Resting state default mode network is associated with wise advising

Default mode network (DMN) may be associated with wisdom (i.e., mature understanding of life featured by perspectival metacognition) when advising from a self-referential perspective due to the involvement of the DMN in reflecting on personal life experiences. After a resting-state functional MRI scan, 52 adults advised some youths going through life dilemmas, half from a second-person perspective and half from a third. After advising each youth, participants indicated the psychological distance they felt between themselves and the youth. The amplitude of low-frequency fluctuation (ALFF) was measured in the DMN during resting states. Moreover, trained raters rated the participants' advice on wisdom criteria (i.e., metacognitive humility (MH), meta-level flexibility, and perspective-taking). The results showed that participants felt a significantly smaller psychological distance from the youth when advising from the second- than the third-person perspective. Moreover, only when advising from the second-person perspective was MH associated with ALFF in regions within the DMN (i.e., right rostral anterior cingulate cortex (ACC) and left dorsomedial prefrontal cortex). The right rostral ACC showed a significantly greater association with MH from the second- than the third-person perspective. Therefore, resting-state DMN activities may be important for self-involved wisdom performance (e.g., giving advice directly to others).

Table 1.Wisdom scores and psychological distance when advising from different perspectives and the corresponding correlation and paired-sample t-test results.N = 52; 2nd: second-person perspective; 3rd: thirdperson perspective; CI: confidence interval for the difference between the means.*, p < 0.05, **, p < 0.01.The p values are for the differences between different perspectives, adjusted by the False Discovery Rate method; N = 52.Figure 1.Significant correlation between resting-state ALFF within the DMN and metacognitive humility when advising from the second-person perspective.Specifically, the ocean blue mask is the regions of the dorsomedial prefrontal cortex (BA 9, 10, 24, 32), the turquoise regions located in the temporal gyrus represent the lateral temporal cortex (BA 21), the yellowish-red mask aside with node "rACC" indicates the ventral medial prefrontal cortex (BA 10, 24, 32).In the central of this image, the brain mask with light red situated under node "PFC" and node "rACC" is the inferior parietal lobule (BA 39, 40); the blue-green brain mask located between two hemispheres is the posterior cingulate cortex (BA significantly stronger than its counterpart when advising from a different perspective (see Table 2 and Fig. 2 for details).

Discussion
The present study supports our hypothesis that resting-state DMN activity is associated with wise advising from the second-person perspective.Moreover, the results revealed that wise advising from the second-person perspective was associated with distinct brain regions from the third-person perspective.Specifically, resting-state activity in the dorsomedial PFC was associated with MH from the second-person perspective (r = 0.555).The dorsomedial PFC is responsible for metacognition during decision-making 36,37 and metacognitive representations of social information 38 .Our results replicate findings from the previous EEG study-demonstrating that resting-state neural oscillations in the frontal lobe are associated with MH from the second-but not the thirdperson perspective 10 .Additionally, one study showed that reflecting on negative personal experiences from the first-person perspective was associated with greater activity in the medial PFC-an important region of the DMN-compared to the third-person perspective 39 .Moreover, the right rostral ACC showed a significantly greater association with MH from the second-(r = 0.574) than the third-person perspective (r = 0.231).The DMN activity may be related to self-referential processing functions involved in wisdom (i.e., metacognitive humility).DMN may be related to greater self-awareness of one's intellectual limitations, especially from a more self-related perspective (i.e., the second-person perspective).Additionally, wise advising from each perspective is associated with resting-state activation in brain regions neighboring the DMN, such as the precuneus and calcarine.These regions were involved in directing attention to real or simulated movement in space 40 .Our results dovetail with previous findings that wisdom is associated with the cuneus and lingual gyrus 18 .Wise individuals may have robust brain systems supporting visual imagery, simulated movements between viewpoints, and greater awareness of the self and others.Furthermore, we demonstrated that the precuneus and calcarine showed a significantly greater association with PT from the second-than the third-person perspective.The resting-state activity in the precuneus and calcarine may be more related to self-related wisdom performance (e.g., imagining oneself problem-solving in social scenarios one has similarly experienced).
We found an association between MH and ACC (rs > 0.57).The ACC is responsible for controlling impulsivity (e.g., premeditation before decision-making) 41 , and detecting errors and conflicts 42 .The ACC may support MH by suppressing one's impulsive judgments and reflecting on mistakes.Wisdom experts have consistently proposed that the ACC is responsible for dealing with uncertainty 34 , an essential feature of MH.Additionally, we found that the rostral ACC was related to MH when advising from the second-person perspective, while the dorsal ACC was related to MH when advising from the third-person perspective.In contrast to the dorsal ACC, the rostral ACC is believed to be associated with more self-related processes such as reward-directed decisionmaking 43,44 .Taken together, the dorsal and rostral ACC may have distinct functions in wisdom performance from different psychological distances.
Overall, the present study has demonstrated a relationship between resting-state brain activity and wise advising from distinct perspectives.Specifically, brain regions in the DMN involved in self-referential processing (i.e., dorsomedial PFC, rostral ACC) were associated with metacognitive humility when advising from the second-person perspective.In contrast, when advising from a third-person perspective, metacognitive humility was associated with the dorsal ACC.Similarly, the precuneus and calcarine showed a significantly greater correlation with perspective-taking from the second-than the third-person perspective.These findings may be related to the fact that the second-person perspective invites greater proximity to the self.Future studies may extend this work by exploring the involvement of the DMN in wise advising across different ages or cultures.
We live in a time when the resting-state is greatly occupied by information technology such as cell-phone.Cell-phone addiction has caused many mental problems (e.g., anxiety, stress, and depression) 45 , and the lack of a resting-state for wisdom development may be a potential mechanism.Our work provides important implications for wisdom interventions.For example, wisdom may be improved by increasing resting-state brain activity, similar to the effect of mindfulness meditation on wisdom 46 .Nevertheless, there are some limitations to this study.A man-focused dilemma scenario may lead to systematic differences in results.Future research should adopt gender/sex-neutral dilemma scenarios to see if the results are similar to this study.

