Oral Presentations

As a science that studies human beings, the success of psychology depends on its ability to draw conclusions on a wide range of cultural contexts. Previous studies indicate an issue of overrepresentation from certain cultural contexts. Yet, the extent and dynamics in national diversity remain poorly understood because 1) the sampled journals typically did not cover the diverse disciplines in psychology; 2) the motivating framework mostly relied on an oversimplified dichotomy of the Western, educated, industrialized, rich, and democratic (WEIRD) societies against the non-WEIRD societies, masking genuine differences within each category; and 3) the focus was on either authors or editorial boards, but not both, making it unclear whether an overrepresentation in editorial boards of a certain cultural context might be numerically justified based on the representation in authorship. 
Here, based on Journal Citation Reports in 2020, the top 10% journals that have an editorial board in 11 different disciplines in psychology were selected. Each journal was coded for the countries of origin of its authors, editors (including editor-in-chief, EIC in short), and owner (typically, an academic society or a publisher). In total, there were 68 journals, 5,654 editorial board members (including 82 EICs), and 37,801 authors. Results show that the journals are from US (75%), UK (13.2%), and other EU countries (11.8%). Likewise, EICs are from US (73.8%), UK (11.6%), other EU countries (9.2%), Canada (3.7%), and Australia (1.8%). Comparing authors and editors, we found that US editors (60.3%) outnumbered US authors (44.1%) by 16.2%, whereas this number was negative for most other countries. Although the journal country of origin is associated with positive editor–author asymmetry for US (20.0%), UK (9.1%), and other EU countries (1.8%), such bias persists to journals from outside the country only for US (4.6%), but not for UK (–2.8%) or other EU countries (–0.9%). These results suggest that, not only are journals, editors, and authors not diverse, but there is a systemic overrepresentation of the US at the expense of other countries.

Everyday life consists of a continuous stream of information, while the human memory system can discrete it into distinct events. These highly structured modes of memory clustering are important for us to remember abundant information. Previous event research mainly focused on examining discrete episodic memory, asking observers to make a binary choice. Although these approaches have been highly successful in examing the consequences for event recognition, they provide us little information about the memory details and how memory is shaped by these event structures. In our current study, we investigated whether and how event structure shapes detailed memory for individual items. We introduced context shift between trains of stimuli to create structured events and investigated how the event structure (Experiment 1) and the statistical summary within an event (Experiment 2) affect precision memory for an individual item. In Experiment 1, we found that event boundaries increased the memory error for items present at the boundary, indicating a boundary cost for memory quality. However, when we introduced a summary statistic (gist mean) among items within an event, this boundary cost disappeared. Instead, we found a robust bias toward the gist mean for both boundary and non-boundary items. More importantly, this bias was adaptive, which increased the memory performance for all items as well as overriding the boundary cost for items at the event boundary. In summary, we found a boundary cost for items’ memory quality, but summary statistical information can be processed and used to compensate for the cost and increase memory performance. These discoveries indicate an adaptive memory system, which makes use of structure information to scaffold our memory.

Writing about deepest feelings and thoughts relating to negative emotional events (i.e., expressive writing, EW) brings psychological and physical well-being improvements. Such beneficial effects are found in Western populations, where emotion expressions are encouraged as they affirm the individuality of the self and allow for emotion regulation. However, small beneficial effects are found in Chinese populations, where emotion expressions, especially negative emotions, are downplayed to maintain interpersonal harmony, and behavior rather than emotion regulation is encouraged. Nonetheless, there may exist individual differences in the efficacy of EW in Chinese context. Due to rapid economic growth and social changes in urban China, some Chinese may embrace a greater sense of independent self and are more emotionally expressive; other Chinese may embrace a greater sense of interdependent self, and are less emotionally expressive. EW may work better for Chinese who hold a greater sense of independent self. This study sought to examine the effects of culture and individual in the efficacy of EW in the Chinese context.
Fifty college students (aged 18-22) completed four weekly sessions of either EW or control writing (CW), with each session lasted for maximum 25 minutes. They completed a one-off self-construal measure and a battery of well-being measures pre- and post-writing. Writing length and emotional and cognitive terms were coded.
EW group wrote longer and referenced more emotional and cognitive terms than CW group. At the cultural level, there was no well-being improvement in EW group, although CW group demonstrated mixed effects (i.e., less negative affect and optimism). At the individual level, independent Chinese (Independent-to-interdependent-self ratio≥1) in EW group showed improved life satisfaction, and interdependent Chinese (Independent-to-interdependent-self ratio<1) in CW group showed decreased negative affect.
Findings underscored the importance of examining not just cultural effects, but also individual effects in understanding the efficacy of EW in Chinese.

We encounter a large amount of sensory information in our daily life. In such a complex environment, objects, people, and events can be detected quickly, identified precisely, and responded to appropriately, which largely benefits from the coordination among different modalities. To optimize such responses, we can attend selectively by focusing on relevant information and filtering out irrelevant information across modalities. Although it is hardly controversial that a multisensory environment does influence our responses, there is less agreement as to whether it helps or hinders and why. To address this issue, we used a modified flanker task, in which adults (N = 69) were asked to attend to either visual or auditory targets and categorize them while ignoring other distracting items. A Multinomial Processing Tree (MPT) model was used to decompose selective attention into focusing and filtering components. We found that both the ability to focus on the target and filter out irrelevant information were needed in the visual and auditory categorization, but the underlying mechanisms were different. Specifically, auditory information tends to be prioritized: We allocate more attention to it automatically, regardless of whether it is a target or a distractor. In contrast, visual information is processed flexibly: We can focus on such information voluntarily and filter it out effectively. These novel findings point to the critical but differential role of selective attention in multisensory categorization, which offers a promising approach to understand the mechanisms of multisensory processing.

