Top Read Articles

Published in last 1 year |

In last 2 years |

In last 3 years |

All

Please wait a minute...


For Selected: Download Citations EndNote Ris BibTeX Toggle Thumbnails

Select

Cognitive outsourcing based on generative artificial intelligence: An Analysis of interactive behavioral patterns and cognitive structural features WANG Fancong, TANG Xiaoyu, YU Shengquan Acta Psychologica Sinica    2025, 57 (6): 967-986.   DOI: 10.3724/SP.J.1041.2025.0967 Abstract （690）   HTML （27）    PDF （3482KB）（358）       Knowledge map    Humans can enhance task efficiency and quality by delegating part of their cognitive tasks to generative artificial intelligence (AI), a process referred to as cognitive outsourcing. However, individuals’ effectiveness in using AI varies. To identify the key characteristics and inherent requirements of effective cognitive outsourcing, this study designed a cognitive outsourcing activity for graduate students. Participants wrote articles on open-ended topics with the assistance of a generative AI system and were divided into high-performance and low-performance groups based on their article scores. Differential analysis of knowledge pre-tests revealed that the high-performance group exhibited significantly higher prior domain knowledge compared to the low-performance group. Through lag sequential analysis and epistemic network analysis of interaction process data, differences in interactive behavioral patterns and cognitive structural features between the two groups were identified: participants in the high-performance group demonstrated more diversified behavioral transitions, forming a pattern characterized by “rapid and autonomous task comprehension and planning, efficient and precise human-computer interaction, selective information extraction and deep processing”; the cognitive structure of the high-performance group was balanced and comprehensive, primarily engaging with higher-level cognitive processing, while the low-performance group's cognitive structure was unbalanced and fragmented, primarily engaging with lower-level cognitive processing. In conclusion, effective cognitive outsourcing is a multifaceted process that necessitates active participation and profound cognitive processing. It demands proficient integration between internal cognitive frameworks and external technological tools. Table and Figures | Reference | Related Articles | Metrics

Select

The impacts of music training and music sophistication on empathy HUA Shan, JIANG Xintong, GAO Yangzhenyu, MU Yan, DU Yi Acta Psychologica Sinica    2025, 57 (4): 544-558.   DOI: 10.3724/SP.J.1041.2025.0544 Abstract （468）   HTML （29）    PDF （694KB）（521）       Knowledge map    Music has long been recognized for fostering social bonds, with potential benefits for prosocial behaviors and empathy. Empathy, a key predictor of prosocial behaviors, encompasses both cognitive and affective components, involving the mentalizing and sharing of others’ emotional states. While musical training has been linked to increased empathy and prosocial behaviors, the influence of musical sophistication—a comprehensive measure of musical experience—on empathy is less well understood. Moreover, the specific components and pathways through which musical experience influences empathy remain unclear, with existing research relying largely on subjective measures and lacking objective behavioral evidence. To address these gaps, we conducted two studies using musical training and musical sophistication as indicators of musical experience to explore their impact on trait and state empathy through questionnaires and a behavioral experiment. In Study 1, we examined the relationship between musical training, musical sophistication and empathy in 130 musicians and 121 non-musicians, using standardized measures including the Goldsmiths Musical Sophistication Index (Gold-MSI) and Interpersonal Reactivity Index (IRI). Musicians scored significantly higher than non-musicians in cognitive empathy components (Perspective Taking and Fantasy). After controlling for gender, musical sophistication was positively correlated with cognitive empathy components (Perspective Taking and Fantasy) and an affective empathy component (Empathic Concern). Furthermore, after controlling for gender, openness, psychological states (depression, anxiety, alexithymia), and subjective social status, path analysis revealed that musical sophistication directly influenced cognitive empathy (Fantasy), while musical training indirectly influenced it via the mediating effect of music sophistication. Study 2 employed a pain empathy paradigm to assess empathic responses in 59 musicians and 61 non-musicians. Musicians demonstrated a higher alignment between ratings of their own and others' pain when observing others in pain, indicating greater empathy. The musicians' empathic response to pain was serially mediated by music sophistication and Fantasy in cognitive empathy. However, musical sophistication alone did not significantly affect pain empathy when the influence of musical training on musical sophistication was controlled. In summary, both long-term musical training and musical sophistication positively impact cognitive empathy. Specifically, musical training indirectly enhances the cognitive component (Fantasy) of trait empathy by improving musical sophistication, thereby fostering greater empathy for others’ pain. In contrast, musical sophistication has a direct and stable relationship with cognitive empathy. These findings support the “social bonding” hypothesis, highlighting music’s role in developing empathy and interpersonal skills. Table and Figures | Reference | Related Articles | Metrics

