Interpersonal Competition in Elderly Couples: A Functional Near-Infrared Spectroscopy Hyperscanning Study

Figure 1. Experimental design. (A) Task design (top) and trial design (bottom). There were two competition blocks, each containing 20 trials (top). Events and time flow in one trial were also depicted (bottom). (B) Optode probe set. The set was placed on the prefrontal and bilateral temporoparietal regions. The integers on the brain surface indicate the recording channels.

Figure 2. Behavioral performance. The difference in response time (DRT) between the two participants of one pair in block 2 was significantly higher than that in block 1 for controls. ** p < 0.01. Error bars indicate one standard error of the means.

Figure 3. Inter-brain synchronization (IBS). (A) The interaction effects (F value) between group and task block for all channel combinations (right). Black dots indicate the position of optode probes (left). (B) Comparisons of the MTC-TPJ IBS. (C) A significant interaction effect between group and task block across several time-lags at the MTC–TPJ IBS. (D) The permutation test. The red line indicates the positions of the interaction effect for the real pairs relative to the distribution of the permutated data. * p < 0.05, ** p < 0.01, *** p < 0.001. Error bars indicate one standard error of the means.

Figure 4. Directional coupling in couples. The G-causality was analyzed in two directions (from men to women, from women to men). Couples showed greater G-causality from men to women (vs. from women to men) in block 1, but no significant difference between the two directions in block 2. * p < 0.05. Black dots indicate the position of optode probes (left). Error bars indicate one standard error of the means (right).

Figure 5. Classification results. (A) The classification accuracy of couple and controls. The dashed line indicates the positions of the classification accuracy for the original data. (B) Scatter plots depicted subjects plotted on a 2D plane using multidimensional scaling. Green and blue background colors presented areas where pairs of participants were classified as couples and controls, respectively, using a KNN classifier that was trained on the whole group. The data were plotted just for visualization and were not subjected to further statistical testing.