Increased or Decreased Interpersonal Neural Synchronization in - 2022 - NeuroIm

NeuroImage 260 (2022) 119448
Contents lists available at ScienceDirect
NeuroImage
journal homepage: www.elsevier.com/locate/neuroimage
Increased or decreased? Interpersonal neural synchronization in group

creation
Zheng Liang a,b,1, Songqing Li a,b,c,1, Siyuan Zhou d,1, Shi Chen a,b, Ying Li a,b,e, Yanran Chen a,b,
Qingbai Zhao a,b,∗, Furong Huang f,∗∗, Chunming Lu d, Quanlei Yu a,b,∗, Zhijin Zhou a,b,∗
a
Key Laboratory of Adolescent Cyberpsychology and Behavior (CCNU), Ministry of Education, Central China Normal University, Wuhan, China
b
Key Laboratory of Human Development and Mental Health of Hubei Province, School of Psychology, Central China Normal University, Wuhan, China
c
College of Electronic Engineering, Naval University of Engineering, Wuhan, China
d
State Key Laboratory of Cognitive Neuroscience and Learning and IDG/McGovern Institute for Brain Research, Beijing Normal University, Beijing, China
e
School of Preschool Education, Changsha Normal University, Changsha, China
f
School of Psychology, Jiangxi Normal University, Nanchang, China
a r t i c l e i n f o a b s t r a c t
Keywords: Group creation is the process by which group members collaborate to produce novel and useful ideas or products,
Group creation including ideas generation and evaluation. However, the interpersonal neural mechanism of group creation dur-
Hyperscanning ing natural communication remains unclear. In this study, two groups of same-sex dyads with similar individual
fNIRS
creativity collaborated to complete the Product Improvement Task (creative condition) and the Item Purchase
Interpersonal neural synchronization
Plan Task (control condition), respectively. Functional near-infrared spectroscopy (fNIRS) was used to record both
members’ neural activity in the left prefrontal (lPFC) and right temporal-parietal junction (rTPJ) regions during
the task. Considering that the role asymmetry of group members may have an impact on interpersonal neural pat-
terns, we identified leaders and followers in the dyads based on participant performance. The results showed that
leaders and followers in the creative condition had significantly lower interpersonal neural synchronization (INS)
in the right superior temporal gyrus-left superior frontal gyrus, right supramarginal gyrus-left superior frontal
gyrus, and right supramarginal gyrus-left middle frontal gyrus than in the control condition. Partial multivariate
Granger causality analyses revealed the influence between dyads was bidirectional but was significantly stronger
from the leaders to the followers than the other direction. In addition, in the creative task, the INS was signifi-
cantly associated with novelty, appropriateness, and conflict of views. All these findings suggest that the ideas
generation and ideas evaluation process in group creation have poor interpersonal neural activity coupling due
to factors such as the difficulty of understanding novel ideas. However, performances may be improved when
groups can better integrate views and reach collective understanding, intentions, and goals. Furthermore, we
found that there are differences in the dynamics of INS in different brain regions. The INS related to the novelty
of the group creation decreased in the early stages, while the INS related to the appropriateness decreased in
the middle stages. Our findings reveal a unique interpersonal neural pattern of group creation processes in the
context of natural communication.
1. Introduction 2016; Senaratne and Gunawardane, 2015). Therefore, understanding

