The effects of positive or negative self-talk on the alteration of brain functional connectivity by performing cognitive tasks

Abstract

Self-talk can improve cognitive performance, but the underlying mechanism of such improvement has not been investigated. This study aimed to elucidate the effects of self-talks on functional connectivity associated with cognitive performance. We used the short form of Progressive Matrices Test (sRPM) to measure differences in performance improvements between self-respect and self-criticism. Participants were scanned using functional magnetic resonance imaging in the following order: baseline, during-sRPM1, post-sRPM1, self-respect or self-criticism, during-sRPM2, and post-sRPM2. Analysis was conducted to identify the self-talks' modulatory effects on the reward-motivation, default mode, and central-executive networks. Increase in sRPM2 score compared to sRPM1 score was observed only after self-criticism. The self-talk-by-repetition interaction effect was not found for during-sRPM, but found for post-sRPM; decreased nucleus accumbens-based connectivity was shown after self-criticism compared with self-respect. However, the significant correlations between the connectivity change and performance change appeared only in the self-respect group. Our findings showed that positive self-talk and negative self-talk differently modulate brain states concerning cognitive performance. Self-respect may have both positive and negative effects due to enhanced executive functions and inaccurate confidence, respectively, whereas self-criticism may positively affect cognitive performance by inducing a less confident state that increases internal motivation and attention.

Full text



Scientic Reports | (2021) 11:14873 | 
www.nature.com/scientificreports
The eects of positive or negative
self‑talk on the alteration
of brain functional connectivity
by performing cognitive tasks
Junhyung Kim1,2, Joon Hee Kwon3, Joohan Kim4, Eun Joo Kim5, Hesun Erin Kim1,
Sunghyon Kyeong1 & Jae‑Jin Kim1,3*
Self‑talk can improve cognitive performance, but the underlying mechanism of such improvement has
not been investigated. This study aimed to elucidate the eects of self‑talks on functional connectivity
associated with cognitive performance. We used the short form of Progressive Matrices Test (sRPM)
to measure dierences in performance improvements between self‑respect and self‑criticism.
Participants were scanned using functional magnetic resonance imaging in the following order:
baseline, during‑sRPM1, post‑sRPM1, self‑respect or self‑criticism, during‑sRPM2, and post‑sRPM2.
Analysis was conducted to identify the self‑talks’ modulatory eects on the reward‑motivation,
default mode, and central‑executive networks. Increase in sRPM2 score compared to sRPM1 score
was observed only after self‑criticism. The self‑talk‑by‑repetition interaction eect was not found for
during‑sRPM, but found for post‑sRPM; decreased nucleus accumbens‑based connectivity was shown
after self‑criticism compared with self‑respect. However, the signicant correlations between the
connectivity change and performance change appeared only in the self‑respect group. Our ndings
showed that positive self‑talk and negative self‑talk dierently modulate brain states concerning
cognitive performance. Self‑respect may have both positive and negative eects due to enhanced
executive functions and inaccurate condence, respectively, whereas self‑criticism may positively
aect cognitive performance by inducing a less condent state that increases internal motivation and
attention.
Self-talk is the systematic use of cue words in a silent or vocalized dialog with one’s self. is process has two
conceptual properties: the form of verbalizations is an essential requirement and the sender of the message is
also the receiver1. Since self-talk has benecial eects on attention2 and emotion regulation3, it is widely used
for performance enhancement in sports1,4, academic engagement5, and regulating anxiety or depression in a
clinic6,7. Self-talk with positive contents can help with promoting positive psychological states and regulating
cognitions8,9, whereas self-talk with negative contents is associated with emotional ill-being10. However, some
studies have presented that negative self-talk can improve physical performance11,12. How negative self-talk
can be benecial in performance improvement has been explained by several hypotheses, such as motivational
interpretation13, reverse reection of condence14, stimulating eorts to avoid a negative outcome15, and viewing
negative self-talk as a challenge12.
Our research group previously reported the modulation eects of positive and negative self-talks on brain
connectivity as measured by functional magnetic resonance imaging (fMRI). For example, posterior cingulate
cortex (PCC)-based and ventromedial prefrontal cortex (VMPFC)-based functional connectivity for investi-
gating the default mode network (DMN) showed that gratitude interventions modulated connectivity among
motivation-related regions, including the nucleus accumbens (NA), whereas resentment interventions made
considerable alteration in the connection with the DMN and task-positive regions16. Self-respect altered only
OPEN

             
   
            
            *

Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2021) 11:14873 | 
www.nature.com/scientificreports/
the PCC-frontoparietal connection, whereas self-criticism changed the wide range of the self-referential, default
mode, and reward-motivation networks17. rough these studies, the modulation eects of self-talk on brain
connectivity have been revealed, but the brain basis of performance improvement due to positive and negative
self-talks remains uncertain.
One of the variables measuring performance improvement is uid intelligence, which is minimally dependent
on language and acquired knowledge18. e Raven’s Progressive Matrices (RPM) test is one of tools for measur-
ing it19. Bilateral frontal (i.e., dorsolateral prefrontal cortex, DLPFC) and parietal (i.e., intraparietal sulcus, IPS)
regions have been associated with uid cognitive processes induced by the RPM task20,21. Performance of uid
intelligence tests can be aected by psychological states, such as depression22 or psychosis23. Another study
reported performance improvement in anagram-solving tasks related to uid intelligence aer interrogative
self-talk24. Taken together, it is worth studying the eects of self-talk on brain networks during psychological
states in terms of changes in uid intelligence-related performance, but this has not been demonstrated yet to
our knowledge.
e eects of task-related cognitive load on functional connectivity have provided useful insights not only
into ongoing processes concerning cognitive functions, but also subsequent processes during post-task resting-
state. For example, studies investigating post-task resting-state have shown that changes in connectivity reect
recent visual/cognitive experience25 and further predict subsequent cognitive performance26,27. Additionally,
post-task resting-state connectivity is associated with experience-induced plasticity28. Other examples include
post-task changes related to cognitive functions, such as episodic memory29 and visual perception30. Taken
together, when investigating the eects of self-talk on cognitive performance, it would be meaningful to evaluate
both during-task and post-task changes.
e present study aimed to elucidate the eects of positive and negative self-talks on alterations in func-
tional connectivity related to performance of uid intelligence tests. For this aim, seed-based connectivity was
investigated on fMRI data, which were obtained while and aer performing the RPM tasks before and aer the
self-respect or self-criticism task. Given the two conceptual properties of self-talk, the verbalized form and the
identity of the sender and receiver of the message, and the experimental requirement that all participants be
given the same conditions, the self-respect and self-criticism tasks consisted of having the participants read and
record the sentences expressing themselves with “I” as the subject in advance, and having them repeat the con-
tents while listening to the recordings in the fMRI experiment. Our hypotheses were that both the self-respect
and self-criticism tasks would induce performance improvement in the RPM tasks, whereas the during-task and
post-task modulations of functional connectivity underlying these improvements would be dierent between
the two self-talk tasks in the reward-motivation network, DMN, and task-positive network in the brain. Based
on these hypotheses, the seeds were dened as the NA and VMPFC in the reward-motivation network, the PCC
in the DMN, and the DLPFC and IPS in the task-positive network.
