ISSN 0439-755X
CN 11-1911/B


    25 October 2019, Volume 51 Issue 10 Previous Issue    Next Issue

    Reports of Empirical Studies
    For Selected: Toggle Thumbnails
    Reports of Empirical Studies
    Aging effect of picture naming in Chinese: The influence of the non-selective inhibition ability
    YANG Qun, ZHANG Qingfang
    2019, 51 (10):  1079-1090.  doi: 10.3724/SP.J.1041.2019.01079
    Abstract ( 3453 )   HTML ( 17 )  
    PDF (601KB) ( 3444 )   Peer Review Comments

    Older speakers frequently report more linguistic dysfluencies, verbose and even anomia in speech production than young speakers. The transmission deficit hypothesis assumes that normal aging reduces the activation transmission between the meaning and the word form of target words, which results in more failures. By contrast, the inhibition hypothesis assumes that the inhibitive ability about irrelevant information in older adults is decreased in comparison with young adults, thus more word retrieval failures in the old than in young adults. On the other hand, semantic interference effect and distractor word frequency effect have been observed in picture-word interference (PWI) task. Researchers interpreted the two effects via the competitive (i.e., lexical selection competition during lemma retrieval) and the non-competitive hypotheses (i.e., response exclusion hypothesis). The present study aims to investigate the influence of non-selective inhibition ability in picture naming by examining distractor frequency effect in young and older native Chinese speakers.

    In the PWI task, participants were instructed to name pictures as quickly and accurately while ignoring distractor words. In experiment 1, we manipulated age (young vs. older), the frequency of distractor words (High vs. Low), and the onset interval between distractors and target pictures (-100 ms, 0 ms, and 100 ms). In experiment 2, we manipulated age, the frequency of distractor words, and the frequency of target names (High, Medium, and Low). The non-selective inhibition ability was measured by stop-signal task in both groups. Distractors and pictures were presented simultaneously.

    Results indicated a target name frequency effect in both young and older groups. Importantly, we found distractor word frequency effects in young adults, but non in older adults. The distractor frequency effect in older adults was absent due to weaker phonological activation of distractor words, and thus support the transmission deficit hypothesis. The correlations between the ability of non-selection inhibition and distractor frequency effect were not significant in both groups, indicating this kind of ability did not affect the magnitude of distractor frequency effect. However, the ability of non-selection inhibition positively correlated with mean naming latencies only in older adults, indicating that the decrease of non-selective inhibition ability influences naming latencies in older adults, while the absence in the young due to the small variations of naming latencies, which need to be investigated further by covering a wide age range (18~80 years).

    References | Related Articles | Metrics
    Effects of task type, family size, and grammatical consistency on the activation of grammatical information of semantic radicals
    ZHANG Yuzhi, ZHANG Jijia
    2019, 51 (10):  1091-1101.  doi: 10.3724/SP.J.1041.2019.01091
    Abstract ( 1552 )   HTML ( 1 )  
    PDF (620KB) ( 1685 )   Peer Review Comments

    In the study of phonetic characters, whether the lexical grammatical information can be independently activated through speech comprehension or production has been a controversial topic. In recent years, increasing evidence has shown that grammar exists independently from semantics and phonetics.

    Chinese characters are the only ideograms in the world. Chinese do not emphasize the grammatical categories of words, like phonetic characters, because Chinese words can represent a variety of word classes without changing the pronunciation and orthography. The complicated grammatical features of Chinese words make them difficult to be explored, but they still arouse researchers’ interest. Researchers have examined the grammatical activation of Chinese from the perspectives of sentence context, idioms, phrases, and Chinese words. Previous studies have also examined the grammatical features of Chinese words at or beyond the level of words. Accordingly, the present research extended the previous studies through the sub-lexical pathway.

    Semantic radicals are the ideographic components and concentrated embodiment of Chinese characters. Relevant statistics and studies have shown that semantic radicals also represent grammar. On the basis of these studies, the present work used the semantic radical priming paradigm to investigate the grammatical activation and related influencing factors of Chinese words from the sub-lexical processing level.