the mechanisms of group creativity is important for many domains.
Employing small groups to solve problems is critical to modern life Group creativity is defined as a group of members working together
(Doboli and Doboli, 2021). Cooperation between members with differ- to produce novel and useful ideas or products (Ristic et al., 2016;
ent knowledge and skills may create novel ideas or products and solve Wang and Zhu, 2011). There are two basic characteristics of group cre-
problems creatively, resulting in greater group benefits (Mao et al., ativity: group structure and creative process. For the group structure,
due to differences such as knowledge utilization preferences and indi-
vidual traits (Ray and Romano, 2013), role asymmetry may exist in the
∗
Corresponding authors at: School of Psychology, Central China Normal University, Wuhan, 430079, China.
∗∗
Corresponding author at: School of Psychology, Jiangxi Normal University, Nanchang 330022, China.
E-mail addresses: [email protected] (Q. Zhao), [email protected] (F. Huang), [email protected] (Q. Yu), [email protected] (Z.
Zhou).
1
The first three authors contribute equally.
https://doi.org/10.1016/j.neuroimage.2022.119448.
Received 29 October 2021; Received in revised form 1 July 2022; Accepted 3 July 2022
Available online 14 July 2022.
1053-8119/© 2022 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/)
Z. Liang, S. Li, S. Zhou et al. NeuroImage 260 (2022) 119448
interaction process of group creativity. Groups tend to emerge with indi- To better understand the interactions that occur naturally in the
viduals who are good at generating novel ideas and individuals who are group creativity process and the characteristics of its interpersonal neu-
willing to understand and coordinate with others (Bolinger et al., 2009). ral mechanism, Mayseless et al. (2019) explored the INS difference be-
The former usually gain status in the group and may become leaders who tween creative tasks (open product design) and control tasks (prescribed
lead ideas. The latter, as followers, exert less influence on others but are 3D model building) in the context of natural communication and its rela-
also important in promoting creativity (Bolinger et al., 2009). This pat- tionship with behavioral indices of creativity. The result suggested that
tern of collaboration may be relatively stable throughout the process dyads’ INS in anterior prefrontal cortex (aPFC)-posterior superior tem-
of group creativity (Mayseless et al., 2019). Similarly, in the leader- poral gyrus (pSTG) and aPFC-TPJ in the creative task was significantly
less group discussions, many studies have also found that individuals greater than in the control task. However, it should be noted that while
may spontaneously form a “leader-follower” relationship (i.e. one per- the creative design task includes the idea generation and idea evalua-
son lead the task and the other one follow) in interactions (Jiang et al., tion, the 3D model building task as the control condition is quite differ-
2015; Konvalinka et al., 2014; Selten and Warglien, 2007). This spon- ent from it. For example, the 3D model building task requires advanced
taneous asymmetrical interpersonal relationship may be beneficial to manual dexterity and spatial skill. However, this interactive process may
group performance (Selten and Warglien, 2007; Wallot et al., 2016). For involve less verbal communication and less complex processing of lan-
the creative process, group creativity involves both divergent and con- guage and emotions (Li et al., 2021), which may affect the strength of
vergent thinking, which is similar to individual creativity (Tan, 2015; INS (Hasson et al., 2012).
Ulrich, 2018). The interacting members not only generate ideas (related Although the above two studies found higher INS in group cre-
to divergent thinking) but also evaluate the ideas (related to convergent ation, there is also some evidence that the INS of group creation tasks
thinking) of others (Chen et al., 2017; Paulus and Yang, 2000; Ray and may be lower than the general tasks. First, the shared neural response
Romano, 2013), with the two stages being constantly recursive and it- may reflect similar representation or thinking (Cetron et al., 2019;
erative (Harvey, 2014; Harvey and Kou, 2013). The former is expected Meshulam et al., 2021; Nguyen et al., 2021; Wass et al., 2020). To gener-
to improve the novelty of ideas and the latter is expected to improve the ate novel ideas, group members often need to inhibit dominant and con-
appropriateness of the group’s creative ideas (Paletz and Schunn, 2010; sensus representations, engage in mental representational change, break
Singh and Fleming, 2010). In addition, due to cognitive diversity and dif- the thinking impasses and diverge in different directions (Huang et al.,
ficulty in understanding the intentions of others’ views, members may 2019). As a result, the INS of group creation tasks may decrease. Sec-
have a conflict of views during interactions, but creativity tends to de- ond, due to the heterogeneity of knowledge, skills, and experience,
velop better when teams are better able to integrate members’ views group members in the process of idea evaluation may have difficulty
(Harvey, 2014; Kohn et al., 2011; Ristic et al., 2016; Xue et al., 2018). understanding each other’s novel ideas and intentions, resulting in dis-
Since group creation depends on interpersonal interaction, in re- agreements or conflicts (Bodla et al., 2018), which may lead to the
cent years, researchers have used functional near-infrared spectroscopy INS of group creation tasks lower than general tasks (Fishburn et al.,
(fNIRS) based hyperscanning technique to understand the interaction 2018; Lu et al., 2021; van Baar et al., 2021). Finally, Mayseless et al.
of creative groups (e.g., Duan et al., 2020; Lu et al., 2019a, 2019b, (2019) found that the originality score in the creative task showed a
2020b; Lu and Hao, 2019; Mayseless et al., 2019). Compared with trend level negative correlation with the INS increase of inferior frontal
single-brain studies, hyperscanning aims to measure the brain activity gyrus (IFG)-pSTG. This seems to imply that the INS may decrease when
of multiple brains simultaneously to meet the needs of studying the in- the interacting members generate some creative responses (Shamay-
terbrain activity patterns of interactive participants (see (Kelsen et al., Tsoory et al., 2019).
2020; Redcay and Schilbach, 2019; Wang et al., 2018) for further de- The present study used a Product (umbrella) Improvement task as
tails). Because fNIRS has high temporal and spatial resolution, and high the group creation task (Torrance, 1966). Participants were required
tolerance for physical activity, it has been widely used as a brain sig- to eventually generate a novel and appropriate solution. This process
nal acquisition modality in hyperscanning studies (Brockington et al., involved both idea generation and evaluation. The item (umbrella) pur-
2018). By analyzing the synchrony of group members’ neural activ- chase plan task, which requires a similar interaction pattern but does
ity during the interaction, termed interpersonal neural synchronization not demand creativity, was used as a general task for control. The simi-
(INS), researchers mainly focused on the influencing factors of the pro- larity of the experimental materials used and interaction patterns in the
cess of group creation. Recent evidence suggests that INS in the process two tasks better reduced the differences in INS due to excessive differ-
of group creation is related to romantic relationships (Duan et al., 2020), ences between experimental tasks. Given the advantages of fNIRS-based
communication mode (Lu et al., 2020b), cooperation (Xue et al., 2018), hyperscanning techniques, this study used fNIRS to continuously record
feedback (Lu et al., 2019a). And the stronger INS may indicate better changes in the cerebral activity of dyads during the interaction. We fo-
team performance. However, the above-mentioned influencing factors cused on the rTPJ and lDLPFC to explore the interpersonal neural mod-
on INS have also been found in other hyperscanning studies of interper- els between the individuals involved in group creation. The reasons are
sonal interaction (Cui et al., 2012; Liu et al., 2019; Long et al., 2021; as follows. First, the rTPJ and lDLPFC have been considered to be key
Zhu et al., 2021). brain regions for cognitive processing in creative tasks (Huang et al.,
To identify the interpersonal neural models specific to group creativ- 2021). The rTPJ as a component of the default-mode network (DMN)
ity, a direct comparison of group creativity tasks with general tasks (i.e. was considered to be related to the generation of creative ideas, while
non-creative tasks) may be needed. Lu et al. (2019b) compared the INS the lDLPFC as a component of the executive control network (ECN) was
difference between the alternative uses task (AUT, demanding divergent considered to be related to the evaluation of creative ideas (Huang et al.,
thinking) and the object characteristic task (OCT, not demanding diver- 2021; Kleinmintz et al., 2019). Moreover, in the context of social inter-
gent thinking) under the conditions of cooperation. They found that the action, the TPJ and PFC regions are also considered to be components
dyads’ INS of right dorsolateral prefrontal cortex (rDLPFC) and right of the mentalizing system (MS), which is responsible for mentalizing,
temporal-parietal junction (rTPJ) in the AUT task was stronger than in theory of mind, and other social cognition. The coupling between the
the OCT task. However, in this study, participants were only asked to two may reflect the state of interaction between individuals (Lu et al.,
continuously generate novel ideas in the turn taking without eventually 2021). In addition, previous hyperscanning studies of group creativity
forming an integrated proposal. So the study focused more on idea gen- have also shown that the lDLPFC and rTPJ are recruited (Lu et al., 2019a;
eration and less involved idea evaluation in group creativity. In addition, Lu and Hao, 2019).
unlike natural communication, turn taking itself is already a cooperative Role asymmetry has been neglected in previous studies on interper-
process that may affect interpersonal neural models related to creativity sonal neural mechanisms of group creativity, and members’ roles are
(Mayseless et al., 2019). considered to be equal and symmetrical. The INS of different brain areas
2
was calculated as the average across all group members in any given con- ment of interest in the task (see Supplementary Material S3 for ques-
dition. For example, the INS between lDLPFC and rTPJ was calculated tionnaires).
from the average of the INS between lDLPFC (Participant 1) and rTPJ
(Participant 2) and the INS between rTPJ (Participant 1) and lDLPFC 2.3.2. General experimental procedure
(Participant 2) (Li et al., 2021). Because role differentiation already ex- The general experimental procedure consisted of an 8 min resting-
hibits unique interpersonal neural characteristics early in natural com- state session, 2 min instruction session, and a 20 min task session (see
munication (Jiang et al., 2015), this calculation method of INS may ob- Fig. 1C).
scure some important information. In the present study, we assessed Participants were asked to sit face to face (see Fig. 1B), and the ini-
individual roles (leader or follower) based on dyads’ interactions before tial 8 min resting-state session served as a baseline. During this session,
data analysis. Then INS was calculated between each channel of the participants were asked to remain as still as possible, with their eyes
leaders and all channels of the followers in each dyad and was not aver- closed and their minds relaxed, and don’t think about specific problems
aged. This study examined the differences and dynamics of INS between or fall asleep (Lu et al., 2019b). Next, in the instruction session, the
creative condition and control condition, the information flow between task and requirements were introduced. Then, during the task session,
roles, and the relationship between INS and behavior indicators. group members cooperated to complete the corresponding experimental
tasks through natural communication. This process was recorded with a
2. Methods video camera with audio. fNIRS data were simultaneously collected dur-
ing both the resting-state session and task session. To ensure the validity
2.1. Participants of the data, the participants were asked to avoid large-scale movement
as much as possible during the experiment. After the discussion, the par-
One hundred and twenty-seven college students were recruited. They ticipants were asked to write down the plan together, and individually
were asked to complete an AUT task through an online questionnaire be- rated the degree of conflict of views during the discussion session from
fore the experiment. The novelty scores of the answers generated by the 1 (not at all) to 5 (very much) points.
participants were used as a measure of individual creativity (Runco and
Acar, 2012) (see details in the supplementary materials S1). 2.3.3. Task performance evaluation and role division
To avoid the influence of gender and creativity level on group cre- Eight graduate students, who were blind to the group assignment,
ation (Cheng et al., 2015; Xue et al., 2018), the dyads were made up of used the consensus assessment technique (Amabile, 1983) to evaluate
two people with the same level of creativity and the same gender (see the novelty and appropriateness of the problem solution on a Likert scale
details in the supplementary materials S1). After excluding participants ranging from 1 (not at all) to 5 (extremely). Four of them evaluated
who disobeyed the experimental instructions (see details in the supple- the creative product improvement task and the other four evaluated
mentary materials S1), there were 17 dyads (6 dyads of men and 11 the purchase plan task. The average of the four scores across two items
dyads of women) under each experimental condition, for a total of 68 was used as the final score for group performance. All the inter-rater
participants (mean age: 20.01 ± 1.91 years old). agreements were satisfactory (ICCs > 0.82).
The members of each dyad did not know each other before the ex- In addition, three graduate students were invited to watch the video
periment. All the participants were right-handed, had no brain disease and evaluate the role of each member of the dyad (leader: who always
or mental illness, and had normal or corrected-to-normal vision. They takes the lead, follower: who more or less follows the other) based on the
were paid a standard experiment participation fee and provided signed naturally emerging discussion. One member was identified as a leader
informed consent before participation. All experimental protocols were if two or more raters marked him or her as a leader. The other member
approved by the Ethics Institute Review Board of Central China Normal of the dyad was marked as a follower. Another four graduate students
University. counted the number of expressed views per individual in each dyad. All
graduate students were blind to the purpose of the experiment.
2.2. Experimental tasks
2.4. fNIRS data acquisition
In the creative condition, dyads were asked to complete the “Prod-
uct Improvement Task”. This task is an item on the Torrance Tests of NIRScout (NIRx Medical Technologies, New York) was used to
Creative Thinking (TTCT), which is typically used to measure individ- record changes in each participant’s oxy-hemoglobin (HbO) and deoxy-
ual creativity (Wei et al., 2014). We slightly changed the task for use in hemoglobin (HbR) concentrations during the experimental task. There
the dyadic context. To make the task relevant to daily life, we replaced were 4 probe sets. Of these, 2 probes had 3 emitters and 4 detectors to
the elephant, the object of the original creative product improvement form a 3∗ 4 probe set, forming a total of 8 measurement channels. The
task, with an umbrella (see Fig. 1A). These dyads were asked to discuss other 2 probe sets had 5 emitters and 4 detectors to form a 5∗ 4 probe set,
form a novel and appropriate improvement plan for the umbrella. In forming a total of 12 measurement channels. Thus, there were a total of
the control condition, dyads were asked to complete a general task of 20 measurement channels. The distance between each channel was up
“Item Purchase Plan”. Compared with the creative task, the general task to 3 cm (fixed with a 3 cm fixing piece).
required little creativity. These dyads were asked to discuss to form an In dyads, each participant had two probe sets on the head: a 3∗ 4
appropriate purchase plan for the umbrella (see supplementary mate- probe set on the left forehead and a 5∗ 4 probe set on the right temporal-
rials S2 for task instructions). After the discussion, the team ultimately parietal joint area. The probe was placed according to the 10-20 inter-
needed to reach a consensus on the improvement plan or purchase plan national system. The emitters on the left forehead covered FP1, AF3,
and then write down the plan together. and F5, with detectors located at FPz, AFz, AF7, and F3. The emitters
on the right temporal-parietal joint area were located at P8, T8, CP6,
2.3. Experimental procedure P4, and C4, with detectors at TP8, P6, C6, and CP4. The positions of
the probes are shown in Fig. 1D. The brain region corresponding to the
2.3.1. Subjective measurement channel was positioned using a three-dimensional locator (NIRx Medi-
Before the general experimental procedure, participants were asked cal Technologies, New York), which to determine the Nz, Cz, Iz, AL, AR
to complete assessments of cooperative preference, the familiarity with points and probe positions. The probabilistic registration method was
umbrellas, and the degree of demand for umbrellas. After the general used to register the fNIRS channel position with the Montreal Neuro-
experimental procedure, participants were asked to complete an assess- logical Institute (MNI) space coordinates to obtain the corresponding
3
Fig. 1. Experiment procedure. (A) Umbrella prototype.