Select

The influence of positive emotion with varying intensities of approach motivation on false memory and its neural mechanisms: A study based on semantic-related false memory ZHANG Huan, QIN Xiquan, LIU Yu, LIN Lin, WU Jie Acta Psychologica Sinica    2025, 57 (3): 349-362.   DOI: 10.3724/SP.J.1041.2025.0349 Abstract （433）   HTML （33）    PDF （3750KB）（425）       Knowledge map    Emotions can influence false memories. Previous research has primarily focused on the effects of emotional valence and arousal on false memories. To date, the motivational dimension of emotions and its neural mechanisms in influencing false memories remain unclear. This study induced participants to experience positive emotions with varying intensities of approach motivation using the facial-expression-gesture method, and combined it with functional near-infrared spectroscopy (fNIRS) to investigate changes in cortical oxyhemoglobin concentration during the learning of DRM word lists under high, medium, and low intensities of approach motivation positive emotions, as well as the impact of these changes on semantically related false memories. The results showed that the high approach condition produced more false memories. Additionally, in some areas of the left frontal and temporal lobes, the brain activation levels under the high approach condition were significantly higher than those under medium and low approach conditions, indicating that high approach conditions elicit greater brain activation in specific regions. Correlation analysis results indicated that under high approach conditions, the activation levels in the left inferior frontal gyrus and temporal lobe were significantly positively correlated with the rate of false memories; under low approach conditions, the activation levels in the left inferior temporal gyrus were significantly negatively correlated with false memories. These results suggest that positive emotions with different intensities of approach motivation affect the generation of false memories, and the intensity of approach motivation affects the strength of activation in semantically related brain regions, with the left inferior temporal gyrus showing a dissociative effect in the process of false memory generation under different intensities of approach motivation positive emotions. Table and Figures | Reference | Related Articles | Metrics

Select

Reciprocal associations between identity confusion and adolescent NSSI: The longitudinal mediation effect of alienation GU Honglei, YU Weiming, CHENG Yufang Acta Psychologica Sinica    2025, 57 (2): 247-259.   DOI: 10.3724/SP.J.1041.2025.0247 Abstract （394）   HTML （26）    PDF （498KB）（661）       Knowledge map    Adolescence is a critical period of identity development, which is accompanied by psychosocial maladjustment (e.g., non-suicidal self-injury). In this study, the random intercept cross-lagged panel model (RI-CLPM) was employed to separate between- and within-person effects, and to test the mediation role of alienation in the reciprocal relations between identity confusion and adolescent non-suicidal self-injury (NSSI). A sample of 1258 Chinese middle school students (50.1% girls; Mage at Wave 1 = 13.81 years) completed self-report questionnaires regarding identity confusion, alienation, and NSSI at three time points (called T1, T2, and T3), with 6-month intervals. Results showed that identity confusion and NSSI were mutually reinforcing at the within-person level when controlling for between-person effects. Specifically, T1 NSSI positively predicted T2 identity confusion, which in turn positively predicted T3 NSSI. T2 NSSI also positively predicted T3 identity confusion. More importantly, T2 alienation longitudinally mediated the association between T1 NSSI and T3 identity confusion. Based on Erikson’s theory of psychosocial development, this study connects adolescent developmental tasks and NSSI on time scales, which has implications for the prevention and intervention of NSSI in adolescents. Table and Figures | Reference | Related Articles | Metrics

Select

The Impact of Narrative Information on Parochial Cooperative and its Mechanisms HE Ning, WANG Ziyi, LIN Jiahao, LI Meng, YOU Xuqun Acta Psychologica Sinica    2025, 57 (4): 513-525.   DOI: 10.3724/SP.J.1041.2025.0513 Abstract （375）   HTML （66）    PDF （212KB）（309）       Knowledge map    Table and Figures | Reference | Related Articles | Metrics

