Archive

25 May 2023, Volume 55 Issue 5 Previous Issue    Next Issue
---
Reports of Empirical Studies


For Selected:

Download Citations EndNote Ris BibTeX

Toggle Thumbnails

Reports of Empirical Studies

Select

Effects of endogenous spatial cue validity on audiovisual integration in older adults GAO Yulin, TANG Xiaoyu, LIU Siyu, WANG Aijun, ZHANG Ming 2023, 55 (5):  671-684.  doi: 10.3724/SP.J.1041.2023.00671 Abstract ( 354 )   HTML ( 43 )   PDF (176KB) ( 197 )   Audiovisual integration is the integration of visual and auditory information into a unified, coherent and stable perceptual process. Although endogenous spatial attention can promote audiovisual integration in youth, studies have found differences in endogenous spatial attention between older adults and youth. It is unclear how endogenous spatial attention affects audiovisual integration in older adults and how audiovisual integration differs between older adults and young adults under endogenous spatial attention conditions. In this study, using the endogenous cue-target paradigm, three experiments investigated how endogenous spatial attention affects the audiovisual integration of elderly individuals under 50% (Experiment 1), 70% (Experiment 2), and 90% (Experiment 3) conditions. A mixed 2 (participant type: elderly vs. young) × 2 (cue type: valid cue vs. invalid cue) × 3 (target stimulus type: A vs. V vs. AV) experimental design was used. The visual stimulus was a 2° × 2° red (RGB: 234, 86, 97) and yellow (RGB: 247, 200, 125) intersecting meta−pattern, the auditory stimulus was a 1600 Hz, 60 dB sinusoidal tone presented by ear headphones, and the audiovisual stimulus was visual and auditory stimuli presented simultaneously on the same side. The gaze screen was presented for 500 ms, followed by a cue screen for 200 ms. The cue was an arrow pointing left or right. After a time interval of 600 ms, the target stimulus (A vs. V vs. AV) was presented in a box on the left or right side for 100 ms. Before the experiment, participants were informed that the cue validity was 50% for Experiment 1, 70% for Experiment 2, and 90% for Experiment 3 and were asked to judge the orientation of the target stimulus and to press the (N/M) key quickly and accurately. From the reaction time and accuracy results in all experiments (Table 1/2/3), it is clear that the reaction time of elderly people is significantly slower than that of younger people. From the relative amount of multisensory response enhancement (rMRE), we can see that (1) audiovisual integration was weaker in older adults than in younger adults regardless of cue validity (Figure 1); (2) at 50% cue validity (Experiment 1), audiovisual integration in the valid cue condition was not significantly different from that in the invalid cue condition for both older and younger adults (Fig. 1a); (3) at 70% cue validity (Experiment 2), audiovisual integration in the valid cue condition was not significantly different from that in the invalid cue condition for older adults, and audiovisual integration in the valid cue condition was significantly higher than that in the invalid cue condition for younger adults (Figure 1b); and (4) at 90% cue validity (Experiment 3), audiovisual integration in the valid cue condition was significantly higher than that in the invalid cue condition for both older and younger adults (Figure 1c). Endogenous spatial attention had different moderating effects on audiovisual integration in older adults under different cue validity conditions and could promote audiovisual integration in older adults under high cue validity conditions. The findings further support the spatial uncertainty hypothesis and deepen the understanding of the interaction between endogenous attention and audiovisual integration. Figures and Tables | References | Related Articles | Metrics

Select

fNIRS evidence for left middle frontal gyrus involved in visual-spatial analysis of Chinese characters CHEN Fakun, CHEN Tian, CAI Wenqi, WANG Xiaojuan, YANG Jianfeng 2023, 55 (5):  685-695.  doi: 10.3724/SP.J.1041.2023.00685 Abstract ( 249 )   HTML ( 22 )   PDF (2354KB) ( 469 )   The left middle frontal gyrus (MFG) is a typical brain region identified in studies of Chinese character reading brain mechanisms, it exhibits specific activation in Chinese character reading. A common interpretation suggests that it is responsible for the unique visual-spatial processing of Chinese characters. However, this interpretation is not supported by direct evidence. This study manipulated the spatial frequency of visually presented Chinese character materials and explored this issue by using functional Near-Infrared Spectroscopy (fNIRS) techniques. By constructing a 3 (character type: Real character, Pseudo character, and Artificial character) × 3 (spatial frequency: Full spatial frequency, Low spatial frequency, and High spatial frequency) repeated-measures experimental design, The blood oxygen concentration changes in the left MFG were recorded as participants completed a one-back task with repeated stimulus detection. It was found that the left MFG showed a significant main effect of character type, that was, Pseudo characters required more MFG activation than Real and Artificial characters. Moreover, the left MFG also showed a significant interaction effect of character type and spatial frequency, with Pseudo characters more activated MFG than Real and Artificial characters in the Low spatial frequency condition, while no significant activation difference of character types was found under the other two spatial frequency conditions. The results suggested that the left MFG was indeed sensitive to the spatial information of Chinese characters, more MFG activation was required especially in the Pseudo characters condition that required more graphemic/orthographic processing and in the processing of low spatial frequency information. The findings provided direct evidence that the left MFG was involved in the visuospatial processing of Chinese characters. Figures and Tables | References | Related Articles | Metrics

Select

Role of executive function in mathematical ability of children in different grades ZHU Xiaoliang, ZHAO Xin 2023, 55 (5):  696-710.  doi: 10.3724/SP.J.1041.2023.00696 Abstract ( 313 )   HTML ( 32 )   PDF (534KB) ( 211 )   A total of 812 children in Grades 3-6 were selected to examine the role of each executive function component in the three mathematical abilities at different grades using correlation analysis and structural equation modelling. The results showed that in the junior grades, working memory span was the most important predictor of mathematical operations, spatial imagination and logical thinking. In the older students, working memory span decreased in predicting the three mathematical abilities, while working memory updating and cognitive flexibility increased in predicting the mathematical abilities. This suggests that there are differences in the predictive effects of the executive function components on different mathematical abilities, and that they change as children progress through the grades. Figures and Tables | References | Related Articles | Metrics

