Abstract:

Both behavioral and electrophysiological studies have demonstrated the future positive bias of individual temporal self-evaluation. However, the characteristic of collective temporal self-evaluation is still an open question. The present study aimed to investigate the similarities and differences between individual and collective temporal self-evaluation using a temporal self-reference paradigm among Chinese undergraduates. In comparison to Westerners, Chinese are more unified in their self-concept, and their temporal self contains more social, relational, and family selves in the context of collectivist culture. Therefore, it was assumed that the collective temporal self-evaluation might be highly similar to individual temporal self-evaluation across time, according to interdependent self-construal theory.
The present study adopted a 3 (time dimension: past vs. present vs. future) × 2 (reference object: self vs. non-self) × 2 (valence: positive vs. negative) within-subjects design and employed an event-related potential (ERP) technique to explore the individual and collective temporal self-evaluation using two experiments. Brain electrical activity was recorded from 64 scalp sites using tin electrodes mounted in an elastic cap according to the international 10-20 system, and late positive components (LPC) were used as the EEG indicators because LPC was closely related to temporal self-evaluation. In Experiment 1, 30 Chinese undergraduates (15 females, 20.97 ± 1.87 years) were recruited, and they were asked to evaluate their and Zhangsan’s past self, present self, and future self (“Zhangsan” was often referred to the name of an ordinary person in China), using positive and negative personality adjectives. In Experiment 2, 23 Chinese undergraduates (15 females, 20.74 ± 1.60 years) were recruited to assess the past, present, and future selves of Chinese and Americans.
In Experiment 1, the means and standard deviations of adjective choice rates and LPC amplitudes in different conditions were shown in Table 1. The choice rates of positive and negative adjectives under the conditions of different reference objects in different time dimensions were shown in Figure 1, and the repeated measures ANOVA showed a significant main effect of reference object, F (1, 29) = 7.90, p = 0.009, η2 p = 0.21, and a significant main effect of valence, F (1, 29) = 246.64, p < 0.001, η2 p = 0.90. The interaction between time and valence was significant, F (2, 58) = 13.49, p < 0.001, η2 p = 0.32, and the interaction between reference object and valence was significant, F (1, 29) = 10.33, p = 0.003, η2 p = 0.26; The interaction between time and reference object was not significant, F (2, 58) = 0.46, p = 0.633. In addition, the three-way interaction of time, reference object, and valence was significant, F (2, 58) = 6.10, p = 0.004, η2 p = 0.17. The simple effects test indicated that the past, present, and future had significantly higher choice rates of positive adjectives than the negative ones (ps < 0.05) in both individual and non-individual self-referential conditions. In individual self-reference condition, positive adjective choice rates were significantly higher in the future than in the past (t = 3.17, p = 0.011, 95% CI [2.80, 25.58]) and the present (t = 4.23, p = 0.001, 95% CI [2.83, 11.32]); There was no significant difference in positive adjective choice rates between the past and the present (p = 0.173). Negative adjective choice rates were significantly higher in the past than in the future (t = 5.22, p < 0.001, 95% CI [8.20, 23.73]) and the present (t = 3.29, p = 0.008, 95% CI [2.41, 18.69]); There was no significant difference between the present and the future (p = 0.128). These behavioral results suggested that Chinese participants tended to use more positive adjectives and fewer negative adjectives to evaluate their future self, compared to their past self and present self; They tended to use more negative adjectives to evaluate their past self, compared to their present self, indicating that individual self-evaluation was positive over time and had a trend to increase from past to future.
The repeated measures ANOVA was performed on LPC amplitudes, and the results demonstrated significant main effects of reference object, F (1, 29) = 15.68, p < 0.001, η2 p = 0.35, and valence, F (1, 29) = 42.68, p < 0.001, η2 p = 0.60. The interaction between time and reference object was significant, F (2, 58) = 3.51, p = 0.036, η2 p = 0.11; However, the interactions between time and valence, and between reference object and valence were not significant (ps > 0.05). The three-way interaction of time, reference object, and valence was marginally significant, F (2, 58) = 2.98, p = 0.058, η2 p = 0.09. The simple effects test indicated that in individual self-reference condition, negative adjectives evoked greater LPC amplitudes than positive ones in evaluating individual past self (p = 0.001, 95% CI [0.35, 1.27]) and in evaluating present individual self (p = 0.001, 95% CI [0.41, 1.44]); There was no significant difference between negative and positive adjectives in evaluating future self (p = 0.213). These ERP data indicated that negative adjectives evoked greater LPC amplitudes than positive ones in the tasks of evaluating past self and present self, while there was no significant difference in future self-evaluation, which demonstrated more “negativity bias” effects in individual past self-evaluation and present self-evaluation, compared to individual future self-evaluation.