(B) Participants’ seating arrangement. (C) Experiment flow
chart. (D) Cap configuration. Red circles indicate emitters;
Blue circles indicate detectors. The measurement channels
are marked by numbers. Measurement channels covered the
frontal, temporal, and parietal cortices.
relationship with the Brodmann area and automatic anatomical labels and the five extraneous variables at the pair level (continuous variables)
(AAL) area (Tsuzuki et al., 2012). were covariates.
The absorption of near-infrared light at two wavelengths (785 and
830 nm) was measured with a sampling rate of 7.8125 Hz. Based on the
modified Beer-Lambert law, changes in the HbO and HbR concentrations 2.6. fNIRS data analysis
were obtained by measuring changes in absorption of fNIRS light after
its transmission through the tissue. Previous studies have shown that 2.6.1. Pair-level analysis
HbO was a sensitive indicator of change in regional cerebral blood flow The fNIRS data collected during the rest session and task session were
(Zheng et al., 2018). Thus, this study focused on the HbO concentrations analyzed. Data from the first 60 s and last 60 s were deleted during pre-
only. processing. Thus, data within the period of steady state were analyzed.
During preprocessing, no filtering or detrending procedures were ap-
plied (Cui et al., 2012). In addition, we did not perform any artifact
2.5. Behavioral data analysis
corrections at the single-subject level, as wavelet transform coherence
(WTC) normalizes the amplitude of the signal according to each time
2.5.1. Correctness check of role division
window and thus was not vulnerable to the transient spikes induced by
Since the leaders may express more views (Wickham and
movements (Nozawa et al., 2016).
Walther, 2007), we conducted a 2 (between-group variable: Task Type:
We employed WTC analysis to estimate INS. A Matlab package was
creative task vs. general task) ∗ 2 (within-group variable: Role: leader
used to perform WTC (Grinsted et al., 2004) as a way to assess the
vs. follower) mixed ANOVA on the number of expressed views to ver-
cross-correlation between the two fNIRS time series generated by each
ify the correctness of the role division. In addition, the communication
pair of participants as a function of frequency and time (Torrence and
skills and competence of members were further evaluated to ensure the
Compo, 1998). For the two participants in one dyad, HbO values were
correctness of the role division (see Supplementary material S4 for de-
obtained in two time series of equal length and aligned. Then, WTC was
tails).
applied to these two aligned time series to find regions in the time fre-
quency space where the two time series co-varied. For more thorough in-
2.5.2. Task performance formation about wavelet coherence, please see Grinsted et al. (2004) and
Before conducting a comparison of task performance between the Chang and Glover (2010). Because there were 20 measurement channels
two tasks, we first tested the control of extraneous variables in the ex- for each participant (leader or follower), 400 (20∗ 20) pairs of time series
periment. A 2 (between-group variable: Task Type: creative task vs. gen- were generated for each dyad, and WTC was thus conducted 400 times
eral task) ∗ 2 (within-group variable: Role: leader vs. follower) mixed (Zheng et al., 2020, 2018). The coherence values were time-averaged
ANOVA was conducted for each of the following dependent variables: across the rest and discussion periods, and converted into Fisher z-
individual creativity, cooperative preference, familiarity with umbrel- values.
las, the degree of demand for umbrellas, and interest in the task. Consistent with previous studies (Dai et al., 2018; Jiang et al., 2012),
As extraneous variables obtained by subjective measurement might we focus on the relative enhancement of INS during the task session
have affected task performance, one-way ANCOVA was conducted to ex- compared to the resting-state session. Thus, we subtracted the coherence
clude possible effects of extraneous variables on task performance. We value of the resting-state session from that of the task session to obtain
separately averaged the five extraneous variables in dyads. For that anal- an index of time-aligned INS increase (Grinsted et al., 2004). At this
ysis, the Task Type (a categorical variable) was an independent variable stage, no specific frequency ranges were selected.
4
Since the group creativity process requires members to continuously conditions, we conducted a time accumulation INS analysis to signifi-
generate and evaluate ideas through verbal communication, this process cant CH combinations respectively (Liu et al., 2019). First, we normal-
involves neurocognitive processing of members’ mutual understanding. ized the discussion session of each dyad into 200 epochs. For each dyad,
The neural coupling between two persons in verbal interaction may not the time-cumulative INS at epoch n was calculated as the sum of the INS
be limited to time alignment. Previous studies have shown that in the ranging from the first epoch to the nth epoch. Second, we performed in-
process of information comprehension, listeners’ neural response may dependent samples t-tests at 200 epochs to compare the difference of
lag behind speakers, resulting in a time-lag INS (Jiang et al., 2021; INS increase between the two conditions. Finally, the resulting p values
Liu et al., 2017; Stephens et al., 2010). Therefore, we added various were FDR corrected.
time-lags to the computation of INS increases to obtain the time-lagged
INS increases (Long et al., 2021; Zheng et al., 2018). For the selection 2.6.2.5. Coupling directionality. We further estimate the magnitude of
of time windows, we referred to previous interpersonal neural studies bidirectional information flow between the leaders and the followers
on verbal comprehension (Jiang et al., 2021; Liu et al., 2020, 2017; in the two tasks by conducting partial multivariate Granger causality
Stephens et al., 2010). Specifically, the time course of the leaders’ brain analyses (PMGCA). Tradition Granger causality analysis (GCA) uses vec-
activity (i.e., HbO) was shifted forward or backward relative to that of tor autoregressive models to measure the causal relationship between
the followers’ brain activity by 1–6 s (step = 1 s). time series in brain data. Since the Granger causality value may indi-
cate the strength of the interpersonal influence during the social inter-
2.6.2. Group-level analysis (time-aligned INS) action (Cheng et al., 2019), it provides a neurobiological suggestion of
2.6.2.1. Task type-related differences in time-aligned INS. The following coupling directionality, i.e., which individual was more actively driving
analyses were conducted to identify the difference in the INS increase another (Barnett and Seth, 2014). However, the exogenous and endoge-
between the two tasks. First, to identify the frequency ranges that were nous influences such as common external stimulus-induced neural re-
specifically associated with the task style, independent sample t-tests sponses or similar intrinsic neural responses may confound the Granger
were conducted on the time averaged coherence value of each CH (chan- causality between individuals and lead to spurious causal inference
nel) combination (400 in total) along the full frequency range (0.01–0.7 (Guo et al., 2008; Roelstraete and Rosseel, 2012; Youssofzadeh et al.,
Hz). Following previous studies (Dai et al., 2018; Jiang et al., 2012), data 2016). PMGCA can better mitigate potentially confounding effects on
above 0.7 Hz were not included to avoid aliasing of higher frequency causal inference by modifying the traditional GCA by adding terms
physiological noise, such as cardiac activity (0.8–2.5 Hz). Data below based on residual correlations between the predicted and the condi-
0.01 Hz were also not used, to remove very low frequency fluctuations. tional variables (see more details in (Guo et al., 2008; Roelstraete and
Finally, data within the frequency range of respiratory activity (0.15– Rosseel, 2012, 2011).
0.3 Hz) were not considered. The t-test results were threshold at p < Consistent with previous studies (Hou et al., 2020; Pan et al., 2021,
0.0005. No further corrections were applied because this analysis was 2018), our PMGCA was based on normalized HbO signals of signifi-
used to identify the pattern along the frequency range rather than to cant CH combinations during the discussion periods. The main steps
obtain the final results (Zheng et al., 2018). It was finally found that the of the PMGCA are as follows: First, in each channel, we converted the
frequency range of interest was 0.315–0.445 Hz, and the coherence val- HbO signals of the task session into z-scores using the mean and the
ues within this frequency range were averaged. Independent samples t- standard deviation of the signals recorded during the resting-state ses-
tests (creative task vs. general task) were performed on the time-aligned sion (Chen et al., 2020; Cheng et al., 2019; Pan et al., 2021). Second,
INS increase of all channels in this band, and FDR correction was per- for each individual, the z-scored time series of significant channels in
formed (Zheng et al., 2020, 2018). The channel combinations exhibit- the same brain regions were averaged. Third, to mitigate the impact
ing significant condition-related INS increase differences were defined of exogenous or latent variables, the time series of the leader corre-
as significant CH combinations. sponding follower’s significant channel and the time series of the fol-
lower corresponding leader’s significant channel were used as moderat-
ing variables (see 3.2.5 for details). Fourth, an R package (FIAR; down-
2.6.2.2. Validation of the INS differences between task types through a per-
load from https://github.com/cran/FIAR) was used to calculate the par-
mutation test. To verify that the group difference in INS increase was
tial multivariate Granger causalities in two directions (Roelstraete and
specific to the original pairing of the leaders and followers, a valida-
Rosseel, 2011): from the leaders to the followers and from the followers
tion approach (i.e., a permutation test) was used (Lu et al., 2020b). For
to the leaders. Finally, we used a one-sample t-test to compare the dif-
each condition, all participants were randomly assigned to new dyads
ferences between each direction and zero in each condition, and then
to recompute the INS and perform a series of independent sample t-tests
the effect of Task Type and Direction was examined by mixed ANOVA.
on the newly formed two sets of data. This permutation process was re-
peated 1000 times to yield a distribution (t value) of all CHs, which was
2.6.3. Group-level analysis (time-lagged INS)
then compared with the original pairing data.
2.6.3.1. Task type-related differences in time-lagged INS. To examine
whether there are differences in time-lagged INS between the two tasks,
2.6.2.3. Linking task performances with time-aligned INS. To determine a series of independent samples t-tests (creative task vs. general task)
the behavioral significance of leader-follower neural coupling, we ex- was applied to each time lag (i.e., -6s (follower precede) to +6s (leader
amined the correlation between INS increase of significant CH com- precede)) in 0.315–0.445 Hz frequency range, and FDR correction was
binations and the task performances, separately, under different con- performed (Long et al., 2021; Zheng et al., 2020).
ditions. Since creativity level, cooperative preference, and interest in
the task may affect the correlation between INS and task performance 2.6.3.2. Linking conflict of views with time-lagged INS. Since the level of
(Kelsen et al., 2020; Lu et al., 2019b; Xue et al., 2018; Zhu et al., 2019), time-lagged INS was considered to represent the degree of individual
we separately averaged the three extraneous variables in dyads. Then, understanding of information in the communication (Jiang et al., 2021;
a Pearson’s partial correlation analysis was applied, which controlled Liu et al., 2017; Stephens et al., 2010), it might be related to the conflict
for the potential effect of the three extraneous variables (creativity, co- of views perceived by the individual during the task (Harvey, 2014). To
operative preference, and interest in the task), and FDR correction was test this hypothesis, we averaged the degree of conflict of views in dyads,
performed. then a Pearson’s correlation analysis was adopted to analyze the rela-
tionships between the degree of conflict of views and the time-lagged
2.6.2.4. Dynamics of the time-cumulative time-aligned INS analysis. To INS of significant CH combinations, separately, under different task con-
identify the earliest time-point where the INS increase differed among ditions, and FDR correction was performed.
5
Table 1
Specific brain areas involved in the INS between leaders and followers.
CH Combinations(leader-follower) Leader Follower
CH3-CH14 right Superior temporal gyrus (rSTG) left Superior frontal gyrus (lSFG)
CH7-CH16 right Supramarginal gyrus (rSMG) left Superior frontal gyrus (lSFG)
CH7-CH18 right Supramarginal gyrus (rSMG) left Middle frontal gyrus (lMFG)
CH15-CH20 left Superior frontal gyrus (lSFG) left Middle frontal gyrus (lMFG)
3. Result 4 significant CH combinations from the 1000 permutations are shown