Testosterone effect on empathy is still in debate [1, 2, 3]. Given that empathic responses to others’ emotional experiences depend strongly upon top–down controlled mechanism of attention [4, 5], it is likely that testosterone effect on empathy is moderated by attention manipulation. To test this hypothesis, this study assessed how testosterone influences empathic responses during pain empathy paradigms in which attention either directed toward or withdrawn from pain of another. Using a double-blind, placebo-controlled within-participant design, we administered a single dose of either testosterone or placebo to 32 healthy male volunteers. In experiment 1, two-choice discrimination tasks, including Pain Judgment and Hands Counting tasks, were employed to assess top–down attention modulation of empathy for pain. During the Pain Judgment task, empathic behavioral and electroencephalographic responses were comparable between two sessions. During the Hands Counting task, participants exhibited lower discrimination accuracy, but greater empathic late positive potential responses to the painful stimulations in the testosterone session than that in the placebo session. It suggests that testosterone enhanced empathic responses to task-irrelevant pain-related features in the stimulations, which empathic response interfered the stimulus discrimination in the task. In experiment 2, we adopted a Pain Rating task to assess testosterone effects on the affective and cognitive empathy. Whereas testosterone administration did not affect empathic ratings, significant empathic effects on the N1 and N2 component of event-related potentials were only observed in the testosterone session. Furthermore, compared with placebo session, spontaneous fronto-central α-oscillation amplitude recorded prior to the onset of empathic stimuli was lower in the testosterone session. Therefore, testosterone selectively enhanced the bottom-up affective processes of empathy for others’ pain. It could be arisen from the altered brain states, e.g., increased vigilance and arousal level after testosterone administration. Our findings have importance for our understanding of the psychobiology of empathy.

Programming and designing autonomous cars (ACs), which passengers would perceive to be safe and reliable, plays a crucial and indispensable role when human beings go further down the hands-free route. Existing literature has highlighted that the acceptance of the AC will increase if it drives in a human-like manner. However, literature presents no human-subject research focusing on passengers in a natural environment that examines whether the AC should behave in a human-like manner. In this paper, we conduct a Turing test of automated driving based on 69 passengers’ feedback in a real scenario. Specifically, after the process in which passengers ride in the AC (driven by either the human driver or AI driver) and infer the intentions of the driver (i.e., mentalizing), we collect passengers’ choices of response, i.e., the confidence in whether the driver was AI algorithm or not. The test results show that Level 4 ACs could pass the Turing test in all three stages with less than 50% accuracy. Furthermore, combining good old-fashioned signal detection theory (SDT) with the cutting-edge pre-trained language models (PLMs), we propose a computational model to understand cognitive mechanisms underlying passengers’ decision-making process in the Turing test. In this model, we use affective variability (AV), tested by modified Differential Emotions Scale (DES-IV) and transformed by PLM, as the signal strength. Though the observations of our current research are limited (no more than 70 observations in each stage), the experimental results and further analysis show that the greater AV that passengers have, the more likely they identify the driver as the AI algorithm. These findings suggest that future automated driving should improve the affective stability of passengers.

Drawing upon the views of humanistic psychology, justice can be a necessity for human growth. This study aimed to examine whether and how procedural justice at workplace is associated with mental health professionals’ psychological flourishing, an indicator of individuals’ general psychological wellbeing, by empirically testing (i) the association between procedural justice and psychological flourishing, (ii) the mediating role of emotional exhaustion, and (iii) the potential moderating effects of emotion regulation on such mediation among 195 Chinese mental health professionals (males = 31%; Mean age = 30 years) in Macao. Path analysis was conducted to test our hypothesized moderated mediation model. Results showed that emotional exhaustion partially mediated the positive association between procedural justice and psychological flourishing, whereas emotion regulation significantly diminished the effects of procedural injustice on emotional exhaustion. Our findings indicated that injustice at workplace might hamper individuals’ overall wellbeing by inducing emotional exhaustion at work. Yet such negative influence might be weakened by individuals’ emotion regulation ability.

Autism spectrum disorder (ASD) has been associated with a lack of pain empathy that is suggested to be grounded in first-hand pain experiences1-2. Autistic traits are subclinical levels of autism-like symptoms that are continuously distributed within the general population3. It is likely that autistic traits are negatively associated with both pain empathy and pain sensitivity among neurotypical adults, and that first-hand pain sensitivity would account for the link between autistic trait and pain empathy. To test these hypotheses, we recruited a sample of healthy college students (n = 100) whose autistic traits were evaluated using Autism Spectrum Quotient (AQ). Their cognitive and emotional empathy for others’ pain was assessed using ratings of perceived pain, personal distress, and empathic concern in response to reading or observing scenarios about another’s physical pain (Figure 1). Their pain sensitivity was assessed by measuring pain thresholds, pain tolerance, and pain-intensity ratings in response to experimental pain (pressure, heat, and cold pain, Figure 1). After controlling for alexithymia, individual variations in autistic traits correlated with both pain empathy and pain sensitivity; individuals with higher AQ scores provided lower cognitive and emotional empathic ratings of others’ pain and exhibited lower sensitivity to first-hand suprathreshold heat and cold pain. Analysis further revealed that first-hand pain sensitivity fully mediated the link between autistic traits and empathy for others’ pain (Figure 2). Given the similarity in behaviors between individuals with ASD and those with high levels of autistic traits across sensory, affective, cognitive, and social domains, our results highlighted the role of first-hand pain sensitivity in the impairment of empathy for others’ pain in ASD. It supports the idea that early sensory dysregulation in ASD could cascade into social deficits across development, and imply the necessity of pain sensitivity assessments in the diagnosis and treatment of ASD.