Results
Participants’ psychological scale scores and task performances. e two cognitive tasks during
fMRI scanning were short forms of the RPM test, which were referred to as sRPM1 and sRPM2. Although a total
of 46 participants were scanned, data from those with the sRPM1 score of seven (two standard deviations lower
than mean) or less were excluded from the analysis because exceptionally low scores of the rst cognitive task
suggesting poor attention might have an excessive and inappropriate impact on the analysis. ree participants
from the self-criticism group met this criterion and were excluded. e nal analysis was conducted on the
self-respect group of 23 participants and the self-criticism group of 20 participants, and there was no signicant
group dierence in age (22.48 ± 2.13years old and 23.90 ± 2.65years old, respectively) and sex (12 males and 13
males, respectively).
Psychological scale scores and sRPM scores are presented in Table1. e Rosenberg Self-Esteem Scale (RSES)
score, Hospital Anxiety and Depression Scale (HADS)—anxiety score, and HADS—depression score did not
signicantly dier between the two groups. Compared to sRPM1 score, a signicant increase in sRPM2 score
Table 1. Summary of psychological assessments and task performances in each self-talk group. Values are
means ± standard deviation. RSES Rosenberg self-esteem scale, HADS hospital anxiety and depression scale,
sRPM short form of Raven’s Progressive Matrices. sRPM increase rate = [(sRPM2 score – sRPM1 score)/sRPM1
score] × 100.
Var i able Self-respect group
(n = 23) Self-criticism group
(n = 20) t p
RSES 30.65 ± 7.21 31.10 ± 5.25 0.23 0.819
HADS
Anxiety score 4.83 ± 2.99 4.80 ± 2.71 − 0.03 0.976
Depression score 5.13 ± 3.36 5.25 ± 2.69 0.13 0.899
sRPM performances
sRPM1 score 12.74 ± 1.79 12.50 ± 2.26 − 0.39 0.701
sRPM2 score 13.30 ± 2.41 14.20 ± 1.70 1.50 0.141
sRPM increase rate (%) 5.61 ± 18.77 17.07 ± 24.77 5.08 0.030
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2021) 11:14873 | 
www.nature.com/scientificreports/
was observed in the self-criticism group (t19 = 2.80, p = 0.011, Cohen’s d = 0.63), but not in the self-respect group
(t22 = 1.29, p = 0.212, Cohen’s d = 0.27). Accordingly, the self-criticism group showed signicantly higher sRPM
increase rate than the self-respect group (F1,40 = 5.08, p = 0.030, η2 = 0.113).
Changes in during‑sRPM state functional connectivity. Results of the seed-based connectiv-
ity analysis for states during the sRPM tasks are presented in Table2. e main eect of self-talk was found
only in DLPFC-based connectivity with the right precentral gyrus (PrCG), in which the connectivity strengths
were signicantly higher in the self-respect group than in the self-criticism group (t41 = 5.72, p < 0.001, Cohen’s
d = 0.89). e main eect of repetition was seen in the connections of NA—right lateral occipital cortex (LOC),
VMPFC—bilateral parietal operculum cortex (POC), and PCC—le PrCG, in which the connectivity strengths
were all signicantly increased during sRPM2 compared with sRPM1 (t41 = 6.10, p < 0.001, Cohen’s d = 0.95;
t41 = 6.00, p < 0.001, Cohen’s d = 0.93; t41 = 5.23, p < 0.001, Cohen’s d = 0.80; and t41 = 7.11, p < 0.001, Cohen’s d = 1.11,
respectively), and in the connection of DLPFC—right middle temporal gyrus (MTG), in which the connectiv-
ity strengths were signicantly decreased during sRPM2 compared with sRPM1 (t41 = − 6.40, p < 0.001, Cohen’s
d = 0.99). Meanwhile, there was no inter-regional connectivity showing the self-talk × repetition interaction
eect.
Changes in post‑sRPM resting‑state functional connectivity. Results of the seed-based functional
connectivity analysis for resting-states aer the sRPM tasks are presented in Table3. e main eect of self-talk
was found only in NA-based connectivity with the le MTG and right LOC. Post-hoc tests showed that the
connectivity strengths of these two were signicantly higher in the self-respect group than in the self-criticism
group (t41 = 5.20, p < 0.001, Cohen’s d = 0.81; and t41 = 5.25, p < 0.001, Cohen’s d = 0.82, respectively). Although no
repetition eect was seen, the self-talk × repetition interaction eect was observed in NA-based connectivity
with the right inferior temporal gyrus (ITG). As shown in Fig.1, post-hoc tests showed that the NA-right ITG
connectivity strengths in the post-sRPM1 resting-state did not dier between the two groups, whereas those in
the post-sRPM2 resting-state were signicantly higher in the self-respect group than in the self-criticism group
(t41 = 5.27, p = 0.001, Cohen’s d = 0.82).
Changes in functional connectivity and increase of sRPM scores. Table4 presents brain regions
that showed the signicant association between changes in inter-regional functional connectivity and sRPM
increase rates. In the self-respect group, signicant correlations between changes in during-sRPM state connec-
tivity and sRPM increase rates were observed in IPS-based connectivity with the bilateral orbitofrontal cortex
(OFC) (positive correlation), right temporal pole (positive correlation), right ITG, and right thalamus (negative
correlation) (Fig.2a). e self-respect group also showed negative correlations between changes in post-sRPM
resting-state connectivity and sRPM increase rates in NA-based connectivity with the le supplementary motor
area (SMA) and le PrCG (Fig.2b). However, the self-criticism group showed no signicant correlations for
both during-sRPM state and post-sRPM resting-state connectivity.
Table 2. Results of the seed-based functional connectivity analysis for brain states while performing the
short form of Raven’s Progressive Matrices (sRPM) in the two dierent self-talk groups: self-respect and
self-criticism. MNI Montreal neurological institute, Nvox numbers of voxels, Zmax maximum z-value within
the cluster, L. le, R. right, NA nucleus accumbens, VMPFC ventromedial prefrontal cortex, PCC posterior
cingulate cortex, DLPFC dorsolateral prefrontal cortex, IPS Intraparietal sulcus.
