Abstract

Introduction

Humans are keen consumers of creativity. From cavemen paintings and hunter-gatherer toolmaking to contemporary arts and technology, the tendency to create new ideas and products out of otherwise disconnected elements is deeply rooted in human cognition and has contributed to immeasurable human accomplishments¹. Whereas creativity is often seen as an individual capacity, scholars are increasingly acknowledging that it is also a social phenomenon that defines various societies². Recent studies examining cross-cultural differences in creativity indicate that societal factors have a significant impact on the formation of new ideas^3,4. Creative endeavors, among them dance, music, brainstorming in industries, and scientific collaborations, take place primarily in social settings⁵. Even when individuals engage in creative activities on their own, their work is often overtly evaluated by peers⁶ or covertly assessed by internalized cultural norms⁷. Yet despite the significant role played by social factors in regulating creativity, most research in this field has focused predominantly on isolated individuals, without considering potential group dynamics⁸. While emerging studies have focused on creativity in dyads^9–15,the question of what neural mechanisms form the basis of group creativity remains unresolved.

Given the influence of “esprit de corps” (group mind)¹⁶ on individuals within groups, here we adopt an approach that prioritizes the group as the primary unit of analysis. Accordingly, we propose two types of group dynamics that contribute to creativity during group interactions: (1) individuals’ inherent tendency to ‘herd’ and act as a coherent social unit, and (2) individuals’ inclination to act independently¹⁷. Herding involves alignment at various levels and includes movement, emotional, and cognitive aspects¹⁸. While herding can influence group creativity by enhancing group cooperation, it may also lead individuals to imitate others rather than be original¹⁹. Therefore, a mindset of herding would be characterized by high number of similar or imitative responses among group members and lower creative outcomes. On the other hand, independent thinking requires flexibility, exploration of alternative ideas, and the ability to consider diverse possibilities²⁰. Flexibility, which is characterized by being open to different ideas and willing to explore novel approaches, is a fundamental capability that underlies creativity^21–23. Studies on group creativity show that that generating creative ideas in a group context requires that members openly share their ideas^24–26.

Based on the approach outlined above, we hypothesized the existence of two distinct neural networks—one supporting a herding mindset and the other supporting a flexible group mindset. We further hypothesized that these networks make differential contributions to group creativity (Hypothesis 1). A flexible group mindset enables group members to be receptive to diverse ideas and express their individual perspectives, whereas a herding mindset enables them to adjust their beliefs, attitudes, and behaviors to align with those of others. Consequently, we propose that a higher level of group creativity is associated with a greater degree of flexibility relative to herding (Hypothesis 2). That is, we hypothesize that the relation between the group mindsets of flexibility and herding promotes group creativity.

A key network that may support a flexible mindset during group creativity is the executive control network, which is involved in set shifting, decision-making and working memory²⁷. Neuroimaging studies on creativity have repeatedly found that while Default Mode Network (DMN) that includes regions such as the medial prefrontal cortex (mPFC), posterior cingulate cortex (PCC), and temporo-parietal junction (TPJ), mediate the generation phase of creative thinking^28–32, the Executive Control Network (ECN) is thought to play a critical role in the cognitive processes of evaluating, shifting and selecting ideas^4,33. The dorsolateral prefrontal cortex (DLPFC), a core region in the ECN, was found to be activated in studies on creative story generation³⁴, and divergent thinking³⁵. While it was suggested that inhibiting the DLPFC may improve creativity³⁶, different studies have suggested that increased activity in this region is related improvisational music playing³⁷ and insightful problem-solving^38,39. Recent research utilizing functional near-infrared spectroscopy (fNIRS) in dyads engaged in dyadic creativity tasks revealed increased interbrain coupling, specifically in the right DLPFC⁴⁰. Additionally, in comparing group creativity among individuals with different levels of creativity, Xue et al., found that dyads composed of two individuals who were less creative exhibited better cooperation and showed more interbrain coupling in both the right DLPFC and the right TPJ than did dyads containing two highly creative individuals⁴¹.Coupled neural activity in time series showing DLPFC brain signals among interacting individuals may represent a coordinated mindset of flexibility and openness. In contrast, interbrain coupling in the observation-execution network may indicate that the group has a herding mindset. Indeed, a recent study showed that interbrain coupling in the inferior frontal gyrus (IFG) is negatively correlated with originality¹⁴. It was suggested that the IFG possibly assesses ideas within the context of social norms⁴. Considering that the IFG is the core region within the mirror neuron system^42,43 and that interbrain coupling in this region was found during tasks requiring synchronized behavior^44,45, it is possible that interbrain coupling in the IFG is essential for coordination among group members during the creative process. Yet because this region is responsible for imitation and alignment⁴⁶, inter-brain coupling should be mitigated to achieve original and creative outcomes. Previous research found that lesions in the IFG are associated with increased creativity⁴⁷ and that cathodal (inhibitory) direct current stimulation (tDCS) targeting this region enhanced creativity^30,48. Moreover, while a recent study did not find that cathodal tDCS of the prefrontal cortex increased the novelty of participants’ responses³⁶, a previous study demonstrated that cathodal tDCS over the left IFG enhanced fluency in a divergent-thinking task that required generating uncommon ideas⁴⁹. On the other hand, it has been demonstrated that anodal tDCS over the left DLPFC facilitates performance in convergent thinking tasks that require creative problem-solving^50,51. Hence it is possible that interbrain coupling in the IFG signals a herding mindset in groups and may diminish levels of group creativity.

