Archive

25 October 2025, Volume 57 Issue 10 Previous Issue    Next Issue
---
Reports of Empirical Studies


For Selected:

Download Citations EndNote Ris BibTeX

Toggle Thumbnails

Reports of Empirical Studies

Select

The multiple image advantage in face identity recognition relies on the formation of facial representation FENG Jun Ye, WANG Zhe, SUN Yu-Hao P. 2025, 57 (10):  1689-1700.  doi: 10.3724/SP.J.1041.2025.1689 Abstract ( 1042 )   HTML ( 281 )   PDF (2907KB) ( 1527 )   Peer Review Comments Presenting multiple face images of the same person has been shown to enhance a person's ability to recognize faces, which is known as multiple image advantage (MIA). However, MIA appears selective as evidenced by it manifesting in delayed face matching tasks but being absent in non-delayed face matching tasks. Two hypotheses are proposed to account for this selectivity. In one hypothesis, the delay after multiple face images are presented facilitates face representation formation, which strengthens participants’ face recognition. In the other hypothesis, it is proposed that the quantity of image information and task difficulty are key variables. That is, the more face images presented, the more information is processed, which improves recognition performance. Additionally, delayed face matching tasks are more difficult than non−delayed ones. Together, these yield MIA in delayed face matching tasks but the ceiling effect in non−delayed tasks. In the current study, we conducted two experiments to investigate the cognitive mechanisms underlying MIA in face identity recognition. Experiment 1A adopted a face-matching paradigm where university students (N = 81) judged whether a target face matched a set of one, two, or three study faces presented simultaneously (non-delayed condition) or sequentially (delayed condition). In the simultaneous condition, target and study faces were displayed concurrently, allowing direct perceptual comparison; in the sequential condition, study faces were presented for 5, 000 ms followed by a 500 ms blank screen and the target face, requiring memory-based matching. In Experiment 1B, the simultaneous matching task was modified to enforce a mandatory 5, 000 ms viewing period for studying faces before allowing responses, which ensured identical exposure duration across both simultaneous and sequential conditions. This eliminated potential confounds arising from differences in processing time. Both experiments utilized unfamiliar face images, counterbalanced across tasks and participants, with familiarity screening applied post-test to exclude recognizable faces. In Experiment 2, we repeated Experiment 1 but inverted all faces to disrupt the integration processes of facial representation in the delay time. The results of Experiment 1 show that: (1) MIA appears only when the images were presented sequentially as participants' discriminability improved when the number of images increased and (2) When presenting three face images, participants' discriminability under the sequential condition was higher than under the simultaneous condition. After controlling for duration of faces presented in Experiment 1B, results remain unchanged. For Experiment 2, (3) regardless of whether faces were single or multiple, participants' discriminability under the sequential condition was lower than under the simultaneous condition, and (4) no MIA appeared in either task. Altogether, these findings suggest that MIA in facial identity recognition in delayed face matching tasks is derived from the formation of facial representation that happens in memory. Figures and Tables | References | Related Articles | Metrics

Select

Musical training enhances the interaction between pitch and time dimensions in auditory working memory ZHOU Linshu, ZHANG Yuqing, CAI Dan-Chao 2025, 57 (10):  1701-1714.  doi: 10.3724/SP.J.1041.2025.1701 Abstract ( 956 )   HTML ( 202 )   PDF (777KB) ( 1262 )   Peer Review Comments Musical training enhances sensitivity to both temporal (e.g., rhythm) and non-temporal (e.g., pitch) structures during auditory processing. However, little is known about the joint processing of these structures in auditory working memory and the influence of musical expertise. This study explored whether pitch and rhythmic structures are processed interactively or independently in auditory working memory, and how musical expertise modulates this relationship. Two experiments were conducted with 36 musicians and 36 nonmusicians. Experiment 1 involved a pitch maintenance task, where participants judged whether two pitch sequences were the same or different while ignoring rhythmic variations, which could act as background interference. The sequences varied in pitch structure (tonal vs. atonal), rhythmic structure (simple vs. complex), and length (five-note vs. seven-note). Experiment 2 examined rhythm maintenance, where participants judged whether two rhythmic sequences were the same or different while ignoring pitch information. While the task was conceptually the reverse of Experiment 1—requiring attention to rhythm rather than pitch—the design similarly manipulated pitch structure, rhythmic structure, and sequence length to compare how participants processed rhythmic information when suppressing irrelevant pitch content. The results of Experiment 1 indicated that nonmusicians tended to process pitch and rhythmic structures independently during the pitch maintenance task. In contrast, musicians exhibited a significant interaction between pitch and rhythm, demonstrating their ability to integrate these structures for more effective processing. A correlation analysis further confirmed the role of musical training in this integration; sensitivity to the interaction of pitch and rhythm was positively correlated with participants' musical sophistication scores. This suggests that individuals with higher musical expertise exhibit more pronounced integration of pitch and rhythmic information, reflecting enhanced multidimensional auditory processing abilities in musicians. In Experiment 2, however, both musicians and nonmusicians processed pitch and rhythm independently during the rhythm maintenance task, with no significant interaction. This contrasts with the findings of Experiment 1, indicating that pitch and rhythmic structures are processed separately during rhythm maintenance. Furthermore, nonmusicians showed no difference in sensitivity between tonal and atonal conditions in the rhythm maintenance task, although they responded faster in the tonal condition than in the atonal condition. These results support dynamic attending theory, which posits that attentional resources align with temporal events, thus facilitating the processing of non-temporal information, such as pitch, when in synchrony with rhythmic structure. While nonmusicians relied more on independent processing strategies across both tasks, musicians showed enhanced cognitive flexibility, modulating their processing strategies in response to task complexity. This suggests that musical training not only improves auditory sensitivity but also allows individuals to adaptively switch between processing strategies based on task demands. Future research may further explore how different types of musical training affect the interaction between temporal and non-temporal auditory information under varying cognitive loads. Figures and Tables | References | Related Articles | Metrics