Select

The interactive effects of parent-child relationship, sensory processing sensitivity, and the COMT Val158Met polymorphism on preschoolers’ prosocial behaviors LIU Qianwen, WANG Zhenhong 2023, 55 (5):  711-725.  doi: 10.3724/SP.J.1041.2023.00711 Abstract ( 299 )   HTML ( 15 )   PDF (290KB) ( 166 )   The present study investigated the interactive effects of parent-child relationship, sensory processing sensitivity (SPS), and the COMT Val158Met polymorphism on preschoolers’ prosocial behaviors. A total of 507 preschoolers (Mage = 4.83, SD = 0.90; 236 girls) were recruited through advertisements at two local kindergartens. Saliva samples for DNA extraction were obtained from preschoolers. Their parents completed questionnaires assessing parent-child relationship, children’s SPS, and prosocial behaviors. The results indicated that the three-way interactive effect of parent-child conflict, SPS, and the COMT gene Val158Met polymorphism on prosocial behaviors was significant, that is, Val/Val genotype carriers with high SPS showed significantly more prosocial behaviors under a low level of parent-child conflict and fewer prosocial behaviors under a high level of parent-child conflict. However, there was no difference in prosocial behaviors of Val/Val genotype carriers with low SPS in a high and low-level conflict. These findings signified the different types of sensitivities (temperament and genes) to the family stressful environment may have a multiplicative effect on preschoolers’ prosocial behaviors and contributed to a further understanding of children’s prosocial behaviors based on the person × environment approach, especially from the perspective of children’s multiple sensitivities. Figures and Tables | References | Related Articles | Metrics

Select

The influence of positive co-experience on teacher-student relationship: The mediating role of emotional bonding DING Yuting, ZHANG Chang, LI Ranran, DING Wenyu, ZHU Jing, LIU Wei, CHEN Ning 2023, 55 (5):  726-739.  doi: 10.3724/SP.J.1041.2023.00726 Abstract ( 483 )   HTML ( 39 )   PDF (332KB) ( 444 )   Based on questionnaire survey, field experiment and laboratory experiment, this paper investigates the influence and mechanism of positive co-experience on adolescent teacher-student relationship. The results show that: (1) positive co-experiences positively affect teacher-student relationship, and different types of experiences (recall, imagination, example) are prominent promoting effect; (2) Positive emotional bonding plays a stable mediating role in the influence of positive co-experiences on teacher-student relationship. This study preliminarily proposed the “co-experience relationship effect model”, which promotes the research on the influence mechanism of teacher-student relationship, and has good ecological validity and practical educational value. Figures and Tables | References | Related Articles | Metrics

Select

Personality subtypes of depressive disorders and their functional connectivity basis LI Yu, WEI Dongtao, QIU Jiang 2023, 55 (5):  740-751.  doi: 10.3724/SP.J.1041.2023.00740 Abstract ( 481 )   HTML ( 43 )   PDF (403KB) ( 195 )   Heterogeneity among mental health issues has always attracted considerable attention, thereby restricting research on mental health and cognitive neuroscience. Additionally, the person-centred approach to personality research, which emphasizes population heterogeneity, has received more attention. On the other hand, the heterogeneity among depressive patients has been a problem that cannot be ignored (most studies ignored the actual situation and directly assumed sample homogeneity). A large number of empirical studies have provided evidence that isolated personality traits are often associated with depression. Only a few studies have considered the probable effect from a taxonomy perspective. Moreover, the neural mechanisms of personality types in depression remain unclear. This study aimed to reveal different personality subtypes of depressive disorders and elucidate subtypes from the perspective of resting-state functional connectivity. Personality and resting-state functional imaging data of 159 depressive patients and 156 controls were collected. Demographic characteristics are shown in Table 1. First, combined with “depression diagnosis”, the personality types in depressive patients and controls were identified through functional random forest. Specifically, neuroticism and extraversion (input features) were fitted with the diagnosis of depression by a random forest model. The random seeds were set to 1234, and 500 decision trees were fitted. The performance of the model was evaluated by tenfold cross-validation. Subsequently, the random forest algorithm generated a proximity matrix that represented the similarity between paired participants. Then, based on the proximity matrix, community detection clustering analysis was conducted on depressive patients and controls, and personality types associated with depression diagnosis were obtained. Finally, we selected the amygdala, hippocampus, insula (AAL atlas) and limbic network, default network, and control network (Schaefer-Yeo template) as regions of interest and calculated the functional connectivity of the subcortical regions to the networks. ANOVA was used to compare resting-state functional connectivity between the personality types. The results showed the following. (1) Depression was more common among individuals with high neuroticism and low extraversion tendencies, but there were also individuals with low neuroticism and high extraversion tendencies. The controls were more likely to be individuals with low neuroticism and high extraversion (see Figure 1). (2) The results of resting-state functional connectivity showed no significant difference between depression and controls. (3) The functional connectivity strength of the left amygdala-limbic network (F(6, 214) = 4.273, p = 0.0004, threshold-controlling FDR at 0.05/6) and left insula-limbic network (F(6, 214) = 4.177, p = 0.0005, threshold-controlling FDR at 0.05/6) was significantly different across personality subtypes. The post-hoc tests are presented in Table 2, Figure 2 and Figure3. In summary, the personality subtypes of depression identified by person-centred perspectives are more in line with reality and individual cognitive patterns, and they have potential clinical adaptive value. The findings of this study enhance the understanding of heterogeneity among depressive disorders. Figures and Tables | References | Related Articles | Metrics