Source Target
MNI coordinate
Nvox Zmax Post-hoc analysisx y z
Main eect of self-talk (self-respect versus self-criticism)
NA/VMPFC/PCC –
DLPFC R. precentral gyrus 36 − 26 54 298 5.14 SR > SC
IPS –
Main eect of repetition (during-sRPM1 versus during-sRPM2)
NA R. lateral occipital cortex 42 − 78 − 22 209 5.68 sRPM1 < sRPM2
VMPFC L. parietal operculum cortex − 54 − 26 20 278 6.46 sRPM1 < sRPM2
R. parietal operculum cortex 54 − 28 26 119 4.24 sRPM1 < sRPM2
PCC L. precentral gyrus − 56 04 16 452 7.36 sRPM1 < sRPM2
DLPFC R. middle temporal gyrus 60 − 32 − 14 230 − 7.05 sRPM1 > sRPM2
IPS –
Interaction eect: self-talk × repetition
NA/VMPFC/PCC/DLPFC/IPS –
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2021) 11:14873 | 
www.nature.com/scientificreports/
Discussion
To identify the dierence in the eects of positive and negative self-talks on functional connectivity concerning
alterations in cognitive performance, we investigated changes in during-sRPM and post-sRPM connectivity
before and aer self-respect versus self-criticism. Behavior data showed that sRPM increase was signicantly
higher aer self-criticism than aer self-respect, suggesting that negative self-talk may be more benecial in the
improvement of cognitive performance than positive self-talk. e modulation eects on various networks and
associations between connectivity alterations and performance changes also diered between the self-talk groups,
suggesting that the modication of brain connectivity may play a mediating role in the eects of self-talks on
the promotion of cognitive performance.
Self‑talks and task repetition. e repetition eect was shown for during-sRPM states, but not for post-
sRPM resting-states. DLPFC-MTG connectivity decreased in during-sRPM2 compared with during-sRPM1.
e DLPFC-based network is necessary for key competencies of intelligence and executive functions31, and
MTG activity may involve increases of task demand32. Task repetition within a short time, similar to the cur-
rent study, degrades performance due to cognitive fatigue33,34. Cognitive fatigue can be motivational fatigue
related to a system that maintains motivation through monitoring internal states35, induce decreases of atten-
tion-related network connectivity36, and reduce the demand for cognitive tasks of the same diculty level34.
erefore, decreased DLPFC-MTG connectivity may reect a decline of cognitive demand associated with cog-
nitive fatigue.
NA-, PCC-, VMPFC-based connectivity increased in during-sRPM2 compared with during-sRPM1. e NA-
related network is engaged in reward prediction associated with motivation37,38. e NA involves not only external
Table 3. Results of the seed-based functional connectivity analysis for resting-states aer performing the
short form of Raven’s Progressive Matrices (sRPM) in the two dierent self-talk groups: self-respect and
self-criticism. MNI Montreal neurological institute, Nvox numbers of voxels, Z maximum z-value within
the cluster, L. le, R. right, NA nucleus accumbens, VMPFC ventromedial prefrontal cortex, PCC posterior
cingulate cortex, DLPFC dorsolateral prefrontal cortex, IPS Intraparietal sulcus, SR self-respect, SC self-
criticism.
Source Target
MNI coordinate
Nvox Zmax Post-hoc analysisx y z
Main eect of self-talk (self-respect versus self-criticism)
NA L. middle temporal gyrus − 66 − 48 04 144 5.71 SR > SC
R. lateral occipital cortex 54 − 68 12 142 5.25 SR > SC
VMPFC/PCC/DLPFC/IPS –
Main eect of repetition (post-sRPM1 versus post-sRPM2)
NA/VMPFC/PCC/DLPFC/IPS –
Interaction eect: self-talk × repetition
NA R. inferior temporal gyrus 52 − 58 − 10 113 5.04 See Fig.1
VMPFC/PCC/DLPFC/IPS –
Figure1. Post-hoc analysis of repeated measure analysis of covariance for resting-state functional connectivity
aer performing the short forms of Raven’s Progressive Matrices (sRPM1 and sRPM2). R. right, NA nucleus
accumbens, ITG inferior temporal gyrus. *p < 0.05, **p < 0.01 for post-hoc comparisons aer Bonferroni
correction.
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2021) 11:14873 | 
www.nature.com/scientificreports/
reward but also novelty of stimulus39, and thus task repetition can induce deactivation of the reward system.
Cognitive fatigue negatively aects motivation35, whereas self-talks provided as self-related information positively
aect individual performance concerning motivation1. erefore, the change in NA-based connectivity seems
to support motivational interpretation of self-talks. Alternatively, it may reect inaccurate condence associated
with familiarity according to task repetition. Inaccurate condence means high condence that does not match
actual accuracy40. Increased NA-based connectivity was observed in the right LOC. Magnetic stimulation of the
occipital cortex reduces task accuracy and, conversely, increases condence41. e increase in condence or lack
of attention increases the variability of the internal signal for stimuli, thereby inducing inaccurate selection42.
Given that the PCC and VMPFC are nodes of the self-referential network and DMN43, increased PCC-PrCG
or VMPFC-POC connectivity is likely to be induced by self-talks rather than task repetition. ere is a recent
report that connectivity of the self-referential network and DMN negatively correlated with cognitive fatigue34.
Table 4. Signicant relationships between score increase rate of the short form of Raven’s Progressive Matrices
(sRPM) and changes of functional connectivity in during-sRPM states and in post-sRPM resting-states in
each of the self-respect and self-criticism groups. MNI Montreal neurological institute, Nvox numbers of
voxels, Zmax maximum z-value within the cluster, L. le, R. right, B. bilateral, NA nucleus accumbens, VMPFC
ventromedial prefrontal cortex, PCC posterior cingulate cortex, DLPFC dorsolateral prefrontal cortex, IPS
intraparietal sulcus.
Group Seed Target
MNI coordinate, mm
Nvox Zmax
x y z
Changes in functional connectivity between during-sRPM1 and during-sRPM2
Self-respect
NA/VMPFC/PCC/DLPFC –
IPS R. orbitofrontal cortex 30 20 − 28 247 7.96
L. orbitofrontal cortex − 32 24 − 24 76 5.72
R. temporal pole 48 04 − 30 116 7.59
R. inferior temporal gyrus 62 − 22 − 20 61 5.41
R. thalamus 20 − 18 20 64 − 7.34
Self-criticism NA/VMPFC/PCC/DLPFC/IPS –
Changes in functional connectivity between post-sRPM1 and post-sRPM2
Self-respect NA B. supplementary motor area − 12 − 36 60 1236 − 8.07
L. precentral gyrus − 20 − 22 62 93 − 5.26
Self-criticism VMPFC/PCC/DLPFC/IPS –
NA/VMPFC/PCC/DLPFC/IPS –
Figure2. Scatter plots showing the relationships between during-task (a) and post-task (b) changes in inter-
regional functional connectivity (FC) and score increase rates of the short form of Raven’s Progressive Matrices
(sRPM) in the self-respect group. rs-FC, resting-state functional connectivity; L. le, R. right, B. bilateral, NA
nucleus accumbens, IPS intraparietal sulcus, SMA supplementary motor area, PrCG precentral gyrus, OFC
orbitofrontal cortex, ITG inferior temporal gyrus, TP temporal pole.