Building on studies reporting that groups of four mammals exhibit complex movement patterns that reflect social interaction dynamics of groups⁵², the present study aimed to investigate the neural mechanisms underlying creativity within groups of four participants. To evaluate group creativity, we employed two divergent thinking tasks that allow assessing group creativity during discussions (see Methods). We randomly divided 88 participants into groups of four. Each group was tasked with holding discussions to generate innovative ideas and solutions for predefined challenges (Fig. 1). The first task involved brainstorming alternative uses for a common object (e.g., a can), while the second task required devising original strategies to prevent an egg from breaking when dropped from a building. Prior to the discussion phase, we assessed each participant’s individual creativity using established metrics (Torrance) to create a baseline for subsequent comparative analysis. We focused on O2Hb (oxygenated hemoglobin) and posited that engaging in creativity-focused discussions would elicit interbrain coupling in neural structures associated with both herding and flexibility processing. Furthermore, we anticipated that coupling in the DLPFC would be higher than coupling in the IFG when creative output is greater. We predicted that interbrain coupling in the IFG will be associated with a herding mindset characterized by high imitation between group members. Moreover, we predicted that interbrain coupling in the DLPFC would be positively correlated with group creativity, whereas interbrain coupling in the IFG would be negatively correlated with group creativity. Additionally, we predicted that the ratio between the DLPFC and the IFG would predict levels of group creativity.

Open in a separate window

Fig. 1

Experimental design.

Participants were scanned with functional near-infrared spectroscopy (fNIRS) during a group divergent thinking task. The paradigm involved collaborative discussions in which groups of four participants collectively generated original uses for common objects and collaboratively devised solutions to complex challenges. Individuals signed an informed consent for publication of this image. N=22 groups.

Consistent with our hypotheses, we observed increased interbrain coupling in both the DLPFC and IFG during brainstorming sessions. Specifically, interbrain coupling in the left DLPFC was significantly higher compared to the left IFG and right IFG. Additionally, our analysis of factors predicting elevated group creativity levels revealed a positive correlation between left DLPFC interbrain coupling and group creativity outcomes, whereas right IFG interbrain coupling was negatively correlated with creativity. These findings align with the concept that low interbrain coupling in the IFG and high interbrain coupling in the DLPFC underlie the dual demands of creative cognition in groups: the propensity to herd with the group and the tendency to generate novel ideas.