The present meta-analytic review examined the associations between different forms of trauma and mental disorders among prisoners and ex-prisoners. Studies published from 1998 to March 31 2021 were identified by searching PsycINFO, PubMed, Medline and Web of Science. Data were meta-analyzed using a random-effect model. Moderator and mediator analyses were conducted. The protocol was registered in PROSPERO (CRD42020181587). We identified 62 studies (50 non-duplicated population) with 15,115 (97.86%) prisoners and 330 (2.14%) ex-prisoners in 16 countries. A multi-level meta-analysis found that overall trauma was positively associated with more diagnoses or symptoms of mental disorders (Zr=0.198, 95% CI=[0.167, 0.229], p<.001). Stronger effect sizes were found between childhood trauma (Zr=0.357, 95% CI=[0.147, 0.568], p<.001) and sexual trauma (Zr=0.326, 95% CI=[0.216, 0.435], p<.001) and stress-related disorders. Multilevel moderator analysis showed that effect size was stronger in imprisonment trauma (β=0.247, 95% CI=[0.177, 0.316], p<.001), mixed trauma (β=0.234, 95% CI=[0.196, 0.272], p<.01), and stress-related disorders (β=0.261, 95% CI=[0.214, 0.307], p<.001). Associations between trauma and mental disorders were mediated by social support but not coping. Our findings provide an evidence base for future research on the impact of trauma and inform assessment and intervention in correctional settings.

The research aims are two. This study compared the social-emotional difficulty, linguistic skills, and reading comprehension of left-behind children (LBC) and non-left-behind children (NLBC), and examined the direct effect of social-emotional difficulty on reading comprehension and the indirect effect of social-emotional difficulty on reading comprehension through linguistic skills. In the present study, 658 Chinese third graders ( 408 LBC, 250 NLBC; 325 girls, 333 boys; mean age = 8.67 years) completed the tests of morphological awareness, syntactic awareness, vocabulary knowledge, and reading comprehension as well as the measures of the fear of negative evaluation, social avoidance and distress, loneliness, parental education, and family income. Manova analysis showed that LBC children lagged behind their NLBC peers on syntactic awareness. Latent means comparison indicated that no significant differences between the two groups in social-emotional difficulty and linguistic skills. Multigroup structural equation analyses to test the direct effect of social-emotional difficulty showed that, for both groups, there was a direct and positive effect (p = .09) of social-emotional difficulty on reading comprehension after controlling the linguistic skills, family income, and parental education. Multigroup structural equation analyses to test the indirect effect of social-emotional difficulty showed that, for both groups, there was indirect and negative effect (pNLBC = .02, pLBC < .001) of social-emotional difficulty on reading comprehension through linguistic skills, while the residual direct effect of social-emotional difficulty is positive (p = .09). The two groups differed in that the social-emotional difficulty more strongly predicted linguistic skills for the LBC than the NLBC (p = .06). The implications and limitations are discussed.

Objective: Using an intensive longitudinal study design, the current five-wave study aimed to investigate the trajectories of depression and anxiety symptoms and the predictive roles of self-compassion among college students across a year during the pandemic.
Methods: In February 2020 (baseline), we invited Chinese college students to participate in an online survey to collect data on depression and anxiety symptoms, self-compassion, and sociodemographic variables. Follow-up studies of depression and anxiety symptoms were conducted in May, August, November 2020, and February 2021, respectively. Finally, 494 college students completed five measurements and provided valid data. Latent growth mixed models were used to explore the trajectories of depression and anxiety symptoms. Then, logistic regression models were used to test the predictive effects of self-compassion on the trajectories after adjusting for sociodemographic variables.
Results: There were four latent trajectories for depression symptoms: healthy group (normal throughout the study, 75.7% of participants), persistence group (moderate throughout the study, 9.9%), remission group (initially moderate then declining to mild, 9.0%), and deterioration group (initially mild then increasing to moderate, 7.2%). Three latent trajectories were identified for anxiety symptoms: healthy group (normal throughout the study, 14.0%), remission group (initially mild then declining to normal, 8.7%), and deterioration group (initially normal then increasing to mild, 6.1%). Compared to the healthy group, participants with lower levels of self-compassion were more likely to appear in the persistence group, the remission group, and the deterioration group of depression symptoms, as well as the remission group and the deterioration group of anxiety symptoms.
Conclusions: Group heterogeneity exists in the depression and anxiety trajectories of college students during the pandemic. Improving the levels of self-compassion may help to prevent and alleviate the development of mental health problems among youth.