Select

The relationship between awe and prosocial behavior: Three-level meta-analysis and meta-analytic structural equation modeling LIN Rongmao, YU Qiaohua, HU Tianxiang, ZHANG Jiumei, YE Yushan, LIAN Rong Acta Psychologica Sinica    2025, 57 (4): 631-651.   DOI: 10.3724/SP.J.1041.2025.0631 Abstract （374）   HTML （19）    PDF （3062KB）（159）       Knowledge map    As two key indicators of human social development, awe and prosocial behavior are important emotional elements and behavioral paths to promote the construction of a Community with a Shared Future for Mankind. Three-level meta-analysis and meta-analytic structural equation modeling were used to investigate the relationship and mechanism between awe and prosocial behavior. Through literature search and screening, a total of 110 articles were included in Study 1, comprising 221 studies and 620 effect sizes, with a total sample size of 2, 961, 227 participants. Study 2 identified 33 articles, which included 42 studies and 42 correlation matrices, involving 30, 045 participants. The results of three-level meta-analysis indicated the correlation between awe and prosocial behavior was significantly positive (r = 0.37), and awe could positively predict prosocial behavior (g = 0.59). Moderator analyses revealed that awe has greater prosocial effect when the cultural context was collectivist, the valence was positive, the elicitor was social, prosocial behavior directed towards the collective level. The results of meta-analytic structural equation modeling showed that awe leads to prosocial behavior via self-diminishment and self-transcendence. The study systematically examines prosocial effects of awe and the conditions that facilitate it. It provides a theoretical foundation for enhancing prosocial behavior through an emotional lens and contributing to the vision of a Community with a Shared Future for Mankind. Table and Figures | Reference | Related Articles | Metrics

Select

The Impact of Temporal Landmarks on the Willingness of Conspicuous Prosocial Behavior* KUAI Ling, WEI Haiying, YAO Qi, XIAO Tingwen, XIE Shengcheng Acta Psychologica Sinica    2025, 57 (4): 526-543.   DOI: 10.3724/SP.J.1041.2025.0526 Abstract （369）   HTML （22）    PDF （442KB）（993）       Knowledge map    The propensity of individuals to engage in conspicuous prosocial behavior is influenced by environmental factors. Drawing on self-awareness theory and costly signaling theory, this study explores the impact of temporal landmarks on the willingness to engage in conspicuous prosocial behavior. Through six experiments, the results confirm that at the start (vs. end) of a period, individuals are more inclined to engage in conspicuous prosocial behavior (Experiments 1a, 1b, and 1c). The underlying mechanism is that the temporal landmarks at the beginning of a period trigger a situational public self-awareness, which influences the behavior (Experiments 2a and 2b). Further analysis reveals that self-monitoring plays a moderating role in this effect; for individuals with low self-monitoring, the activation of temporal landmarks has a limited impact on their willingness to engage in conspicuous prosocial behavior (Experiment 3). This study extends the research on the relationship between temporal landmarks and conspicuous prosocial behavior and provides practical guidance for charitable organizations or businesses in planning public welfare marketing activities at critical temporal junctures. Table and Figures | Reference | Related Articles | Metrics

Select

When AI “possesses” personality: Roles of good and evil personalities influence moral judgment in large language models JIAO Liying, LI Chang-Jin, CHEN Zhen, XU Hengbin, XU Yan Acta Psychologica Sinica    2025, 57 (6): 929-946.   DOI: 10.3724/SP.J.1041.2025.0929 Abstract （365）   HTML （46）    PDF （836KB）（210）       Knowledge map    At the intersection of technology and morality, a critical question arises: Can large language models (LLMs) simulate good and evil personalities, and does this capacity influence their performance in moral judgment tasks? This study investigated the moral judgment characteristics of LLMs when simulating different good and evil personalities, as well as the similarities and differences between these patterns and those of humans. Across two studies, we analyzed moral judgment data generated by two LLMs—ERNIE 4.0 and GPT-4 (N = 4,832)—alongside responses from human participants (N = 370). The results revealed that: (1) LLMs are capable of successfully simulating varying levels of good and evil personalities; (2) the personality configuration significantly affects the moral judgments made by LLMs; and (3) a personality hierarchy emerges in the alignment between human and LLMs’ responses: good personality plays a more critical role than evil personality (inter-personality hierarchy), and within the good personality, conscientiousness and integrity dimension exerts the strongest influence (intra-personality hierarchy). This research constructed a theoretical model of good and evil personalities in LLMs under moral judgment tasks, contributing to a deeper understanding of how simulated personalities function in AI moral reasoning. The findings provided a theoretical foundation for promoting moral alignment in artificial intelligence systems. Table and Figures | Reference | Related Articles | Metrics