Select

Influence and mechanisms of common ingroup identity on competitive victimhood in doctor-patient relationships DENG Xun, LONG Siyi, SHEN Yilin, ZHAO Huanhuan, HE Wen 2023, 55 (5):  752-765.  doi: 10.3724/SP.J.1041.2023.00752 Abstract ( 196 )   HTML ( 10 )   PDF (170KB) ( 105 )   Competitive victimhood is a psychological phenomenon pervasive on both sides of an intergroup conflict; it implies that one person believes their group suffers more than the other does. As one of the most prominent and growing barriers to positive intergroup relations globally, competitive victimhood has gradually attracted the attention of researchers. However, little research has been conducted on competitive victimhood in Chinese hospitals, where the doctor-patient relationship is increasingly tense. The common ingroup identity model holds that by reconstructing social identity and breaking the boundaries of conflict groups, members can develop a common ingroup identity. This identity improves attitudes toward outer groups, which may help reduce competitive victimhood. The need-based model argues that power and morality are ingroups and outgroups’ basic needs. Members of both sides are threatened by power or morality and are motivated to restore their identities, affecting competitive victimhood. Therefore, it is worth studying whether common ingroup identity can effectively reduce competition victimization between doctors and patients and what roles power and moral needs play. In Study 1, 90 doctors and nurses and 86 patients and their families from three hospitals in Shanghai and Sichuan were selected in a 2 (group: doctors vs. patients) × 2 (common identity: control group vs. common ingroup identity) design. A brief story about doctors and patients fighting disease together was used to improve common ingroup identity. Then we used a questionnaire about competitive victimhood to investigate whether common ingroup identity affected competitive victimhood between doctors and patients. In Study 2A, another group of participants was selected, including 71 doctors and nurses along with 73 patients and their families from three hospitals in Shanghai and Sichuan. Participants underwent the same procedure as in Study 1, then completing the power and moral needs questionnaires. Study 2A aimed to investigate the influence of common ingroup identity on victimhood between doctors and patients, as well as the roles of power and moral needs. To further test the hypothesized model, we selected 54 medical students with hospital internship experience and 54 non-medical students with recent treatment experience in Study 2B, where we activated common ingroup identities using a re-categorization strategy. The main results were as follows. In Study 1, ANOVA results showed that compared with control group (M = 5.15, SD = 0.96), group with common ingroup identity (M = 4.58, SD = 0.93) reported lower competitive victimhood, F(1, 172) = 16.58, p < 0.001, ηp2 = 0.088. And there was no interaction between group identity and common ingroup identity, F(1, 172) = 2.66, p = 0.105. Study 2A showed that power need mediated the relationship between common ingroup identity and competitive victimhood. ANOVA results showed that main effect of common ingroup identity was significant, F(1, 140) = 13.54, p < 0.001, ηp2 = 0.088, group with common ingroup identity(M = 4.60, SD = 0.92) reported lower competitive victimhood than control group (M = 5.19, SD = 1.06). The indirect effect of power need was significant and the indirect effect of moral need was not significant (see Figure 1). Study 2B reconfirmed the model for doctors but not for the patients group. ANOVA results showed that main effect of common ingroup identity was not significant, F(1, 104) = 1.58, p= 0.212, and main effect of group identity was significant, F(1, 104) = 26.52, p < 0.001, ηp2 = 0.203, the competitive victimhood of doctors (M = 4.99, SD = 1.00) was higher than patients (M = 4.11, SD = 0.83), p < 0.001, 95% CI = [0.87, 1.83]. And there was significant interaction between group identity and common ingroup identity, F(1, 104) = 7.43, p = 0.008, ηp2 = 0.067. For doctors, group with common ingroup identity (M = 4.65, SD = 1.05) reported lower competitive victimhood than control group (M = 5.33, SD = 0.83), p = 0.006, and the indirect effect of power need was also significant, but for patients there was no difference, p = 0.301 (see Figure 2). Based on the common ingroup identity model, this study proposed and confirmed the applicability of this model in the doctor-patient field in China. Additionally, the study proposed new methods and perspectives on the doctor-patient relationship. In the future, researchers should focus on other mediators, such as empathy and trust in different groups. Figures and Tables | References | Related Articles | Metrics

Select

Undervaluing the advantages of displaying skills in front of an expert QIU Tian, JIANG Nan, LU Jingyi 2023, 55 (5):  766-780.  doi: 10.3724/SP.J.1041.2023.00766 Abstract ( 238 )   HTML ( 16 )   PDF (411KB) ( 166 )   Job candidates and competitors aim to earn admission or high ratings. People tend to avoid displaying their skills in front of an expert due to the prediction that they will be rated unfavorably because the expert can accurately evaluate their level of skill. However, is this prediction accurate? The present research proposes a misprediction: candidates will undervalue the advantages of showing skills in front of an expert. This is because evaluators partially base their evaluations on the pride elicited by alluding to their expertise, whereas candidates base their predictions on whether their competence will be accurately evaluated but neglect evaluators’ pride. Eight studies (N = 1888) demonstrated the proposed misprediction and tested its underlying mechanism. In Study 1, we assigned the participants to the candidate or the evaluator condition. The candidates made an incentive-compatible prediction on how they would be more likely to be admitted by displaying their skills in front of an expert or a non-expert. The evaluators admitted one between a candidate displaying skills in the evaluators’ area of expertise and a candidate displaying skills outside the evaluators’ area of expertise. The results showed that 63.10% of the evaluators preferred the candidate who showcased skills in the evaluators’ area of expertise. However, only 44.60% of the candidates chose to do so (χ2(1, N = 130) = 4.46, p = 0.035, φ = 0.19). These results revealed that the candidates undervalued the advantages of showing skills in front of an expert. Studies 2 and 3 replicated the results in Study 1 with different competition forms (promotion or elimination) and in