Methods
The current study was preregistered on October 24, 2019, prior to data collection (https:// osf.io/ jkug8/).This study was performed following the principles of the Declaration of Helsinki.The ethics committee at the Hangzhou Normal University approved all procedures used in the current study.Informed consent was obtained from all the participants after they were provided with a brief description of the study.Software G*Power 3.1 was used to perform a statistical power analysis, which indicated that a sample size of 46 with a power of 0.80 (0.05 alpha error probability rate) was required to detect the minimum effect (i.e., r = 0.36)-as found in the previous EEG study measuring the association between resting-state neural oscillations and wise advising 10 .
Participants.We recruited participants from sponsoring universities through recruitment posters on campus and collected the fMRI data in a lab at the university's affiliated hospitals.We recruited 55 participants aged 18 to 26 (M ± SD = 20.80 ± 1.99; 38 males, 17 females).Two participants did not complete the advising tasks, and one showed significant artifacts from the fMRI (i.e., the head motion was more than 3 mm, while none of the other participants' frame-wise displacement calculated by Jenkinson's method 47 was more than 3 mm).Thus, the final data included 52 participants (M ± SD = 20.90 ± 1.95; 36 males, 16 females).Participants were all nativeborn healthy Chinese participants without current or historical psychiatric, neurological, or medical conditions.