    Experiment 1 includes two small experiments, which examined whether the activation of the semantic radical’s grammatical information under the radical priming paradigm was specific to the task. Experiments 1a and 1b adopted a 2 (semantic radical priming/control priming) × 2 (G+ Chinese characters/G- Chinese characters) two-factor within-subject design. Experiment 1a used lexical decision as the reaction task, whereas Experiment 1b replaced it with grammatical classification. The comparison of results indicate that under the grammatical classification task, the grammatical information of semantic radicals was easier to activate. The discovery also set the foundation for the selection of the reaction tasks for Experiment 2.

    Experiment 2 further investigated the influence of family size and grammatical consistency on the activation of the semantic radical’s grammatical information. Experiment 2a adopted a 2 (large family size/small family size) × 2 (G+ Chinese characters/G- Chinese characters) two-factor within-subject design to examine the influence of family size on the syntax activation of semantic radicals. The results of this experiment discovered the advantages of a big family. Experiment 2b replaced the family size factor with grammatical consistency, and the results also revealed the advantages of high grammatical consistency.

    The overall results show that grammatical information can be activated at the sub-lexical level of Chinese words and the semantic radical marks the grammatical information. However, the activation of the semantic radical’s grammatical information is specific to the task. When the task is directly targeted to the grammar of the characters, the grammatical information can be easily activated. Family size and grammatical consistency are important factors that affect the activation of the semantic radical’s grammatical information. The results further manifest the promotion effect of a large family size and high grammatical consistency.

    References | Related Articles | Metrics
    The influence of dopaminergic genetic variants and maternal parenting on adolescent depressive symptoms: A multilocus genetic study
    CAO Yanmiao, ZHANG Wenxin
    2019, 51 (10):  1102-1115.  doi: 10.3724/SP.J.1041.2019.01102
    Abstract ( 4168 )   HTML ( 21 )  
    PDF (805KB) ( 5141 )   Peer Review Comments

    For decades, there is increasing evidence for the importance of single-gene by environment interactions (G × E) in understanding the etiology of depression. However, several concerns have been raised about the ignoring the polygenic traits of depression when conducting G × E research using single loci. Within this context, the multilocus genetic profile score (MGPS) have recently emerged as an approach of capturing polygenic nature across multiple genes. In line with the monoamine deficiency hypothesis, recent research has begun to show that the combined effects of multiple dopaminergic genetic variants are stronger than the influence of any single gene examined in isolation. Additionally, genes related to the functioning of the dopaminergic system, which coordinates individual’s response to stress. However, existing G × E research has largely focused on adverse family environments (i.e., maltreatment, maternal unresponsiveness) and to a lesser extent on positive environment, such as positive parenting. Therefore, the present study aimed to examine the interaction between dopaminergic genetic variants and maternal parenting on adolescent depressive symptoms, by adopting the approach of multilocus genetic profile score.

    Participants were 1052 mother-offspring (adolescents mean age 12.31 ± 0.37 years old at the first time point, 50.2% females) dyads recruited from the community. Youth completed assessments twice with an interval of one year. Saliva samples, self-reported depressive symptoms and mother-reported parenting were collected. All measures showed good reliability. Genotyping in three dopaminergic genes were performed for each participant in real time with MassARRAY RT software version and analyzed using the MassARRAY Typer software version 3.4 (Sequenom). To examine whether multilocus genetic profile score moderates the effects of parenting on adolescent depressive symptoms and whether this potential moderating effect act in a diathesis-stress or differential susceptibility manner, hierarchical regression analyses were conducted. We also tested above questions by recoding into categorical variables and re-conducted analyses.

    The results found that multilocus genetic profile score was a significant risk factor of depression, with higher dopamine genetic risk scores (indicating lower dopaminergic neurotransmission) predicting higher levels of depression. After controlling for gender and prior depressive symptoms, the G × E effect with positive and negative parenting were also significant, suggesting that G × E interaction significantly predicted change in depression level between Time 1 and Time 2. Specifically, adolescents with higher MGPS exhibited higher risk for depression when encountered with lower levels of positive parenting and higher levels of negative parenting, compared to their counterparts with lower MGPS. The results support the diathesis-stress model and highlight the complex ways that genes and environment interact to influence development.

