Relations between family cohesion and adolescent-parent’s neural synchrony in response to emotional stimulations

Behavioral results

Table 1 showed the average valence ratings and average arousal ratings of the parent-adolescent dyads when viewing positive, negative and neutral emotional films.

Table 1 Average Valence Ratings and Arousal Ratings between Low and High Family Cohesion Parent-adolescent Dyads in Different Conditions

We employed a 2 (Group: HFCs vs. LFCs) × 3 (Valence: positive vs. negative vs. neutral) × 2 (Dyad: parent vs. adolescent) repeated measures ANOVA to analyze the differences in the average valence rating. The main effect of Valence was significant, F(2, 108) = 63.27, p< .001, ηp2= .54. The valence ratings of the positive film were higher than the negative and neutral films (both p< .001). The main effect of Dyad was significant, F(1, 54) = 4.86, p = .032, ηp2 = .83. The valence ratings of parents were significantly lower than the valence ratings of adolescents (p = .032). The main effect of Group was not significant, F(1, 54) = .21, p = .645, ηp2 = .00.The interaction of Dyad and Valence was significant, F (2,108) =6.65, p=.005, ηp2=.11.The valence ratings of parents were significantly lower than the valence ratings of adolescents in the negative conditions. (p = .001). However, there was no significant difference between parents and adolescents in the positive (p = .280) and neutral conditions (p = .810). The interaction of Group and Valence was not significant, F (2,108) =.77, p=.433, ηp2=.010. The interaction of Group and Dyad was not significant, F (1,54)=.21, p=.646, ηp2=.00. The interaction of Group, Dyad, and Valence was not significant, F (2,108)=.45, p=.585, ηp2=.01.

We employed a 2 (Group: HFCs vs. LFCs) × 3 (Valence: positive vs. negative vs. neutral) × 2 (Dyad: parent vs. adolescent) repeated measures ANOVA to analyze the differences in the average arousal rating. The main effect of Valence was significant, F (2, 108) = 63.10, p< .001, ηp2= .54. The arousal ratings of the positive films were higher than the negative and neutral films (both p<.001). The arousal ratings of the negative films were higher than neutral films (p<.001). The main effect of Dyad was not significant, F(1, 54) = 1.26, p = .267, ηp2 = .02. The main effect of Group was significant, F(1, 54) = 4.74, p = .034, ηp2 = .81.The arousal ratings of LFCs were significantly lower than the arousal ratings of HFCs (p = .034). The interaction of Dyad and Valence was significant, F (2,108)=5.89,p=.004, ηp2=.10. The arousal ratings of adolescents were significantly lower than the arousal ratings of parents in the positive conditions. (p = .009). However, there was no significant difference between parents and adolescents in the negative (p = .144) and neutral conditions (p = .113). The interaction of Group and Valence was not significant, F (2,108)=.43, p=.651, ηp2=.01.The interaction of Group and Dyad was not significant, F (1,54)=.07, p=.800, ηp2=.00.The interaction of Group, Dyad, and Valence was not significant, F (2,108)=.30, p=.745, ηp2=.01.

To examine the similarity in valence and arousal ratings between the parents and adolescents, Pearson’s correlation coefficients were calculated between the valence ratings and arousal ratings of the parents and adolescents. Fisher’s z transformation was conducted for the Pearson’s correlation coefficients before the t-test for rating similarities. Independent t-tests were conducted to examine the differences in the similarity in valence and arousal ratings between high and low family cohesion dyads. As show in Table 2, there was no significant difference in the similarity in valence (p = .964) and arousal ratings (p = .748) between high and low family cohesion dyads.

Table 2 Similarity in the Valence and Arousal Ratings between High and Low Family Cohesion DyadsNeural results

To examine the interbrain synchrony between the parent-adolescent dyads when viewing positive, negative and neutral emotional films, we calculated the interbrain phase-locking-value (PLV) (as shown in Table 3) which has been developed to measure whether the signals from the two interacting individuals are perfectly phase-locked across time.