Select

A comparative study on human or AI delivering negative performance feedback influencing employees’ motivation to improve performance WANG Guoxuan, LONG Lirong, LI Shaolong, SUN Fang, WANG Jiaqing, HUANG Shiyingzi Acta Psychologica Sinica    2025, 57 (2): 298-314.   DOI: 10.3724/SP.J.1041.2025.0298 Abstract （350）   HTML （23）    PDF （193KB）（332）       Knowledge map    Table and Figures | Reference | Related Articles | Metrics

Select

Perceived Robot Threats Reduce Pro-Social Tendencies XU Liying, ZHANG Yuyan, YU Feng Acta Psychologica Sinica    2025, 57 (4): 671-699.   DOI: 10.3724/SP.J.1041.2025.0671 Abstract （331）   HTML （15）    PDF （2073KB）（101）       Knowledge map    The entry of robots into society may pose a psychological threat to human beings, and such a threat can bring challenges to interpersonal relationships. Through eight studies, combining archival database backtracking, questionnaires, contextual experiments, and offline surveys, the article explores the effects of perceived robot threat on pro-social tendencies, as well as its underlying mechanisms and boundary conditions. The results found that: perceived robot threat reduces people's pro-social tendency (Study 1~7); the mechanism is mediated by collective anxiety, i.e., perceived robot threat increases collective anxiety, which reduces pro-social tendency (Study 2~ 4); this effect is moderated by in-group and out-group, i.e., perceived robot threat reduces pro-social tendency for out-group members (Study 5); at the same time, the effect is moderated by moral comparison tendency, i.e., perceived robot threat reduces pro-social tendency for out-group members (Study 6); at the same time, this effect is moderated by This effect is moderated by the moral comparison tendency, i.e., perceived robot threat mainly reduces the pro-social tendency of downward moral comparators (Study 6). The findings reveal the negative impact of perceived robot threat on interpersonal relationships and extend existing research on the social impact of robots. Table and Figures | Reference | Related Articles | Metrics

Select

The Self-reference Effect in Prospective Memory and Its Cognitive Mechanisms in Elementary School Students Aged 7 to 11 Years XIN Cong, CHEN Youzhen, TIAN Mi, LIU Guoxiong Acta Psychologica Sinica    2025, 57 (5): 775-791.   DOI: 10.3724/SP.J.1041.2025.0775 Abstract （324）   HTML （18）    PDF （993KB）（173）       Knowledge map    This study employed the ownership paradigm for the first time to investigate the existence of the self-reference effect (SRE) in prospective memory (PM) among elementary school students aged 7 to 11. Through three experiments, the findings revealed that: (1) Regardless of whether ownership was actively selected (Experiment 1) or passively assigned (Experiment 2), elementary school students exhibited the SRE in PM, with older students demonstrating a stronger effect. (2) Under a high cognitive load condition, the SRE in PM was absent in the salient and non-salient target conditions among elementary school students. Conversely, under a low cognitive load condition, the SRE was observed in the non-salient PM targets but not in the salient targets. Additionally, individuals flexibly allocated cognitive resources before and after the appearance of PM targets in self- and other-referenced conditions based on task demands. When cognitive resources were insufficient, individuals prioritized self-reference PM targets (Experiment 3). In conclusion, a stable SRE can be observed in the PM of elementary school students when employing the ownership paradigm, with large effects in the upper elementary school grades. The generation of the SRE in PM requires cognitive resources, which are deployed flexibly based on task demands and showcase a dynamic processing characteristic. The finding supports the notion that the SRE in PM involves cognitive resource-intensive dynamic processing, which can further enrich the dynamic processing theory of PM. Table and Figures | Reference | Related Articles | Metrics