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2021) 11:14873 | 
www.nature.com/scientificreports/
Connectivity between the DMN and PrCG is related to associative learning or motivational assignments to the
ongoing motor task demands44. POC activity involves retrieving recently learned information45. erefore, the
self-referential network and DMN activated by self-talks may provide an environment that can lead to perfor-
mance improvement despite cognitive fatigue.
Eects of self‑respect. e main eect of self-talk was observed in DLPFC-PrCG connectivity for dur-
ing-sRPM states and in NA-MTG and NA-LOC connectivity for post-sRPM resting-states. Self-respect lead to
stronger connectivity in all of these connections than self-criticism. Given a key role of the DLPFC in executive
functions31 and responsibility of the PrCG for implementing corrective strategies46, robust DLPFC-PrCG con-
nectivity suggests that self-respect may be more benecial for executive functions than self-criticism. e results
of NA-based connectivity are almost unheard of, making it dicult to interpret their meaning. MTG activity is
inuenced by subjective condence in one’s accuracy of tasks47,48, and LOC activity is associated with an event-
sequence that aects the reward system49. Condence is an environmental factor in the computational model
of motivation50,51. erefore, NA-based connectivity for post-sRPM resting-states suggests that individuals who
have experienced self-respect may be more condent than those who experienced self-criticism. Alternatively,
these results may reect the inference that motivational interpretation of positive self-talk may be related to
an induced environmental factor, such as more enhanced condence. Considering that there is an association
between external stimulus on the occipital cortex and increased inaccurate condence41, robust NA-LOC con-
nectivity in the self-respect group may involve inaccurate condence that can negatively aect cognitive perfor-
mance with impulsiveness.
Only the self-respect group showed signicant correlations between connectivity changes and performance
changes for both during-sRPM and post-sRPM states. However, since the sRPM scores were not changed in this
group, the connectivity changes were not large enough to appear as a behavioral change. Alternatively, it can
be because self-respect has both positive and negative eects on cognitive performance. Specically, changes in
IPS-OFC connectivity during s-RPM tasks positively correlated with sRPM increase. e parietal network plays
an essential role in cognitive reasoning33 and is modulated by psychological interventions52. e OFC is engaged
in coordination and synthesis of visual and motor representations and in performance on processing speed53.
erefore, our result may be associated with altered brain states that are benecial for potential performance
improvement induced by self-respect. In contrast, sRPM increase negatively correlated with changes in NA-
based connectivity aer cognitive tasks, suggesting that increased NA-based connectivity by self-respect may
negatively aect cognitive performance. About NA-based connectivity associated with condence, self-respect
may adversely aect cognitive performance by increasing impulsiveness, similar to risk behaviors in associa-
tion with inaccurate condence dissociated from actual results54. Taken together, the eects of self-respect on
cognitive performance seem both negative, due to impulsivity related to inaccurate condence, and positive, due
to performance improvement related to enhanced executive functions. Since there are various methods other
than self-respect for positive self-talk, additional studies using other self-talk tasks are needed to understand its
eects on cognitive performance.
Eects of self‑criticism. Compared to sRPM1 score, sRPM2 score was signicantly increased in the self-
criticism group, but not in the self-respect group, and thus sRPM increase rate was signicantly higher in the
self-criticism group than in the self-respect group. Increased sRPM score in the self-criticism group is consistent
with previous ndings for the benecial eect of negative self-talk on enhancing performance12,55. is eect
may be because negative self-talk has a signicant inuence on attention. Negative stimuli increase attention to
a subsequent stimulus compared with positive stimuli56. Alternatively, given that motivation is a critical factor
in maintaining attention57, self-criticism may reduce cognitive fatigue-related inattention by being interpreted
more motivational. is motivational interpretation may be either because individuals try to avoid negative
results on their own through negative self-talk15 or accept it as a challenge12. Despite performance improvement
aer self-criticism, it was not correlated with connectivity change, maybe due to the ceiling eect as most par-
ticipants showed an increase in performance.
is behavioral result is supported by the self-talk eects on DLPFC-PrCG connectivity for during-sRPM
states and NA-MTG and NA-LOC connectivity for post-sRPM resting-states, which should be considered con-
trary to self-respect. In particular, considering that less condent state can induce motivation58, these ndings
suggest that condence lowered by self-criticism and subsequent motivational interpretation can lead to per-
formance improvement. is is also supported by NA-ITG connectivity in post-sRPM resting-states, which
showed no group dierence before self-talk, but was decreased aer self-criticism. Since enhanced ITG activity
is involved in more condent states40,59 and the less robust ITG activity is associated with the greater internal
motivation60, decreased NA-ITG connectivity can reect decreased condence and increased motivation induced
by self-criticism.
Although self-criticism was better at increasing sRPM scores than self-respect, it cannot be generalized that
negative self-talk will have a superior eect on performance improvement than positive self-talk. e eect of
self-talk decreases as repeated over time61, and long-term exposure to negative self-talk has harmful eects1.
erefore, our ndings on the eects of negative self-talk should be interpreted only from a short-term perspec-
tive. Further studies are needed on the long-term eects of negative self-talk on changes in brain connectivity
that underlie cognitive performance changes.
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2021) 11:14873 | 
www.nature.com/scientificreports/
Limitations
ere are some limitations to our study, which can constrain the interpretations. First, our study samples con-
sisted of young, healthy, college students who were likely of higher intellectual capacity than average. us, it
is uncertain whether the results will be similar in the general population. Second, sRPM scores represented
mainly uid intelligence, not overall cognitive performance, and the type and diculty of cognitive tasks were
not considered. ird, as the current study compared two groups divided according to the type of self-talk tasks,
there is a possibility that confounding factors may be involved. In fact, it might be desirable to see the eect of
performing both self-respect and self-criticism in a single group. To do this, however, the experimental time
given to one participant would be too long, and the sRPM sets would have to be doubled. ese could lead to
other confounding factors, and thus we had no choice but to choose the current two-group design. Fourth, the
current study design did not include non-self-reective neutral control task, and thus analysis for common eects
of repetitions was inevitably lacking. Finally, the current study did not monitor physiological data, including
heart rate, which can aect cognitive performance.