the case where candidates were assigned to display skills in or outside the evaluators’ area of expertise. In Study 2, we adopted a 2 (role: candidate or evaluator) by 2 (competition form: promotion or elimination) between-subjects design. Note that the evaluators in Study 1 might be more familiar with their area of expertise (vs. non-expertise) and could easily evaluate the candidate who showcased skills in this area, which could also lead to our proposed effect. Therefore, we controlled for the evaluators’ time spent on the area of their expertise and non-expertise to ensure that the evaluators were equally familiar with the two areas. The results revealed that neither the main effect of role (B = −0.12, SE = 0.25, Wald χ2= 0.23, p = 0.630, Exp(B) = 0.89) nor the main effect of competition form (B = −0.34, SE = 0.25, Wald χ2= 1.82, p = 0.178, Exp(B) = 0.71) was significant. More crucially, the interaction between role and competition form was significant (B = 3.45, SE = 0.55, Wald χ2= 39.67, p < 0.001, Exp(B) = 31.62, 95% CI = [10.80, 92.63]; see Figure 1). Study 2 showed that the proposed effect held across different competition forms and ruled out the alternative explanation that the evaluators could easily evaluate the candidate in their area of expertise because they were familiar with this area. In Study 3, we adopted a 2 (role: candidate or evaluator) by 2 (domain assignment: expertise or non-expertise) between-subjects design. In previous studies, the evaluators probably believed that the candidates who chose to perform in evaluators’ area of expertise shared the same interest with the evaluators. As a result, the evaluators preferred these candidates. To rule out this alternative explanation, in Study 3, the candidates were assigned to a domain instead of choosing by themselves. The results showed that the main effect of role was significant (B = −1.50, SE = 0.30, Wald χ2= 25.43, p < 0.001, Exp(B) = 0.22, 95% CI = [0.13, 0.40]). The candidates underestimated the possibility of being admitted regardless of the domain they were assigned to. There was no significant main effect of domain assignment (B = −0.20, SE = 0.28, Wald χ2= 0.50, p = 0.479, Exp(B) = 1.22), nor interaction (B = −0.49, SE = 0.59, Wald χ2= 0.67, p = 0.413, Exp(B) = 0.62; see Figure 2). Study 3 thus replicated the effect and ruled out the similar-interest account. Studies 4 and 5 manipulated the candidates’ motivation to win the competition and their level of competence, respectively, to test whether they avoided displaying skills in front of experts due to the concern that their competence could be evaluated accurately by experts. The results indicated that the candidates showed a stronger misprediction and were less likely to showcase skills in front of experts when they highly (vs. less) desired to win the competition or had a lower (vs. moderate and higher) competence. In Study 4, candidates with a higher motivation to win (44.44%) were less likely to display skills in the evaluators’ area of expertise than candidates with a lower motivation (64.71%) did (χ2(1, N = 140) = 5.79, p = 0.016, φ = 0.20). As a result, candidates with a high motivation more strongly underestimated the evaluators’ choices (84.38%, χ2(1, N = 136) = 23.24, p < 0.001, φ = 0.41) than candidates with a low motivation did (χ2(1, N = 132) = 6.67, p = 0.010, φ = 0.23, see Figure 3). In Study 5, the main effects of role (B = −1.38, SE = 0.23, Exp(B) = 0.25, 95% CI = [0.16, 0.40], Wald χ2= 35.77, p < 0.001) and level of competence were significant (Wald χ2= 28.66, p < 0.001). The interaction was also significant (Wald χ2= 8.21, p = 0.016). Specifically, candidates with a high level of competence could accurately predict the evaluation (candidate: 77.14%, evaluator: 82.86%, χ2(1, N = 140) = 0.71, p = 0.398), while candidates with a moderate or low level of competence underestimated the evaluation (moderate level of competence: candidate: 50.00%, evaluator: 81.43%, χ2(1, N = 140) = 15.34, p < 0.001, φ = 0.33; low level of competence: candidate: 27.14%, evaluator: 72.86%, χ2(1, N = 140) = 29.26, p < 0.001, φ = 0.46; see Figure 4). Study 6 prompted the candidates to empathize with evaluators. We asked the candidates to think about their feelings when others made references to their expertise. As a result, compared to the control group (candidate: 40.00%, evaluator: 72.86%, χ2(1, N = 140) = 15.37, p < 0.001, φ = 0.33), the prompted candidates were more aware of their pride and made a more accurate prediction (candidate: 58.57%, evaluator: 72.86%, χ2(1, N = 140) = 3.17, p = 0.075, see Figure 5). Study 7 manipulated the evaluators’ pride by changing their achievement and thereby tested whether their preference for the candidate was due to that they felt proud when their expertise was referred to. The results revealed that the main effect of role was significant (B = −1.75, SE = 0.27, Wald χ2= 41.60, p < 0.001, Exp (B) = 0.17, 95% CI = [0.10, 0.30]). The main effect of evaluators’ achievement was not significant (B = 0.04, SE = 0.27, Wald χ2= 0.03, p = 0.874, Exp(B) = 1.04). Most importantly, the interaction was significant (B = 1.65, SE = 0.56, Wald χ2= 8.56, p = 0.003, Exp(B) = 5.52, 95% CI = [1.73, 15.77]). The evaluators with a lower (vs. higher) achievement were less likely to admit the candidates who displayed skills in the evaluators’ area of expertise (lower achievement: 63.77%, higher achievement: 81.67%, χ2(1, N = 129) = 5.11, p = 0.024, φ = 0.20, see Figure 6). In Study 8, we recorded the participants’ real-time thoughts during their decision making (see Table 1). The results again showed that the candidates focused on their competence during their decision-making process, whereas the evaluators’ preferences were affected by their pride. As a result, 76.39% of the evaluators preferred the candidate who showcased skills in the evaluators’ area of expertise, while only 17.65% of the candidates chose to do so (χ2(1, N = 140) = 48.36，p < 0.001，φ = 0.59). In addition, the real-time thoughts led to the underestimation about the benefits of displaying skills in front of an expert (z = −3.20，p = 0.001). We reveal that people fail to accurately predict the effect of a self-presentation strategy. Candidates undervalue the strategy of displaying skills in front of experts due to the empathy gap that they neglect the pride experienced by experts. Consequently, candidates mistakenly avoid displaying skills in front of experts and thus miss the chance to earn admission. Besides, we offer a feasible approach to reduce such a bias. Our findings encourage candidates to empathize with evaluators and strategically perform to experts. Figures and Tables | References | Related Articles | Metrics