In Experiment 2, the means and standard deviations of adjective choice rates and LPC amplitudes in different conditions were shown in Table 2. The choice rates of positive and negative adjectives under the conditions of different reference objects in different time dimensions were shown in Figure 2, and the repeated measures ANOVA showed a significant main effect of reference object, F (1, 22) = 9.59, p = 0.005, η2 p = 0.30, and a significant main effect of valence, F (1, 22) = 89.85, p < 0.001, η2 p = 0.80. The interaction between time and valence was significant, F (2, 44) = 3.61, p = 0.035, η2 p = 0.14, and the interaction between reference object and valence was significant, F (1, 22) = 35.82, p < 0.001, η2 p = 0.62; The interaction between time and reference object was not significant, F (2, 44) = 0.72, p = 0.492. In addition, the three-way interaction of time, reference object, and valence was significant, F (2, 44) = 4.19, p = 0.022, η2 p = 0.16. The simple effects test indicated that in collective self-reference condition, the choice rates of positive adjectives in the past (p < 0.001, 95% CI [48.57, 77.83]), present (p < 0.001, 95% CI [47.22, 78.20]), and future (p < 0.001, 95% CI [63.98, 88.38]) were significantly higher than negative ones; However, there was no significant difference between positive and negative adjectives in different time dimensions (ps > 0.05). Compared to evaluating Americans, participants tended to use more positive adjectives (p < 0.001, 95% CI [20.06, 37.40]; p < 0.001, 95% CI [25.46, 47.63]; p < 0.001, 95% CI [22.16, 47.02]) and fewer negative words (p = 0.015, 95% CI [−29.62, −3.61]; p < 0.001, 95% CI [−39.08, −14.32]; p < 0.001, 95% CI [−40.04, −16.78]) in the past, present and future to evaluate Chinese, indicating that participants had a stable and consistent positive bias toward their collective temporal self.
The repeated measures ANOVA was performed on LPC amplitudes, and the results demonstrated significant main effects of reference object, F (1, 22) = 9.44, p = 0.006, η2 p = 0.30, and valence, F (1, 22) = 13.19, p = 0.001, η2 p = 0.38. The interaction between time and reference object was significant, F (2, 44) = 3.25, p = 0.048, η2 p = 0.13; However, the interactions between time and valence, and between reference object and valence were not significant (ps > 0.05). The three-way interaction of time, reference object, and valence was marginally significant, F (2, 44) = 3.06, p = 0.057, η2 p = 0.12. The simple effects test indicated that in collective self-reference condition, negative adjectives evoked greater LPC amplitudes than positive ones in evaluating past collective self (p = 0.017, 95% CI [0.13, 1.17]), and in evaluating present collective self (p = 0.002, 95% CI [0.39, 1.45]); There was no significant difference between negative and positive adjectives in evaluating future collective self (p = 0.198). These ERP data illustrated that greater LPC amplitudes were evoked for negative adjectives than positive ones in evaluating collective past self and present self, while there was no significant difference in evaluating collective future self, which indicated that participants had more “negativity bias” effects in collective past self-evaluation and present self-evaluation, compared to collective future self-evaluation.
In conclusion, our study suggested that both individual self-evaluation and collective self-evaluation had a positive bias across time. Moreover, there existed a “negativity bias” effect at the neurophysiological level in individual self-evaluation and collective self-evaluation. These results demonstrated that individual temporal self-evaluation and collective temporal self-evaluation among Chinese college students had similar cognitive and neurophysiological characteristics, to some extent, providing supporting evidence for the theory of interdependent self-construal.

Key words: individual temporal self-evaluation, collective temporal self-evaluation, temporal self-reference paradigm, event-related potential