in Fig. 2C. Compared with the distribution generated by the permuta-
3.1. Behavioral result tion procedure, the t-values of the actual dyads were in the 1% areas
of the distribution generated by the permutation procedure. These re-
3.1.1. Correctness check of role division sults suggest that the significant INS increase was specific to the original
The mixed ANOVA results on the number of expressed views re- dyads.
vealed a significant main effect of Role, F (1, 32) = 76.79, p < 0.001,
𝜂 2 partial = 0.71, the leaders (27.71 ± 1.45) expressed more views than 3.2.3. Relationship between task performances and INS
the followers (22.50 ± 1.31). In addition, there was a significant in- A significant positive correlation was found between the time-
teraction between Task Type and Role, F (1, 32) = 4.96, p = 0.033, aligned INS increase of CH3-CH16 (rSTG-lSFG) and the novelty score
𝜂 2 partial = 0.13. The simple effect for Role at creative task was signifi- under the creative condition (partial r = 0.62, corrected p = 0.027) (see
cant, F (1, 32) = 60.40, p < 0.001, 𝜂 2 partial = 0.65, the leaders (28.88 ± Fig. 2D). There was also a significant positive correlation found between
2.05) expressed more views than the followers (22.35 ± 1.85). The sim- the INS increase of CH7-CH18 (rSMG-lMFG) and the appropriateness
ple effect for Role at general task was also significant, F (1, 32) =21.35, score under the creative condition (partial r = 0.64, corrected p = 0.026)
p < 0.001, 𝜂 2 partial = 0.40, the leaders (26.53 ± 2.05) expressed more (see Fig. 2E). The correlation between INS and task performances was
views than the followers (22.65 ± 1.85). That is, the difference in the not found in the general task (corrected ps > 0.05).
number of expressed views between leaders and followers was greater
in the creative task compared to the general task. The main effect for
Task Type was not significant, F (1, 32) = 0.15, p = 0.705. The results 3.2.4. Dynamic INS difference between two tasks
suggested that the evaluators’ judgment on the role of participants is The time accumulation INS analysis explored how time-aligned INS
reasonable. dynamically changes over the course of tasks. The result showed that
from the 5th epoch (about 23-29s), the INS increase of CH3-CH15 and
3.1.2. Task performance CH3-CH16 showed a continuous and stable significant difference among
There was no significant main effect or interaction effect for individ- conditions (corrected ps < 0.05). Starting from the 57th (about 319–
ual creativity, cooperative preference, familiarity with umbrellas, the 324s) epoch and the 60th (about 336–342s) epoch, respectively, the
degree of demand for umbrellas, and interest in the task (ps > 0.05). INS increase of CH7-CH16 and CH7-CH18 showed continuous and sta-
The ANCOVA results showed a significant main effect of Task Type ble significant differences among conditions (corrected ps < 0.05) (see
on novelty score (F (1, 27) = 15.95, p < 0.001, 𝜂 2 partial = 0.37). Specif- Fig. 3).
ically, the novelty score of dyads in the creative task (3.06 ± 0.81)
was higher than that of the general task (2.29 ± 0.78). The main 3.2.5. Coupling directionality
effect of Task Type on appropriateness score was not significant (F PMGCA was used to measure the directional information flow be-
(1, 27) = 3.43, p = 0.075; creative task (3.06 ± 0.52), general task tween two members. Leaders contain two time series: rSTG (CH3) and
(3.57 ± 0.55)). rSMG (CH7), and followers contain two time series: lSFG (the average
of CH15 and CH16) and lMFG (CH18). In addition, two time series of
3.2. Time-aligned INS result lSFG (the average of CH15 and CH16) and lMFG (CH18) from leaders
and two time series of rSTG (CH3) and rSMG (CH7) from followers were
3.2.1. Task type-related differences in time-aligned INS used as condition variables.
The differences between the conditions on the time-aligned INS in- In creative condition, the G-causalities of both directions were
crease of all channel combinations in the 0.315–0.445 HZ frequency significantly higher than zero: from the leaders to the followers (t
were tested. The results of t-tests showed that the INS for the chan- (16) = 10.29, p < 0.001) and from the followers to the leaders (t
nel combinations (leader-follower) of CH3-CH15 (t (32) = -4.10, cor- (16) = 17.80, p < 0.001). The PMGCA result of the control condition
rected p = 0.026, Cohen’s d = 1.41), CH3-CH16 (t (32) = -4.40, cor- was similar to the creative condition, with both directions identified
rected p = 0.020, Cohen’s d = 1.51), CH7-CH16 (t (32) = -4.75, corrected significant increases in the G-causality relative to zero: from the leaders
p = 0.016, Cohen’s d = 1.63) and CH7-CH18 (t (32) = -4.30, corrected to the followers (t (16) = 8.88, p < 0.001) and from the followers to the
p = 0.020, Cohen’s d = 1.47) under the creative condition were signif- leaders (t (16) = 12.26, p < 0.001).
icantly lower than under the control condition (see Fig. 2A, B). These Mixed ANOVA revealed a significant main effect of Task Type, F
four channel combinations are defined as significant CH combinations (1, 32) = 4.82, p = 0.036, 𝜂 2 partial = 0.13, with the larger G-causalities
for subsequent analysis, and the specific brain they involved are shown in the control condition (0.0049 ± 0.0020) compared to the creative
in the Table 1. condition (0.0038 ± 0.0013) (see Fig. 4). This result suggested that a
stronger interpersonal influence between group members in the gen-
3.2.2. Validation of the INS differences between task types eral task condition. The main effect of Direction was also significant,
After random permutation, the results did not show any significant F (1, 32) = 4.62, p = 0.039, 𝜂 2 partial = 0.13. G-causalities from the
group differences in time-aligned INS increase for any CH combination leaders to the followers (0.0047 ± 0.0021) was significantly greater
at the 0.315–0.445 Hz frequency. Validation results (t-values) for the than that from the followers to the leaders (0.0040 ± 0.0013). How-
6
Fig. 2. Time-aligned INS result. (A) Time-aligned INS matrix at 0.315–0.445 HZ. The color indicates the t value. The significantly different CH combinations are
marked by the red frames. (B) The location of significant CH combinations on the cerebral cortex. (C) The distribution of the t value was calculated by 1000 random
dyads for the significant CH combinations. The dotted lines denote the positions of the t values on the actual dyads. (D) The partial correlation between time-aligned
INS increase of CH3-CH16 and novel score. (E) The partial correlation between time-aligned INS increase of CH7-CH18 and appropriateness score.
ever, we did not find a significant interaction effect, F (1, 32) = 0.01, 3.3.2. Relationship between conflict of views and time-lagged INS
p = 0.914. When the brain activity of the leaders precedes the followers, the
correlation analysis found that the INS increase of the CH3-CH14 at 1s
3.3. Time-lagged INS result and 2s time lags was significantly negatively correlated with the degree
of conflict of views in the creative condition (1-s time lag: rCH3-CH14 = -
3.3.1. Task type-related differences in time-lagged INS 0.73, corrected pCH3-CH14 = 0.002; 2-s time lag: rCH3-CH14 = -0.71, cor-
The time-lag results showed that the INS increase of CH3-CH14 (t rected pCH3-CH14 = 0.002) (see Fig. 5E, 5F). The correlation between
(32) = -4.70, corrected p = 0.019, Cohen’s d =1.66) and CH15-CH20 (t time-lagged INS increase and conflict of views was not found in CH15-
(32) = -4.22, corrected p = 0.038, Cohen’s d = 1.49) were significantly CH20 or the general task (corrected ps > 0.05).
lower in the creative condition than in the control condition when the
leaders’ brain activity preceded that of the followers by 1s. In addition, 4. Discussion
when the leaders’ brain activity preceded that of the followers by 2s, the
INS increase of CH3-CH14 (t (32) = -4.42, corrected p = 0.043, Cohen’s The present study explored the difference between the group creativ-
d = 1.56) was significantly lower in the creative condition than in the ity task and the general task and unveiled the underlying interpersonal
general task condition. See Table 1 for specific brain areas involved in neural correlates, using the fNIRS-based hyperscanning technique.
the CH3-CH14 and CH15-CH20. No significant results were found when Our findings first confirm that the “leader-follower” role asymmetry
the followers’ brain activity preceded that of the leaders at any time- occurred spontaneously in groups during natural communication. Lead-
lags, at any CH combinations (corrected p > 0.05) (see Fig. 5A-5D). ers expressed more views, and there was a greater information flow
7
Fig. 3. The dynamics of the time-cumulative INS. The time-cumulative INS increase of 200 normalized epochs in the creative condition and control condition. The
red and blue shaded areas denote the standard error at each epoch. The yellow color indicates the INS significant difference between these two conditions, ∗ p < 0.05,
FDR corrected.
Research on the emergence of leadership has found that the higher