Select

The cognitive characteristics and neural mechanisms of multisensory category learning: EEG and drift-diffusion model evidence WU Jie, CHE Zixuan 2025, 57 (10):  1715-1728.  doi: 10.3724/SP.J.1041.2025.1715 Abstract ( 842 )   HTML ( 219 )   PDF (6364KB) ( 2451 )   Peer Review Comments Category learning in multisensory environments, which is a fundamental human cognitive ability, has significant implications for understanding cross-modal knowledge representation. This study systematically examines the cognitive characteristics and neural mechanisms of multisensory category learning by integrating event-related potential (ERP) techniques and drift-diffusion modeling (DDM). We established three experimental groups— the early-stage, middle-state and later-stage groups—in which participants acquired the ability to discriminate four categories of multisensory stimuli through corrective feedback. During the learning process, we simultaneously recorded electroencephalographic (EEG) data and employed a multimodal analytical approach integrating neural oscillation with computational modeling by using the DDM. This combined methodology enabled us to systematically examine how varying degrees of learning proficiency modulate the neurocomputational mechanisms underlying multisensory category acquisition. From a behavioral perspective, the middle- and later-stage learning groups demonstrated significantly greater accuracy, reaction time and drift rates than the early-stage learning group, along with a decision threshold bias toward correct responses. At the neural level, middle- and later-stage learning elicited amplified amplitudes in the N1, P1, and LPC components while decreasing the amplitude of the N250-FSP complex. Time‒frequency analyses demonstrated significant power reductions in the theta, alpha, and delta frequency bands. Regression analyses identified distinct neural predictors: variations in drift rates were jointly explained by reductions in N250-FSP amplitude and theta oscillations, whereas decision threshold biases were predicted by coordinated activity in early perceptual processing (P1), feature discrimination (N250-FSP), and memory retrieval (LPC) components. These findings reveal a dual-mechanism framework through which learning sufficiency optimizes decision efficiency. (1) Enhanced information accumulation rates are associated with reduced N250-FSP amplitudes and theta-band reorganization, reflecting streamlined feature integration and conflict resolution. (2) Decision threshold shifts result from the synergistic interplay of sensory encoding (P1), categorical feature discrimination (N250-FSP), and postretrieval monitoring (LPC). Notably, the dissociation between theta-mediated drift rate modulation and fronto-posterior ERP dynamics in threshold adjustment offers compelling evidence for parallel neural pathways that govern distinct decision parameters. This study advances multisensory learning theories by elucidating the neurocognitive mechanisms underlying learning optimization, thereby providing insights regarding the development of targeted interventions in adaptive learning systems and cross-modal training paradigms. These findings highlight the pivotal role of learning duration in shaping both the neurocomputational architecture of decision-making processes and the efficiency of cross-modal knowledge consolidation. Figures and Tables | References | Related Articles | Metrics