The COVID-19 pandemic brings about various stressors for adolescents (e.g., school closure), highlighting coping (i.e., one’s efforts to regulate emotion, cognition, physiology, behavior, and the environment in reaction to stressful situations) as a critical factor that may contribute to individual differences among adolescents in emotional maladjustment during the challenging time of the pandemic. The current study investigated the implications of adolescents’ coping with pandemic-related stress for their emotional maladjustment and the role of perceived parental reactions to adolescent’s negative emotions in foreshadowing adolescents’ coping. Two hundred and thirteen Chinese adolescents (103 females and 110 males; mean age = 12.18 years) completed a survey one year before (Wave 1 in December 2018) and five months after the COVID-19 outbreak in China (Wave 2 in May 2020). Path analysis revealed that after controlling for adolescents’ emotional maladjustment at Wave 1, perceived parental supportive reactions to adolescents’ negative emotions at Wave 1 predicted adolescents’ greater use of approach coping (e.g., problem reappraisal, problem solving, or support seeking) and less use of avoidance coping (e.g., denying, ignoring, emotional venting, or wishful thinking) at Wave 2, which in turn, was associated with less emotional maladjustment at Wave 2; conversely, perceived parental nonsupportive reactions at Wave 1 predicted adolescents’ greater use of avoidance coping at Wave 2, which in turn, was associated with greater emotional maladjustment at Wave 2. These results were similar for mothers and fathers. Theoretically, the findings support the applicability of the approach/avoidance coping model among Chinese adolescents and demonstrate the important role of parental emotion socialization in shaping adolescents’ coping that may further make significant differences in adolescents’ emotional well-being. Practically, two-folded interventions that both cultivate adaptive coping in adolescents and optimize parents’ reactions to adolescents’ negative emotions are suggested for endeavors to uphold adolescents’ well-being during and after the pandemic.

Global climate change is proposed as one of the greatest threats for humanity now. Thus, how to increase individual pro-environmental behavior to mitigate climate change is one of the collective agendas in many countries. Understanding and promoting pro-environmental behaviors should consider not only the role of personal characteristics but also the influence of the surrounding contexts such as the family and cultural context. The present study proposed a global view to explore the dyadic influence of parents and adolescents in each other’s pro-environmental behaviors across 14 societies. We evaluated whether their own (actor effects) and each other’s (partner effects) self-efficacy in climate change are linked with their pro-environmental behaviors between parents and their children. We also explored the moderating role of two important societal level moderators, i.e., societal power distance and societal relational mobility, which are closely related to family relationships, on the parent-child dyadic effects. We tested our hypothesis with an international dataset, i.e., Program for International Student Assessment 2018, which involved 185,468 participants (i.e., 92,734 parent-child dyads) from 14 societies in the present study. Conducting a series of multilevel Actor-Partner Interdependence Models (APIM) and their cross-level interaction with societal power distance and societal relational mobility, we found that within a family context, parents’ and children’s self-efficacy in climate change was positively related to their own pro-environmental behaviors; meanwhile, parents’ self-efficacy in climate change was positively related to children’s pro-environmental behaviors and it is also true vice versa. More importantly, these patterns are stronger among societies with low levels of power distance and relational mobility. The present study indicates the importance of cultivating self-efficacy in climate change to increase pro-environmental behaviors across societies varying in different characteristics. It also makes a call for examining both family and cultural contexts for better understanding and promoting individual pro-environmental behavior worldwide.

The natural environment is rapidly deteriorating (UNEP, 2018; WHO, 2018), which is a growing threat to humanity. Micro-level evidence showed that the increasing environmental degradation was driven by rising economic inequality globally (e.g., Boyce, 1994; Holland et al., 2009; Scruggs, 1998). It was suggested that economic inequality might affect the environment by changing individuals’ behaviors because the micro-level behavior was a major contributor to macro-level environmental degradation (Berthe & Elie, 2015; Hamann et al., 2018; Islam, 2015). However, few of the previous studies examined how, why, and when economic inequality affected individuals’ pro-environmental behaviors. In this research, I suggest that economic inequality increases feelings of relative deprivation, which in turn decreases pro-environmental behaviors. Besides, the effect of economic inequality varies across different cultures, which can be explained by the lay theories of agency.
Six studies are planned to test the hypotheses. Study 1 and Study 2 preliminarily verified the negative associations between economic inequality and pro-environmental behaviors, as well as perceived social norms in 30 countries (Study 1) and in the United States (Study 2). And the negative associations were weaker in less collectivistic countries or states. Study 3 found that economic inequality causally decreased environmental donation intentions through inducing relative deprivation. Study 4 further demonstrated that the effects of economic inequality on relative deprivation, responsibility diffusion, and pro-environmental behaviors were weaker for individuals endorsing lower group (vs. individual) agency belief. Studies 5 and 6 are designed to test whether there are significant cross-cultural differences, and whether such differences can be explained by the lay theories of agency. To this end, I will test the moderating role of the lay theories of agency in the effect of economic inequality on pro-environmental behaviors.