Conclusions
e current study is the rst study that directly compared the eects of positive and negative self-talks concern-
ing both cognitive performance and functional connectivity. By identifying brain responses to self-talks, our
study presented that both types of self-talks can enhance cognitive performance through dierent brain changes
related to motivation. In summary, the eects of self-respect on cognitive performance seem both negative, due to
impulsivity related to inaccurate condence, and positive, due to performance improvement related to enhanced
executive functions. On the other hand, self-criticism may induce an increase in cognitive performance, maybe
due to a less condent state that elevates internal motivation and attention. Additional studies are needed to
elucidate the modulation of condence and motivation concerning both self-talk and cognitive performance.
Moreover, further studies need to address the long-term eect of positive and negative self-talks on changes in
brain connectivity that underlie cognitive performance changes.
Materials and methods
Participants. Participants were 46 healthy college student volunteers (23.17 ± 2.39years old, 25 males and
21 females) with no past or present history of major neurological or psychiatric disorders and medical diseases
that can cause dysfunctions in cognitive performance and no experience of any form of psychological interven-
tions including self-talks. All of them were right-handed as assessed with the Annett Handedness Inventory62.
is study was approved by the Institutional Review Board of Yonsei University Severance Hospital and carried
out in accordance with the Declaration of Helsinki. All participants voluntarily signed written informed consent,
and received the same amount of Korean money 50,000 won in exchange for participation in the experiment.
Psychological assessments. Before conducting fMRI experiments, all participants completed two self-
report questionnaires. e rst was the RSES, comprised of 10 items and four-point Likert scale for measuring
an individual’s self-esteem63. e second was the HADS, comprised of 14 items (seven for anxiety and seven for
depression) and four-point Likert scale for measuring an individual’s level of depression and anxiety64.
Audiovisual stimuli and assignment of participants. Based on our previous study, which presented
dierent patterns between alterations in brain connectivity by self-respect and self-criticism17, we prepared
scripts for the two types of self-talks, which were intended to facilitate participants to focus on the feeling of self-
respect or self-criticism by telling themselves in their minds how much they respect or criticize themselves. Full
scripts of the text are provided in Supplementary Material S1. A 5-min audiovisual stimulus for the self-talk task
was produced 1week before the fMRI scan. We made an audio stimulus of participants’ own voice by recording
their script reading and then combined it with a visual stimulus, in which the scripts were visually presented in
black letters on a gray background. Assigning participants to the self-respect or the self-criticism group was done
through simple randomization using a computerized random number generator.
Meanwhile, we also prepared two sets of 5-min cognitive tasks (sRPM1 and sRPM2) produced by selecting
20 questions dierently out of the 60 questions of the RPM test19 and reducing the answer options to four. In
these tasks, visual stimuli were presented on the screen for 15s with a question placed in the center and the
answer options placed in the bottom. e diculty levels of the two tasks were set to be as similar as possible,
and in a preliminary study of 10 participants other than those who participated in this fMRI experiment, the
average score of sRPM1 and sRPM2 showed no statistical dierence (13.50 ± 1.58 and 13.50 ± 1.72, respectively;
t9 = 0.00, p = 1.00).
Experimental procedure and behavioral analyses. As shown in Fig.3a, the experimental procedure
consisted of six 5-min sessions of the fMRI scanning in the following order: baseline resting-state, rst sRPM
task (sRPM1), second resting-state (post-sRPM1), self-talk task, second sRPM task (sRPM2), and third resting-
state (post-sRPM2). In the self-talk task sessions, the participants were instructed to focus on mental images of
self-respect or self-criticism, respectfully, according to their assigned groups. roughout these tasks, the audio-
visual guidance instructed the participants to focus on the narrated scripts once, and then recite them silently,
sentence by sentence, in their mind. In the session of sRPM1 or sRPM2, the participants chose their answers by
pressing one of the four buttons with their index and middle ngers of both hands. e order of these two cogni-
tive tasks was counterbalanced among the participants. During the three resting-state sessions, the participants
were instructed to stare at a xation cross presented on the screen, relax, and think of nothing in particular. For
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2021) 11:14873 | 
www.nature.com/scientificreports/
enhancing the eectiveness of the tasks, a minute-long audiovisual guides instructing how to breathe and relax
or how to solve the problem were provided during a short break prior to the self-talk task and the cognitive tasks,
respectively. ere was no break for instruction between the sRPM task and following resting-state.
Responses for the sRPM tasks were reported as the total number of correct answers, named the sRPM1 and
sRPM2 scores. For comparing the modulation eect of two self-talk tasks, we calculated the sRPM increase rate
with the formula of [(sRPM2 scores − RPM1 score)/sRPM1 score] × 100 (%). We performed paired t tests for
the sRPM1 and sRPM2 scores in each group. In addition, the eect of two self-talk tasks on the sRPM increase
rate was compared using analysis of covariance (ANCOVA) with controlling for age, which is a factor related to
the decline of uid intelligence in adults18.
Imaging data acquisition and preprocessing. Images were acquired using a 3.0T MR scanner (Ingenia
CX, Philips, Best, the Netherlands) with a 32-channel dS head coil. For each participant, we acquired fMRI
scans using the multiband SENSitivity Encoding (SENSE) sequence (matrix size, 96 × 93; eld of view, 216mm;
number of slices, 60; slice order, bottom-up and interleaved; slice thickness, 2.4mm; echo time, 30ms; repetition
time, 800ms; ip angle, 52°; MB factor, 6; and SENSE factor, 1). We acquired additional fMRI scans of the same
parameters with two opposite phases encoding directions (anterior to posterior and posterior to anterior) to
correct the geometric distortion of the multi-band fMRI data. Anatomical images were obtained in the coronal
direction using a 3D T1-weighted fast gradient echo sequence (matrix size, 224 × 224; eld of view, 224mm;
number of slices, 220; slice thickness, 1mm; echo time, 4.6ms; repetition time, 9.9ms; and ip angle, 8°).
All fMRI scans were corrected for susceptibility-induced geometric distortions using the FSL TOPUP
tool65,66, and the rst 10 scans were discarded for magnetic eld stabilization. Preprocessing of the functional
data was carried out in Montreal Neurological Institute (MNI)-space using CONN functional connectivity
toolbox (ver.19.c, http:// www. nitrc. org/ proje cts/ conn) and Statistical Parametric Mapping 12 (SPM12, http://
www. l. ion. ucl. ac. uk/ spm). e remaining 375 functional scans for each run were realigned to the rst scan,
and temporal misalignment between dierent slices was corrected using the slice-timing correction procedure.