Select

The effect of temporal focus on implicit space-time mapping in a life history strategy framework WANG Yue, WANG Xiaoyu, SONG Ying, LI Ying 2023, 55 (5):  781-791.  doi: 10.3724/SP.J.1041.2023.00781 Abstract ( 169 )   HTML ( 12 )   PDF (175KB) ( 97 )   It is a fundamental feature of human cognition to understand the abstract concept of time through space. As far as the direction of metaphorical mapping is concerned, space-time mappings in the 'front-back' direction are the most common. At present, most studies have confirmed the psychological reality of spatiotemporal mapping. However, research on implicit space-time mapping in mental thinking and its influencing factors are still controversial. The Temporal Focus Hypothesis (TFH) holds that language is not the only factor affecting the implicit space-time mapping, and that the attention individuals pay to past or future time, namely, Temporal Focus, is the key factor shaping the implicit space-time mapping. Temporal focus refers to the extent to which individuals allocate their attention to past, present and future time periods, and is influenced by a variety of factors such as culture, individual differences, and cognitive training. Life history strategy is a stable pattern of psychological behavior. It is a variable belonging to the category of individual differences. Research has shown that temporal orientation is the signature difference between life history strategies. Slow life history strategy individuals are more inclined to make long-term plans. They are less likely to choose risky behaviors for immediate benefits, exhibit a higher sense of control in pursuit of a better future, and are more quality-conscious about the upbringing of their offspring. However individuals with the fast life history strategy prefer present gratification, tend to take risky behaviors regardless of the future, and place more importance on quantity than quality of offspring. So those with a slow life history strategy attach more importance to events related to the future and show a preference for “future thinking”, while those with a fast life history strategy focus on the present and have no obvious preference for “future thinking” or “past thinking”. This study adopts a time-focus questionnaire, a time-diagramming task and a time-word categorization task to explore the temporal focus and implicit temporal mapping preferences of subjects with different life history strategies, and the effects of life history strategies and time-focus on implicit temporal mapping. Study 1 used a 2 (life history strategy type: slow strategy vs. fast strategy) × 2 (temporal focus: future vs. past) two-factor mixed experimental design. The relationship between life history strategies and temporal focus was examined using the life history strategies questionnaire and the temporal focus scale. 304 subjects were selected (173 females and 131 males, M age= 22.6 years). Based on the mean score of 27.45 on the life history strategy scale as a cut-off, scores of 28 and above were classified as slow life history strategy and scores of 27 and below were classified as fast life history strategy. There were 169 subjects in the slow life history strategy and 135 subjects in the fast life history strategy. An independent samples t-test was used to test the validity of the grouping, and the results showed that the fast life history strategy group scored significantly lower (16.79 ± 7.44) than the slow life history strategy group (36.14 ± 5.58), t(302) = −22.23, p < 0.001, and the grouping was valid. One subject was removed for guessing the purpose of the experiment, and a total of 303 data were analyzed. Results of the Study 1 repeated measures ANOVA indicated that the life history strategy main effect was not significant, F(1, 302) = 2.28, p > 0.05. The time focus main effect was significant, F(1, 302) = 27.74, p< 0.001, η2p = 0.08. The time focus and life history strategy interaction was significant, F(1, 302) = 55.81, p< 0.001, η2p = 0.16. The simple effects analysis indicated that the slow life history strategy subjects rated the future (M = 5.27, SD = 0.85) significantly higher than their ratings of the past (M = 4.18, SD = 0.94), p < 0.001. There was no significant difference between the ratings of the future (M = 4.51, SD = 1.02) and the past (M = 4.70, SD = 1.17) for subjects with the fast life history strategy, p > 0.05 (see Table 1). Study 2 explored the relationship between life history strategy type and implicit temporal mapping orientation through two sub-experiments using a time-diagramming task (see Figure 1) and a time-word categorization task, respectively. Study 2a used a two-factor mixed experimental design of 2 (life history strategy type: slow strategy vs. fast strategy) × 2 (temporal mapping type: future ahead vs. past ahead). A time-diagramming task was used to examine the relationship between life history strategies and implicit temporal mappings. The experimental subjects were the same as in Study 1. Study 2b used a mixed experimental design of 2 (life history strategy: fast strategy vs. slow strategy) × 2 (response type: congruent vs. incongruent). The aim was to test the relationship between life history strategies and implicit temporal mappings using a time-word classification task. Undergraduate and graduate students were recruited at a university and 150 life history strategy questionnaires were distributed. The extreme grouping method was used to screen the 20% with the highest scores totaling 30 as the slow life history strategy group and the 20% with the lowest scores totaling 30 as the fast life history strategy group. The results of Experiment 2a are as follows. Fisher's exact test on a 2 × 2 four-compartment table revealed a significant interaction between subject type and spatio-temporal mapping direction, χ2 (1) = 19.78, p < 0.001. Sign test analysis showed that there was no significant difference between the rate of selecting “future in front” and “past in front” for the fast life history strategy, p > 0.05. The rate of selecting “future in front” was significantly higher than that of “past in front” for the slow life history strategy, p < 0.001. There was no significant difference in the “future ahead” and “past ahead” rates, p > 0.05. The slow life history strategy had a significantly higher rate of “future ahead” than “past ahead”, p < 0.001 (see Table 2). The results of Experiment 2b are as follows. The statistical analysis showed that the error rate of the subjects was low and evenly distributed, less than 5%, so the error rate was not analyzed in the subsequent results. All incorrect responses and data beyond 2.5 standard deviations were removed, accounting for 4.1% of the total data. A repeated measures ANOVA for response time indicated a significant main effect of response type, F(1, 58) = 5.77, p= 0.02, ηp2 = 0.09. Response time in the response consistency condition (1026.37 ms) was significantly faster than that in the inconsistency condition (1048.88 ms). Subject type main effect was not significant, F(1, 58) < 1, p> 0.05. Subject type and response type interaction was significant, F(1, 58) = 17.11, p< 0.001, ηp2 = 0.23. Simple effects analysis showed that there was no significant difference in response times between consistent and inconsistent conditions for fast strategy subjects, F(1, 58) = 1.51, p> 0.05. Slow strategy subjects had a significantly faster reaction time in the consistent condition (1011.57 ms) than in the inconsistent condition (1072.85 ms), F (1, 58) = 21.37, p < 0.001 (see Table 3, Figure 2). Study 3 further tested the stability of the temporal focus hypothesis by initiating different temporal focuses on individuals with fast and slow life history strategies through two sub-experiments, respectively. Experiments 3a used a two-factor mixed experimental design of 2 (intervention direction: focus on the past, focus on the future) × 2 (response type: consistent with the intervention direction, inconsistent with the intervention direction).Two hundred life history strategy questionnaires were distributed, and the extreme grouping method was used to select the lower 30% of scores, a total of 60, as fast life history strategy subjects and the higher 30% of scores, a total of 60, as slow life history strategy subjects. Experiment 3a was conducted with 60 fast life history strategy subjects. Experiment 3b was conducted with 60 slow strategy subjects. Others were consistent with Experiment 3a. The results of Experiment 3a are as follows. Using the temporal focus scale to examine scores on future and past focus, the intervention direction was that subjects in the focus on past group scored significantly greater on the past temporal focus (M past = 4.98) than on the future temporal focus (M future = 4.16), t(58) = −3.00, p = 0.004. The intervention direction was that subjects in the focus on future group scored significantly greater on the future temporal focus (M future= 5.24) than on the past temporal focus. The intervention was successful for subjects with different temporal focus. ANOVA results showed a significant main effect of response type, F(1, 58) = 47.9, p< 0.001, ηp2 = 0.45, with subjects responding significantly faster in the consistent with the intervention direction condition (950.76 ms) than in the inconsistent with the intervention direction condition (1002.41 ms). Intervention direction main effect was not significant, F (1, 58) = 1.72, p > 0.05. Response type and intervention direction interaction was not significant, F(1, 58) = 1.25, p > 0.05 (see Table 4). The results of study 3b were as follows. The scores of future and past focuses were examined using the temporal focus scale, and the results showed successful manipulation of subjects' temporal focus. The intervention direction was that subjects in the past focus group scored significantly greater in the past temporal focus (M past = 4.71) than in the future temporal focus (M future = 4.08), t(58) = −2.43, p = 0.018. The intervention direction was that subjects in the future focus group scored significantly greater in the future focus (M future = 5.25) than in the past temporal focus (M past = 4.60), t (58) = 2.9, p = 0.005. ANOVA results indicated that the main effect of response type was not significant, F(1, 58) = 1.41, p= 0.24. The main effect of intervention direction was not significant, F (1, 58) < 1, p > 0.05. The interaction of response type and intervention direction was not significant, F (1, 58) < 1, p > 0.05 (see Table 5). The results showed that the fast-strategy subjects had no obvious preference for temporal focus and implicit temporal mapping, while the slow-strategy subjects preferred future temporal focus and “future-front, past-back” implicit temporal mapping associations. The intervention of temporal focus shaped the implicit space-time mappings direction of fast-strategic individuals with a metaphorical consistency effect, whereas for slow-strategic subjects, the intervention had a limited effect. The entire study demonstrates that life history strategies can influence temporal focus and implicit space-time mapping from an evolutionary adaptive perspective. And the temporal focus hypothesis has boundary conditions. The findings suggest that we can facilitate individuals' processing of temporal events in daily life by shaping temporal focus, such as decision-making behavior, and encouraging students to focus on future development. It can also help people to better understand the temporal metaphors in their lives. Figures and Tables | References | Related Articles | Metrics