One of the basic goals of raising socialist cultural-ethical standards in China is to cultivate citizens’ ideals, moral virtues, culture, and discipline. The four aspects of cultivation may refer to an attainment of eudaimonia that focuses on higher-order personal concepts, such as self-actualization and meaning. The present research aims to investigate the effect of ostracism, a ubiquitous daily interpersonal experience, on eudaimonia. Because ostracism thwarts people’s belonging need and makes them vulnerable to threats, it urges ostracized people to distribute available resources in the service of belonging and self-protection. Consequently, ostracized people may lack sufficient psychosocial resources to seek higher-order needs, thereby impairing their pursuit of eudaimonia.
Adopting a multimethod approach, six studies (total valid N = 1334), including one cross-sectional study, one 3-wave longitudinal study, three experiments, and one 7-day daily diary study, were conducted to examine the effect of ostracism on three categories of eudaimonia (i.e., motivation, behavior, and well-being). The results showed that ostracism reduces psychosocial resources, thereby lowering eudaimonic motivation. Moreover, future time perspective moderates the effect of ostracism on eudaimonic motivation, such that the effect is only observed among individuals low in future time perspective. Furthermore, post-ostracism decreases in eudaimonic motivation have implications on eudaimonic behavior (i.e., inconspicuous benevolence) and well-being. Specifically, ostracism decreases inconspicuous benevolence and eudaimonic well-being through reduced eudaimonic motivation.
The present findings provide the first empirical evidence that ostracism lowers eudaimonic motivation, thereby decreasing eudaimonic behavior and well-being. They also illustrate higher future time perspective as the protective factor and psychosocial resources as the psychological mechanisms underlying the effect of ostracism on eudaimonia. These findings advance existing theories and work by explicating how ostracism influences people’s pursuit of eudaimonia and identifying the potential antecedent of eudaimonia and ways for enhancing eudaimonic motivation, behavior, and well-being following ostracism.

Until recently, researchers in the field of objectification have primarily focused on its outcomes from targets’ perspectives (i.e., the objectified person). However, we know little about the consequences of being a source of objectification (i.e., the objectifier). In the current six studies (total N = 1048), we examined the hypothesis that objectification increases the source’s sense of power. The first experimental study identified that people who objectified others are perceived as more powerful by others (Study 1). Studies 2-6 have consistently demonstrated the link between objectification and power self-attribution. The correlational study revealed that people’s willingness to objectify others was positively associated with their desire for power (Study 2). In four experimental studies, we found that people who objectified others have a stronger subjective sense of power (Studies 3-6). Moreover, Study 5 showed that the initial objectification leads to an increased sense of power, which in turn promotes continued objectification. As for the outcomes of objectification, Study 6 revealed that the executors of objectification were more likely to engage in self-serving behavior. These results demonstrate a clear bi-directional relationship between objectification and power. Taken together, our findings deepen our understanding of objectification from the source’s perspective.

Substantial evidence indicates that feedback processing is not only determined by the valence of the feedback, but also highly dependent on contextual factors. However, whether the process of outcome evaluation would be influenced by prior outcome history has been widely disregarded.To investigate this issue, we conducted two experiments to test the different influence of preceding feedback on brain responses of following feedback processing between inner-trial and inter-trial. We designed a modified version of gambling task that each trial was associated with two consequences. In experiment 1, participants played a gambling task, in which each choice was associated with two consequences indicating their performance from two dimensions within a trial respectively. In experiment 2, participants were firstly instructed to finish two gambling tasks within a trial and then received two subsequent feedbacks indicating their performance from two tasks respectively within a trial. The behavioral results in both of two experiments showed that participants tend to repeat their decision as in the previous choice if its feedback was positive and switch to choose another option if the feedback was negative. Moreover, the ERP results in both experiments revealed that RewP amplitude was more positive during the processing of win feedback when it was preceded by wins at feedback1 compared to win feedback preceded by losses at feedback1 in inner-trial condition. However, in inter-trial condition, no influence of preceding feedback was found on RewP activity in experiment1. Taken together, our study revealed that the prior feedback will lead to different motivation of next outcome and thus influencing the perception of the next outcome in our mind in inner-trial condition. This effect can be manipulated by changes in the experiment conditions. Further, the findings demonstrate that neural systems involved in reward processing dynamically and continuously integrate preceding feedback for the evaluation of present feedback.

A core symptom of borderline personality disorder (BPD) is impairment in intimacy, which is marked by unstable and conflicted romantic relationships together with fear of abandonment. BPD has long been viewed as a female predominate mental illness, though epidemiological evidence from community samples suggested that the prevalence rates of BPD are the same amongst males and females. Few studies have investigated gender differences in the behavioral manifestations in BPD’s romantic dysfunctions. This study focused on responsiveness, a crucial variable in relationship research that reflects the extent to which an individual’s behaviors express understanding, validation, and caring for a partner, as previous studies indicated that males and females enact responsiveness towards partners in different ways. We examined the gender differences in the association between people’s BPD features and their responsiveness during romantic conflict in a dyadic design. Dating couples (N = 84 pairs) engaged in a 10-min videotaped discussion about an unsolved conflict and then reviewed the taped discussion. For each 30-second segment of interaction, participants rated responsiveness and insecurity (i.e., I/my partner felt insecure) of themselves and their partners. Trained observers also rated each participant’s displayed responsiveness and insecurity in each segment. The results showed that though in both males and females, high-BPD-feature individuals experienced more insecurity during romantic conflict than low-BPD-feature individuals, actor effects of BPD features on responsiveness only exist in females. To be specific, self-report, partner-report, and observer-report outcomes consistently revealed that BPD features are associated with reduced enacted responsiveness in females, but not in males. Meanwhile, BPD features are also associated with reduced perceived partner responsiveness in females, but not in males. This study provided evidence for the gender differences in manifestations of borderline personality pathology and implied the significance of gender effect in psychopathology.

