Neural correlates underlying change in state self-esteem

Abstract

State self-esteem, the momentary feeling of self-worth, functions as a sociometer involved in maintenance of interpersonal relations. How others' appraisal is subjectively interpreted to change state self-esteem is unknown, and the neural underpinnings of this process remain to be elucidated. We hypothesized that changes in state self-esteem are represented by the mentalizing network, which is modulated by interactions with regions involved in the subjective interpretation of others' appraisal. To test this hypothesis, we conducted task-based and resting-state fMRI. Participants were repeatedly presented with their reputations, and then rated their pleasantness and reported their state self-esteem. To evaluate the individual sensitivity of the change in state self-esteem based on pleasantness (i.e., the subjective interpretation of reputation), we calculated evaluation sensitivity as the rate of change in state self-esteem per unit pleasantness. Evaluation sensitivity varied across participants, and was positively correlated with precuneus activity evoked by reputation rating. Resting-state fMRI revealed that evaluation sensitivity was positively correlated with functional connectivity of the precuneus with areas activated by negative reputation, but negatively correlated with areas activated by positive reputation. Thus, the precuneus, as the part of the mentalizing system, serves as a gateway for translating the subjective interpretation of reputation into state self-esteem.

Figure 1

Schematic of the experiment. (A) An example condition sequence is shown. The feedback, button-press, and rest trials were presented in a pseudorandom order. (B) Visual stimuli in feedback trials showing evaluations. A photograph of the evaluation target (participant) was presented throughout the feedback trials. Adjectives were presented in the lower part of the display. Participants were told that ostensible evaluators selected the adjectives suited for the participants. Participants were required to rate the perceived pleasantness of the presented evaluation. Based on the results of pleasantness ratings, the feedback conditions were categorized as positive, negative, or intermediate. (C) Visual stimuli in the button-press trials are shown. Here, ‘XXX’ is displayed instead of adjectives. Participants were required to move the rating toward the side where the star is presented; in this example, participants were required to move the circle to the far left. (D) In the rest trial, a fixation cross was presented at the center of the screen for 5 s. (E) Visual stimuli of self-esteem trials are shown. Participants were required to rate perceived self-esteem at that time using a visual analog scale ranging from 0 to 100.

Figure 2

Relationship between the pleasantness rating and change in state self-esteem. (A) Correlation coefficients between the pleasantness rating and change in state self-esteem over two runs with 28 participants. The blue square shows data from 21 participants, which exhibited significant correlations over two runs. Light red triangles show data from two participants that exhibited significant correlation over one run. The red triangles show data from five participants that did not exhibit significant correlations. (B) Scatter diagram of perceived pleasantness and change in state self-esteem of a typical participant. (C) Trial-by-trial standardized value of perceived pleasantness (blue) and change in state self-esteem (red) for a typical participant.

Figure 3

Significant activation for positive evaluation (PE) > negative evaluation (NE). Four significant clusters are shown. We did not show a significant cluster (located at (44, −16, 56)), which showed negative beta value in PE condition. Activation threshold was set at an uncorrected p < 0.001 at the voxel level and a family-wise error (FWE) corrected p < 0.05 at the cluster level. Peaks of significant clusters in the right occipital cortex (R OC) (A), the left middle insula (L MI) (B), the R MI (C) and orbitofrontal cortex (OFC)/ventral striatum (VS) (D), overlaid onto the mean normalized T1 image of 26 participants, are shown in the upper area. Average beta values in spheres of 4-mm radius located at peaks of significant clusters for the two conditions, in comparison with the control condition (positive evaluation [PE] – control [CN], negative evaluation [NE] – CN), are shown in the lower area.

Figure 4

Significant activation for negative evaluation (NE) > positive evaluation (PE). Four significant clusters are shown. Activation threshold was set at an uncorrected p < 0.001 at the voxel level and a family-wise error (FWE) corrected p < 0.05 at the cluster level. Peaks of significant clusters in medial prefrontal cortex (mPFC)/dorsal anterior cingulate cortex (dACC) (A), right anterior insula (R AI)/right inferior frontal gyrus (R IFG) (B), right superior temporal sulcus (R STS) (C), left AI/IFG (L AI/IFG) (D), right temporo-parietal junction (R TPJ) (E), overlaid onto the mean normalized T1 image of 26 participants, are shown in the upper area. Average beta values in spheres of 4-mm radius located at peaks of the five significant clusters for the two conditions, in comparison with the control condition (positive evaluation [PE] – control [CN], negative evaluation [NE] – CN), are shown in the lower area.

Figure 5

Precuneus activation related to evaluation effects positively correlated with evaluation sensitivity. Cluster of the left precuneus for (average of positive evaluation [PE] and negative evaluation [NE] > control [CN]) was overlaid onto the mean normalized T1 image of 19 participants. Scatter diagram of evaluation sensitivity and average beta-values of (0.5 × (PE + NE) > CN) in spheres of 4-mm radius located at the peak (−10, −78, 28) of the left precuneus cluster are shown at lower side. Activation threshold was set at an uncorrected p < 0.001 at the voxel level and a family-wise error (FWE) corrected p < 0.05 at the cluster level.

Figure 6

Significant functional connectivity within evaluation representation and evaluation sensitivity-related areas during resting state. (A) Functional connectivity within positive evaluation–related regions and evaluation sensitivity–related area (left precuneus), which exhibited significant correlation with evaluation sensitivity, is shown. (B) Functional connectivity within negative evaluation–related regions and evaluation sensitivity–related area (left precuneus), which exhibited significant correlation with evaluation sensitivity, is shown. Blue and green lines indicate positive and negative significant correlation, respectively. Threshold was false discovery rate (FDR) corrected p < 0.05. VS = ventral striatum; OFC = orbitofrontal cortex; R = right; L = left; MI = middle insula; OC = occipital cortex; mPFC = medial prefrontal cortex; dACC = dorsal anterior cingulate cortex; AI = anterior insula; STS = superior temporal sulcus; TPJ = temporoparietal junction.