Select

Working memory load modulates the attentional bias of individuals with social anxiety: Evidence from steady-state visual evoked potentials YANG Xinchao, CHEN Yanling, ZHENG Junmeng, ZHANG Rongxiang, MU Xiao, XU Qiang 2025, 57 (10):  1729-1744.  doi: 10.3724/SP.J.1041.2025.1729 Abstract ( 1202 )   HTML ( 190 )   PDF (3687KB) ( 1623 )   Peer Review Comments The primary characteristic observed in individuals with social anxiety is the atypical processing of social stimuli, which has been identified as a crucial factor in the development and persistence of symptoms. According to attentional control theory, impaired attentional control results in attentional vigilance and difficulty in disengagement from threatening social stimuli. Although previous studies have provided evidence, the neural mechanisms between social anxiety and attentional bias remain unclear. A robust method for continuously quantifying attentional resources is needed. Therefore, attentional control was manipulated by setting the working memory (WM) load in this study. Steady-state visual evoked potentials (SSVEPs) were used as a neural index of attention allocation to examine the attentional biases toward threatening social stimuli in individuals with social anxiety. Through the Liebowitz Social Anxiety Scale (LSAS) and the Social Interaction Anxiety Scale, participants (N = 38) were recruited and aligned into high social anxiety (HSA; Mage = 20.68, n = 19) and low social anxiety (LSA; Mage = 21.42, n = 19) groups. SSVEPs were elicited by a frequency-tagging paradigm in which stimuli depicting angry, happy, and neutral expressions were presented at distinct frequencies. The participants were required to focus on the facial stimuli and judge whether the number (or number set) had been presented at the beginning. Attentional bias indices were derived by calculating the difference in mean SSVEP amplitudes between emotional and neutral face conditions. Specifically, negative (positive) attentional bias indices were computed by subtracting the mean SSVEP amplitudes elicited by neutral faces from those elicited by angry (happy) faces. The behavioral results showed faster response times in the low-WM-load conditions than in the high-WM-load conditions, indicating that the attentional control was effectively manipulated through WM load indirectly in the task. The negative and positive attentional bias indices were separately analyzed through a three-way repeated measures ANOVA, containing the between-subject factor group (HSA, LSA) and the within-subjects factors WM load (high, low) and time window (180~500 ms, 500~1000 ms, 1000~2000 ms). The results of negative attentional bias showed that in the high-WM-load condition, higher negative attentional bias indices were observed in the HSA group than in the LSA group during the 180~500 ms and 500~1000 ms time windows. By contrast, no significant differences were identified between HSA and LSA groups in the low-WM-load condition. Simple linear regression analyses were used to examine the relationship between social anxiety levels and attention bias indices. The results showed that LSAS scores positively predicted negative attention bias indices (180~1000 ms). In the low-WM-load condition, lower positive attentional bias indices were observed in the HSA group compared with those in the LSA group. The negative attentional bias in social anxiety was modulated by WM load in this study. Individuals with HSA showed enhanced and maintained attentional vigilance toward threatening social stimuli when the attentional control was suppressed experimentally. Meanwhile, individuals with HSA might exhibit insufficient attentional processing toward positive social stimuli. In conclusion, impaired attentional control could be a risk factor for the development of social anxiety. The continuous spectrum perspective would be suitable for further revealing the dynamic attentional bias in individuals with social anxiety. Figures and Tables | References | Related Articles | Metrics