Considering the pervasive impact of head motion on measures of functional connectivity67–69, correcting for
motion by regressing out both motion parameters and specic frames with motion outliers was performed using
the Artifact Rejection Toolbox (ART; htt p:// www. nitrc. org/ proje cts/ artif act_ detect/) implemented in CONN for
outlier detection and scrubbing to create confound regressors for motion parameters (global-signal Z value = 9;
subject motion = 2mm). Functional and structural data were normalized into standard MNI space and segmented
into tissue classes of grey matter, white matter, and cerebrospinal uid using SPM12 unied segmentation and
normalization procedure70. To increase signal-to-noise ratio and reduce the inuence of variability in functional
Figure3. Experimental procedures of resting-state functional magnetic resonance imaging (rs-fMRI) and
fMRI during the two short forms of Raven’s Progressive Matrices (sRPM 1 and sRPM2) in the self-respect and
self-criticism groups (a), and the diagrams for statistical analysis (b). e order of sRPM1 and sRPM2 was
counterbalanced across participants in each group. Analysis of covariance (ANCOVA) was performed for self-
talk (self-respect vs. self-criticism) × repetition (during-sRPM states in Analysis 1 and post-sRPM resting-states
in Analysis 2).
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2021) 11:14873 | 
www.nature.com/scientificreports/
data and gyral anatomy across subjects, functional smoothing was conducted using spatial convolution with a
Gaussian kernel of 6mm full width at half maximum. Functional data were then temporally band-pass ltered
(0.009–0.08Hz) to remove low-frequency dri while minimizing the inuence of physiological, head-motion,
and other noise sources71.
Seed‑based functional connectivity. Based on our hypothesis that the NA and VMPFC in the reward-
motivation network, the PCC in the DMN, and the DLPFC and IPS in the task-positive network would involve
changes in the performance of cognitive tasks depending on the two contrasting self-talk tasks, a seed-based
whole-brain approach was conducted using these ve as the regions of interest (ROIs). eir MNI coordinates
(x/y/z) were determined by referring to the results of previous studies: the NA, ± 12/8/− 872, the VMPFC, 9/51/16
and PCC, 1/− 26/3173, and the DLPFC, ± 42/24/24 and IPS, ± 36/− 54/3974. e ROIs were dened as a sphere of
3-mm radius around the selected MNI coordinates.
In the rst-level analysis, the levels of functional connectivity between each ROI and every voxel in the brain
were computed as the Fisher-transformed bivariate correlation coecients between the time series of functional
data. Potential confounding factors including cerebral white matter and cerebrospinal areas, estimated subject-
motion parameters, identied outlier scans, constant and rst-order linear session eects were estimated and
regressed out using CONN’s default denoising pipeline implement an anatomical component-based noise cor-
rection procedure (aCompCor). We conducted two repeated-measures ANCOVA controlling for age, gender,
and head motion parameters to explore any signicant dierences in functional connectivity related to cognitive
performs changes according to two self-talk tasks (Fig.3b). In ANCOVA model I, we considered two self-talk
tasks and two states during cognitive tasks: 2 (self-talk, self-respect versus self-criticism) × 2 (repetition, sRPM1
versus sRPM2). In the ANCOVA model II, we considered two self-talk tasks and two resting-states aer cogni-
tive tasks: 2 (self-talk, self-respect versus self-criticism) × 2 (repetition, post-sRPM1 versus post-sRPM2). Direct
comparison of fMRI data obtained during the self-respect and self-criticism task sessions was excluded from
the analysis because this issue did not t the purpose of the study to elucidate the eect of self-talks on cognitive
performance and was addressed more intensively in our previous study17. Statistical inferences for identifying
brain regions showing main and interaction eects were performed at a threshold of the cluster-level false-
discovery-rate-corrected p (pFDR) < 0.05 with the cluster-forming threshold at the voxel level of uncorrected
p < 0.001. Post-hoc two-sample t- or paired t tests were conducted to compare the mean beta values of all voxels
in the signicant clusters, and signicant results were identied based on Bonferroni-corrected p < 0.05.
In addition, to identify the brain regions that demonstrated signicant associations between changes before
and aer self-talk tasks in during-sRPM state or post-sRPM resting-state functional connectivity and changes in
the performance of cognitive tasks, functional connectivity dierence maps of each ROI calculated by extract-
ing during-sRPM1 from during-sRPM2 and extracting post-sRPM1 from post-sRPM2 were applied for linear
regression analysis using the sRPM increase rates as a dependent variable. Voxelwise-analyses were performed,
and the signicance was considered at pFDR < 0.05 among clusters at a cluster-dening threshold of uncorrected
p < 0.001. Next, we computed the Pearson correlation between the mean beta values of all the signicant clusters
in functional connectivity dierence maps and the sRPM increase rates.
Received: 5 March 2021; Accepted: 9 July 2021
References
1. Latinjak, A. T., Hatzigeorgiadis, A., Comoutos, N. & Hardy, J. Speaking clearly … 10 years on: e case for an integrative perspec-
tive of self-talk in sport. Sport Exerc. Perform. Psychol. 8, 353–367 (2019).
2. Hatzigeorgiadis, A. & Galanis, E. Self-talk eectiveness and attention. Curr. Opin. Psychol. 16, 138–142 (2017).
3. Moser, J. S. et al. ird-person self-talk facilitates emotion regulation without engaging cognitive control: Converging evidence
from ERP and fMRI. Sci. Rep. 7, 4519–4529 (2017).
4. Hatzigeorgiadis, A., Zourbanos, N., Galanis, E. & eodorakis, Y. Self-talk and sports performance: A meta-analysis. Perspect.
Psychol. Sci. 6, 348–356 (2011).
5. Callicott, K. J. & Park, H. Eects of self-talk on academic engagement and academic responding. Behav. Disord. 29, 48–64 (2003).
6. Chakhssi, F., Kraiss, J. T., S ommers-Spijkerman, M. & Bohlmeijer, E. T. e eect of positive psychology interventions on well-being
and distress in clinical samples with psychiatric or somatic disorders: A systematic review and meta-analysis. BMC Psychiatry 18,
211 (2018).
7. Pietrowsky, R. & Mikutta, J. Eects of positive psychology interventions in depressive patients—A randomized control study.
Psychology 3, 1067–1073 (2012).
8. Sheldon, K. M. & Lyubomirsky, S. How to increase and sustain positive emotion: e eects of expressing gratitude and visualizing
best possible selves. J. Posit. Psychol. 1, 73–82 (2006).
9. Walsh, S., Cassidy, M. & Priebe, S. e application of positive psychotherapy in mental health care: A systematic review. J. Clin.
Psychol. 73, 638–651 (2017).
10. Tennen, H. & Aeck, G. Blaming others for threatening events. Psychol. Bull. 108, 209–232 (1990).
11. DeWolfe, C. E. J., Scott, D. & Seaman, K. A. Embrace the challenge: Acknowledging a challenge following negative Self-Talk
improves performance. J. Appl. Sport Psychol. https:// doi. org/ 10. 1080/ 10413 200. 2020. 17959 51 (2020).