Materials.
Wise advising questionnaire.The questionnaire consisted of four advising tasks; two were randomly assigned to a second-person perspective, and the other two were assigned to a third-person perspective.We printed an equal number of different versions of questionnaires (e.g., "task1-2nd perspective, task2-3rd perspective, task3-3rd perspective, task4-2nd perspective", "task1-3rd perspective, task2-2nd perspective, task3-3rd perspective, task4-2nd perspective"), ensuring that each task was evenly distributed to different perspectives across participants.For each advising task, a brief description of the life dilemma of a protagonist was provided from a second-or third-person perspective.For example, a dilemma in the second-person condition includes: "I told my good friend my most important secret, but I didn't expect the friend to reveal the secret to others.For this, I am very distressed."In contrast, a dilemma in the third-person condition includes: "A young person told the best friend the most important secret, but the person did not expect this good friend would reveal the secret to others.For this, the young person was very distressed."Subsequently, participants were given the following instructions for advising from a second-person perspective: "Now please close your eyes and imagine this young person is right in front of you and needs your advice.Please try to think about how these events will develop.Why did it develop like this?What should this person do?Please refer to this person using the second-person term 'you' ."Similarly, participants were instructed from a third-person perspective as follows: "Now please close your eyes and imagine this young person needing your advice.Please try to think about how these events will develop.Why did it develop like this?What should the protagonist do?Please refer to this person using the thirdperson term 'he' ."The questionnaire states that "you are expected to provide advice to some young people going through trouble in their lives."Moreover, the scenarios are based on "troubles" in real life of Chinese youths.See the Supplementary information for a comprehensive overview of the questionnaire in English and Chinese.
Procedure.Participants' resting-state brain activity was scanned before completing the wise advising questionnaire.Participants gave advice on some life dilemmas from the second-person perspective and some from the third (randomly allocated).
Before entering the MRI scanner, written consent was obtained from the participants.Additionally, participants completed the International Interview Examination for neuropsychiatric disorders (MINI) and Beck Depression Inventory (BDI).Participants with mental disorders, claustrophobia, metal implants, or brain injury within the past year were excluded to prevent personal injury and extraneous variability in brain activity.
All participants removed any metal item(s) from their bodies (e.g., jewelry, wallets) to avoid injury and interference from the fMRI.Once participants entered the scanner, they were instructed to keep their eyes closed and lay motionless to reduce head motion during scanning.A sponge filling was used to fill the space outside the head to reduce movement and the impact of noise on participants.Advising.After receiving an eight-minute fMRI scan, participants randomly drew a questionnaire and completed the advising tasks on a pen-and-paper questionnaire.After advising each protagonist, the participants indicated the psychological distance they felt between themselves and the protagonist on a 7-point Likert Scale (0-no distance, 6-furthest distance).
Data analysis.Rating of wise advising.Following the procedures of previous studies on WAP [7][8][9][10] , two raters naive to the purpose of the experiment were trained on sample advising transcripts until they reached high inter-rater reliability.These trained raters then rated of the participants' advice on the four life dilemmas with each of the wisdom subcomponents: metacognitive humility (MH), meta-level flexibility (MF), and perspective-taking (PT).
Analyses of fMRI data.Functional images were preprocessed with the DPARSF toolbox in MATLAB 2014b 48 .Initially, we translated data formats from DICOM to NIFTI and discarded the first ten time-point slices to improve the data quality.Next, slice-timing corrections were conducted with the middle slice as the reference slice.Then, the data was reoriented and realigned to the first slice.Participants with maximum displacement in any direction larger than 3.0 mm or head motion bigger than 3.0 mm were excluded.We used realign function of DPABI combined with the algorithm of FD-Jenkinson to calculate the participants' head motion data with XYZ translation and rotation 49 .After we excluded undesirable data (i.e., translation more than 3 mm, rotation more than 3 degrees), we used realign function to correct the head motion.Moreover, we checked the head motion (i.e., lower than 3 mm) and data quality (i.e., no artifacts) for each participant to ensure the quality of images was acceptable for analyses.Subsequently, the data were spatially normalized to the standard Montreal Neurological Institute (MNI) space using EPI (i.e., echo planar imaging) template and re-sliced to 3 × 3 × 3 mm voxels.Besides, the ALFF of each voxel is divided by the average ALFF of all voxels in the whole brain to obtain the mALFF for each voxel.Subsequently, the data were smoothed using a 4-mm full width at half maximum (FWHM) isotropic Gaussian kernel.Additionally, several sources of spurious variances were removed by the regression algorithm, including the head motion parameters, linear drift, global BOLD signals, and BOLD signals in white matter and cerebrospinal fluid.Finally, we calculated the resting-state ALFF following the procedure in previous studies 25 .Specifically, the time series of any given voxels was transformed through the Fast Fourier Transform (FFT), then the ALFF was calculated as the average square root of power in the low-frequency band (0.01-0.1 Hz).
In the DPABI software, we combined the AAL template with the BA area and then resliced the mask to define DMN.Previous studies indicate that the ventral medial prefrontal cortex (BA 10, 24, 32), posterior cingulate cortex (BA 23, 29, 30, 31), inferior parietal lobule (BA 39, 40), lateral temporal cortex (BA 21), and dorsomedial prefrontal cortex (BA 9, 10, 24, 32) are essential regions for a DMN mask 12,14 .Thus, we viewed the AAL template mask in the Viewer function and optioned these regions in DAPBI.Then we constructed these regions as the grey mask and resliced it into a 3 × 3 × 3 mask.
We calculated the whole brain ALFF first, then used a DMN mask to extract the DMN-ALFF value and get the ALFF within the DMN.For the whole-brain analysis, the whole-brain grey matter mask was applied.After getting the whole-brain grey matter, we checked the mask quality visibly using MRIcro software.Then we renormalized and resliced the mask to guarantee the mask could be used in later analyses.Subsequently, a correlation analysis between ALFF within the DMN and wisdom scores (i.e., MF, MH, PT) was conducted in the DAPAI, with gender, age, and scanner as covariates.Additionally, the Gauss Random Field (GRF) correction (two-tailed) was used for multiple corrections.Specifically, we selected the voxels with p < 0.001 and the clusters with p < 0.05.Similarly, the relationship between ALFF within the whole brain and wisdom scores (i.e., MF, MH, PT) was investigated with a correlation analysis (with GRF correction, two-tailed).Finally, for each significant correlation between the ALFF and wisdom score, we analyzed its difference with the corresponding correlation when advising from the other perspective.

Figure 2 .
Figure 2. Significant correlation between resting-state ALFF within the whole brain and wisdom performance when advising from different perspectives.To be specific, (A) is for the significant correlation between PT of second-person perspective and resting-state ALFF; (B) is for the significant correlation between MH of thirdperson perspective and resting-state ALFF; (C) is for the significant correlation between MH of second-person perspective and resting-state ALFF within the DMN.MH: metacognitive humility.PT: perspective-taking.ACC: anterior cingulate cortex.PFC: prefrontal cortex (Note: circled brain regions indicate significant areas after the Gauss Random Field correction, two-tailed.See the un-thresholded map in the Supplementary Information).

Table 2 .
Significant correlations between wise advising (from 2nd or 3rd person perspective) and restingstate ALFF after Gauss Random Field correction (two-tailed).N: number of the voxels; 2nd: second-person perspective; 3rd: third-person perspective; t and p values are for the differences between correlations when advising from different perspectives; PFC prefrontal cortex ACC anterior cingulate cortex.*, p < 0.05, **, p < 0.01.Bold font indicates results based on the correlation analyses within the DMN.Italic font indicates results based on the correlation analyses within the whole brain).