Select

Preliminary development and evaluation of the Chinese self-conscious emotions nonverbal behavior expression stimulus set, and its application in research YIN Huazhan, LIU Pengyu, ZHANG Li 2025, 57 (10):  1745-1761.  doi: 10.3724/SP.J.1041.2025.1745 Abstract ( 699 )   HTML ( 113 )   PDF (1756KB) ( 1728 )   Peer Review Comments The nonverbal expression of self-conscious emotions plays a unique role in the social process. However, currently there is a lack of sufficient amount of emotional images with multiple emotional dimensions in standardized self-conscious emotions nonverbal behavior expression stimulus set in domestic and foreign researches, which may lead to problems such as the inability to control extraneous variables in psychology and cognitive neuroscience research and the lack of comparability of different laboratory research results. The purpose of this study is to develop a standardized Chinese self-conscious emotions nonverbal behavior expression stimulus set with multiple emotional dimensions, including three recognizable self-conscious emotions (pride, shame and embarrassment) and explore the effects of shame, pride, embarrassment on time perception. Study 1 encompassed three sequential stages: preparatory phase, image production, and image evaluation. Stage 1 involved structured interviews and development of self-conscious emotion-eliciting scenarios. In Stage 2, twenty models (N = 20, 10 female) were participated in picture taking comprised three photography sub-stages: (1) models understood and felt emotions which were induced by the neutral and three self-conscious emotions pictures presented in the computer, (2) neutral pictures were taken photo when model was very calm, and (3) self-conscious emotions pictures were taken photo including three steps. Firstly, model imagined the emotional situation which induced their self-conscious emotions and to express. Secondly, the recordings of emotional situation stories were played in sequence (three stories of each emotion), and model felt and expressed. Thirdly, self-conscious emotions pictures compiled in domestic and foreign studies were presented on the computer screen successively, supplemented by nonverbal behavior expression clues, and model imitated and expressed. Finally, four hundred and seventy-nine pictures (N = 479) were preliminary filtered and produced. In Stage 3, one hundred and three participants (N = 103, 55 female), excluded from state anxiety and depression, were rated the pictures in two sessions, each lasting about an hour. The emotional classification and the valence of all pictures were rated for the first session. Arousal and dominance of all the pictures were rated for the second session. The results of emotion classification showed that: (1) A total of Three hundred and eighty-nine pictures (N = 389) were generated, including 36 neutral pictures, 124 shame pictures, 107 pride pictures and 122 embarrassment pictures; (2) discovered new hand actions posed by models, such as hand pulling the hem of close (shame), scratching the head with one hand (embarrassment), and raising above the head with one hand (pride); (3) the internal consistency coefficient of the stimulus set was high; (4) compared with neutral pictures, pride was characterized by higher pleasure, higher arousal, higher dominance; Shame was characterized by lower pleasure, higher arousal, lower dominance; Embarrassment was characterized by lower pleasure, higher arousal, lower dominance. In study 2, One hundred and twelve participants (N = 112, 56 female) were recruited to complete the time replication task and the picture evaluation task for the main experimental session, aiming to explore the effects of shame, pride, embarrassment on time perception. The results showed that relative to neutral picture, nonverbal behavior expressions picture of both shame and pride elicited significant temporal underestimation, whereas embarrassment nonverbal behavior expressions picture showed no significant temporal distortion. In conclusion, this study developed a multi-dimensional and standardized Chinese self-conscious emotions nonverbal behavior expression stimulus set with good reliability, which provides the emotion-inducing picture material for the future research. Meanwhile, shame and pride led to underestimation of time perception, while embarrassment did not significantly distort time perception. These results were interpreted within the theoretical framework of the scalar timing model. Figures and Tables | References | Related Articles | Metrics

Select

The power of confiding: Negative emotional self-disclosure facilitates peer prosocial behavior WANG Shuqi, CHENG Feng, GUO Bing, WANG Qiao, CHENG Xiaojun 2025, 57 (10):  1762-1776.  doi: 10.3724/SP.J.1041.2025.1762 Abstract ( 3087 )   HTML ( 621 )   PDF (1457KB) ( 5658 )   Peer Review Comments In response to emotional distress, individuals often turn to confiding—the act of disclosing negative emotions to others—to seek social support and alleviate internal burdens. While confiding offers emotional relief, it may also carry interpersonal benefits, such as enhancing empathy and fostering social bonds. However, little is known about how negative emotional self-disclosure influences prosocial behavior between peers, or what psychosocial mechanisms underlie this effect. Additionally, linguistic cues such as the use of first-person (“I”) versus third-person pronouns (“He/She”) may influence the impact of emotional disclosure by altering how disclosure is perceived and processed. The present study examines whether disclosing or listening to negative emotions promotes prosocial behavior in peer contexts, and how personal pronoun framing moderates this effect. Addressing these questions contributes to a deeper understanding of the social value of emotional expression, particularly in moments of vulnerability that invite connection and support. Two experiments were conducted. Experiment 1 examined the discloser’s perspective, using a 2 (emotional disclosure: negative vs. neutral; within-subject) × 2 (personal pronoun: first-person vs. third-person; between-subject) mixed design. Participants took on the role of the discloser, narrating either negative or neutral emotional experiences in the first-person or third-person to a peer (played by an experimenter). They then reported their prosocial intentions, empathy, and liking toward their peers, and engagement with the disclosure process. Experiment 2 shifted to the listener’s perspective, using a fully within-subject 2 (emotional disclosure: negative vs. neutral) × 2 (personal pronoun: first-person vs. third-person) design. Participants watched videos of peers narrating their emotional (negative or neutral) experiences framed with either first- or third-person pronouns, and then reported their prosocial intentions, empathy, and liking toward the discloser. Both experiments were conducted in controlled laboratory settings with adult participants, using validated self-report measures to assess prosociality. Experiment 1 showed that participants reported significantly greater engagement when sharing negative emotions compared to neutral ones. Importantly, disclosing negative emotions led to increased prosocial behavior, but only in the first-person pronoun condition. The increased prosocial behavior was closely related to the enhanced interpersonal liking and a stronger sense of meaning during the disclosure. No such effect emerged in the third-person condition, indicating that linguistic self-involvement strengthens the social impact of confiding. Experiment 2 found that participants expressed higher levels of prosocial behavior, empathy, and liking after hearing negative (vs. neutral) emotional disclosures, regardless of pronoun type. Crucially, the prosocial gains were closely associated with elevated empathy and liking, suggesting these variables as potential mediators. While the effect of pronoun reference was not significant for listeners, the emotional valence of the disclosure played a decisive role. Together, these findings suggest that confiding—either as discloser or listener—can foster peer prosociality, but the underlying mechanisms may differ: disclosers benefit from personalization (via first-person framing), while listeners respond primarily to emotional authenticity. This study reveals the positive social consequences of negative emotional self-disclosure. For disclosers, using first-person pronouns enhances this effect by increasing the sense of meaning, and interpersonal liking. For listeners, the content of the emotional experience—rather than linguistic framing—drives prosocial responses, which are closely linked to feelings of empathy and liking. These findings contribute to understanding the interpersonal value of emotional expression and offer practical insights into how linguistic and emotional cues can be harnessed to build social support, especially in emotionally challenging situations. Figures and Tables | References | Related Articles | Metrics