Select

An examination of configural effects of employees’ proactive behavior: A process perspective LI Liyuan, GAO Xiangyu, ZHENG Xiaoming 2023, 55 (5):  792-811.  doi: 10.3724/SP.J.1041.2023.00792 Abstract ( 271 )   HTML ( 11 )   PDF (1289KB) ( 220 )   Employees’ proactive behavior refers to future-oriented, self-initiated behaviors taken by employees to bring about change for organization or for their own (Grant & Ashford, 2008; Parker et al., 2010). Prior research has found that employees' proactive behavior can have a series of positive effects on employees and organizations, such as improving employees' job performance (e.g., Thomas et al., 2010), increasing employees’ organizational identity and job satisfaction (e.g., Seibert et al., 2001). However, an increasing body of evidence suggests that proactive behavior is not consistently effective (Li & Huang, 2021; Parker et al., 2019). A summary of previous research conducted by Parker et al. (2019) on the effectiveness of proactive behavior reveals that it has been examined from four perspectives: the forms of proactive behavior, aspects of the proactive person, aspects of the situation, and the person-situation interaction. For example, if an individual's supervisor exhibits high levels of openness, the individual's proactive behavior is more likely to be accepted and produce desirable outcomes (Tucker & Turner, 2015). Previous research has contributed to a deeper understanding of the effectiveness of proactive behavior, however, significant research gaps remain. Among these prior studies, researchers have primarily focused on the impact of overt behavioral element on various individual and organizational outcomes, while neglecting the influence of covert behavioral elements and the synergistic effect of all elements in a proactive process. Proactive behavior is, in fact, a self-regulatory process including both overt behavioral element (i.e., enacting) and covert behavioral elements (i.e., envisioning, planning, and reflecting) (e.g., Bindl et al., 2012). This narrow focus leads to an incomplete understanding of proactive behavior. To address this problem, we adopt a process perspective and utilize a configural approach and action regulation theory (Frese & Zapf, 1994; Zacher & Frese, 2018) to investigate: 1) the configural effects of all behavioral elements (i.e., envisioning, planning, enacting, and reflecting) in a proactive process on employees’ job performance and emotional exhaustion; 2) the impact of task context (i.e., environmental uncertainty) and social context (i.e., felt trust) on the configural effects of all behavioral elements; and 3) incremental effect of configuration membership on employees’ outcomes after controlling for the unique effects of each element and individual difference (i.e., proactive personality and neuroticism). By using fuzzy set qualitative comparative analysis (fsQCA) and regression analysis, we conducted a longitudinal study to test our hypotheses and examine research questions. We examined the configural effects of all four elements on employees’ job performance and emotional exhaustion, and then examined the impacts of task and social context and the incremental effect of these configurations. 426 full-time employees and their direct supervisors from an education and training company in China participated in a questionnaire survey and reported data at two time points two weeks apart. Supervisors reported their subordinates’ job performance at time 2. The final sample size was 383. The descriptive statistics and correlations of all variables are presented in Table 1. In accordance with the requirements of fsQCA, the raw data for each condition was calibrated to membership scores ranging from 0 to 1, with the calibration anchors reported in Table 2. The subsequent analysis included both necessity and sufficiency analysis, which were performed by constructing truth tables. The results of the fsQCA are shown in Tables 3 and 4. Table 3 shows the configural effect of proactive process on employees’ job performance and emotional exhaustion, revealing four configurations that were sufficient for high or low outcomes. Specifically, configuration 1a (high envisioning, high planning, high enacting, and high reflecting) is sufficient for high job performance. Configuration 1b (low envisioning, low planning, low enacting, and low reflecting) is sufficient for low job performance. Configuration 1c (low envisioning, low planning, high enacting, and low reflecting) is sufficient for high emotional exhaustion. Configuration 1d (high envisioning, high planning, low enacting, and high reflecting) is sufficient for low emotional exhaustion. Table 4 presents the configural effect of proactive process and contextual factors on employees’ job performance and emotional exhaustion, revealing five configurations that were sufficient for high or low outcomes. Specifically, configuration 2a (high envisioning, high planning, high enacting, and high reflecting) and 2b (high envisioning, high planning, high enacting, and high environmental uncertainty) are sufficient for high job performance. Configuration 2c (low envisioning, low planning, low enacting, and low reflecting) is sufficient for low job performance. Configuration 2d (low envisioning, low planning, high enacting, and low reflecting) is sufficient for high emotional exhaustion. Configuration 2e (high envisioning, high planning, high reflecting, and high felt trust) is sufficient for low emotional exhaustion. Further, we combined fsQCA and regression analysis to explore the incremental effect of configuration membership on employees’ outcomes after controlling for the unique effects of each element and individual difference. The regression analysis results are shown in Table 5 and 6. Model 1-4 in Table 5 show that enacting (model 1: B = 0.13, p = 0.016) and reflecting (mode 1: B = 0.15, p = 0.013) are both positively related to job performance. When considering the effect of the four elements of proactive process, proactive personality, and configuration 1a on employees’ job performance, proactive personality is positively related to employees’ job performance (model 3: B = 0.20, p = 0.005); configuration 1a is positively related to employees’ job performance (model 3: B = 0.42, p = 0.007), and the relationship between enacting and employees’ job performance is no longer significant. Similarly, when considering the effect of the four elements of proactive process, proactive personality, and configuration 1b on employees’ job performance, proactive personality is positively related to employees’ job performance (model 4: B = 0.23, p = 0.001); configuration 1b is negatively related to employees’ job performance (model 4: B = -0.66, p < 0.001); enacting is still positively related to employees’ job performance (model 4: B = 0.10, p = 0.046); the relationship between reflecting and employees’ job performance is no longer significant. These results suggest that the proactive process configuration can better predict employees’ job performance than the individual elements of proactive process. Besides, the proactive process configuration can predict employees’ job performance after controlling for proactive personality. Model 5-8 in Table 5 show that, when considering the effect of the four elements of proactive process, neuroticism, and proactive process configuration on emotional exhaustion, neuroticism is positively related to employees’ emotional exhaustion (model 7: B = 0.32, p < 0.001; model 8: B = 0.32, p < 0.001). However, the individual elements of proactive process and proactive process configuration cannot significantly impact employees’ emotional exhaustion. Model 3-5 in Table 6 reveal the effect of the four elements of proactive process, environmental uncertainty, proactive personality, and proactive process configuration on employees’ job performance. The results indicate that the individual elements and environmental uncertainty cannot be significantly related to employees’ job performance. However, proactive personality is positively related to employees’ job performance (model 3: B = 0.21, p = 0.003; model 4: B = 0.20, p = 0.004; model 5: B = 0.30, p = 0.001). Additionally, configuration 2a (model 3: B = 0.42, p = 0.007) and configuration 3b (model 4: B = 0.43, p = 0.003) are both positively related to employees’ job performance. Conversely, configuration 2c is negatively related to employees’ job performance (model 5: B = -0.38, p < 0.001). Model 8-9 in Table 6 present the effect of the four elements in proactive process, felt trust, neuroticism, and proactive process configuration on employees’ emotional exhaustion. The results indicate that the individual elements of proactive process cannot be significantly related to employees’ emotional exhaustion. However, felt trust is negatively related to employees’ emotional exhaustion (model 8: B = -0.27, p < 0.001; model 9: B = -0.20, p < 0.001). Additionally, neuroticism is positively related to employees’ emotional exhaustion (model 8: B = 0.30, p < 0.001; model 9: B = 0.29, p < 0.001). Besides, configuration 2a is negatively related to employees’ emotional exhaustion after controlling for other variables (model 9: B = -0.43, p = 0.043). The findings reveal that: 1) a configuration with high level of enacting alone is not sufficient for generating high job performance; 2) a configuration with low (high) levels on all four elements is sufficient for producing low (high) job performance regardless of the level of environmental uncertainty; 3) when environmental uncertainty is high, a configuration with high level of envisioning, planning, and reflecting is sufficient for producing high job performance regardless of the level of enacting; 4) a configuration with high level of enacting and low level of envisioning, planning, and reflecting is sufficient for generating high emotional exhaustion; 5) when felt trust is high, a configuration with high level of envisioning, planning, and reflecting is sufficient for producing low emotional exhaustion regardless of the level of enacting; 6) in general, after controlling for the unique effects of each element and individual difference, the configuration membership can still predict employees' job performance and emotional exhaustion whereas elements in a proactive process cannot. Our research contributes to the literature on proactive behavior. First, our research introduces a novel perspective on the process of proactive behavior. Prior studies on the effectiveness of proactive behavior have focused solely on the overt behavioral element of proactive behavior without considering the impact of covert behavioral elements and the synergistic effect of all elements in a proactive process. The process perspective of proactive behavior provides a fresh window onto why the consequences of proactive behaviors vary between employees and studies. Second, our research advances the understanding of proactive behavior by revealing the configural effects of four elements in a proactive process on employees’ outcomes, and further develops the theory about proactive process by incorporating contextual factors. The configural perspective emphasizes the holistic nature, and scholars have always emphasized that individuals interact with their environment (e.g., Bandura, 2001; Frese & Zapf, 1994; Zacher & Frese, 2018). Our findings highlight the significance of considering contextual factors, such as environmental uncertainty and felt trust, in the theoretical framework of proactive process and demonstrate their impact on the configural effects of the proactive process. Lastly, our research demonstrates the value of a configural approach in studying process-related issues in proactive behavior research and lays the foundation for future research on proactive process. This approach provides a more comprehensive understanding of proactive behavior and its outcomes. Figures and Tables | References | Related Articles | Metrics

