Morimoto, Koji Jimura,
tional magnetic resonance imaging. Human
Tomoki Asari, Junichi Chikazoe, Ken-ichiro
brain mapping, 21(4):247–256, 2004.
Yamashita, Satoshi Hirose, and Yasushi
Miyashita. Differential superior prefrontal ac- [203] Tomoki Asari, Seiki Konishi, Koji Jimura, and
tivity on initial versus subsequent shifts in
Yasushi Miyashita. Multiple components of
naive subjects. Neuroimage, 41(2):575–580,
lateral posterior parietal activation associated
2008.
with cognitive set shifting. Neuroimage, 26(3):
694–702, 2005.
[195] Margit Jehna, Christian Langkammer, Mirja
Wallner-Blazek, Christa Neuper, Marisa Loit- [204] Chuh-Hyoun Lie, Karsten Specht, John C Marfelder, Stefan Ropele, Siegrid Fuchs, Michael
shall, and Gereon R Fink. Using fmri to decomKhalil, Aga Pluta-Fuerst, Franz Fazekas, et al.
pose the neural processes underlying the wisCognitively preserved ms patients demonstrate
consin card sorting test. Neuroimage, 30(3):
functional differences in processing neutral and
1038–1049, 2006.
emotional faces. Brain imaging and behavior, 5
[205] Eveline A Crone, Kiki Zanolie, Linda
(4):241–251, 2011.
Van Leijenhorst, P Michiel Westenberg, and
[196] Matthew L Dixon and Kalina Christoff. The
Serge ARB Rombouts. Neural mechanisms
decision to engage cognitive control is driven by
supporting flexible performance adjustment
expected reward-value: neural and behavioral
during development. Cognitive, Affective, &
evidence. PloS one, 7(12):e51637, 2012.
Behavioral Neuroscience, 8(2):165–177, 2008.
[197] F Simard, L Monetta, A Nagano-Saito, and [206] France Simard, Yves Joanette, Michael
O Monchi. A new lexical card-sorting task
Petrides, Thomas Jubault, Cécile Madjar, and
for studying fronto-striatal contribution to proOury Monchi.
Fronto-striatal contribution
cessing language rules. Brain and language, 125
to lexical set-shifting. Cerebral Cortex, page
(3):295–306, 2013.
bhq182, 2010.
32
[207] Emiko Aizawa, Yasuhiro Sato, Takanori [216] Laura Menenti, Karl Magnus Petersson, René
Kochiyama, Naohiro Saito, Masahiro IzuScheeringa, and Peter Hagoort. When elemiyama, Joe Morishita, Motoyori Kanazawa,
phants fly: differential sensitivity of right and
Keisetsu Shima, Hajime Mushiake, Michio
left inferior frontal gyri to discourse and world
Hongo, et al. Altered cognitive function of preknowledge. Journal of cognitive neuroscience,
frontal cortex during error feedback in patients
21(12):2358–2368, 2009.
with irritable bowel syndrome, based on fmri
[217] Roshan Cools, Luke Clark, Adrian M Owen,
and dynamic causal modeling. Gastroenteroland Trevor W Robbins. Defining the Neural
ogy, 143(5):1188–1198, 2012.
Mechanisms of Probabilistic Reversal Learning
Using Event-Related Functional Magnetic Res[208] Yasuhiro Sato, Naohiro Saito, Atsushi Utsumi,
onance Imaging. The Journal of Neuroscience,
Emiko Aizawa, Tomotaka Shoji, Masahiro Izu22(11):4563–4567, 2002.
miyama, Hajime Mushiake, Michio Hongo, and
Shin Fukudo. Neural basis of impaired cognitive flexibility in patients with anorexia nervosa. PloS one, 8(5):e61108, 2013.
[218] Vinod Goel and Oshin Vartanian. Dissociating
the roles of right ventral lateral and dorsal lateral prefrontal cortex in generation and maintenance of hypotheses in set-shift problems. Cere[209] Russell A Poldrack, Paul C Fletcher, Richard N
bral cortex (New York, N.Y. : 1991), 15(8):
Henson, Keith J Worsley, Matthew Brett, and
1170–1177, 2005. ISSN 1047-3211.
Thomas E Nichols. Guidelines for reporting an
fmri study. Neuroimage, 40(2):409–414, 2008. [219] Seiki Konishi, Toshihiro Hayashi, Idai Uchida,
Hideyuki Kikyo, Emi Takahashi, and Yasushi
Miyashita. Hemispheric asymmetry in human
[210] Uwe Friese, Roland Rutschmann, Markus
lateral prefrontal cortex during cognitive set
Raabe, and Franz Schmalhofer. Neural Indicashifting. Proceedings of the National Academy
tors of Inference Processes in Text Comprehenof Sciences of the United States of America, 99
sion : An Event-related Functional Magnetic
(11):7803–8, 2002. ISSN 0027-8424.
Resonance Imaging Study. Journal of cognitive
neuroscience, 20(11):2110–2124, 2008.
[220] Christopher H Chatham, Eric D Claus, Albert
Kim, Tim Curran, Marie T Banich, and Yuko
[211] Sandra Virtue, Jason Haberman, Zoe Clancy,
Munakata. Cognitive control reflects context
Todd Parrish, and Mark Jung Beeman. Neumonitoring, not motoric stopping, in response
ral activity of inferences during story compreinhibition. PloS one, 7(2):e31546, 2012. ISSN
hension. Brain research, 1084(1):104–14, 2006.
1932-6203.
ISSN 0006-8993.
[212] Gina R. Kuperberg, Balaji M. Lakshmanan,
David N. Caplan, and Phillip J. Holcomb. Making sense of discourse: An fMRI study of causal
inferencing across sentences. NeuroImage, 33
(1):343–361, 2006. ISSN 1053-8119.
[221] Adam R Aron, Trevor W Robbins, and Russell a Poldrack. Inhibition and the right inferior frontal cortex: one decade on. Trends in
cognitive sciences, 18(Box 1):1–9, 2014. ISSN
1879-307X.
[222]
[213] Ben a Parris, Gustav Kuhn, Guy a Mizon, Abdelmalek Benattayallah, and Tim L Hodgson.
Imaging the impossible: an fMRI study of impossible causal relationships in magic tricks.
NeuroImage, 45(3):1033–9, 2009. ISSN 10959572.
[223]
Marjaana Lindeman, Annika M Svedholm,
Tapani Riekki, Tuukka Raij, and Riitta Hari.
Is it just a brick wall or a sign from the universe? an fmri study of supernatural believers
and skeptics. Social cognitive and affective neuroscience, page nss096, 2012.
Jonathan a. Fugelsang and Kevin N. Dunbar.
Brain-based mechanisms underlying complex
causal thinking. Neuropsychologia, 43(8):1204–
1213, 2005. ISSN 0028-3932.
[214] V Goel, C Buchel, C Frith, and R J Dolan.
Dissociation of mechanisms underlying syllogistic reasoning. NeuroImage, 12(5):504–14, 2000.
ISSN 1053-8119.
[224] Simon B Eickhoff, Danilo Bzdok, Angela R
Laird, Florian Kurth, and Peter T Fox. Ac[215] Vinod Goel and Raymond J Dolan. Explaining
tivation likelihood estimation meta-analysis remodulation of reasoning by belief. Cognition,
visited. Neuroimage, 59(3):2349–2361, 2012.
87:B11–B22, 2003.
33