    These finding underscores complex polygenic underpinnings of depression and lends support for the mulitlocus genetic profile scores-environment interactions implicated in the etiology of depressive symptoms.

    References | Related Articles | Metrics
    Classification of test-anxious individuals using Event-Related Potentials (ERPs): The effectiveness of machine learning algorithms
    ZHANG Wenpei, SHEN Qunlun, SONG Jintao, ZHOU Renlai
    2019, 51 (10):  1116-1127.  doi: 10.3724/SP.J.1041.2019.01116
    Abstract ( 1999 )   HTML ( 19 )  
    PDF (781KB) ( 2406 )   Peer Review Comments

    Individuals with test anxiety always treat tests/examinations as a potential threat. This cognitive mode impairs these individuals’ cognition, attention and emotions. A traditional method classifying subjects either as high or low on test anxiety (i.e., HTA or LTA, respectively) relies on questionnaire data. Questionnaire data may be unstable due to the subjective nature of participants’ attitudes, implying a reduced classification accuracy. In search for higher levels of (data) stability and classification accuracy a new classification approach is proposed. This new approach overcomes subjective data’s negative impact on classification accuracy by relying on event-related potential (EPR) data (also referred to as ERPs), objective (multivariate, longitudinal) data which adequately capture participants’ reactions to relevant stimuli (over time). However, as ERP data may still be somewhat unstable due to individual differences between participants, (machine) learning algorithms are adopted as their ‘learning’ feature may increase both the stability of ERP data and classification accuracy.

    This study recruited 57 HTA participants and 25 LTA participants based on: (a) Test Anxiety Scale (TAS) scores, and (b) (two) specialists’ psychological diagnostic results on a single participant. Reliance on the emotional Stroop (ES) paradigm in combination with ERP technology enabled the assessment of participants’ cognitive mode related to test anxiety. In ES, the information on the ERP components P1, P2, N2, P3 and LPP ERP were selected as input for seven commonly used machine learning algorithms: Convolutional Neural Network (CNN), Logistic Regression (LR), K Nearest Neighbors (KNN), Support Vector Machine (SVM), Random Forest (RF), Artificial Neural Network (ANN), and Recurrent Neural Network (RNN). To compare the classification accuracy of these algorithms (using the complete sample of HTA and LTA subjects) important indexes (i.e., accuracy and F1-score) were calculated and compared across these algorithms.

    The results showed that: (a) the ERPs data collected in ES allow effective differentiation between HTA and LTA (P1: F(1, 80) = 11.68, p < 0.001, η 2 = 0.13; P2: F(1, 80) = 14.10, p < 0.001, η 2 = 0.15; N2: F(1, 80) = 28.55, p < 0.001, η 2 = 0.26; P3: F(1, 80) = 22.41, p < 0.001, η 2 = 0.22; LPP: F(1, 80) = 16.92, p < 0.001, η 2 = 0.18); (b) classification on the basis of ERP data using machine learning algorithms shows high accuracy and stability, that is the classification accuracy of all seven algorithms is found to be high as evidenced by an accuracy index of 71.8% or higher (CNN: 86.5%, LR: 80.3%, KNN: 71.8%, SVM: 79.0%, RF: 73.1%, ANN: 82.7%, and RNN: 79.2%) and an F1-score of 0.814 or higher (CNN: 0.911, LR: 0.868, KNN: 0.817, SVM: 0.865, RF: 0.814, ANN: 0.882, and RNN: 0.870); (c) CNN outperforms the other six common machine learning algorithms showing both the highest accuracy index and F1-score. Moreover, as over and above this (relative) superiority CNN combines the (technical) property known as ‘shift invariance’ and robustness to noise, the algorithm may be considered ideal for effectively classifying test anxious individuals using ERP data.

    It is concluded that: (a) as manifested by its ‘discriminatory’ nature and stable classification performance (as evidenced by all machine learning algorithms’ favorable values for all important indices) reliance on the ES paradigm enables machine learning leading up to effective diagnosis of test anxiety; and (b) participants’ classification into HTA and LTA by relying on ERP data which are subsequently analyzed by means of the machine learning algorithm CNN is (most) effective (i.e., as benchmarked against six other commonly used machine learning algorithms). Consequently, using ES in combination with ERP technology and the CNN machine learning algorithm can be conceived as an ideal method for diagnosing test anxiety.