Select

Measuring love: Relational mobility promotes calculative mindset in mate choice LI Yawen, CAI Pan, ZUO Shijiang, WANG Fang, DONG Yu, ZHOU Yang, HUANG Niwen 2025, 57 (10):  1777-1790.  doi: 10.3724/SP.J.1041.2025.1777 Abstract ( 1912 )   HTML ( 428 )   PDF (1262KB) ( 2491 )   Peer Review Comments In today's China, people tend to break down their own and others' profiles into numbers and ranks during the blind date process, using these indexes to find perfect matches. For example, a female with 7 points might be matched with a male with 7 points. This study aims to address the following questions: What social mentality is reflected in this phenomenon, and why does it occur? Firstly, the tendency is conceptualized as a calculative mindset in mate choice—a cognitive process that translates dating information into a quantitative index. Thereafter, the study proposes that individuals tend to adopt this strategy in mate choice due to the increasing relational mobility in contemporary Chinese society, viewed from a socio-ecological perspective. With the rise in relational mobility, individuals are endowed with the freedom to choose their soulmate as they wish. However, this also imposes a greater cognitive load on the decision-making process. Therefore, the calculative mindset, which transforms qualitative profiles into comparable and matchable quantitative numbers, becomes prominent. Hypotheses were tested through studies combining both macro and micro levels, including nationwide large-scale online data, self-report questionnaires, and experimental manipulations in the lab. Study 1 (N = 44, 703) utilized province-level statistical data from the nationwide statistical yearbook and online dating websites to preliminarily test our hypothesis on a macro scale across 31 provinces and regions. Specifically, the divorce-to-marriage rate was calculated to represent relational mobility, and the calculative mindset index was obtained based on self-introductions and mating requirements using text analysis with Python. Study 2 (N = 311) explored the questions at the individual level through questionnaires. It collected data on individuals' perceptions of relational mobility in their current environment and their inclination to use a calculative mindset for cognitively processing both others' profiles and their own in the mate choice process. In Study 3A (N = 130) and Study 3B (N = 132), which aimed to establish a causal relationship between variables, participants were induced with high or low relational mobility mindsets. Subsequently, they were asked to express their preferences regarding the use of graded profiles for assessing others and presenting themselves separately. As for the results, using hierarchical linear modeling (HLM), Study 1 revealed that regional relational mobility positively and significantly predicted individuals' tendency to use quantified words in self-introductions and mating requirements, even after controlling for age, gender, education level, regional population, regional GDP, and regional residential mobility. Study 2 confirmed the hypotheses and identified a positive relationship between individuals’ perception of relational mobility and a calculative mindset in assessing others and themselves for aiding mating decisions. Lastly, Study 3 found that individuals primed with a high relational mobility mindset (versus a low relational mobility mindset) in laboratory settings were more inclined to use high grading profiles for choosing others (Study 3A) and presenting themselves (Study 3B), further confirming causality. In summary, the main hypothesis was confirmed through four studies conducted at varying levels, providing robust support for explaining the role of relational mobility in shaping a calculative mindset during mate choice. This study initially introduced the concept of a calculative mindset in mate choice and revealed the driving forces of social transformation behind this phenomenon. It not only offers a novel perspective for comprehending the prevalent issue of mate selection in China but also enriches the theoretical understanding of the relational mobility research domain. Figures and Tables | References | Related Articles | Metrics