What is the manipulation unit of working memory (WM)? This question is essential for understanding how the mind works by revealing the working nature of WM. As two important theories in the field of visual attention and WM storage, object theory and Boolean map theory draw forth two hypotheses on this format question separately. (a) The object-based hypothesis predicts that it takes longer to manipulate more objects in WM, while (b) the Boolean-map-based hypothesis expects longer times when manipulating more Boolean maps, a data structure that can be consciously perceived in one instant. In Experiment 1, participants were asked to dye a subset of stimuli in their WM into two different colors then finish a change detection task. The results showed that manipulating colors from one Boolean map was faster than that from two different Boolean maps, supporting the Boolean-map-based hypothesis. In Experiments 2-3, participants were asked to move a subset of stimuli in WM. The results support the Boolean-map-based hypothesis that participants’ manipulating times differed when manipulating different numbers of Boolean maps. In conclusion, we adopted three novel manipulating tasks and found that Boolean maps are the manipulation unit of WM, instead of objects.

Autobiographical memories (AMs) are memories of one’s personally significant life experiences. AMs vary in phenomenology, including length, specific details and valence. AMs also vary in social functions, including building intimacy with others, teaching and informing others and eliciting from and/or providing empathy to others. Individuals tend to engage in conversations and share AMs with social others. Face-to-face conversations have been the conventional mode of memory-sharing. Yet in this digital age, it is increasingly common for memory-sharing to be conversed via mobile messaging applications such as WeChat, especially in China. Sharing memories via texting on WeChat may require greater cognitive load and motivational effort than conversing face-to-face. Accordingly, memory-sharing via WeChat may be shorter, with less specific details and more positive compared to face-to-face. More importantly, social functions may be served during memory-sharing be it via WeChat or face-to-face. Although, the social functions may be served to a lesser extent during memory-sharing via WeChat compared to face-to-face. This study examined the phenomenology and social functions of memory-sharing in the face-to-face and WeChat contexts.
Participants were 40 college students (aged 18-25; 33 females) and his/her friend (aged 18-25; 30 females). Participant-friend pairs either shared two AMs via WeChat (20 pairs) or face-to-face (20 pairs). After memory-sharing, participant-friend pairs responded to questionnaires assessing phenomenology and social functions of memory-sharing. Memory-sharing via WeChat was shorter and with less specific details than face-to-face. Memory-sharing via WeChat and face-to-face were equally positive and served empathy function. Although, memory-sharing via WeChat served intimacy and teaching/informing functions to a lesser extent than face-to-face.
Findings underscore that the act of human memory-sharing is taking place beyond the conventional face-to-face context to include the digital context. Albeit less than optimal phenomenological characteristics and social functions served compared to face-to-face memory-sharing, digitized memory-sharing does serve some important social functions that matter.

Visual working memory (vWM) performance is enhanced when a memorized object is cued after encoding. This so-called retro-cue effect is typically observed with a predictive (80% valid), retrospective cue. In cases where the cue conveys social signals, such as gaze, a nonpredictive (50% valid), retrospective cue can similarly enhance internal memory representations. However, it remains unclear whether this effect of gaze on vWM simply reflects reflexive attention shift, or intention processing. To investigate this issue, we place a baffle on each side of gaze cues during the retention interval of a change-detection task, and examine whether gaze cues can still enhance working memory even when the baffle blocks the view of gaze. In experiment 1, we found that with baffles there was no difference in participants’ performance between valid gaze cues and invalid gaze cues. But when the baffles became transparent, the enhancement effect of gaze cues occured, suggesting that intention processing was necessary for this effect. Furthermore, the retro-cue effect of gaze disappeared when the face were inverted, also indicating the necessity of social information processing (experiment 2). And this enhancement effect was not caused by low-level motion cues either (experiment 3). In experiment 4, we examined whether emotion could modulate this retro-cue effect of gaze and found the same trend of exp 1. In sum, these results clearly showed that the retro-cue effect reflected intention processing of gaze rather than reflexive attention shift.

Background: Overweight and obesity among children and adolescents are a major public health problem worldwide. Obesity is the result of an unbalanced energy intake and expenditure many and is related to factors including genetic, metabolic, behavioral components. One of the factors, recently proposed to be closely related to obesity is appetitive traits. Previous studies used regression or correlation to explore the relationships between appetitive traits and weight status, however, limited research has explored the co-occurrences and interactions between appetitive traits, meanwhile, the interactions of appetitive traits on predicting childhood overweight and obesity among Chinese community haven’t been confirmed.
Methods: Appetitive traits were measured using the Chinese version of Adult Eating Behavior Questionnaire (C-AEBQ). Surveys from 2685 adolescents (1204 boys and 1481 girls) were retained in the data collection, with a mean age of 13.93 years (SD = 1.51). A decision tree model was trained and tested using machine learning method to identify the relative importance of appetitive traits in adolescents for predicting obesity.
Results: Satiety responsiveness, slowness in eating, and food responsiveness were the three most important appetitive traits for detecting obesity in Chinese adolescents. Meanwhile, co-occurrences and interactions in appetitive traits were found for detecting obesity, and difference between Chinese adolescent male and female were found in the model. The overall model based on appetitive traits had a sensitivity of .58 and a specificity of .62 for boys, while a sensitivity of.62 and a specificity of .53 for girls, in the test subsample.
Conclusions: The most important appetitive traits for predicting obesity for Chinese adolescents were satiety responsiveness, slowness in eating and food responsiveness. The results of this study can be further extended to provide guidance for identifying adolescents at potential risk of obesity and designing intervention at obesity.