    References | Related Articles | Metrics
    Good and evil in Chinese culture: Personality structure and connotation
    JIAO Liying, YANG Ying, XU Yan, GAO Shuqing, ZHANG Heyun
    2019, 51 (10):  1128-1142.  doi: 10.3724/SP.J.1041.2019.01128
    Abstract ( 5869 )   HTML ( 57 )  
    PDF (1030KB) ( 7293 )   Peer Review Comments

    The ideas of good and evil characteristics have a long history in Chinese language and culture, with these characteristics considered to be two aspects of the Chinese personality. Abundant descriptions of good and evil characteristics can be found in human nature. However, the specific structures of good and evil concepts have not yet been delineated clearly and detailed studies on these concepts are limited. A careful definition is important with regard to what comprises a good or evil personality and to understand the psychological structure of the two concepts.

    A lexical approach was applied to determine constitutes a good or evil personality in Chinese culture. In Study 1, a set of 3, 240 good and evil personality descriptors was selected from The Contemporary Chinese Dictionary and accompanied by daily life expressions. The list was condensed into 62 good and 65 evil items following evaluation and selection by 12 psychology graduates. In Study 2, we explored the dimensions of a good personality using exploratory factor analysis (EFA) with Sample 1 (n = 313) and confirmatory factor analysis (CFA) with Sample 2 (n = 518). Results revealed that good personality contained four dimensions: conscientiousness and integrity, altruism and dedication, benevolence and amicability, and tolerance and magnanimity. We refined the four-dimensions good personality lexical rating scale, which contained 27 items. In Study 3, we used the same method as in Study 2 to explore the structure of an evil personality (EFA: Sample 1, n = 367; CFA: Sample 2, n = 269). Results showed the evil personality contained four dimensions: atrociousness and mercilessness, mendacity and hypocrisy, calumniation and circumvention, and faithlessness and treacherousness. Thus, we employed a 28-item evil personality lexical rating scale. The internal consistency, reliability, and criterion-related validity of the two scales were verified.

    The study found four-factor structures of good and evil personalities using the lexical approach and psychometrics methods, which contributes to knowledge on personality. In China, Confucius regarded benevolence as the highest moral principle for a person. Therefore, good and evil traits are embedded naturally in the personality structure of Chinese people. Understanding the structure of good and evil is useful for researchers interested in Chinese culture and for studies in Chinese indigenous psychology.

    References | Related Articles | Metrics
    The behavioral patterns of stereotype activation among four different warmth-competence social groups based on Stereotype Content Model
    YANG Yaping, XU Qiang, ZHU Tingting, ZHENG Xutao, DONG Xiaoye, CHEN Qingwei
    2019, 51 (10):  1143-1156.  doi: 10.3724/SP.J.1041.2019.01143
    Abstract ( 2810 )   HTML ( 17 )  
    PDF (940KB) ( 3547 )   Peer Review Comments

    Stereotypes are vital for social interaction by facilitating social decision making as well as conserving limited time and cognitive resources. Previous studies on stereotype activation mainly focus specific social groups, such as gender, race, etc. However, exactly how stereotypes are activated among various social groups remains unknown.

    To fill this gap, we classified social groups into four clusters according to stereotype content model in the present study, measuring for perceptions of warmth and competence. These clusters form a four-way axis, that is, high warmth-high competence (HW-HC), high warmth-low competence (HW-LC), low warmth-high competence (LW-HC) and low warmth-low competence (LW-LC). Two experiments were conducted to investigate the behavioral patterns of stereotype activation among these four clusters. We predicted that the stereotype activation pattern would be similar among these four clusters.

    In the first experiment, we employed a sequential priming paradigm to explore stereotype activation explicitly. The prime stimuli were 24 social groups equally attributed to these four clusters and the target stimuli were stereotype trait words of these 24 social groups. All of the prime stimuli and target stimuli were obtained from pilot study. The participants were instructed to judge whether the target word was consistent with the stereotypes of the prime social group. Fifty undergraduates (35 female, 19~25 years old, M = 20.68, SD = 2.08) were recruited for this experiment.