Select

Influence of knowledge stock on improvisation under time pressure WANG Yongyue, ZHANG Fanying, YUE Fengkai, XIE Jiangpei 2025, 57 (10):  1791-1812.  doi: 10.3724/SP.J.1041.2025.1791 Abstract ( 1093 )   HTML ( 197 )   PDF (686KB) ( 1195 )   Peer Review Comments Improvisation, defined as “creative and spontaneous management of unforeseen situations” is thought to be creativity under time pressure. It embodies two key elements: spontaneity and creativity. Prior research has largely focused on the creative aspects of improvisation, emphasizing the effects of various individual and organizational factors. However, this exclusive focus limits a comprehensive and precise understanding of the fundamental nature of improvisation. To construct a robust theoretical model that elucidates the generative mechanism of improvisation, both aspects must be considered. Accordingly, this study examines the essential characteristics of improvisation to reveal its underlying mechanisms. Drawing on the knowledge linking theory and activation theory, we hypothesized that knowledge stock has a positive impact on individual improvisation via knowledge transformation. Furthermore, we expected that the relationship between knowledge stock and knowledge transformation will be stronger under medium time pressure, ultimately promoting individual improvisation. By taking a multimethod approach, we validated our hypotheses across two experiments (Studies 1, 2) and a field survey (Study 3). Prior to conducting formal experiments, time limits for high and medium time pressure groups were established through a preliminary experiment. In Study 1, we conducted a 2 (knowledge stock: high vs. low) × 3 (time pressure: high vs. medium vs. low) between-subjects design and recruited 163 students. The aim of this experiment was to examine the main effects of knowledge stock on knowledge transformation and improvisation, along with the moderating effect of time pressure. In Study 2, we employed a job-related organizational scenario to evaluate the entire model, involving 163 full-time employees. In Study 3, we collected dyadic data from 201 leader−employee pairs at two time points. At Time 1, employees reported knowledge stock, perceived time pressure, and demographic information. At Time 2, employees reported knowledge transformation, while leaders reported the improvisation behavior of employees. We applied analysis of variance, confirmatory factor analysis, path analysis, and Monte Carlo method to analyze the data. The results were as follows: Firstly, knowledge stock was positively related to both knowledge transformation and improvisation. Secondly, knowledge transformation played a mediating role in the relationship between knowledge stock and improvisation. Additionally, time pressure moderated both the direct effects between knowledge stock and knowledge transformation, as well as the indirect effects of knowledge stock on improvisation through knowledge transformation. Specifically, when time pressure is medium, the positive effect of knowledge stock on improvisation is stronger via knowledge transformation, while such relationship weakened when time pressure is high or low. It is interesting to note that time pressure demonstrated an inverted U-shaped moderating effect. Essentially, under conditions of medium time pressure, knowledge stock exhibited a more pronounced and positive impact on knowledge transformation, consequently enhancing improvisation. This study has the following theoretical contributions. First, it thoroughly examines the mechanism of improvisation, integrating both spontaneity and creativity, thus expanding the theoretical foundations of the field. By grounding the analysis in fundamental concepts and their core characteristics, this approach offers a valuable reference for future research. Second, we provide novel insights into the inverted U-shaped moderating effect of time pressure on the generation mechanism of individual improvisation, thus enriching and advancing research in the field. Third, this study evaluates the effectiveness of knowledge stock from a knowledge management perspective and elucidates the mechanisms through which it influences individual improvisation, addressing its substantive characteristics. Overall, our detailed examination of the complexity and underlying dynamics of improvisation contributes to a profound understanding of this phenomenon and its implications. Figures and Tables | References | Related Articles | Metrics