Results

Analysis of interbrain coupling

To assess interbrain coupling between individuals in each group, we applied Wavelet Transform Coherence (WTC) to the two neural signal time-series generated from the two regions, implemented using the wavelet coherence package in MATLAB⁵³. These values ranged from 0 (no coherence) to 1 (complete coherence). For each ROI in a group of four participants (A, B, C, D), there were six pairs (AB, AC, AD, BC, BD, CD). We computed the WTC for each pair. To derive a single measure of interbrain coupling for the group for each ROI, we then averaged the WTC values across all pairs. To test whether interbrain coupling in the sample was greater than chance, we created 22 pseudo-groups by randomly pairing fNIRS recordings of four participants, each from a different group. We calculated interbrain coupling of each pseudo-group in a manner that resembled the analysis of the real group.

ANOVA comparison of interbrain coupling between the real and pseudo groups revealed a significant effect for group (F (1,42)=23.22, p<0.001, partial η2=0.36). The interbrain coupling in the real groups (M=0.32, SD=0.01) was significantly higher than in the pseudo groups (M=0.30, SD=0.01). There was also a significant effect for brain area (F (1,42)=16.55, p<0.001, partial η2=.28). Interbrain coupling in the DLPFC (M=0.32, SD=0.02) was significantly higher than interbrain coupling in the IFG (M=0.31, SD=0.01). Lastly, the analysis revealed a significant three-way interaction: group type (real, pseudo) by brain area (IFG, DLPFC) by hemisphere (left, right) interaction (F (1,42)=4.27, p<0.05, partial η2=0.09).

To examine the source of the interaction, we first explored between-group differences in the various brain regions using one-way multivariate analysis of variance (MANOVA), with group type as the between-subjects independent variable and interbrain coupling in R.IFG, L.IFG, R.DLPFC, and L.DLPFC as the dependent variables. The results revealed a significant difference between real groups and pseudo groups in the R.IFG (F (1,42)=7.65, p<0.01, partial η2=0.15), L.IFG (F (1,42)=15.90, p<0.001, partial η2=0.24), R.DLPFC (F (1,42)=6.95, p<0.05, partial η2=0.14), and L.DLPFC (F (1,42)=18.20, p<0.001, partial η2=0.30). Interbrain coupling for the real groups was significantly higher than interbrain coupling for the pseudo groups in all regions (see Fig. 2).

Open in a separate window

Fig. 2

Group differences in interbrain coupling in the different regions.

Interbrain coupling for the real groups was significantly higher than interbrain coupling for the pseudo groups in all regions. Asterisks indicate significant difference. N=22 groups.

Next, we explored differences between the brain regions separately for each group, using two repeated-measures ANOVAs, with the four brain regions as the between-subjects independent variable and interbrain coupling as the dependent variable. The results revealed a significant effect only for the real groups (F (3,19)=5.77, p<0.01, partial η2=0.48) but not for pseudo groups (F (3,19)=2.86, p=0.064, partial η2=.31). Post hoc comparisons with Bonferroni correction showed a significant difference between the L.DLPFC and L.IFG (p<0.05), and between the L.DLPFC and R.IFG (p<0.01) only for real groups. Interbrain coupling in the L.DLPFC was significantly higher than interbrain coupling in the L.IFG and R.IFG (see Fig. 3).

Open in a separate window

Fig. 3

Interbrain coupling in the right and left IFG and DLPFC within real groups.

The color intensity is proportional to the interbrain coupling. Red represents lower interbrain coupling. The legend color displays the interbrain coupling value and the corresponding colors. Interbrain coupling in the L.DLPFC was significantly higher than interbrain coupling in the L.IFG and R.IFG. N=22 groups.

Brain and behavior correlation

We thought to initially examine whether a herding mindset is predicted by interbrain coupling in the IFG. We calculated the number of imitative responses and used stepwise linear regression to identify the specific brain regions that best explain the variance in herding, with herding as the dependent variable and interbrain coupling in the brain regions as the independent variables. The analysis involved iteratively adding variables to the regression equation based on their statistical significance and contribution to the model’s goodness of fit. The results revealed that only the R.IFG (β=0.55, t=2.93, p<0.01) significantly contributed to the model’s goodness of fit. The final regression model including R.IFG was significant (F(1, 20)=8.61, p<0.01), with adjusted R2=0.27. These results suggest that higher levels of interbrain coupling in R.IFG (positive correlation) predict higher levels of herding (see Fig. 4).