    In the second experiment, to validate the findings of Experiment 1, we utilized a lexical decision task to further investigate the stereotype activation patterns among the four clusters implicitly, using the same stimuli from Experiment 1. Pseudowords were also added, corresponding to the target words of Experiment 1. Participants were asked to identify whether the target word as a real word or pseudoword. Forty eight undergraduates (32 female, 19~25 years old, M = 20.64, SD = 1.93) participated in Experiment 2.

    Four (Social groups: HW-HC, HW-LC, LW-HC, LW-LC) × 2 (Consistency: consistent vs. inconsistent) repeated measure ANOVA were examined for response time and accuracy in both experiments. The results of Experiment 1 revealed classical stereotype activation patterns for HW-HC, HW-LC, LW-HC social groups while showing a reverse pattern for LW-LC social groups. Specifically, the participants responded more quickly and more accurately on consistent condition than on inconsistent condition for the former three clusters. However, when the prime stimuli were LW-LC social groups, the reverse was true; faster and more accurate response was shown for inconsistent condition rather than consistent condition. In Experiment 2, only real word trails were analyzed. The results of Experiment 2 replicated the findings of Experiment 1. Therefore, classical stereotype activation patterns for HW-HC, HW-LC, LW-HC social groups and the reverse pattern of stereotype activation for LW-LC were relatively robust, both explicitly and implicitly, demonstrating the great differences among the stereotype activation patterns among these four clusters.

    The results of these two experiments partially support our hypothesis, while revealing an unforeseen reverse pattern of stereotype activation for LW-LC social groups. We hypothesize that this may be due to disgust elicited by LW-LC social groups. The present study expanded the research framework of stereotype activation and provided new behavioral evidence for the specificity of LW-LC. The mechanism underlying the reverse pattern of stereotype activation for LW-LC should be examined in the future.

    References | Related Articles | Metrics
    Cognitive control strategies from the perspective of perceptual conflict: An example of stereotyped information and counterstereotyped information
    CUI Yichen, WANG Pei, CUI Yajuan
    2019, 51 (10):  1157-1170.  doi: 10.3724/SP.J.1041.2019.01157
    Abstract ( 2167 )   HTML ( 10 )  
    PDF (829KB) ( 2378 )   Peer Review Comments

    The question of whether the activation of stereotype is automatic activation (automatic activation) or controlled (inhibition) has been controversial. With the in-depth study, “stereotype activation is a cognitive process which is influenced by many cognitive factors, and the conditional automation process” has attracted more and more attention. Among them, the most discussed factors are category information, prompt clue, directional attention and so on. In addition, it also involves interpersonal trust, perspective taking, cognitive psychology, imagination, perception of their age and other factors. In recent years, researchers have explained how social categories start stereotyped information from the perspective of perceived conflict. Stereotype information processing can help activate stereotype. So, does stereotype activation affect cognitive control strategies? When stereotype information is disturbed by other information and even affected by counter stereotyped information, does the stereotype associated with it automatically activate? When the two types of conflict information are subliminal priming, does cognitive control, which regulates the formation of impressions, have an effective inhibitory effect on stereotype activation? In response to these questions, the researchers began to focus on the influence of other people’s visual cues on impression formation to others. These studies controlling the stimulus presentation time and masking stimuli, so that part of the subliminal stimuli presented in the other part, stimuli presented in the above threshold. As a result, we can compare the similarities and differences between the effects of subliminal cues and subliminal cues on cognitive control. Studies have shown that when the participants are initiated by relevant cues, they acquire attributes about a social group in the long-term memory system. This information processing process is a rapid and implicit process of perceptual preparation, which consists of two stages: the social classification based on stereotyped information and the stereotype activation based on the extracted social categories. The two stages of information processing are divided into two parallel processing cognitive systems: the previous stage involves the intentional operating system, and the latter stage relates to the automatic monitoring system. Can the two cognitive systems be regarded as the carriers of active control and reactive control in the whole process of impression formation? At present, few studies have directly related to the relationship between stereotype processing and cognitive control strategies. In view of this, this study assumes that: (1) When there is a perceived conflict stereotype information and counterstereotype information, and the perceptual load is high, the impression formed by the effects of the reactive control to activate counterstereotype; and when the perceptual load is low, the impression formed by the proactive control of the role of prone to stereotype bias. (2) Cognitive control takes a “double-edged sword” model for conflict information of different intensity (stereotyped information vs. counterstereotyped information). That means that when he processing method is controlled by the proactive control of the intentional operating system (the conscious, conscious processing) is performed and when the two classes of information start simultaneously at the threshold, the processing mode is controlled by the reactive control of the automatic monitoring system (the processing of the unconscious and unconscious control).