a member’s participation and contribution, the greater the chance of
being considered a leader (Barge, 1989). That is, the amount of commu-
nication predicts becoming a leader (Van Vugt, 2006). Our behavioral
results also confirm the above views. Individuals evaluated as leaders in
creative tasks and general tasks expressed more views than individuals
evaluated as followers. In addition, we found that the difference in the
number of views expressed by leaders and followers was greater in cre-
ative tasks compared to general tasks, which suggests that more novel
ideas may be proposed by leaders in creative tasks.
GCA was originally used to track the direction of information flow
between different areas within a brain, and recently has been expanded
to track the direction of information flow across interlocutors’ brains
during social interactions (Schippers et al., 2010). The G-causalities
could be used as an indicator of the influence exerted by one individ-
Fig. 4. The G-causalities from leader (L) to follower (F) were significantly
ual on another individual at the brain level (Cai et al., 2018). Consis-
greater than vice versa. The G-causalities of the creative condition were sig- tent with previous interpersonal communication research (Jiang et al.,
nificantly weaker than the control condition. ∗ p < 0.05. 2015), our PMGCA results firstly showed that the influence between in-
dividuals in each dyad was bidirectional. However, the influence from
the leaders to the followers was significantly stronger than the other
from the leaders to the followers. Moreover, we reveal the unique in- direction. That is, although the leaders and followers were mutually in-
terpersonal neural models of group creation. The time-aligned INS and fluenced, the primary information flow was from leaders to followers.
time-lagged INS between leaders and followers in the creative condition Specifically, the past neural activity of the leaders’ rSTG and rSMG can
were significantly lower than in the control condition, and the former better predict the future neural activity of the followers’ lSFG and lMFG.
positively predicted the performances of group creation, and the latter These results may suggest that although both participants were actively
negatively predicted the conflict of views in the group creation. In ad- engaged in the interaction, the leaders may entrain the followers’ neural
dition, differences between groups emerged earlier for INS related to activity more during the interaction (Pan et al., 2018).
novelty and later for INS related to appropriateness. The above results provide further evidence of role asymmetry in
group creation of natural communication. Since the INS may reflect the
role relationship within the group (Zheng et al., 2020), the previous cal-
4.1. Role asymmetry in group creation culation form of averaging between the two directions in the dyads may
lose valuable information, and likely hinder us from exploring the in-
Leadership is a universal feature of social species (Jiang et al., 2015). terpersonal neural mechanisms of group creation at a finer granularity.
Leader-follower relationships emerge spontaneously among participants (Hamilton, 2021; Jiang et al., 2021).
in leaderless group discussions, even though they have no differences in
natural or social status (Jiang et al., 2021, 2015). However previous in-
terpersonal neural studies on group creativity did not take into account 4.2. INS characteristics of group creation
group structure (i.e. role asymmetry). In this study same-sex pairs of
college students who had similar individual creativity scores worked on The present study used a product improvement task that included
either a group creativity task or a general task. Based on their inter- ideas generation and ideas evaluation phases as the group creation task
actions, one member was identified as the leader and the other as the and the item purchase task that did not require creativity as the control
follower. condition. The former required dyads to end up with novel and appropri-
8
Fig. 5. Time-lagged INS result. (A) Time course of the INS increase of CH3-CH14 and CH15-CH20 from -6s (follower precede) to +6s (leader precede). The y-axis
shows the t value (i.e., independent sample t-tests on the INS increase, two-tailed), ∗ p < 0.05. (B) The location of significant CH combinations on the cerebral cortex.
(C, D) Time-lagged INS matrix at 0.315–0.445 HZ, when the leaders’ brain activity preceded that of the followers by 1s and 2s. The color indicates the t value. The
significantly different CH combinations are marked by the red frames. (E) The correlation between INS increase of CH3-CH14 at 1s time lags and conflict of view
score. (F) The correlation between INS increase of CH3-CH14 at 2s time lags and conflict of view score.
ate solutions, while the latter only required dyads to end up with appro- uation of ideas (Ellamil et al., 2012; Huang et al., 2021). These results
priate solutions. After controlling for extraneous factors, the behavioral may reflect a unique pattern of synergistic activation between DMN and
results showed that the novelty score of the creative task was signif- ECN among leaders and followers during group creation (Li et al., 2021).
icantly higher than the general task, while the appropriateness score Compared with the traditional single-brain functional connectiv-
was not significantly different between the two tasks. This result is con- ity analysis, the functional correlations between subjects (e.g., INS)
sistent with the performance characteristics of both creative and general have a higher signal-to-noise ratio and interpersonal interaction sen-
tasks and demonstrates the validity of our experimental manipulation. sitivity (Pan et al., 2020). According to the cross-brain functional in-
The fNIRS results showed significant differences in the time-aligned tegration hypothesis of INS, INS may ‘reflect a reciprocal and dy-
INS of rSTG-lSFG, rSMG-lSFG, and rSMG-lMFG between the creative namic interplay between the neural states of socially interacting con-
condition and control condition. The STG and SMG belong to the DMN specifics’ (Holroyd, 2022). This hypothesis holds that multiple brains
(Abe et al., 2019; Marron et al., 2018), which are considered to be can come together to act jointly as a functional unit, much like mod-
the core brain regions for creativity and are related to novel associ- ules within a single brain can coordinate their activities to accomplish
ation, idea generation, and divergent thinking (Benedek et al., 2014; tasks (Holroyd, 2022; Valencia and Froese, 2020). The coupling between
Huang et al., 2021; Kleibeuker et al., 2017; Wei et al., 2014; Wu et al., DMN and ECN has been widely found in the study of individual creativ-
2016). The lSFG and lMFG are located in lDLPFC (Kikinis et al., 2010), a ity (Beaty et al., 2016, 2015; Ellamil et al., 2012; Kleinmintz et al., 2019;
key node in the ECN (Huang et al., 2021). The lDLPFC is associated with Pinho et al., 2015), which seems to reflect that the ECN can top-down
working memory and verbal comprehension (Klaus and Schutter, 2018; monitor and direct the DMN’s idea generation process in the form of idea
Kleinmintz et al., 2019), and was found to be recruited during the eval- evaluation to meet the task goals (Beaty et al., 2016). Thus, when the
9
ideas generation and ideas evaluation of individual creation is mapped negatively correlated with the conflict degree of views in the creative
to group creation, it may manifest as the interpersonal coupling of DMN condition.
and ECN. We did not find significant results of time-lagged INS when the fol-
Unlike previous studies, we found that the INS of the creative con- lower’s brain activity preceded that of the leaders. The results suggest
dition was significantly lower than the control condition. One possible that the leaders played a dominant role in the interaction during group
explanation for the results may be that the ideas generated by the lead- creation. In this case, the brain activity of the leaders might always be
ers in the creative tasks were more novel and unique than the general ahead of the followers in time (Jiang et al., 2015).
tasks, and it might be difficult for the followers with different knowledge
and experience to understand their intentions or form shared represen- 4.3. Temporal dynamics of INS in group creation
tation. And many group creation studies have indeed found that group
members may react negatively when evaluating the novel ideas of oth- The INS may not be temporal stationarity during measurements
ers (Harvey, 2014; Harvey and Kou, 2013; Mueller et al., 2012). The (Li et al., 2021). We examined the INS dynamics of the creative task.
above factors might lead to the difficulty of good cooperation between Since cumulative INS may be a better dynamic indicator than moment-
the DMN and ECN in dyads, resulting in lower INS (Fishburn et al., to-moment INS (Jiang et al., 2015), we used cumulative INS to assess
2018; Jiang et al., 2021). In addition, the PMGCA results show that the the time point at which INS differences emerged between the creative
G-causalities of creative tasks are also significantly lower than the con- and control conditions. The results of the time-course analysis showed
trol condition, which seems to indicate that the interpersonal influence that about half a minute after the beginning of the problem-solving pro-
between leader and follower is lower in the creative task than in the cess, the difference in time cumulative INS of rSTG and lSFG between
general task for the above reasons (Cheng et al., 2019). the two conditions became significant and persisted until the end of the
However, there are other possible explanations (not necessarily al- interactive process. However, the stable difference in the time cumu-
ternative) for the above results. Although we consider the STG and SMG lative INS of rSMG-lSFG and rSMG-lMFG between the two conditions
as being in the DMN and the SFG and MFG as being within the ECN, it appeared later, about 5–6 minutes after the beginning of the problem-
does not imply in any way that these regions contribute exclusively to solving process.
those networks. For example, STG and SMG are also subcomponents of Our results found that the INS of rSTG-lSFG was related to nov-
rTPJ, which is related to social cognitive processes such as perspective elty and the INS of rSMG-lMFG was related to appropriateness. The