Individuals’ implicit theories about the changeability of their own and others’ personal traits affect their social attitudes and behaviors. The present research tested the influence of individuals’ implicit theories on their prosocial behaviors. Study 1A (N = 1229) and 1B (N = 179) found a positive correlation between entity belief and trait empathy. But entity belief only positively correlated to the score of self-reported prosocial behaviors scale in Study 1A, not to the allocated amount in a dictator game in Study 1B. Study 2A (N = 167) and 2B (N = 124) examined the causal relationship by manipulating participants’ implicit theories with scientific articles. The results corroborated those in Study 1A and 1B, showing that entity theorists had higher intention to help in imaginary prosocial scenarios but didn’t allocate more in the dictator game. Study 3A (N = 67) and 3B (N = 1737) found that the effect of implicit theories on prosocial behaviors was moderated by prosocial cost. The differences between entity and incremental theorists’ help intention were larger in high-cost scenarios but smaller in low-cost scenarios. Across all studies, the mechanisms of empathy and trait inference were tested. Entity theorists had higher prosociality because they were more empathetic towards the victims and tended to view their behaviors as representative of their intrinsic moral character.

The aim of this study was to translate and validate a Chinese version of the revised Palatable Eating Motives Scale (C-PEMS-R) among Chinese adults. The revised Palatable Eating Motive Scale accesses four motives (Social, Coping, Enhancement, and Coping) of eating palatable food and drinks. We also investigated the relationships between the C-PEMS-R and BMI and disordered eating. A total of 773 participants including both community and college men and women were recruited from the Internet. The measures used the Palatable Eating Motives Scale – revised (PEMS-R), Short Form of the Eating Disorder Examination Questionnaire (EDE-QS), Loss of Control Over Eating Scale – brief (LOCES-B), and Kessler Psychological Distress Scale (K6). The Cronbach’s α for the C-PEMS-R was .93, .85, .88, .88, .85 for the total scale, social, coping, enhancement, and conformity subscales, respectively. From a sample of 66 participants who took the survey again, the intraclass correlation coefficient was .77 for the full scale, indicating adequate test-retest reliability. The four-factor structure of the PEMS-R was replicated in the current study with confirmatory factor analysis. The C-PEMS-R was found invariant across genders. The total scores of the C-PEMS-R were positively and significantly correlated with the scores of the EDE-QS (r = .43), the K6 (r = .36), and the LOCES-B (r = .59), indicating good convergent validity. However, there was no significant relationship between the C-PEMS-R and BMI. This study found adequate reliability and validity of the C-PEMS-R in Chinese adults, and the findings also suggest palatable eating motives as potentially important contributing factors to eating disturbances.

Researchers in psychology, education, and organizational behavior often encounter multilevel data with hierarchical structure (e.g., participants may cluster within communities, classes, or clinics). Ignoring the hierarchical structure of data would lead to a violation of the independence assumption of some models. To solve the statistical problems caused by this kind of observations, multilevel modeling is often conducted. However, in real studies, the sample sizes of level 1 and level 2 are often small in multilevel modeling. Traditional frequentist-based maximum likelihood (ML) approach, which relies on large-sample theory, might lead to problems in parameter estimation and model convergence in multilevel modeling with small samples. In contrast, Bayesian approach is often more advantageous in small samples, but it is also more susceptible to the subjective specification of priors. To investigate the potentially detrimental effects of inaccurate prior information on Bayesian approach and compare their performance with the traditional approaches, a series of simulations were conducted under the multilevel model framework with different dependent variable types, sample sizes and intraclass correlation coefficients (ICCs). In sum, the results revealed the devastating impacts of inaccurate prior information on Bayesian estimation, especially in the cases of larger ICC, smaller level 2 sample size, and smaller prior variance. When the dependent variable was non-normal or binary, these negative effects were more obvious. The present study investigated the impacts of inaccurate prior information on Bayesian estimation and provided advice on the specification of priors.

Keywords: Multilevel Modeling; Small Samples; Bayesian Modeling; Inaccurate Prior Information

Questionnaires are often applied to measure individual’s personality, intelligence, and other latent variable on the research fields of psychology, education and organization behavior. Confirmatory factor analysis (CFA) is often conducted to investigate the relationships between items in the questionnaire and the latent factors. Local independence is an essential assumption in latent variable modeling. But in reality, due to the settings of questionnaires’ content and restriction of testing environment, there are other factors influencing the relations between items, which is referred as local dependence. Bayesian structural equation modeling (BSEM) is a newly developed method, which can be used to deal with local dependence. Inverse Wishart distribution is a commonly used prior for the covariance matrix. To explore the effects of different prior variances in detecting the different levels of local dependence, a series of simulation studies and empirical examples are conducted for continuous data and dichotomous data respectively under the one-factor model with different sample sizes, item numbers and factor loadings. To sum up, for continuous data, larger prior variance of residual covariance results in less Type I error in detecting the local dependence. And for dichotomous data, prior variance of residual covariance mainly affects the estimation accuracy of the residual covariance parameters with real local dependence. In applied researches, different prior variances should be chosen cautiously according to different situations. The present study investigates the effect of BSEM with different prior variance in detecting different local dependence in depth, in the hope of providing a guidance for applied researchers.