Open in a separate window

Fig. 4

Herding mindset is predicted by interbrain coupling in the R.IFG.

Higher levels of interbrain coupling in R.IFG predict higher levels of herding. N=22 groups.

To further identify the set of the brain regions that best explain the variance in creativity, we employed stepwise linear regression, with creativity as the dependent variable and interbrain coupling in the brain regions as the independent variables. The analysis involved iteratively adding variables to the regression equation based on their statistical significance and contribution to the model’s goodness of fit. The results revealed that L.DLPFC (β=0.73, t=3.84, p<0.01), R.IFG (β=−0.42, t=−2.64, p<0.05), and L.IFG (β=−0.43, t=−2.24, p<0.05) significantly contributed to the model’s goodness of fit. The final regression model including L.DLPFC, R.IFG, and L.IFG was significant (F(3, 18)=7.34, p<0.01), with adjusted R2=0.48. These results suggest that higher levels of interbrain coupling in L.DLPFC (positive correlation) and lower levels of interbrain coupling in R.IFG and L.IFG (negative correlation) predict higher levels of creativity (see Fig. 5).

Open in a separate window

Fig. 5

The brain regions that predict the variance in creativity.

Higher levels of interbrain coupling in L.DLPFC and lower levels of interbrain coupling in R.IFG predict higher levels of creativity. N=22 groups.

Next, we examined our hypothesis that higher creativity / lower herding within a group would be associated with higher / lower interbrain coupling of the R/L.DLPFC relative to the interbrain coupling of the R.IFG and L.IFG.

To identify which specific ratio explains the variance in herding, we employed stepwise linear regression, with herding as the dependent variable and the ratios as the independent variables. The analysis involved iteratively adding variables to the regression equation based on their statistical significance and contribution to the model’s goodness of fit. The results revealed that only the L.DLPFC/R.IFG ratio (β=−0.50, t=−2.59, p<0.05) significantly contributed to the model’s goodness of fit. The final regression model including the L.DLPFC/R.IFG ratio was significant (F(1, 20)=6.73, p<0.05), with adjusted R2=0.21. The results suggest that higher levels of balance between L.DLPFC and R.IFG predict lower levels of herding (see Fig. 6).

Open in a separate window

Fig. 6

The variance in herding is explained by L.DLPFC/R.IFG ratio.

Higher levels of balance between L.DLPFC and R.IFG predict lower levels of herding. N=22 groups.

To identify which ratio explains the variance in creativity, we employed stepwise linear regression, with creativity as the dependent variable and the ratios as the independent variables. The analysis involved iteratively adding variables to the regression equation based on their statistical significance and contribution to the model’s goodness of fit. The results revealed that only the L.DLPFC/R.IFG ratio (β=0.63, t=3.65, p<0.01) significantly contributed to the model’s goodness of fit. The final regression model including the L.DLPFC/R.IFG ratio was significant (F(1, 20)=13.32, p<0.01), with adjusted R2=0.37. The results suggest that higher levels of balance between L.DLPFC and R.IFG predict higher levels of creativity (see Fig. 7).

Open in a separate window

Fig. 7

The variance in creativity is explained by L.DLPFC/R.IFG ratio.

Higher levels of balance between L.DLPFC and R.IFG predict higher levels of creativity. N=22 groups.

Next, we examined whether the mean individual creativity in each group explains the variance of group creativity using Linear regression model. The regression model, with group creativity as the dependent variable and the mean individual creativity in each group as the independent variable revealed no significant effect on group creativity (F(1, 20)=0.15, p=0.704).

Lastly, we examined whether interbrain coupling explain the variance of individual creativity using Linear mixed (LME) model. The LME model, with individual creativity as the dependent variable, interbrain coupling in the brain regions as the independent variables, and group ID as random effect, also showed no significant effect of L.DLPFC (F(1, 17)=0.33, p=0.576), R.DLPFC (F(1, 17)=0.00, p=0.978), R.IFG (F(1, 17)=0.34, p=0.568), and L.IFG (F(1, 17)=0.13, p=0.725) on individual creativity.