Select

Human advantages and psychological transformations in the era of artificial intelligence WU Michael Shengtao, PENG Kaiping Acta Psychologica Sinica    2025, 57 (11): 1879-1884.   DOI: 10.3724/SP.J.1041.2025.1879 Abstract （316）            Knowledge map    In the era of artificial intelligence (AI), the boundaries between humans and machines have become blurred, and re-understanding and developing humanity's unique advantages are increasingly prominent and urgent. Meanwhile, with the rapid development of technology and scientific paradigm, a broad psychology encompassing the minds and behaviors of humans, animals, and machines is emerging. Recent researchers have conducted a series of studies on the psychology and governance of AI, from the perspectives of impacts of AI, new human-machine relationships, AI methods, and interdisciplinary empowerment. Future psychology researchers should focus on human society and future development, and reflect on the status of humanity and human dignity under the impact of AI, especially the unique advantages derived from human evolution as well as the expansions of human nature and identity; truly master and utilize AI technologies to empower the development of psychology, making mind research on the black box of human consciousness and complex social behavior more precise and efficient, and promoting AI-based mind computation and intervention across time and space scales and personalized interventions. More important, they must consider how psychology (with strengths in studying human nature, social relations, and ethical values) could empower the development of AI, by exploring AI cognition and its comparison with humans and animals, which is critical for promoting the AI application and governance in a human-machine symbiotic society. Reference | Related Articles | Metrics

Select

Category and semantic distance modulate the impact of prediction on memory DAI Jiaojian, SUN Mingze, WANG Dongfang, MAO Xinrui, GUO Chunyan Acta Psychologica Sinica    2026, 58 (1): 1-14.   DOI: 10.3724/SP.J.1041.2026.0001 Abstract （315）   HTML （32）    PDF （942KB）（92）       Knowledge map    Table and Figures | Reference | Related Articles | Metrics

Select

The topological structure of adolescents’ internet adaptation: A longitudinal tracking study DONG Wanghao, ZHANG Jie, MENG Sujie, JIA Min, WANG Weijun Acta Psychologica Sinica    2025, 57 (3): 415-427.   DOI: 10.3724/SP.J.1041.2025.0415 Abstract （306）   HTML （18）    PDF （1437KB）（224）       Knowledge map    As the saying goes, “Survival of the fittest”. Nowadays, the Internet has become a critical channel for information acquisition, social interaction, and educational learning. Adolescents’ internet adaptation capabilities must be continuously improved to adapt to this rapidly developing information age. Internet adaptation is inherently a “multidimensional system” encompassing various stages and dimensions. However, there remains a gap in the research exploring the internal topological characteristics and functional mechanisms of internet adaptation. Consequently, this study aims to employ network analysis techniques to elucidate the core characteristics, internal structure, dynamic evolution, and relationships with external variables of adolescents’ internet adaptation through network analysis. This approach will offer a comprehensive framework for understanding adolescents’ successful adaptation in the digital age and provide scientific insights for preventing and intervening in adolescent internet addiction. This study collected all data through paper-and-pencil questionnaires. At Time 1, valid data were obtained from 5783 participants (Males for 37.4%, Mage = 17.20 years, SD = 2.62). Five months later, data from 1235 of these participants were tracked (Males for 38%, Mage = 14.98 years, SD = 1.66). Based on the research objectives, we conducted cross-sectional network analysis, network comparison, and cross-lagged network analysis. All cross-sectional and cross-lagged network analyses were primarily conducted using R (V.4.3.2). Network visualizations were created with the qgraph package (version 1.9.5). The accuracy of edge estimates was assessed by performing 1000 bootstrap iterations to construct 95% non-parametric bootstrap confidence intervals for each edge. In the cross-sectional network of internet adaptation, “Internet curiosity” is the node with the highest strength (1.18). Network comparison results indicate no significant difference in the overall strength between the T1 (3.52) and the T2 network (3.79) (p = 0.120), although the network invariance test result is significant (p < 0.001). The cross-lagged network analysis shows that “Internet self-efficacy” has the strongest out-expected influence (0.60), “Internet learning ability” and “Internet information searching” has the strongest in-expected influence (0.31 & 0.30). Additionally, the cross-lagged network analysis of internet adaptation and internet addiction reveals that “Internet information protection capability” exhibits the strongest outgoing predictive ability. The main conclusions are as follows: (1) Adolescent internet adaptation is characterized by its dynamic and staged nature; (2) Adolescents’ internet curiosity plays a multifaceted role in their internet adaptation process: insufficient curiosity can lead to low internet self-efficacy, while excessive curiosity can result in poor internet self-control; (3) Internet self-efficacy has the most significant impact on the overall development of internet adaptation, serving as the “primary driving force”; (4) Internet learning ability and internet information search receive the most internal influence, constituting the main “landing point” of adolescents’ internet adaptation. (5) Internet information protection is the strongest predictor of cross-cluster outgrowth of internet addiction networks, acting as a “guardian” of adolescents’ internet adaptation. Table and Figures | Reference | Related Articles | Metrics