Select

“Good pressure, bad pressure?” The double-edged sword effect of algorithmic regulatory pressure on service performance GAO Xueyuan, ZHANG Zhipeng, XIE Baoguo, LONG Lirong, YIN Kui 2025, 57 (10):  1813-1831.  doi: 10.3724/SP.J.1041.2025.1813 Abstract ( 1013 )   HTML ( 143 )   PDF (648KB) ( 3005 )   Peer Review Comments With the continuous increase in the number of gig workers, work pressure has become a significant public concern. Gig workers experience algorithmic regulatory pressure from platforms through automatic task allocation, real-time guidance, and tracking evaluation, which permeate the entire work process. Coping strategies adopted under such pressure directly influence workers’ physical and mental health, as well as their work outcomes. However, prior research has only explored the conceptual nature of algorithmic regulatory pressure and its potential impact on individual well-being and work behavior. The mechanism through which this pressure affects service performance remains unclear. To address this gap, the present study applies the job demand-resource (JD-R) theory to investigate the impact of algorithmic regulatory pressure on gig workers' service performance. JD-R theory identifies two broad categories of working conditions—job demands and job resources. Job resources trigger motivational processes, whereas job demands lead to health-impairment processes. Moreover, job resources can buffer the adverse effects of job demands. Based on this framework, the study hypothesizes that algorithmic regulatory pressure exerts a double-edged sword effect on service performance by inducing distinct job crafting behaviors, particularly when two key resources—algorithmic transparency and online community support—are present. Two studies were conducted to test the hypotheses. Study 1 employed a scenario-based experiment to examine the causal relationship between algorithmic regulatory pressure and job crafting behaviors. A total of 377 take-away riders were recruited via an online survey platform (Credamo) and randomly assigned to either a high- or low-pressure scenario. Participants provided demographic data, viewed an experimental video, and responded to manipulation checks and job crafting measures. The final valid sample comprised 358 riders. Study 2 involved a three-wave, multi-source survey to test the proposed model, incorporating objective service performance metrics. Each wave was spaced four weeks apart. A total of 450 ride-hailing drivers were recruited through Credamo. At Time 1, participants reported demographics, algorithmic regulatory pressure, time pressure, alienation pressure, physical and mental pressure, proactive personality, algorithmic transparency, and online community support. At Time 2, they reported their approach and avoidance job crafting behaviors. At Time 3, drivers’ service performance was obtained from platform-generated metrics. The final sample comprised 350 drivers. Across both studies, SPSS and Mplus were used to conduct ANOVA, linear regression, confirmatory factor analysis, and Bayesian estimation. To test moderating effects, the Johnson-Neyman technique was applied in R, with plots illustrating the moderation. Results confirmed the double-edged sword effect of algorithmic regulatory pressure. Specifically, algorithmic regulatory pressure positively affected service performance through approach job crafting and negatively via avoidance job crafting. These effects were amplified by algorithmic transparency and online community support. The study offers several theoretical contributions. First, it advances understanding of the dual nature and impact of algorithmic regulatory pressure in the gig economy. This pressure embodies both hindering and challenging job demands, thereby exerting a dual influence on service performance. This perspective enriches existing frameworks on job stress. Second, using the JD-R model, the study examines the underlying mechanisms of this dual effect, revealing the mediating role of job crafting and challenging prevailing assumptions about autonomy loss in digital labor. Third, by examining the moderating effects of algorithmic transparency and online community support, the study identifies key contextual resources that regulate whether the outcomes of pressure are constructive or detrimental. Finally, the findings extend the JD-R theory’s applicability to digital labor, demonstrating its relevance in emerging work arrangements and contributing to the evolution of this theoretical model. Figures and Tables | References | Related Articles | Metrics