Missing data is a common occurrence in mediation research. As such, the methods used to generate confidence intervals around the indirect effect should take into account missing data. While there are existing methods to handle missing data in mediation models, for example, bootstrapping coupled with full information maximum likelihood or multiple imputation, they are computationally intensive and may be difficult to use when dealing with complex mediation models. Previous studies have shown that the Monte Carlo method performs as well as nonparametric bootstrapping confidence intervals for the indirect effect in data with complete cases (see MacKinnon et al., 2004; Preacher & Selig, 2012; Tofighi & MacKinnon, 2016). We propose a simple, fast, and accurate two-step approach for generating confidence intervals for the indirect effect in the presence of missing data using the Monte Carlo method. In the first step, an appropriate method, for example, full information maximum likelihood, is used to estimate the regression coefficients and standard errors in a mediation model. In the second step, the Monte Carlo method is used to generate a simulated sampling distribution of the indirect effect of interest using the estimates in the first step. A confidence interval is constructed from the resulting sampling distribution. Results from a simulation study shows that the proposed method performs just as good as if not better the more computationally intensive alternatives. Implications of the results for applied research are discussed.

Background and aims: Despite the increasing attention being paid to the research about adolescents’ gambling disorder and its negative consequences worldwide, research about the psychometric evaluation of the screening tools for gambling disorder among Chinese adolescents is still lacking. This present study aimed to address this gap by evaluating the psychometric properties and cutoff score of the SOGS-RA among Chinese adolescent gamblers.

Methods: This study was conducted in secondary schools in Macao, China, with 258 cases (55.8% male; age = 11 to 19 years). Evaluation of the psychometric properties of the Chinese version of the SOGS-RA included confirmatory factor analysis and reliability as well as validity analysis. In addition, the cutoff score of this scale was also analyzed according to the DSM-5 diagnostic criteria for gambling classification, with the overall consideration of screening efficacy indices.

Results: The results showed the 12-item SOGS-RA had a unidimensional structure, acceptable internal consistency as well as test-retest reliability, and this scale was also positively correlated with impulsivity and functional impairment in physical, mental, and social activities. Screening with a cutoff score of 4 is recommended to identify possible cases of adolescent gambling disorder.

Conclusions: This study is the first to evaluate the unidimensional structure, reliability, validity, and cutoff score of the SOGS-RA among Chinese adolescent gamblers. Results suggested that the Chinese version of the SOGS-RA was a valid and reliable instrument to screen gambling disorder among Chinese adolescents.

Generosity usually declines as social distance increases, which is called social discounting. Previous literature has found that prosociality increases with age, but it remains unclear whether older adults are more or less generous as social distance increases compared with younger adults. The present study aimed to examine the effect of social distance on generosity among younger and older adults by further investigating the effect of frameworks and generativity.
Study 1 used the choice titration paradigm to explore whether hiding zero information of others or not would influence individuals’ choices. A total of 88 younger adults and 88 older adults participated in this study. The results showed that compared with younger adults, older adults were more generous overall and exhibited a smaller discount rate, which meant that older adults were also generous even to distant others. Besides, older adults cared more about others’ loss in the condition of showing zero information of others.
Study 2 also applied a choice titration paradigm. A total of 81 younger adults and 91 older adults participated in a social discounting task (tapping both generosity and social distance) with different gain and loss frameworks. The generativity tendencies were also measured with self-reported questionnaires. It was revealed that compared to younger adults, older adults were less sensitive to social distances and showed a higher level of generosity toward socially distant others regardless of gain or loss frames. Moreover, generativity partially mediated the relationship between age and overall generosity.
These findings suggest that older adults care less about self-loss in the context of donation. Their tendencies about contributing to others increase their generosity. The present study extends our understanding of the underlying mechanism of age-related changes in prosociality toward others.

Keywords: Social Discounting, Generosity, Age Differences, Frameworks, Generativity

People as third-party observers, may punish norm violators even without direct self-interest. However, third-party judgment and the follow-up punishment might be susceptible to the way we frame (i.e.,verbally describe) a norm violation. We conducted a behavioral and a functional magnetic resonance imaging (fMRI) experiment to investigate the above phenomenon, which we call the “third-party framing effect”. The paradigm combined the “social framing” task with third-party punishment in which participants observed an anonymous perpetrator deciding whether to keep her/his economic benefit while exposing a victim to a painful shock (described as “shocking others” in harm frame condition and “not helping others” in the help frame condition), then they had a chance to punish that perpetrator at their own cost. Our results showed that the participants were more willing to execute third-party punishment under the harm frame compared to the help frame, manifesting as a framing effect. In the behavioral experiment, self-reported anger toward perpetrators mediated the relationship between empathy toward victims and the framing effect. In the fMRI experiment, we found that the mid-cingulate cortex (MCC), and insula were both significantly sensitive to the main effect of frame; the functional connectivity between these regions could significantly predict the framing effect size. Also, insular activation mediated the relationship between mid-cingulate cortex activation and the framing effect. Overall, this study reveals that third-party punishment is susceptible to decision frames, which may help understand why moral standards and moral actions could be flexible and relativistic. Besides, our findings may broaden the understanding of the framing effect in general.