    In this study, we used stereotype information and counterstereotyped information as an example and the experimental tasks were divided into two levels: high or low perceptual load or threshold, and subliminal priming by using the word sense Stroop paradigm and masked version of goal priming paradigm. Experiment 1 explored the cognitive control strategy in the context of conflict between stereotype information and counterstereotyped information perception. Using 2 (perceived load: high vs. low) × 2 (gender specific words and gender attributes word perception: conflict vs. compatibility) mix design. Using word meaning Stroop paradigm of to divided manipulation of perceptual load into two categories: gender traits (target) displayed above the same gender attribute words (such as “grumpy-male”) in the low perceptual load task and gender words each side presents a gender attribute words and are compatible with the gender words or conflict (such as male-grumpy-female) in the high perceptual load task. Experiment 2 explored the conflict information intensity of stereotype information and counterstereotyped information impacted on the cognitive control strategy. A hybrid design using 2 (conflict information intensity: subliminal priming vs. subliminal priming) × 2 (two types of gender trait words perception: conflict vs. compatibility). Using masked version of the start-target paradigm and affective error attribution program to make a part of other people’s information is shown in subliminal manner, while another part of others’ information is presented in a threshold manner by controlling the presentation time of two kinds of information and masking stimuli. So that we can investigate how the cognitive control strategy of stereotype formation is affected by conflict intensity.

    In Experiment 1, the dynamic changes of perceptual load determine the individual’s cognitive control strategy of stereotyped information and counterstereotyped information. The results showed the dynamic change of conflict information intensity and perceived load determines individual’s cognitive control strategy of stereotyped information and counterstereotyped information. Reactive control consumes less cognitive resources, to make processing of stereotype consistent information more quickly, to make processing of stereotype conflict information more slowly, and to activate counterstereotype easily. On the contrary, proactive control consumes such more cognitive resources as to slower processing of stereotype consistent information, to quicken processing of stereotype conflict information and to prone to stereotype bias easily. In Experiment 2, cognitive control takes a “double-edged sword” model of stereotyped information and counterstereotyped information. When stereotyped information and counterstereotyped information start simultaneously on threshold, the processing mode is controlled by the initiative of the intentional operating system which is the processing of involuntary, unconscious control. And when the two types of information start simultaneously at the threshold, the processing mode is controlled by the automatic monitoring system which is the processing of unconscious and unconscious control.

    The activation and expression of stereotype are closely related to the monitoring and coordination of cognitive control. In particular, people need stronger cognitive control in their daily life to avoid biased judgments caused by stereotyped information. As the saying goes, “Standers-by see more than gamesters.” Actor/observer effects show that the processing of the impression of others is determined by their own cognitive control mechanism. This phenomenon can be explained by experiment 1 and Experiment 2 respectively. These results indicate that the individual, individuals can flexibly balance among the two cognitive control systems (intentional operating system vs. automatic monitoring system) and adjust their weights (either initiate active control or bias reactive control), thus forming the most favorable impression processing strategy for others. Even under the threshold of perception, individuals can process the impression of others through unconscious cognitive control.

    According to Gestalt theory, group entity makes stereotyped group impression stereotype by influencing people’s conformity processing of group members’ information. Subsequently, stereotyped group impression has strong social significance because of social factors and group classification. So they can be maintained and strengthened continuously, and eventually become stereotyped. Then, whether group entity is a moderator variable contained in the cognitive control mechanism, and how is it related to cognitive control? In this regard, follow-up studies can further develop the study of the dynamic construction of impression formation cognitive control.