taking, mentalization, and theory of mind (Lu et al., 2020). The rTPJ difference in the temporal dynamics of INS may reflect the character-
and the dlPFC (i.e., where SFG and MFG are located) together form istics of the group creation process. In the early stages of group cre-
the MS. This system can help individuals understand others’ intentions ation, unconstrained divergent thinking is a key factor of group cre-
based on gestures, behaviors, and facial expressions (Mayseless et al., ativity (Rosing et al., 2018). Since the group has not yet determined
2019; Wang et al., 2018). Thus, another possible explanation for these the direction of a novel solution, members can explore a variety of dif-
results is that because of the novelty of the ideas talked about in the ferent approaches and directions. The group tends to generate more
creative task, it may be more difficult for group members to figure out unique and original ideas during this process, and these initial nov-
each other’s intentions. elty ideas generated often influence the novelty of the final solution
Consistent with previous research (Duan et al., 2020; Lu et al., (Puccio et al., 2020). Therefore, the novelty-related INS (e.g., rSTG-
2019a), we also found that INS in group creation was positively cor- lSFG) showed inter-condition differences earlier. As the discussion pro-
related with final task performance (i.e., novelty and appropriateness). gresses, beginning in the middle to late stages of group creation, mem-
Due to factors such as the heterogeneous structure of the group and bers may gradually begin to consider the implementation of creative so-
cognitive diversity, groups may generate different views during the cre- lutions, focusing on their feasibility (Rosing et al., 2018). At this point,
ation process. The dialectical model of group creativity argues that when the group may consider whether the solution is appropriate when devel-
groups are able to actively process the ideas of others and creatively oping its ideas. Therefore, the appropriateness-related INS (e.g., rSMG-
synthesize opposing views, they may have the opportunity to achieve lMFG) showed inter-condition differences later.
high-quality creative solutions (Harvey, 2014). The integration of views
emphasized by this theory is a process of building similarities within 4.4. Limitations
different perspectives and shaping collective attention and collective
understanding, which may improve INS (Cirelli, 2018; Fishburn et al., The study has several limitations. First, gender may affect interper-
2018; Gvirts and Perlmutter, 2020). Therefore, if leaders and follow- sonal interaction (Cheng et al., 2015), and the participants in this study
ers integrate their views into a shared view in a cyclical generation- were all same-sex dyads. In the future, researchers should consider ex-
evaluation process, the DMN and ECN between the two may be better ploring the cooperation mechanism in group creation of opposite-sex
coupled, which will help the group develop a good goal directed and im- members, as leadership and cooperation with the group may vary by
prove task performance (Harvey, 2014; Huang et al., 2021; Paulus et al., gender (Lu et al., 2020a; Mu et al., 2018). Second, in addition to the
2012; Paulus and Brown, 2007). generation and evaluation of novel ideas, DMNs and ECNs have multi-
Previous hyperscanning studies have found that the brain activ- ple functions. For example, the DMN also has the role of making social
ity of listeners tends to lag behind that of speakers (Liu et al., 2020; predictions in interpersonal interactions (Barrett, 2016). This implies
Stephens et al., 2010; Zheng et al., 2018). This delay phenomenon of that there may not be only one interpretation for our findings and fur-
INS is explained by the fact that interactive speech processing between ther detailed exploration is needed in the future. Third, the interper-
individuals takes a certain amount of time to reach mutual understand- sonal neural activity of group creativity may not be limited to rTPJ and
ing, and the strength of the time-lagged INS is also considered as the lDLPFC, and future research should study more brain areas.
level of understanding (Jiang et al., 2021). Our results found that the
time-lagged INS of rSTG-lSFG was significantly lower in the creative 5. Conclusion
condition than in the control condition when the leaders’ brain ac-
tivity was preceded by 1s or 2s to the followers. This may indicate In the present study, we identified the roles in the leaderless group
that followers had difficulty understanding when evaluating the novel discussion and found that leaders expressed more views and influenced
ideas generated by the leaders. Interpersonal incomprehension has long followers more. Compared with the control condition, the interpersonal
been considered an important factor in the generation of conflict in influence between leaders and followers in the creative condition was
group creation (Hu et al., 2017). This view is also confirmed by our weaker, and the time-aligned INS between leaders’ DMN and followers’
findings that the time-lagged INS of rSTG-lSFG was significantly and ECN was lower. However, when these two brain regions of the dyads
10
could be better coupled, the task performances of group creation would Barge, J.K., 1989. Leadership as medium: a leaderless group discussion model. Commun.
be improved. The time-lagged INS, in which the leaders’ brain activity Q. 37, 237–247. doi:10.1080/01463378909385547.
Barnett, L., Seth, A.K., 2014. The MVGC multivariate Granger causality toolbox: a
was preceded to the followers, was lower in the creative condition than new approach to Granger-causal inference. J. Neurosci. Methods 223, 50–68.
the control condition, and it may reflect the creative groups’ conflict of doi:10.1016/j.jneumeth.2013.10.018.
views. In addition, the INS related to the novelty of the group creation Barrett, L.F., 2016. The theory of constructed emotion: an active inference account of inte-
roception and categorization. Soc. Cogn. Affect. Neurosci. doi:10.1093/scan/nsw154.
decreased in the early stages, while the INS related to the appropriate- Beaty, R.E., Benedek, M., Barry Kaufman, S., Silvia, P.J., 2015. Default and exec-
ness decreased in the middle stages. Our findings provide interpersonal utive network coupling supports creative idea production. Sci. Rep. 5, 10964.
neural evidence for group creative interactions in the context of natural doi:10.1038/srep10964.
Beaty, R.E., Benedek, M., Silvia, P.J., Schacter, D.L., 2016. Creative cognition and
communication and increase our understanding of the nature of group
brain network dynamics. Trends Cogn. Sci. 20, 87–95. doi:10.1016/j.tics.2015.
creativity. 10.004.
Benedek, M., Jauk, E., Fink, A., Koschutnig, K., Reishofer, G., Ebner, F., Neubauer, A.C.,
2014. To create or to recall? Neural mechanisms underlying the generation of creative
Credit authorship contribution statement new ideas. Neuroimage 88, 125–133. doi:10.1016/j.neuroimage.2013.11.021.
Bodla, A.A., Tang, N., Jiang, W., Tian, L., 2018. Diversity and creativity in cross-national
Zheng Liang: Conceptualization, Formal analysis, Visualization, teams: the role of team knowledge sharing and inclusive climate. J. Manage. Organ.
24, 711–729. doi:10.1017/jmo.2016.34.
Writing - original draft, Writing - review & editing. Songqing Li:
Bolinger, A.R., Bonner, B.L., Okhuysen, G.A., 2009. Sticking together: the glue role and
Conceptualization, Investigation, Data curation, Project administration, group creativity. In: Mannix, A., E, Goncalo, A., J, A., Neale, M. (Eds.), Creativity in
Writing - original draft. Siyuan Zhou: Methodology, Formal analysis, Groups, Research on Managing Groups and Teams. Emerald Group Publishing Lim-
ited, pp. 267–289. doi:10.1108/S1534-0856(2009)0000012013.
Writing - original draft. Shi Chen: Investigation, Writing - review & edit-
Brockington, G., Balardin, J.B., Zimeo Morais, G.A., Malheiros, A., Lent, R., Moura, L.M.,
ing. Ying Li: Investigation, Writing - review & editing. Yanran Chen: Sato, J.R., 2018. From the laboratory to the classroom: the potential of functional
Methodology, Formal analysis. Qingbai Zhao: Supervision, Writing - re- near-infrared spectroscopy in educational neuroscience. Front. Psychol. 9, 1840.
view & editing, Funding acquisition. Furong Huang: Supervision, Writ- doi:10.3389/fpsyg.2018.01840.
Cai, Y.X., Zhang, D.L., Liang, B.S., Wang, Z.J., Li, J.C., Gao, Z.N., Gao, M.X., Chang, S.,
ing - review & editing. Chunming Lu: Supervision. Quanlei Yu: Super- Jiao, B.Q., Huang, R.W., Liu, M., 2018. Relation of visual creative imagery ma-
vision, Writing - review & editing. Zhijin Zhou: Supervision, Writing - nipulation to resting-state brain oscillations. Brain Imaging Behav. 12, 258–273.
review & editing. doi:10.1007/s11682-017-9689-8.
Cetron, J.S., Connolly, A.C., Diamond, S.G., May, V.V., Haxby, J.V., Kraemer, D.J.M.,
2019. Decoding individual differences in STEM learning from functional MRI data.
Ethics statement Nat. Commun. 10, 2027. doi:10.1038/s41467-019-10053-y.
Chang, C., Glover, G.H., 2010. Time–frequency dynamics of resting-
state brain connectivity measured with fMRI. Neuroimage 50, 81–98.
All experimental protocols were approved by the Ethics Institute Re- doi:10.1016/j.neuroimage.2009.12.011.
view Board of Central China Normal University. Chen, M.-H., Chang, Y.-Y., Chang, Y.-C., 2017. The trinity of entrepreneurial team dy-
namics: cognition, conflicts and cohesion. IJEBR 23, 934–951. doi:10.1108/IJE-
BR-07-2016-0213.
Data and code availability statement Chen, Q., Beaty, R.E., Qiu, J., 2020. Mapping the artistic brain: Common and distinct
neural activations associated with musical, drawing, and literary creativity. Hum.
The data and code that support the findings of this study are not pub- Brain Mapp. 41, 3403–3419. doi:10.1002/hbm.25025.
Cheng, X., Li, X., Hu, Y., 2015. Synchronous brain activity during cooperative exchange
licly available due to research data sharing restrictions from the univer- depends on gender of partner: a fNIRS-based hyperscanning study: synchronous brain
sity, but can be available from the corresponding author by submitting activities. Hum. Brain Mapp. 36, 2039–2048. doi:10.1002/hbm.22754.
a formal project outline. Cheng, X., Pan, Y., Hu, Yinying, Hu, Yi, 2019. Coordination elicits synchronous brain
activity between co-actors: frequency ratio matters. Front. Neurosci. 13, 1071.
doi:10.3389/fnins.2019.01071.