12. Hamilton, R. A., Scott, D. & MacDougall, M. P. Assessing the eectiveness of self-talk interventions on endurance performance.
J. Appl. Sport Psychol. 19, 226–239 (2007).
13. Hardy, J., Oliver, E. & Tod, D. A framework for the study and application of self-talk within sport. In Advances in Applied Sport
Psychology: A Review (ed. Mellalieu, S. D.) 37–74 (Routledge, 2009).
14. Van Raalte, J. L., Brewer, B. W., Rivera, P. M. & Petitpas, A. J. e relationship between observable self-talk and competitive junior
tennis players’ match performances. J. Sport Exerc. Psychol. 16, 400–415 (1994).
15. Goodhart, D. E. e eects of positive and negative thinking on performance in an achievement situation. J. Personal. Soc. Psychol.
51, 117–124 (1986).
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol:.(1234567890)
Scientic Reports | (2021) 11:14873 | 
www.nature.com/scientificreports/
16. Kyeong, S., Kim, J., Kim, D. J., Kim, H. E. & Kim, J. J. Eects of gratitude meditation on neural network functional connectivity
and brain-heart coupling. Sci. Rep. 7, 5058 (2017).
17. Kyeong, S. et al. Dierences in the modulation of functional connectivity by self-talk tasks between people with low and high life
satisfaction. Neuroimage 217, 116929 (2020).
18. Schretlen, D. et al. Elucidating the contributions of processing speed, executive ability, and frontal lobe volume to normal age-
related dierences in uid intelligence. J. Int. Neuropsychol. Soc. 6, 52–61 (2000).
19. Raven, J. e Raven’s progressive matrices: Change and stability over culture and time. Cogn. Psychol. 41, 1–48 (2000).
20. Kroger, J. K. et al. Recruitment of anterior dorsolateral prefrontal cortex in human reasoning: A parametric study of relational
complexity. Cereb. Cortex 12, 477–485 (2002).
21. Prabhakaran, V., Smith, J. A. L., Desmond, J. E., Glover, G. H. & Gabrieli, J. D. E. Neural substrates of uid reasoning: An fMRI
study of neocortical activation during performance of the Raven’s Progressive Matrices Test. Cogn. Psychol. 33, 43–63 (1997).
22. Paelecke-Habermann, Y., Pohl, J. & Leplow, B. Attention and executive functions in remitted major depression patients. J. Aect.
Disord. 89, 125–135 (2005).
23. B arch, D. M., Yodkovik, N., Sypher-Locke, H. & Hanewinkel, M. Intrinsic motivation in schizophrenia: Relationships to cognitive
function, depression, anxiety, and personality. J. Abnorm. Psychol. 117, 776–787 (2008).
24. Puchalska-Wasyl, M. M. When interrogative self-talk improves task performance: e role of answers to self-posed questions.
Appl. Cogn. Psychol. 28, 374–381 (2014).
25. Albert, N. B., Robertson, E. M. & Miall, R. C. e resting human brain and motor learning. Curr. Biol. 19, 1023–1027 (2009).
26. Gregory, M. D. et al. Resting state connectivity immediately following learning correlates with subsequent sleep-dependent
enhancement of motor task performance. Neuroimage 102, 666–673 (2014).
27. Vahdat, S., Darainy, M., Milner, T. E. & Ostry, D. J. Functionally specic changes in resting-state sensorimotor networks aer motor
learning. J. Neurosci. 31, 16907–16915 (2011).
28. Guerra-Carrillo, B., MacKey, A. P. & Bunge, S. A. Resting-state fMRI: A window into human brain plasticity. Neuroscientist 20,
522–533 (2014).
29. Murty, V. P., Tompary, A., Adcock, R. A. & Davachi, L. Selectivity in postencoding connectivity with high-level visual cortex is
associated with reward-motivated memory. J. Neurosci. 37, 537–545 (2017).
30. Sarabi, M. T. et al. Visual perceptual training recongures post-task resting-state functional connectivity with a feature-represen-
tation region. PLoS ONE 13, e019866 (2018).
31. Barbey, A. K. Network neuroscience theory of human intelligence. Trends Cogn. Sci. 22, 8–20 (2018).
32. Drummond, S. P. A., Brown, G. G., Salamat, J. S. & Gillin, J. C. Increasing task diculty facilitates the cerebral compensatory
response to total sleep deprivation. Sleep 27, 445–451 (2004).
33. Vakhtin, A. A., Ryman, S. G., Flores, R. A. & Jung, R. E. Functional brain networks contributing to the Parieto-Frontal Integration
eory of Intelligence. Neuroimage 103, 349–354 (2014).
34. Wylie, G. R., Yao, B., Genova, H. M., Chen, M. H. & Deluca, J. Using functional connectivity changes associated with cognitive
fatigue to delineate a fatigue network. Sci. Rep. 10, 1–12 (2020).
35. Müller, T. & Apps, M. A. J. Motivational fatigue: A neurocognitive framework for the impact of eortful exertion on subsequent
motivation. Neuropsychologia 123, 141–151 (2019).
36. Lim, J. et al. Imaging brain fatigue from sustained mental workload: An ASL perfusion study of the time-on-task eect. Neuroimage
49, 3426–3435 (2010).
37. Mannella, F., Gurney, K. & Baldassarre, G. e nucleus accumbens as a nexus between values and goals in goal-directed behavior:
A review and a new hypothesis. Front. Behav. Neurosci. 7, 135 (2013).
38. Pessiglione, M., Seymour, B., Flandin, G., Dolan, R. J. & Frith, C. D. Dopamine-dependent prediction errors underpinreward-
seeking behaviour in humans. Nature 442, 1042–1045 (2006).
39. Gershman, S. J. Dopamine, inference, and uncertainty. Neural Comput. 29, 3311–3326 (2017).
40. Chua, E. F., Schacter, D. L. & Sperling, R. A. Neural correlates of metamemory: A comparison of feeling-of-knowing and retrospec-
tive condence judgments. J. Cogn. Neurosci. 21, 1751–1765 (2009).
41. Rahnev, D. A., Maniscalco, B., Luber, B., Lau, H. & Lisanby, S. H. Direct injection of noise to the visual cortex decreases accuracy
but increases decision condence. J. Neurophysiol. 107, 1556–1563 (2012).
42. Pleskac, T. J. & Busemeyer, J. R. Two-stage dynamic signal detection: A theory of choice, decision time, and condence. Psychol.
Rev. 117, 864–901 (2010).
43. Northo, G., Qin, P. & Nakao, T. Rest-stimulus interaction in the brain: A review. Trends Neurosci. 33, 277–284 (2010).
44. Vatansever, D., Menon, D. K., Manktelow, A. E., Sahakian, B. J. & Stamatakis, E. A. Default mode network connectivity during
task execution. Neuroimage 122, 96–104 (2015).