Select

Threshold Effects of Distinctiveness: Psychological Mechanisms Underlying Group Identity KE Wenlin, WEN Fangfang, ZUO Bin Acta Psychologica Sinica    2025, 57 (5): 820-837.   DOI: 10.3724/SP.J.1041.2025.0820 Abstract （305）   HTML （8）    PDF （1304KB）（63）       Knowledge map    Finding a balance between the self and the group is a central issue in human social life. Within the context of intergroup comparison, the Optimal Distinctiveness Model proposes an inverted U-shaped relationship between distinctiveness and group identification. However, how individuals balance the self and the group in an intragroup context remains unclear. Through three experiments, this paper systematically explores the impact of self-group distinctiveness on group identification across different levels of distinctiveness and group attributes. The findings reveal that, overall, distinctiveness exerts a non-linear, inverted S-shaped negative effect on group identification, with a distinctiveness threshold at the moderate level where the decline in group identification is most pronounced. Once distinctiveness surpasses this threshold, individuals exhibit complete disidentification from the group (Experiment 1). Both the level and nature of distinctiveness moderate this threshold effect (Experiment 1, 2, and 3). Additionally, intragroup cognitive dissonance mediates this relationship, whereby distinctiveness positively predicts intragroup cognitive dissonance, which in turn negatively predicts group identification (Experiment 3). The identification threshold identified in this study not only aids individuals in maintaining psychological boundaries but also serves as an effective warning signal for organizational management and social harmony. Table and Figures | Reference | Related Articles | Metrics

Select

The effects of rumination on social anxiety: The role of negative self-beliefs GENG Li, FENG Qiuyang, LI Yu, QIU Jiang Acta Psychologica Sinica    2025, 57 (5): 792-804.   DOI: 10.3724/SP.J.1041.2025.0792 Abstract （295）   HTML （19）    PDF （256KB）（421）       Knowledge map    Table and Figures | Reference | Related Articles | Metrics

Select

Moderated Mediation Analyses of Intensive Longitudinal Data FANG Jie, WEN Zhonglin, WANG Huihui, GU Honglei Acta Psychologica Sinica    2025, 57 (5): 915-928.   DOI: 10.3724/SP.J.1041.2025.0915 Abstract （293）   HTML （7）    PDF （338KB）（160）       Knowledge map    Intensive longitudinal data (ILD) is increasing in fields such as psychology and management, yet research on analytical methods for ILD remains relatively scant. Traditionally, the ILD is statistically modeled as a two-level structure, with Level 1 being the time and Level 2 being individuals. Existing analytical methods treat longitudinal moderated mediation as multilevel moderated mediation, without considering the lagged relationship between variables. This paper describes in detail how to construct four intensive longitudinal moderated mediation (ILMM) models with dynamic structural equation model (DSEM). A simulation study is conducted to examine the estimation accuracy of the 1-1-1 intensive longitudinal mediation model moderated by a level 2 moderator. An example is employed to demonstrate how to conduct ILMM analysis with DSEM by Mplus. Mplus codes for analyzing all these ILMM models are provided (The complete dataset, Mplus syntax files, and analysis outputs can be downloaded at https://osf.io/e273c/). Table and Figures | Reference | Related Articles | Metrics