Funding Cirelli, L.K., 2018. How interpersonal synchrony facilitates early prosocial behavior. Curr.
Opin. Psychol. 20, 35–39. doi:10.1016/j.copsyc.2017.08.009.
Cui, X., Bryant, D.M., Reiss, A.L., 2012. NIRS-based hyperscanning reveals increased in-
This study was supported by Self-Determined Research Funds of terpersonal coherence in superior frontal cortex during cooperation. Neuroimage 59,
CCNU from The Central Colleges’ Basic Research and Operation of MOE 2430–2437. doi:10.1016/j.neuroimage.2011.09.003.
(Grant No. CCNU19TD019 and CCNU19ZN022). Dai, B., Chen, C., Long, Y., Zheng, L., Zhao, H., Bai, X., Liu, W., Zhang, Y., Liu, L.,
Guo, T., Ding, G., Lu, C., 2018. Neural mechanisms for selectively tuning in
to the target speaker in a naturalistic noisy situation. Nat. Commun. 9, 2405.
Data and code availability statement doi:10.1038/s41467-018-04819-z.
Doboli, A., Doboli, S., 2021. A novel agent-based, evolutionary model for expressing the
dynamics of creative open-problem solving in small groups. Appl. Intell. 51, 2094–
The data and code that support the findings of this study are not pub- 2127. doi:10.1007/s10489-020-01919-6.
licly available due to research data sharing restrictions from the univer- Duan, H., Yang, T., Wang, X., Kan, Y., Zhao, H., Li, Y., Hu, W., 2020. Is the creativity of
sity, but can be available from the corresponding author by submitting lovers better? A behavioral and functional near-infrared spectroscopy hyperscanning
study. Curr. Psychol. 1–14. doi:10.1007/s12144-020-01093-5.
a formal project outline. Ellamil, M., Dobson, C., Beeman, M., Christoff, K., 2012. Evaluative and genera-
tive modes of thought during the creative process. Neuroimage 59, 1783–1794.
Declaration of Competing Interest doi:10.1016/j.neuroimage.2011.08.008.
Fishburn, F.A., Murty, V.P., Hlutkowsky, C.O., MacGillivray, C.E., Bemis, L.M., Mur-
phy, M.E., Huppert, T.J., Perlman, S.B., 2018. Putting our heads together: interper-
The authors declare no conflict of interest. sonal neural synchronization as a biological mechanism for shared intentionality. Soc.
Cogn. Affect. Neurosci. 13, 841–849. doi:10.1093/scan/nsy060.
Grinsted, A., Moore, J.C., Jevrejeva, S., 2004. Application of the cross wavelet transform
Supplementary materials and wavelet coherence to geophysical time series. Nonlin. Process. Geophys. 11, 561–
566. doi:10.5194/npg-11-561-2004.
Supplementary material associated with this article can be found, in Guo, S., Seth, A.K., Kendrick, K.M., Zhou, C., Feng, J., 2008. Partial Granger causal-
ity—eliminating exogenous inputs and latent variables. J. Neurosci. Methods 172,
the online version, at doi:10.1016/j.neuroimage.2022.119448. 79–93. doi:10.1016/j.jneumeth.2008.04.011.
Gvirts, H.Z., Perlmutter, R., 2020. What guides us to neurally and behaviorally align with
References anyone specific? A neurobiological model based on fNIRS hyperscanning studies. Neu-
roscientist 26, 108–116. doi:10.1177/1073858419861912.
Abe, M.O., Koike, T., Okazaki, S., Sugawara, S.K., Takahashi, K., Watanabe, K., Sadato, N., Hamilton, A.F., de, C., 2021. Hyperscanning: beyond the hype. Neuron 109, 404–407.
2019. Neural correlates of online cooperation during joint force production. Neuroim- doi:10.1016/j.neuron.2020.11.008.
age 191, 150–161. doi:10.1016/j.neuroimage.2019.02.003. Harvey, S., 2014. Creative synthesis: exploring the process of extraordinary group creativ-
Amabile, T.M., 1983. A consensual technique for creativity assessment. In: The So- ity. AMR 39, 324–343. doi:10.5465/amr.2012.0224.
cial Psychology of Creativity. Springer, New York, New York, NY, pp. 37–63. Harvey, S., Kou, C.-Y., 2013. Collective engagement in creative tasks: the role
doi:10.1007/978-1-4612-5533-8_3.
11
of evaluation in the creative process in groups. Adm. Sci. Q. 58, 346–386. and performance on a complex task. PLoS One 11, e0153048. doi:10.1371/jour-
doi:10.1177/0001839213498591. nal.pone.0153048.
Hasson, U., Ghazanfar, A.A., Galantucci, B., Garrod, S., Keysers, C., 2012. Brain-to-brain Marron, T.R., Lerner, Y., Berant, E., Kinreich, S., Shapira-Lichter, I., Hendler, T., Faust, M.,
coupling: a mechanism for creating and sharing a social world. Trends Cogn. Sci. 16, 2018. Chain free association, creativity, and the default mode network. Neuropsy-
114–121. doi:10.1016/j.tics.2011.12.007. chologia 118, 40–58. doi:10.1016/j.neuropsychologia.2018.03.018.
Holroyd, C.B., 2022. Interbrain synchrony: on wavy ground. Trends Neurosci. Mayseless, N., Hawthorne, G., Reiss, A.L., 2019. Real-life creative problem solv-
doi:10.1016/j.tins.2022.02.002. ing in teams: fNIRS based hyperscanning study. Neuroimage 203, 116161.
Hou, Y., Song, B., Hu, Yinying, Pan, Y., Hu, Yi, 2020. The averaged inter-brain coherence doi:10.1016/j.neuroimage.2019.116161.
between the audience and a violinist predicts the popularity of violin performance. Meshulam, M., Hasenfratz, L., Hillman, H., Liu, Y.-F., Nguyen, M., Norman, K.A.,
Neuroimage 211, 116655. doi:10.1016/j.neuroimage.2020.116655. Hasson, U., 2021. Neural alignment predicts learning outcomes in students
Hu, N., Chen, Z., Gu, J., Huang, S., Liu, H., 2017. Conflict and creativity in inter- taking an introduction to computer science course. Nat. Commun. 12, 1922.
organizational teams: the moderating role of shared leadership. IJCMA 28, 74–102. doi:10.1038/s41467-021-22202-3.
doi:10.1108/IJCMA-01-2016-0003. Mu, Y., Cerritos, C., Khan, F., 2018. Neural mechanisms underlying interpersonal coor-
Huang, F., Song, Y., Jiang, Y., Zhao, Q., Luo, J., 2021. Where and how are original and dination: a review of hyperscanning research. Soc. Personal. Psychol. Compass 12.
valuable ideas generated? tDCS of the generation-related posterior temporal lobe and doi:10.1111/spc3.12421.
the executive control-related prefrontal cortex. Cerebral Cortex 261. doi:10.1093/cer- Mueller, J.S., Melwani, S., Goncalo, J.A., 2012. The bias against creativ-
cor/bhab261, bhab.. ity: why people desire but reject creative ideas. Psychol. Sci. 23, 13–17.
Huang, P.-S., Liu, C.-H., Chen, H.-C., 2019. Examining the applicability of representational doi:10.1177/0956797611421018.
change theory for remote associates problem-solving with eye movement evidence. Nguyen, M., Chang, A., Micciche, E., Meshulam, M., Nastase, S.A., Hasson, U., 2021.
Think. Skills Creat. 31, 198–208. doi:10.1016/j.tsc.2018.12.001. Teacher-student neural coupling during teaching and learning. Soc. Cogn. Affect. Neu-
Jiang, J., Chen, C., Dai, B., Shi, G., Ding, G., Liu, L., Lu, C., 2015. Leader emergence rosci. doi:10.1093/scan/nsab103.
through interpersonal neural synchronization. Proc. Natl. Acad. Sci. USA 112, 4274– Nozawa, T., Sasaki, Y., Sakaki, K., Yokoyama, R., Kawashima, R., 2016. Interper-
4279. doi:10.1073/pnas.1422930112. sonal frontopolar neural synchronization in group communication: an exploration
Jiang, J., Dai, B., Peng, D., Zhu, C., Liu, L., Lu, C., 2012. Neural synchronization dur- toward fNIRS hyperscanning of natural interactions. Neuroimage 133, 484–497.
ing face-to-face communication. J. Neurosci. 32, 16064–16069. doi:10.1523/JNEU- doi:10.1016/j.neuroimage.2016.03.059.
ROSCI.2926-12.2012. Paletz, S.B., Schunn, C.D., 2010. A social-cognitive framework of multidisciplinary team
Jiang, J., Zheng, L., Lu, C., 2021. A hierarchical model for interpersonal verbal commu- innovation. Top. Cogn. Sci. 2, 73–95. doi:10.1111/j.1756-8765.2009.01029.x.
nication. Soc. Cogn. Affect. Neurosci. 16, 246–255. doi:10.1093/scan/nsaa151. Pan, Y., Dikker, S., Goldstein, P., Zhu, Y., Yang, C., Hu, Y., 2020. Instructor-learner brain
Kelsen, B.A., Sumich, A., Kasabov, N., Liang, S.H.Y., Wang, G.Y., 2020. What has social coupling discriminates between instructional approaches and predicts learning. Neu-
neuroscience learned from hyperscanning studies of spoken communication? A sys- roimage 211, 116657. doi:10.1016/j.neuroimage.2020.116657.
tematic review. Neurosci. Biobehav. Rev. doi:10.1016/j.neubiorev.2020.09.008. Pan, Y., Guyon, C., Borragán, G., Hu, Y., Peigneux, P., 2021. Interpersonal brain syn-
Kikinis, Z., Fallon, J.H., Niznikiewicz, M., Nestor, P., Davidson, C., Bobrow, L., chronization with instructor compensates for learner’s sleep deprivation in interactive
Pelavin, P.E., Fischl, B., Yendiki, A., McCarley, R.W., Kikinis, R., Kubicki, M., learning. Biochem. Pharmacol. 191, 114111. doi:10.1016/j.bcp.2020.114111.
Shenton, M.E., 2010. Gray matter volume reduction in rostral middle frontal Pan, Y., Novembre, G., Song, B., Li, X., Hu, Y., 2018. Interpersonal synchronization of
gyrus in patients with chronic schizophrenia. Schizophr. Res. 123, 153–159. inferior frontal cortices tracks social interactive learning of a song. Neuroimage 183,
doi:10.1016/j.schres.2010.07.027. 280–290. doi:10.1016/j.neuroimage.2018.08.005.
Klaus, J., Schutter, D.J.L.G., 2018. The role of left dorsolateral pre- Paulus, P.B., Brown, V.R., 2007. Toward more creative and innovative group idea gener-
frontal cortex in language processing. Neuroscience 377, 197–205. ation: a cognitive-social-motivational perspective of brainstorming: cognitive-social-
doi:10.1016/j.neuroscience.2018.03.002. motivational view of brainstorming. Soc. Personal. Psychol. Compass 1, 248–265.
Kleibeuker, S.W., Stevenson, C.E., van der Aar, L., Overgaauw, S., van Duijvenvoorde, A.C., doi:10.1111/j.1751-9004.2007.00006.x.
Crone, E.A., 2017. Training in the adolescent brain: an fMRI training study on diver- Paulus, P.B., Dzindolet, M., Kohn, N.W., 2012. Collaborative creativity—group creativity
gent thinking. Dev. Psychol. 53, 353–365. doi:10.1037/dev0000239. and team innovation. In: Handbook of Organizational Creativity. Elsevier, pp. 327–
Kleinmintz, O.M., Ivancovsky, T., Shamay-Tsoory, S.G., 2019. The two-fold model of cre- 357. doi:10.1016/B978-0-12-374714-3.00014-8.
ativity: the neural underpinnings of the generation and evaluation of creative ideas. Paulus, P.B., Yang, H.-C., 2000. Idea generation in groups: a basis for creativity in organi-
Curr. Opin. Behav. Sci. 27, 131–138. doi:10.1016/j.cobeha.2018.11.004, Creativity. zations. Organ. Behav. Hum. Decis. Process. 82, 76–87. doi:10.1006/obhd.2000.2888.
Kohn, N.W., Paulus, P.B., Choi, Y., 2011. Building on the ideas of others: an ex- Pinho, A.L., Ullén, F., Castelo-Branco, M., Fransson, P., de Manzano, Ö., 2015. Addressing
amination of the idea combination process. J. Exp. Soc. Psychol. 47, 554–561. a paradox: dual strategies for creative performance in introspective and extrospective
doi:10.1016/j.jesp.2011.01.004. networks. Cereb. Cortex 26, 3052–3063. doi:10.1093/cercor/bhv130.