45. Elman, J. A., Cohn-Sheehy, B. I. & Shimamura, A. P. Dissociable parietal regions facilitate successful retrieval of recently learned
and personally familiar information. Neuropsychologia 51, 573–583 (2013).
46. Carter, C. S. et al. Parsing executive processes: Strategic vs. evaluative functions of the anterior cingulate cortex. Proc. Natl. Acad.
Sci. U. S. A. 97, 1944–1948 (2000).
47. Molenberghs, P., Trautwein, F. M., Böckler, A., Singer, T. & Kanske, P. Neural correlates of metacognitive ability and of feeling
condent: A large-scale fMRI study. Soc. Cogn. Aect. Neurosci. 11, 1942–1951 (2016).
48. White, T. P., Engen, N. H., Sørensen, S., Overgaard, M. & Shergill, S. S. Uncertainty and condence from the triple-network per-
spective: Voxel-based meta-analyses. Brain Cogn. 85, 191–200 (2014).
49. Akitsuki, Y. et al. Context-dependent cortical activation in response to nancial reward and penalty: An event-related fMRI study.
Neuroimage 19, 1674–1685 (2003).
50. Lee, D., Seo, H. & Jung, M. W. Neural basis of reinforcement learning and decision making. Annu. Rev. Neurosci. 35, 287–308
(2012).
51. Schultz, W. Neuronal reward and decision signals: From theories to data. Physiol. Rev. 95, 853–951 (2015).
52. Taren, A. A. et al. Mindfulness meditation training and executive control network resting state functional connectivity: A rand-
omized controlled trial. Psychosom. Med. 79, 674–683 (2017).
53. Kringelbach, M. L. e human orbitofrontal cortex: Linking reward to hedonic experience. Nat. Rev. Neurosci. 6, 691–702 (2005).
54. Nigg, J. T. Annual research review: On the relations among self-regulation, self-control, executive functioning, eortful control,
cognitive control, impulsivity, risk-taking, and inhibition for developmental psychopathology. J. Child Psychol. Psychiatry Allied
Discip. 58, 361–383 (2017).
55. Van Raalte, J. L., Brewer, B. W., Lewis, B. P., Linder, D. E. & Al, E. Cork! e eects of positive and negative self-talk on dart throw-
ing performance. J. Sport Behav. 18, 50 (1995).
56. Tartar, J. L., de Almeida, K., McIntosh, R. C., Rosselli, M. & Nash, A. J. Emotionally negative pictures increase attention to a sub-
sequent auditory stimulus. Int. J. Psychophysiol. 83, 36–44 (2012).
57. Engelmann, J. B. & Pessoa, L. Motivation sharpens exogenous spatial attention. Emotion 7, 668–674 (2007).
58. Wulf, G. & Lewthwaite, R. Optimizing performance through intrinsic motivation and attention for learning: e OPTIMAL theory
of motor learning. Psychon. Bull. Rev. 23, 1382–1414 (2016).
59. Kim, H. & Cabeza, R. Common and specic brain regions in high- versus low-condence recognition memory. Brain Res. 1282,
103–113 (2009).
Content courtesy of Springer Nature, terms of use apply. Rights reserved


Vol.:(0123456789)
Scientic Reports | (2021) 11:14873 | 
www.nature.com/scientificreports/
60. Linke, J. et al. Motivational orientation modulates the neural response to reward. Neuroimage 49, 2618–2625 (2010).
61. Bar wood, M. J., Corbett, J., Wagsta, C. R. D., McVeigh, D. & elwell, R. C. Improvement of 10-km time-trial cycling with moti-
vational self-talk compared with neutral self-talk. Int. J. Sports Physiol. Perform. 10, 166–171 (2015).
62. Annett, M. A classication of hand preference by association analysis. Br. J. Psychol. 61, 303–321 (1970).
63. Rosenberg, M. Society and the adolescent self-image (Princeton University Press, 1965).
64. Zigmond, A. S. & Snaith, R. P. e hospital anxiety and depression scale. Acta Psychiatr. Scand. 67, 361–370 (1983).
65. Andersson, J. L. R., Skare, S. & Ashburner, J. How to correct susceptibility distortions in spin-echo echo-planar images: Application
to diusion tensor imaging. Neuroimage 20, 870–888 (2003).
66. Andersson, J. L. R. & Sotiropoulos, S. N. An integrated approach to correction for o-resonance eects and subject movement in
diusion MR imaging. Neuroimage 125, 1063–1078 (2016).
67. Satterthwaite, T. D. et al. Impact of in-scanner head motion on multiple measures of functional connectivity: Relevance for studies
of neurodevelopment in youth. Neuroimage 60, 623–632 (2012).
68. Van Dijk, K. R., Sabuncu, M. R. & Buckner, R. L. e inuence of head motion on intrinsic functional connectivity MRI. Neuroim-
age 59, 431–438 (2012).
69. Zeng, L. L. et al. Neurobiological basis of head motion in brain imaging. Proc. Natl. Acad. Sci. U. S. A. 111, 6058–6062 (2014).
70. Ashburner, J. & Friston, K. J. Unied segmentation. Neuroimage 26, 839–851 (2005).
71. Hallquist, M. N., Hwang, K. & Luna, B. e nuisance of nuisance regression: Spectral misspecication in a common approach to
resting-state fMRI preprocessing reintroduces noise and obscures functional connectivity. Neuroimage 82, 208–225 (2013).
72. Gu, H. et al. Mesocorticolimbic circuits are impaired in chronic cocaine users as demonstrated by resting-state functional con-
nectivity. Neuroimage 53, 593–601 (2010).
73. Dosenbach, N. U. et al. Prediction of individual brain maturity using fMRI. Science 329, 1358–1361 (2010).
74. Bishop, S. J., Fossella, J., Croucher, C. J. & Duncan, J. COMT val158met genotype aects recruitment of neural mechanisms sup-
porting uid intelligence. Cereb. Cortex 18, 2132–2140 (2008).
Acknowledgements
is work was supported by the Technology Development Program of MSS (S2800407) and the National Research
Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (No. NRF-2016R1A2A2A10921744).
Author contributions
Jo.K., E.J.K., and J.J.K. designed research; J.H.K. and H.E.K. performed research; Ju.K. and S.K. analyzed data;
Ju.K., Jo.K., E.J.K., and J.J.K. wrote the paper.
Competing interests
e authors declare no competing interests.
Additional information
Supplementary Information e online version contains supplementary material available at https:// doi. org/
10. 1038/ s41598- 021- 94328-9.
Correspondence and requests for materials should be addressed to J.-J.K.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Open Access is article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the
Creative Commons licence, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.
© e Author(s) 2021
Content courtesy of Springer Nature, terms of use apply. Rights reserved

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com