Select

Effects of coworker anger expression on leader emergence: The mediating roles of perceived warmth and competence and the compensating effect of anger apology JIANG Xuting, WU Xiaoyue, FAN Xueling, HE Wei 2023, 55 (5):  812-830.  doi: 10.3724/SP.J.1041.2023.00812 Abstract ( 395 )   HTML ( 23 )   PDF (165KB) ( 279 )   Although previous research has paid much attention to examining whether leader anger expression is effective in enhancing leadership effectiveness, the social consequences of employee anger expression are underexamined. Integrating the stereotype content model with implicit leadership theory, we propose that appropriate anger expression, compared with suppressed anger, has ambivalent effects on leader emergence by increasing coworkers’ perceived competence of the expresser while decreasing coworkers’ perceived warmth of the expresser. In addition, appropriate anger expression, compared with deviant anger expression, is theorized to positively affect leader emergence by increasing coworkers’ perceived competence and warmth of the expresser. We further propose that apology after anger expression (anger apology) is likely to benefit leader emergence by repairing coworkers’ perceived warmth of the expresser. We conducted two online scenario-based experiments (Study 1 and 3) and two field surveys (Study 2 and 4) to test our research hypotheses. In Study 1, we employed a two (type of anger expression: expressed vs. deviant) by two (anger apology: yes vs. no) between-subjects experimental design, with a silent anger condition (i.e., no anger expression and thus no anger apology) as the control group. The sample consisted of 279 full-time Chinese employees recruited via an online survey panel (Sojump.com). To replicate the findings in Study 1, we conducted a critical incident technique study (Study 2), with a sample of 200 full-time employees recruited via the same panel used in Study 1. Participants were asked to recall and describe a workplace incident of coworker anger expression and then to evaluate their perceptions of competence, warmth, and the likelihood of leader emergence of the expresser. To reconcile some controversial findings in the two studies, we conducted Study 3 (a sample of 354 full-time employees recruited online) to provide a more nuanced examination of the effects of different types of anger expression. Specifically, we employed a three (type of anger expression: muted anger, appropriate anger expression vs. deviant anger expression) by two (anger account: other-orientation vs. self-interest) between-subjects experimental design with an additional condition of silent anger. To further replicate our findings and enhance the external validity, we conducted a field study (Study 4) by collecting two-wave data from 248 full-time employees from a Western online survey panel (Prolific.com). In total, empirical results from four studies (see Table 1, 2, and 3)1 suggested that, compared with deviant anger expression, appropriate anger expression positively affected the likelihood of the expresser’s leader emergence by enhancing observers’ perceived competence and warmth of the expresser. However, the effects of appropriate anger expression, compared with anger suppression, were found to be contingent upon the type of anger suppression (silent vs. muted anger) and the cultural context (Chinese samples in Studies 1-3 vs. Western sample in Study 4). Moreover, anger apology was demonstrated as an effective relationship repair strategy that can increase observer perceived leader emergence of the anger expresser. These findings contribute to anger expression literature by shifting the focus from how leader anger expression affects leadership effectiveness to the social consequences of employees’ anger expression on leader emergence. Additionally, we contribute to implicit leadership theory and the dual threshold model of anger by testing and extending their core theoretical arguments in the context of coworker anger expression in the workplace. Finally, we develop a new construct of anger apology and examine its compensating effects for anger expression, providing new avenues for future research on the social functions of anger expression. Figures and Tables | References | Related Articles | Metrics