Konvalinka, I., Bauer, M., Stahlhut, C., Hansen, L.K., Roepstorff, A., Frith, C.D., Puccio, G.J., Burnett, C., Acar, S., Yudess, J.A., Holinger, M., Cabra, J.F., 2020. Creative
2014. Frontal alpha oscillations distinguish leaders from followers: mul- problem solving in small groups: the effects of creativity training on idea genera-
tivariate decoding of mutually interacting brains. Neuroimage 94, 79–88. tion, solution creativity, and leadership effectiveness. J. Creat. Behav. 54, 453–471.
doi:10.1016/j.neuroimage.2014.03.003. doi:10.1002/jocb.381.
Li, R., Mayseless, N., Balters, S., Reiss, A.L., 2021. Dynamic inter-brain synchrony in real- Ray, D.K., Romano, N.C., 2013. Creative problem solving in GSS groups: do creative styles
life inter-personal cooperation: a functional near-infrared spectroscopy hyperscanning matter? Group Decis. Negot. 22, 1129–1157. doi:10.1007/s10726-012-9309-3.
study. Neuroimage 238, 118263. doi:10.1016/j.neuroimage.2021.118263. Redcay, E., Schilbach, L., 2019. Using second-person neuroscience to eluci-
Liu, J., Zhang, R., Geng, B., Zhang, T., Yuan, D., Otani, S., Li, X., 2019. Interplay date the mechanisms of social interaction. Nat. Rev. Neurosci. 20, 495–505.
between prior knowledge and communication mode on teaching effectiveness: in- doi:10.1038/s41583-019-0179-4.
terpersonal neural synchronization as a neural marker. Neuroimage 193, 93–102. Ristic, I., Skorc, B., Mandic, T., 2016. Novelty and coherence in group creative processes.
doi:10.1016/j.neuroimage.2019.03.004. Psihologija 49, 213–229. doi:10.2298/psi1603213r.
Liu, L., Zhang, Y., Zhou, Q., Garrett, D.D., Lu, C., Chen, A., Qiu, J., Ding, G., 2020. Roelstraete, B., Rosseel, Y., 2012. Does partial Granger causality really eliminate the in-
Auditory–articulatory neural alignment between listener and speaker during verbal fluence of exogenous inputs and latent variables? J. Neurosci. Methods 206, 73–77.
communication. Cereb. Cortex 30, 942–951. doi:10.1093/cercor/bhz138. doi:10.1016/j.jneumeth.2012.01.010.
Liu, Y., Piazza, E.A., Simony, E., Shewokis, P.A., Onaral, B., Hasson, U., Ayaz, H., Roelstraete, B., Rosseel, Y., 2011. FIAR: an R package for analyzing functional integration
2017. Measuring speaker–listener neural coupling with functional near infrared spec- in the brain. J. Stat. Softw. 44, 1–32. doi:10.18637/jss.v044.i13.
troscopy. Sci. Rep. 7, 43293. doi:10.1038/srep43293. Rosing, K., Bledow, R., Frese, M., Baytalskaya, N., Johnson Lascano, J., Farr, J.L., 2018.
Long, Y., Zheng, L., Zhao, H., Zhou, S., Zhai, Y., Lu, C., 2021. Interpersonal neural syn- The temporal pattern of creativity and implementation in teams. J. Occup. Organ.
chronization during interpersonal touch underlies affiliative pair bonding between Psychol. 91, 798–822. doi:10.1111/joop.12226.
romantic couples. Cereb. Cortex 31, 1647–1659. doi:10.1093/cercor/bhaa316. Runco, M.A., Acar, S., 2012. Divergent thinking as an indicator of creative potential. Null
Lu, K., Hao, N., 2019. When do we fall in neural synchrony with others? Soc. Cogn. Affect. 24, 66–75. doi:10.1080/10400419.2012.652929.
Neurosci. 14, 253–261. doi:10.1093/scan/nsz012. Schippers, M.B., Roebroeck, A., Renken, R., Nanetti, L., Keysers, C., 2010. Mapping the
Lu, K., Qiao, X., Hao, N., 2019a. Praising or keeping silent on partner’s ideas: information flow from one brain to another during gestural communication. Proc.
leading brainstorming in particular ways. Neuropsychologia 124, 19–30. Natl. Acad. Sci. USA 107, 9388–9393. doi:10.1073/pnas.1001791107.
doi:10.1016/j.neuropsychologia.2019.01.004. Selten, R., Warglien, M., 2007. The emergence of simple languages in an ex-
Lu, K., Qiao, X., Yun, Q., Hao, N., 2021. Educational diversity and group cre- perimental coordination game. Proc. Natl Acad. Sci. 104, 7361–7366.
ativity: evidence from fNIRS hyperscanning. Neuroimage 243, 118564. doi:10.1073/pnas.0702077104.
doi:10.1016/j.neuroimage.2021.118564. Senaratne, S., Gunawardane, S., 2015. Application of team role theory
Lu, K., Teng, J., Hao, N., 2020a. Gender of partner affects the interaction pat- to construction design teams. Archit. Eng. Des. Manage. 11, 1–20.
tern during group creative idea generation. Exp. Brain Res. 238, 1157–1168. doi:10.1080/17452007.2013.802980.
doi:10.1007/s00221-020-05799-7. Shamay-Tsoory, S.G., Saporta, N., Marton-Alper, I.Z., Gvirts, H.Z., 2019. Herding Brains:
Lu, K., Xue, H., Nozawa, T., Hao, N., 2019b. Cooperation makes a group be more creative. a core neural mechanism for social alignment. Trends Cogn. Sci. 23, 174–186.
Cereb. Cortex 29, 3457–3470. doi:10.1093/cercor/bhy215. doi:10.1016/j.tics.2019.01.002.
Lu, K., Yu, T., Hao, N., 2020b. Creating while taking turns, the choice Singh, J., Fleming, L., 2010. Lone inventors as sources of breakthroughs: myth or reality?
to unlocking group creative potential. Neuroimage 219, 117025. Manage. Sci. 56, 41–56. doi:10.1287/mnsc.1090.1072.
doi:10.1016/j.neuroimage.2020.117025. Stephens, G.J., Silbert, L.J., Hasson, U., 2010. Speaker-listener neural coupling un-
Mao, A., Mason, W., Suri, S., Watts, D.J., 2016. An experimental study of team size derlies successful communication. Proc. Natl. Acad. Sci. 107, 14425–14430.
doi:10.1073/pnas.1008662107.
12
Tan, A.-G., 2015. Convergent Creativity: From Arthur Cropley (1935-) onwards. Creativity Wass, S.V., Whitehorn, M., Marriott Haresign, I., Phillips, E., Leong, V., 2020. Interper-
Res. J. 27, 271–280. doi:10.1080/10400419.2015.1063892. sonal neural entrainment during early social interaction. Trends Cogn. Sci. 24, 329–
Torrance, E.P., 1966. The torrance tests of creative thinking-norms-technical manual re- 342. doi:10.1016/j.tics.2020.01.006.
search edition-verbal tests. Forms A and B-figural tests, forms A and B. Wei, D., Yang, J., Li, W., Wang, K., Zhang, Q., Qiu, J., 2014. Increased resting functional
Torrence, C., Compo, G.P., 1998. A practical guide to wavelet analysis. Bull. Amer. Meteor. connectivity of the medial prefrontal cortex in creativity by means of cognitive stim-
Soc. 79, 61–78. doi:10.1175/1520-0477(1998)079<0061:APGTWA>2.0.CO;2. ulation. Cortex 51, 92–102. doi:10.1016/j.cortex.2013.09.004.
Tsuzuki, D., Cai, D., Dan, H., Kyutoku, Y., Fujita, A., Watanabe, E., Dan, I., 2012. Stable and Wickham, K.R., Walther, J.B., 2007. Perceived behaviors of emergent and assigned leaders
convenient spatial registration of stand-alone NIRS data through anchor-based prob- in virtual groups. Int. J. e-Collab. 3, 1–17. doi:10.4018/jec.2007010101.
abilistic registration. Neurosci. Res. 72, 163–171. doi:10.1016/j.neures.2011.10.008. Wu, C.L., Zhong, S.Y., Chen, H.C., 2016. Discriminating the difference between remote
Ulrich, F., 2018. Exploring divergent and convergent production in idea evalua- and close association with relation to white-matter structural connectivity. PLoS One
tion: implications for designing group creativity support systems. CAIS 101–132. 11. doi:10.1371/journal.pone.0165053.
doi:10.17705/1CAIS.04306. Xue, H., Lu, K., Hao, N., 2018. Cooperation makes two less-creative in-
Valencia, A.L., Froese, T., 2020. What binds us? Inter-brain neural synchronization and its dividuals turn into a highly-creative pair. Neuroimage 172, 527–537.
implications for theories of human consciousness. Neurosci. Conscious 2020, niaa010. doi:10.1016/j.neuroimage.2018.02.007.
doi:10.1093/nc/niaa010. Youssofzadeh, V., Prasad, G., Naeem, M., Wong-Lin, K., 2016. Temporal information of
van Baar, J.M., Halpern, D.J., FeldmanHall, O., 2021. Intolerance of uncertainty modulates directed causal connectivity in multi-trial ERP data using partial granger causality.
brain-to-brain synchrony during politically polarized perception. Proc. Natl. Acad. Sci. Neuroinform 14, 99–120. doi:10.1007/s12021-015-9281-6.
USA 118, e2022491118. doi:10.1073/pnas.2022491118. Zheng, L., Chen, C., Liu, W., Long, Y., Zhao, H., Bai, X., Zhang, Z., Han, Z., Liu, L., Guo, T.,
Van Vugt, M., 2006. Evolutionary origins of leadership and followership. Pers. Soc. Psy- Chen, B., Ding, G., Lu, C., 2018. Enhancement of teaching outcome through neu-
chol. Rev. 10, 354–371. doi:10.1207/s15327957pspr1004_5. ral prediction of the students’ knowledge state. Hum. Brain. Mapp. 39, 3046–3057.
Wallot, S., Mitkidis, P., McGraw, J.J., Roepstorff, A., 2016. Beyond synchrony: joint ac- doi:10.1002/hbm.24059.
tion in a complex production task reveals beneficial effects of decreased interpersonal Zheng, L., Liu, W., Long, Y., Zhai, Y., Zhao, H., Bai, X., Zhou, S., Li, K., Zhang, H., Liu, L.,
synchrony. PLoS One 11, e0168306. doi:10.1371/journal.pone.0168306. Guo, T., Ding, G., Lu, C., 2020. Affiliative bonding between teachers and students
Wang, M.-Y., Luan, P., Zhang, J., Xiang, Y.-T., Niu, H., Yuan, Z., 2018. Concur- through interpersonal synchronisation in brain activity. Soc. Cogn. Affect. Neurosci.
rent mapping of brain activation from multiple subjects during social interac- 15, 97–109. doi:10.1093/scan/nsaa016.
tion by hyperscanning: a mini-review. Quant. Imaging Med. Surg. 8, 819–837. Zhu, Y., Leong, V., Pan, Y., Hou, Y., Zhang, D., Hu, Y., 2021. Instructor-learner neu-
doi:10.21037/qims.2018.09.07. ral synchronization during elaborated feedback predicts learning transfer. bioRxiv.
Wang, P., Zhu, W., 2011. Mediating role of creative identity in the influence of transfor- doi:10.1101/2021.02.28.433286.
mational leadership on creativity: is there a multilevel effect? J. Leadership Organiz. Zhu, Y., Pan, Y., Hu, Y., 2019. Learning desire is predicted by similar neural processing of
Stud. 18, 25–39. doi:10.1177/1548051810368549. naturalistic educational materials. eNeuro 6. doi:10.1523/ENEURO.0083-19.2019.
13

Increased or Decreased Interpersonal Neural Synchronization in - 2022 - NeuroIm

Uploaded by

Copyright:

Available Formats

Increased or Decreased Interpersonal Neural Synchronization in - 2022 - NeuroIm

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Increased or Decreased Interpersonal Neural Synchronization in - 2022 - NeuroIm

Uploaded by

Copyright:

Available Formats

NeuroImage 260 (2022) 119448

Contents lists available at ScienceDirect

Increased or decreased? Interpersonal neural synchronization in group

1. Introduction 2016; Senaratne and Gunawardane, 2015). Therefore, understanding

Fig. 1. Experiment procedure. (A) Umbrella prototype.

CH Combinations(leader-follower) Leader Follower

3. Result 4 signiﬁcant CH combinations from the 1000 permutations are shown

Research on the emergence of leadership has found that the higher

You might also like