Discussion

In the current study we aimed to examine interbrain coupling in areas associated with a group mindset of flexibility and herding and to investigate how interbrain coupling contributes to group creativity. The neuroimaging findings revealed an increase in inter-brain coupling among pairs across regions. This phenomenon was identified by contrasting the interbrain coupling in real vs. pseudo dyads where interaction was lacking. This approach allowed ruling out the possibility interbrain coupling merely reflects general brain activity stemming from task engagement, irrespective of actual interactive participation. Interbrain coupling in the left and right IFG in the creativity condition for the real groups compared to the pseudo groups is particularly interesting in light of the IFG’s established association with synchronization^54,55 and cooperation¹⁴. Neuroimaging studies of imitation⁵⁶ and emotional empathy⁵⁷ have consistently exhibited activation in the IFG. Although studies of synchronization have largely focused on physical movement synchronization^45,58, Shamay-Tsoory et al.¹⁸ suggested that emotion-based synchronization (e.g., emotional contagion) may be conceptually similar to physical synchronization. The recruitment of the IFG during group discussions gives some credence to the notion that interbrain coupling in the IFG supports different types of synchronization.

If interbrain coupling in the IFG reflects the group members’ herding mindset during group discussions, we might expect interbrain coupling in this region to correlate negatively with the subsequent creative outputs. Indeed, at the group level, those groups with stronger interbrain coupling in the IFG had lower creativity scores. Our findings are in line with a previous study conducted by Mayseless et al.¹⁴, which also reported a negative correlation between interbrain coupling in the IFG and creativity scores. Similarly, Gelfand et al.⁵⁹ found that groups instructed to engage in synchronized walking demonstrated lower levels of creativity than did unsynchronized groups. Therefore, our results support the notion that synchronized behavior may have a detrimental effect on group creativity. Notably, the involvement of the right IFG is consistent with its recognized roles in imitation, coordination, intention coding, and perception-action matching⁶⁰. Distinctly, interbrain coupling in the left IFG has been associated with verbal synchronization during verbal tasks⁶¹. whereas the right IFG has been linked to non-verbal coordination⁶². This suggests that elevated interbrain coupling in the right IFG may predominantly enhance non-verbal imitation which could detrimentally impact group creativity.

In contrast to the IFG, the DLPFC has been repeatedly linked to cognitive processes such as flexibility and executive control^63,64. Thus, the interbrain coupling in the DLPFC observed here during the creativity discussions may be related to a flexible mindset of the group, allowing multiple divergent ideas to emerge. Higher cognitive demands may be placed on participants in order for them to overcome the strong tendency to imitate each other. The positive correlation we found between creativity and interbrain coupling in the left DLPFC supports this interpretation, suggesting that interbrain coupling in the left DLPFC is associated with facilitating creative thinking by enabling cognitive flexibility and overcoming imitative tendencies. The lack of correlation between creativity and interbrain coupling in the right DLPFC may be attributed to the specific nature of the creativity task used in this study. The study employed the AUT, which is a verbal divergent thinking task. A meta-analysis showed that the left DLPFC is associated with verbal divergent thinking, while the right DLPFC is linked to visual divergent thinking⁶⁵.

Interestingly, our finding suggest that interbrain coupling is not linked to individual creativity. This suggest that during group interaction the group mindset contributes more to collective outcomes than individual creativity. In exploring the neural correlates of creativity, both individual and group dynamics offer distinct insights. The DMN, known to mediate associative thinking³³ is crucial for generating potential ideas from long-term memory. Concurrently, the ECN facilitates cognitive processes involving the transition between and selection of these creative thoughts⁴. During individual creativity, the DLPFC is particularly critical, promoting cognitive flexibility necessary for selection and transition between ideas. In group settings, DLPFC activity, when coupled across individuals, facilitates a synchronized, flexible mindset through interbrain coupling, enhancing collective creativity. Meanwhile, the Observation-Execution system, prominently involving the IFG, is crucial for evaluating ideas and aligning ideas with internalized norms during individual creativity and externally presented group norms during group creativity. However, while interbrain coupling in the IFG supports social coherence, it may also suppress novelty, leading to mimicry rather than innovation. Thus, we hold that optimal group creativity is achieved by balancing interbrain connectivity: enhanced DLPFC coupling fosters flexible idea navigation, whereas diminished IFG interbrain coupling reduces conformity, promoting originality and diversity in creative outputs.