Select

The effects of daily supervisor negative feedback on employee creativity DONG Niannian, YIN Kui, XING Lu, SUN Xin, DONG Yanan 2023, 55 (5):  831-843.  doi: 10.3724/SP.J.1041.2023.00831 Abstract ( 256 )   HTML ( 17 )   PDF (279KB) ( 236 )   Previous findings regarding the impact of supervisor negative feedback on employee creativity have seemingly been inconsistent. Researchers have reported positive, negative, and nonsignificant relationships between supervisor negative feedback and employee creativity. The present study aims to explore the possibility that supervisor negative feedback has short-lived impacts on employee creativity. Drawing from feedback intervention theory, we propose that proving goal orientation moderates the indirect effect of daily supervisor negative feedback on employee next-day creativity through problem-solving pondering at night such that this effect is stronger for individuals with higher levels of proving goal orientation. In addition, we suggest that avoiding goal orientation moderates the indirect effect of daily supervisor negative feedback on employee next-day creativity through affective rumination at night such that this effect is stronger for individuals with higher levels of avoiding goal orientation. We conducted a field study using experience sampling methodology to collect data from employees of a design institute in northern China. The questionnaire survey process included an initial one-time entry survey and daily surveys administered over a period of two weeks. One week before the start of the daily surveys, participants reported their proving goal orientation, avoiding goal orientation, and demographic information. During the two-week daily survey period, participants assessed daily supervisor negative feedback and daily creativity at 5:30 p.m. and rated problem-solving pondering and affective rumination at 8:30 p.m. each evening. The final sample included 716 usable observations collected from 95 employees. To test the proposed hypotheses, we conducted two-level path-analyses using Mplus 8.0 and performed a Monte Carlo simulation procedure using R software. Table 1 reports means, standard deviations, and correlations for all studied variables. As shown in Table 2, the cross-level daily supervisor negative feedback x proving goal orientation interaction term was positively related to problem-solving pondering at night (γ = 0.16, SE= 0.06, p= 0.005). Figure 1 illustrates this interaction pattern. The relationship between daily supervisor negative feedback and problem-solving pondering at night was positive when proving goal orientation was high (γ = 0.14, t= 2.37, p= 0.018). However, this relationship was not significant when proving goal orientation was low (γ = −0.08, t= −1.47, p= 0.14). Furthermore, for employees with high levels of proving goal orientation, daily supervisor negative feedback promoted their creativity the next day by activating their problem-solving pondering at night (indirect effect = 0.022, 95% CI [0.003, 0.050]). However, for employees with low levels of proving goal orientation, this indirect effect was not significant (indirect effect = −0.012, 95% CI [−0.033, 0.004]). In addition, the cross-level daily supervisor negative feedback x avoiding goal orientation interaction term was positively related to affective rumination at night (γ = 0.14, SE= 0.08, p= 0.069). Figure 2 illustrates this interaction pattern. The relationship between daily supervisor negative feedback and affective rumination at night was positive when avoiding goal orientation was high (γ = 0.17, t= 2.10, p= 0.035). However, this relationship was not significant when avoiding goal orientation was low (γ = −0.02, t= −0.27, p= 0.79). For employees with high levels of avoiding goal orientation, daily supervisor negative feedback inhibited their creativity the next day by eliciting their affective rumination at night (indirect effect = −0.017, 90% CI [−0.036, −0.002]). However, for employees with low levels of avoiding goal orientation, this indirect effect was not significant (indirect effect = 0.002, 90% CI [−0.009, 0.014]). The current study makes several theoretical contributions. First, we adopt a dynamic perspective to capture the within-person variance in creativity resulting from daily fluctuations in supervisor negative feedback. Second, this study enriches feedback intervention theory by exploring the mediating roles of problem-solving pondering and affective rumination in the link of supervisor negative feedback with employee creativity. Third, the present study reconciles the conflicting findings of previous research by demonstrating the differential effects of daily supervisor negative feedback on employees with different goal orientations. Figures and Tables | References | Related Articles | Metrics

Select

Reinforcement and extinction of unethical pro-supervisor behavior: Based on the perspective of supervisor response FU Bo, PENG Jian, LIANG Xiaojie, CHEN Lifang, YU Guilan 2023, 55 (5):  844-860.  doi: 10.3724/SP.J.1041.2023.00844 Abstract ( 191 )   HTML ( 12 )   PDF (304KB) ( 175 )   Unethical pro-supervisor behavior refers to actions that are intended to promote the effective functioning of leaders and violate core societal values, mores, laws, or standards of proper conduct. Although subordinates’ unethical pro-supervisor behavior protects the personal interest of supervisors in the short term, it can be detrimental to the reputation of the supervisor and the company in the long term, thus hindering the high-quality development of the organization. Existing research has devoted considerable efforts to the antecedents of unethical pro-supervisor behavior. However, few studies have explored the consequences of unethical pro-supervisor behavior, which leads to an unanswered research question: will a subordinate engaging in unethical pro-supervisor behavior persist in this behavior in the future? This study aims to investigate supervisors’ responses to subordinates’ unethical pro-supervisor behavior and how these responses shape subsequent unethical pro-supervisor behavior. Drawing on the “Bao” theory, we proposed that supervisors have two paradoxical responses (gratitude-driven resource rewards versus guilt-driven punishment) to their subordinates’ unethical pro-supervisor behavior, which depends on supervisors’ integrity. Supervisors with high levels of integrity will respond to their subordinates who engage in unethical pro-supervisor behavior with guilt-driven punishment (a negative “Bao”), which reduces subordinates’ subsequent unethical pro-supervisor behavior. In contrast, supervisors with low levels of integrity will respond to their subordinates who engage in unethical pro-supervisor behavior with gratitude-driven resource rewards (a positive “Bao”), which increases subordinates’ subsequent unethical pro-supervisor behavior. We conducted three studies. In Study 1, we conducted a scenario-based experiment to explore initial evidence for our hypotheses. In the scenario-based experiment, 120 pairs of subjects played the supervisor role and employee role. Table 1 showed the result of descriptive statistics and correlations for Study 1, which initially supported our research hypotheses. The result of Table 2 showed that when the level of supervisors’ integrity was lower, subordinates’ unethical pro-supervisor behavior increased supervisors’ resource rewards through supervisors’ gratitude, which, in turn, increased subordinates’ subsequent unethical pro-supervisor behavior (i.e., the positive change in UPSB). However, when the level of supervisors’ integrity was higher, subordinates’ unethical pro-supervisor behavior increased supervisors’ punishment through supervisors’ guilt, which, in turn, reduced subordinates’ subsequent unethical pro-supervisor behavior. Study 1 established the internal validity of our findings. However, its external validity is limited. Hence, we conducted a cross-sectional survey study (Study 2: four-wave data from 277 supervisor-subordinate dyads) and a diary survey study (Study 3: data from 87 supervisor-subordinate dyads over 10 working days). Table 3 and Table 5 showed the the result of descriptive statistics and correlations for Study 2 and Study 3 respectively. And Table 4 and Table 6 showed the result of path analysis for Study 2 and Study 3 respectively. Mplus 8.0 was used to analyze the data. Based on the results above, our hypotheses were supported again. This research has several theoretical implications. First, we introduced the perspective of supervisor response (i.e., supervisors’ emotional and behavioral responses) to examine the consequences of subordinates’ unethical pro-supervisor behavior, which advances the literature on unethical pro-supervisor behavior. Second, based on Bao’s theory, we explain how supervisors paradoxically respond to subordinates’ unethical pro-supervisor behavior. In doing so, this research contributes to the development of Chinese indigenous management theory. Third, we identified that the moral quality of supervisors (supervisors’ integrity) plays an important role in determining supervisors’ responses to subordinates’ unethical pro-supervisor behavior. Figures and Tables | References | Related Articles | Metrics