Select

Suicidal ideation data augmentation and recognition technology based on large language models ZHANG Yanbo, HUANG Feng, MO Liuling, LIU Xiaoqian, ZHU Tingshao Acta Psychologica Sinica    2025, 57 (6): 987-1000.   DOI: 10.3724/SP.J.1041.2025.0987 Abstract （287）   HTML （9）    PDF （384KB）（80）       Knowledge map    Suicide has become a global public health challenge. Traditional methods for identifying suicidal ideation primarily rely on patients actively seeking help, while automated identification models based on text analysis are limited by the scarcity of annotated data. This study innovatively proposes a data augmentation method based on large language models (LLMs) to improve the accuracy of suicidal ideation text recognition. The research employs a two-stage design: Study 1 focuses on data augmentation, and Study 2 validates the enhancement effect. In Study 1, ChatGLM3-6B and Qwen-7B-Chat were selected as the underlying models, combining supervised learning strategies with zero-shot and few-shot learning methods to optimize training dataset quality. Through eight rigorous comparative experiments, the results show that the two self-developed models demonstrated excellent performance in data augmentation, with comprehensive scores of 0.90 and 0.92 for their processed datasets, significantly outperforming baseline models (p < 0.001). Study 2 further evaluated the impact of data augmentation on recognition model performance, showing that the enhanced models comprehensively outperformed the best baseline models in terms of recognition accuracy and true negative rate (p < 0.001). This study not only validates the effectiveness of LLM-based data augmentation methods in improving the performance of suicidal ideation recognition models but also opens new directions for artificial intelligence applications in the field of mental health. This approach has the potential to provide timely and effective early warning of suicide risk while protecting user privacy, offering important technical support and research ideas for suicide prevention work. Future research could focus on expanding data heterogeneity, optimizing prompt engineering design, and introducing human-computer interaction paradigms to further extend the application of this method in promoting clinical psychological diagnosis. Table and Figures | Reference | Related Articles | Metrics

Select

The effect of a social robot on the sharing behavior of 3- to 5-year-old children LI Hui, LIU Siyi, PANG Yi Acta Psychologica Sinica    2025, 57 (4): 573-583.   DOI: 10.3724/SP.J.1041.2025.0573 Abstract （275）   HTML （16）    PDF （246KB）（443）       Knowledge map    This study investigates the influence of social robots on the prosocial behaviors of children aged three and five. Experiment 1 explored the impact of different observers (human, social robot, no) on the sharing behaviors of these children. The results indicated that 5-year-olds shared significantly more stickers than 3-year-olds in the absence of an observer. For 3-year-olds, the presence of human and robot observers led to significantly more sharing than when no observer was present, with no significant differences between the human and robot conditions. Experiment 2 manipulated the psychological agency of social robots (with, without, control group) and found that 3-year-olds shared more stickers under the observation of robots with psychological agency compared to the other two conditions, showing increased prosocial behaviors. This research suggests that the prosocial behaviors of 3-year-olds can be influenced by the psychological agency of robots, providing perspectives and evidence for future applications and studies in human-robot interaction. Table and Figures | Reference | Related Articles | Metrics

Select

Influence of Sustained Visual Attention on the Prioritization of Visual Working Memory LIAN Haomin, ZHANG Qian, GU Xuemin, LI Shouxin Acta Psychologica Sinica    2025, 57 (2): 191-206.   DOI: 10.3724/SP.J.1041.2025.0191 Abstract （270）   HTML （26）    PDF （854KB）（130）       Knowledge map    Using behavioral experiments and the simultaneous acquisition technique of event-related potentials and event-related optical signals (ERP-EROS), we manipulated the probe probability of visual working memory (VWM) items to explore whether the impact of sustained visual attention on VWM prioritization is modulated by working memory resources, and to investigate the neural mechanisms underlying this prioritization. The behavioral results showed that during the VWM maintenance phase, when a task that consumed visual attention was introduced, impairment occurred for non-prioritized items when one item was prioritized, while the prioritized item remained unaffected. However, when two items were prioritized, both prioritized and non-prioritized items were impaired. The results from ERP and EROS showed that during the VWM maintenance phase, prioritizing an item, compared to the no- prioritization condition, elicited larger late positive components and negative slow waves, accompanied by increased activation in the frontal and occipital cortices. This suggests that the effect of sustained visual attention on VWM prioritization is modulated by working memory resources. The underlying mechanism for prioritization involves activation of the frontal and occipital cortices during the maintenance phase and a greater allocation of working memory resources to enhance the stability of prioritized item representations. Table and Figures | Reference | Related Articles | Metrics