    References | Related Articles | Metrics
    Effect and underlying mechanism of refutation texts on the trust and moral judgment of patients
    LYU Xiaokang, FU Chunye, WANG Xinjian
    2019, 51 (10):  1171-1186.  doi: 10.3724/SP.J.1041.2019.01171
    Abstract ( 2662 )   HTML ( 39 )  
    PDF (1400KB) ( 2595 )   Peer Review Comments

    Mistrust between doctors and patients is a worldwide concern, especially in China. Among all possible driving factors, the knowledge gap between the two groups plays a key role in the formation of disparate cognitive styles between professional medical workers and laymen. If patients can be educated effectively and prompted to think like experts, they can become more compliant during medical treatment and more tolerant of unexpected diagnosis or treatment results, thus maintaining their trust in doctors. One possible effective way to make patients think like doctors is to conduct knowledge revision in health and medicine, a method that is counterintuitive to ordinary people but familiar to doctors. This process can be facilitated by the adoption of refutation texts, which state previously acquired but incorrect knowledge and then directly refute the wrong information while providing the correct knowledge.

    In the present set of experiments, we systematically examined the effect and underlying mechanism of refutation texts under medical situations. Refutation texts were constructed in the pilot study. The texts consisted of five items that were familiar to doctors but peculiar and interesting to laymen. After asking 103 college students to guess “True” or “False” on each question, correct answers were provided for the participants. Subsequently, they were asked to assess if these answers prompted them to reflect that their previous beliefs were incorrect. At least 97 participants (94.2%) made one error in the test, and 81 participants (78.6%) admitted that this process prompted a reflection on their previous beliefs, a result that proved the effectiveness of refutation texts.

    Study 1 was designed to test the effect of refutation texts constructed in the pilot study on patient’s trust and moral judgment on doctors using a 2 (intervention: with or without refutation texts) × 2 (compliance: following doctors’ instruction or not) × 2 (check result: common fever or leukemia) between-subject design. Participants were randomly assigned those eight conditions. First, participants were required to read a scenario of doctor-patient dialogue discussing whether bone marrow puncture (BMP) should be applied to a child. Then, they were required to guess the compliance of the child’s father before they were given the different combinations of the father’s decision and medical results. Participants were required to judge the intention of the doctor’s suggestion of BMP and assess how much trust they have in doctors. Results confirmed the significant effect of refutation texts on the participants’ trust and moral judgment. Participants who were provided refutation texts had higher trust scores and were more likely to judge the doctor’s intention positively and unselfishly compared with those who were not provided such texts. The medical inspection results influenced the participants’ judgment style and trust level. When the inspection showed a minor symptom, participants tended to show lower trust in the doctor and had negative attitude toward the doctor’s suggestion.

    Study 2 tested the chain mediating role of intolerance of uncertainty and tolerance on medical workers. The same procedure as in Study 1 was used, except that participants’ scores were collected using the Intolerance of Uncertainty Scale and Tolerance on Medical Worker’s Questionnaire. Data were analyzed by using Process Plugin. Results showed a chain mediation effect on moral judgment but failed to confirm the same effect on trust. Moreover, Study 3 confirmed the effectiveness of refutation texts under non-medical-related situations and the chain mediating role of intolerance of uncertainty and tolerance on medical workers’ trust.

    Our findings demonstrated the effectiveness of refutation texts in prompting laymen’s knowledge revision and reflection on their existing health beliefs. This result will improve the tolerance on unexpected treatment consequences and the doctors’ possible misdiagnosis, thus sustaining trust in the doctor. On the basis of our results, we recommend that odd but interesting knowledge of the human body and counter-intuitive medical facts can be employed as routine material in patient education to foster a reflexive attitude on possible unsatisfactory diagnosis or treatment results. Future research may construct more useful refutation text items and further explore the possible hindering effect of previous health beliefs and negative emotions of medical refutation texts.

    References | Related Articles | Metrics