Motivated by the hypothesis that interbrain coupling in the DLPFC may be competing with the IFG in creativity, we examined the relationship between interbrain coupling in the IFG and the DLPFC during creativity discussions by measuring the ratio between the left DLPFC and the right IFG. We found that the ratio between interbrain coupling in the right IFG and the left DLPFC was the most significant predictor of creativity performance. This suggests that the creative outcomes of groups may reflect the relative coupling of these regions, with greater DLPFC coupling biased toward high creativity and greater interbrain coupling in the IFG biased toward low creativity. These findings are highly relevant to understanding how group dynamics affect creativity during group interactions. A related study by Choi et al.⁶⁶ reported that groups marked by collective value orientation and independent self-representation generate more original ideas than groups that are manipulated to be collective but interdependent. Therefore, creating an environment that encourages both collective value orientation and independent thinking with greater emphasis on independent thinking may be a promising strategy to boost group creativity, as suggested by our results showing a balance between interbrain coupling in the IFG and the DLPFC.

There are several limitations that need to be acknowledged in this study. First, the majority of participants in our sample were female. Previous research has pointed to potential gender differences in creativity performance⁶⁷ as well as in neuroimaging investigations of creative thinking⁶⁸. It is important for future research to explore our brain model of group creativity across diverse samples, including a wider range of gender representation, in order to validate and extend the findings to different populations. Finally, note that the duration of the task in our study was relatively short, lasting eight minutes. Exploring the impact of different task durations on group dynamics and creativity could provide valuable insights into the temporal aspects of group creative processes.

Our results are in line with the notion that the IFG and DLPFC represent the twin demands of creative cognition in groups: the tendency to herd with the group and the cognitive goal of generating novel ideas. We believe that these findings, and those of related future studies, will provide a more detailed characterization of specific social elements, their neural substrates, and the social circumstances that affect group creativity. Therefore, not only do our results provide direct empirical support for models of creativity that acknowledge the influence of social interactions, they also serve as a first step toward understanding the neural underpinnings of group dynamics.

Materials and methods

Measures

Measurement of creativity

Group creativity

Two divergent creative tasks were administered to assess group creativity.

1. The Alternate Uses Task (AUT⁶⁹): The experimenter named six common objects (pencil, key, button, tire, can, and shoe) to the groups, one at a time, together with the most common use for each object. They were also given an example of an alternate use for an object (making holes in a bottle of water and transforming it into a sprinkler). The groups were then given 90 seconds⁷⁰ to collectively generate alternative uses for each object. The participants were explicitly instructed to suggest alternative uses that had not been previously considered. To prevent an order effect (response change due to fatigue, practice, carry over effect or other factors⁷¹) two versions of this task were created entailing the same six objects presented in different order. Each version was administered to half of the groups.

2. Egg task⁷²: The groups were given two minutes to collectively think of original ways to prevent an egg from breaking after falling from a ten-meter-high building. To make sure the participants understood the task, they were given the example of taming an eagle to catch the egg when it falls.

In both creativity tasks, the participants were informed that in the realm of appropriate and possible suggestions there were no correct or incorrect answers, and they were encouraged to generate as many responses as possible. Additionally, to ensure the active participation of all participants in the tasks, each round began with each individual offering a response. After this initial contribution, the brainstorming session was open for free collaboration among the group members. Participants were explicitly instructed to disregard the presence of the experimenter and to engage in group brainstorming while generating alternative uses (see Fig. 8).

Open in a separate window

Fig. 8

Creativity tasks.

The creativity tasks were counterbalanced between groups. In the AUT, objects were counterbalanced between groups. N=22 groups.

Scoring: The verbal responses of participants within each group were captured via audio recording and subsequently transcribed verbatim to facilitate the coding process. The transcription was undertaken to precisely attribute each response to its respective participant within the group dynamics.