Select

Faking modeling for forced choice measures in personality assessment based on RES theoretical framework HE Cuiting, PENG Siwei, ZHU Yian, WANG Daxun, CAI Yan, TU Dongbo 2025, 57 (10):  1832-1848.  doi: 10.3724/SP.J.1041.2025.1832 Abstract ( 726 )   HTML ( 65 )   PDF (784KB) ( 787 )   Peer Review Comments Although forced-choice (FC) assessments with social desirability matching reduce faking compared to Likert scales, the desirability of items may shift when matched in blocks and vary across contexts. Consequently, faking remains a persistent issue in FC assessments, compromising measurement accuracy and fairness. To address this, we propose a statistical model for detecting and mitigating faking in FC assessments, integrating Böckenholt’s (2014) model of RES faking theory with the Thurstone Item Response Theory (TIRT) model (Brown & Maydeu-Olivares, 2011). Our approach aims to minimize the adverse effects of faking and enhance the robustness of FC measures. Two simulation studies were conducted to evaluate the proposed RES-TIRT model. Simulation Study 1 examined model performance under varying conditions (sample size, FC scale format, item direction, trait correlation, and dimensionality). Results indicated optimal estimation accuracy when using 3-item blocks, 3 dimensions, a correlation of 0 between dimensions, and a mix of positive and negative item descriptions. Simulation Study 2 compared trait estimation accuracy between TIRT and RES-TIRT models under increasing faking prevalence. While the TIRT model performed better in faking-free conditions, its accuracy declined more sharply than the RES-TIRT model as faking increased—particularly for item parameters—demonstrating the RES-TIRT model’s superior resistance to faking. An empirical analysis further validated the model’s applicability in real-world settings. Comparing honest responses with faked responses (simulating lawyer job applications), we found that applicants strategically inflated traits like openness, agreeableness, and extraversion to meet job requirements. The RES-TIRT model effectively detected these distortions, showing significant discrepancies in these dimensions compared to the TIRT model. Additionally, the RES-TIRT model effectively captured response distortion tendencies, as evidenced by significantly elevated latent trait values θjE under faking conditions compared to honest responses. This indicates that, the faking behavior of the applicants can be successfully captured by the RES-TIRT model. Moreover, the difficulty parameter βEim triggering fake answers can be observed to determine whether a FC block is prone to faking. These empirically derived parameters enable targeted refinements in FC measure development, allowing test constructors to strategically modify or eliminate items with low faking thresholds, thereby enhancing the scale's overall resistance to response biases. In conclusion, both simulation and empirical studies have demonstrated that the RES-TIRT model is a viable alternative to the TIRT model. It can be employed to address the issue of faking in FC scales, particularly in high-stakes situations such as talent selection. Figures and Tables | References | Related Articles | Metrics

Select

A unified design method of the simplest complete Q matrix for cognitive diagnostic tests TANG Xiaojuan, MAO Mengmeng, LI Yu, DING Shuliang, PENG Zhixia 2025, 57 (10):  1849-1866.  doi: 10.3724/SP.J.1041.2025.1849 Abstract ( 300 )   HTML ( 37 )   PDF (592KB) ( 286 )   Peer Review Comments The quality of cognitive diagnose test (CDT) directly influences the performance of the diagnostic results and its remedial function. Thus, CDT design plays a vital role in the cognitive diagnosis process. A key challenge is maximizing differentiation among participants while minimizing the number of items. Although CDT is commonly applied in contexts involving polytomous attributes and responses, prior research often yields fragmented results, limiting practical application and theoretical development. A complete Q matrix represents a valuable tool that can be employed as a high-quality test for various types of tests. The (quasi-) reachability matrix is widely utilized in various testing procedures. This study aims to develop a unified, simplified testing design that is simpler than (quasi-) reachability matrix and improves usability and applicability across various cognitive diagnosis scenarios. The simplest complete Q-matrix (SCQM) is a key concept in test design, offering a straightforward yet effective way to organize test items using fewer questions. This study proposes a unified construction model of SCQM applicable to various cognitive diagnostic test types. The Unstructured SCQM (USCQM) is developed from the Structured SCQM (SSCQM) to enhance organization. The proposed unified SSCQM model is based on the item attribute total score and selecting the SSCQM in the (quasi-) reachability matrix, specifically: (1) retaining the maximum column from each group; (2) keeping columns in each group with the distinct highest average attribute scores from the maximum columns. Following Tang Xiaojuan et al. (2024), a unified USCQM model for dichotomous attributes and polytomous responses is also proposed. A simulation study was conducted to evaluate the effects of the number of attributes, attribute hierarchy, attribute levels, and complete Q matrices on the classification capabilities of SSCQM and USCQM in both long and short tests. Their performance was compared with the (quasi-) reachability and identity matrices. Results showed that classification accuracy generally declined with increasing attribute hierarchy, number of attributes, and attribute levels, except in long tests with multilevel attributes. Accuracy improved with more complete Q matrices. For short tests, SSCQM performed better with dichotomous responses, while USCQM was superior with polytomous responses. In long tests, the trend reversed. When attribute levels were 3 or 4, tests with polytomous attributes and responses achieved the highest accuracy. In other cases, binary attribute tests performed better. Empirical findings confirmed SSCQM and USCQM outperform the (quasi-) reachability matrix. Polytomous attribute and response tests provide richer diagnostic information and higher accuracy, making them valuable in educational assessments. However, research on their design remains limited and fragmented. This study introduces a unified design model of the Simplest Complete Q Matrix (SSCQM and USCQM) for cognitive diagnostic tests based on dichotomous attributes and polytomous responses. Classification accuracy was assessed in both long and short tests. Simulation and empirical results show SSCQM and USCQM outperform or match the (quasi-) reachability matrix, with especially strong performance in short test scenarios. Figures and Tables | References | Related Articles | Metrics