Table 3 Gamma Interbrain Phase-locking-value (PLV) between Low and High Family Cohesion Parent-adolescent Dyads in Different Conditions at Fz, Cz and Pz

Results of the repeated measures ANOVA on the interbrain phase synchrony in the gamma band showed that the main effect of Group was not significant, F(1,27)= .04, p=.852, ηp2= .001. The main effect of Valence was not significant, F (2, 54)=1.43, p=.249, ηp2= .05. The main effect of Electrode was also not significant, F(2, 54)=.79, p=.430, ηp2=.03.The interaction of Group and Valence was not significant, F (2,54)=.07, p=.933, ηp2=.00. The interaction of Group and Electrode was not significant,F(2, 54)=1.40, p=.254, ηp2=.05. The interaction of Valence and Electrode was also not significant,F(4, 108)=.81, p=.520, ηp2=.03.

The interaction of Group ×Valence ×Electrodewas significant, F(4, 108)= 3.00, p=.022, ηp2=.10.As shown in Figure 2, greater gamma interbrain synchrony was observed in the HFCs than the LFCs in the positive conditions at Pz(t(27)= 3.522, p=.002). At Pz, higher gamma interbrain synchrony was observed in the positive conditions than in the negative conditions in the HFCs. (t(27)= 3.319, p=.005).In the HFCs, higher gamma interbrain synchrony was observed at the Pz than at the Cz (t(13)= 2.368, p=.044) and Fz (t(13)= 2.039, p=.037) in positive conditions. There were no significant differences between different conditions at other electrode in the LFCs and HFCs (ps>.05).

Fig. 2figure 2

Gamma Interbrain Phase-locking-value (PLV) in the Parietal areas (Pz) between Low and High Family Cohesion Parent-adolescent Dyads (LFCs and HFCs) in Different Conditions. Note: *p < 0.05, **p < 0.01, ***p < .001

To demonstrate the existence of neural synchronization between parents and adolescents when experiencing different emotion experience together, surrogate date was created similar to prior interbrain research [4]. Shuffling was performed by randomly forming dyads with parents and adolescents who were not from the same family. In this way, new 29-dyad samples were created. Mean PLV of the surrogate data was recalculated (M = .222 , SD = .005). In this case, we obtained an index of the gamma interbrain synchrony level that would be expected by chance. Using Wald-Wolfowitz test, we compared the real and surrogate data to assess the control distribution of the experimental effect. The PLV values in the Wald-Wolfowitz test were averaged across groups, valence conditions, and electrodes before the comparison. Results showed that there was significant difference in the distribution of the interbrain synchrony in the real parent-adolescent dyads and the random pairs (p = .017).

Moreover, to validate the significant interaction effect in real dyads in the repeated measures ANOVA, a validation approach of the permutation test was applied. As mentioned, parent and adolescent dyads were randomly assigned to form new pairs who did not actually in a family. Repeated measures ANOVAs were performed for the surrogate data. Permutation test was conducted 5000 times to yield a distribution (F value), which was then compared with the original data. As shown in Figure 3, compared with the distribution generated by the permutation procedure, the interaction effect between Group × Valence × Electrode reached significance at the p< 0.05 level for the original pairs of the parent and adolescent dyads (p=.024).

Fig. 3figure 3

The results of the permutation test, showing the distribution of the interaction effect (F value). The interaction effect (red line) in real dyads was significant within the 5% area. The x-axis represents the F value, and the y-axis represents the number of the samples

Next, to validate the significant simple effects in real dyads in the above repeated measures ANOVA, a validation analysis was added by recruited pair randomization permutation test. The parent and adolescent dyads were shuffled as mentioned above. The permutation was conducted 5000 times to yield a null distribution of the PLV values for each group in different emotional conditions separately. Significant levels (p< 0.05) were assessed by comparing the PLVs from the original dyads with 5000 renditions of random pairs. As shown in Figure 4, results showed that the significant simple effects in real dyads in the repeated measures ANOVA survived after the random permutation test which was repeated 5000 times (group comparison at Pz under positive condition, HFC>LFC, p < .001; valence comparison at Pz for the HFC group, positive > negative, p = .002; electrode comparison under positive condition for the HFC group, Pz > Cz, p = .02). These results validated the existence of the differences in the neural synchronization between LFCs and HFCs, positive and negative conditions, and parietal and central regions when experiencing different emotion experience together.

Fig. 4figure 4

The results of the permutation test, showing the distribution of the simple effects (t values). The simple effects (red line) in real dyads were significant within the 5% area. The x-axis represents the t values, and the y-axis represents the number of the samples

留言 (0)

沒有登入
gif