Creativity scores were assessed by expert raters according to two dimensions: flexibility (number of categories) and originality (unique responses). Originality was scored by considering statistically infrequent responses. The raters underwent training that involved familiarization with the assessment criteria and calibration exercises to align their ratings. Frequency analysis was conducted for each response across all participants. Responses given by more than 10% of the participants were assigned 0 points, responses given by 5% to 10% were assigned 1 point, and responses provided by less than 5% were assigned 2 points. The originality score was then calculated as the cumulative number of points across all responses. The originality and flexibility scores were calculated for each task. The total creativity score for each task was computed as the mean of the originality and flexibility scores. The final creativity score used for each group in the analysis was the sum of the total creativity scores on both tasks. This decision was influenced by our prior identification of a significant correlation between these tasks, both of which involve divergent thinking⁷³.

Table 1 shows the means (and standard deviations) of the originality, flexibility, and total creativity scores of the AUT and EGG tasks for the groups.

Table 1

Means (and standard deviations) of the creativity measures

Variables	AUT	EGG	Total
Originality	15.91 (7.41)	0.45 (0.8)	16.36 (7.74)
Flexibility	34.23 (7.62)	3.95 (1.17)	38.18 (7.76)
Total creativity	25.07 (6.17)	2.20 (0.73)	27.27 (6.38)

Open in a separate window

Note. The table displays means and standard deviations (in parentheses) for the creative tasks and a total creative score of the groups. N=22 groups.

fNIRS data acquisition

We measured interbrain coupling between the brains of the four members of each group (group) using the Brite24 Functional near-infrared spectroscopy (fNIRS) system (Artinis Medical Systems; Elst, The Netherlands). This system is used to simultaneously measure changes in oxygenated hemoglobin (O2Hb) and deoxygenated hemoglobin (HHb) concentrations in the cortex. Both O2Hb and HHb signals can be employed to measure changes in cerebral blood flow. Here we focused mainly on the O2Hb signal, since research has shown it to be more sensitive to changes in cerebral blood flow^12,61,76. The system employs near-infrared light transmission at two wavelengths, 760 and 850nm.

Measurements were taken at a 50Hz rate. The system consists of a flexible probe unit (cap) with 18 optodes:10 transmitters and 8 receivers. Each pairing of a transmitter and receiver forms a channel, resulting in 12 channels in each hemisphere. Optode location was chosen using the international EEG 10–20 system, partially covering the bilateral prefrontal cortex (see Fig. 9a). Channels were later clustered into regions of interest (ROIs) of each transmitter with all its receivers, based on the estimated Broadmann’s area over which they were placed. The bilateral ROIs included the Premotor Cortex (PMC), the opercular part of Inferior Frontal Gyrus (IFG’s pars opercularis; BA44), the triangular part of the IFG (pars triangularis; BA45), the Dorsolateral Prefrontal Cortex’s (DLPFC) Broadmann’s area 46, and the DLPFC’s Broadmann’s area 9 (see Fig. 9b).

Open in a separate window

Fig. 9

fNIRS Channels configuration.

a Optode montage relative to the EEG 10/20 system. T represent a transmitter and R represent a receiver. T 2 and 3 were considered the left IFG; T 9 and 8 were considered the right IFG. T 6 was considered the left DLPFC; T 1 was considered the right DLPFC. b Illustration of the Brite24 fNIRS system channels montage. Red represents sources and blue represents detectors; together they form 24 channels. Black lines are the channels, optode projections from scalp to cortex. N=22 groups.

Acknowledgements

Author contributions

Conceptualization: S.S.T., N.F., H.P. Methodology: S.S.T., N.F., H.P. Investigation: D.Z., N.F., H.P. Visualization: N.F., H.P. Supervision: S.S.T. Writing—original draft: S.S.T., N.F., H.P. Writing—review & editing: S.S.T., N.F., H.P.

Peer review

Data availability

The datasets analyzed during this study is available at 10.5281/zenodo.12200031 (https://zenodo.org/records/12200032). The source data behind the graphs and table in the paper is available in Supplementary Data 1. Other data will be available from the corresponding author on reasonable request.

Code availability

The codes generated during this study will be available from the corresponding author upon request.

Competing interests

The authors declare no competing interests.

Footnotes

References