The Relationship Between Negative Attention Bias, Interpretation Bias, and Depression in College Students

XU Xiliang¹, LIU Mingfan²*, CHENG Guo³, YANG Lihao³

¹College of Arts and Sciences, Hubei Normal University, Huangshi 435109, China
²School of Psychology, Jiangxi Normal University, Nanchang 330022, China
³School of Education, Xinyang College, Xinyang 464000, China

Corresponding author: LIU Mingfan, Professor; E-mail: lmfxub@sina.com

Abstract

Background In China, college students have become a high-risk group for depression. Maladaptive attentional bias and interpretation bias are significant factors contributing to the onset and maintenance of depression. Previous studies have examined the characteristics of attentional and interpretation biases in depressed individuals, yet the mechanisms through which these biases influence depression remain unclear. Objective This study aimed to investigate the relationships between negative attentional bias, interpretation bias and depression among college students, as well as to explore the mediating and moderating effects of interpretation bias in the association between attentional bias and depression. Methods Using a random sampling method, 66 college students from Xinyang College were selected as research participants between January and February 2023. The Beck Depression Inventory (2nd Edition) (BDI-Ⅱ) was used to assess the participants' depressive symptoms. An eye tracker was employed to record the participants' total gaze time bias scores, and bias interpretation materials were distributed to collect interpretation bias scores from the participants. The analysis of correlation was conducted using Spearman's correlation analysis, the mediation effect was tested using PROCESS, and the moderating effect of interpretation bias was examined through linear regression. Results Depression scores were positively correlated with total gaze duration bias scores (r=0.688, P<0.01) and negatively correlated with relative interpretation bias scores (r=-0.731, P<0.01). Additionally, total gaze duration bias scores and relative interpretation bias scores showed a negative correlation (r=-0.580, P<0.01). The indirect effect of interpretation bias was 0.278, accounting for 42% of the total effect (95%CI: 0.148–0.453), confirming its mediating role. Furthermore, the interaction between total gaze duration bias and relative interpretation bias was significant (β=-3.479, P<0.05), indicating a moderating effect of interpretation bias. Conclusions Negative attentional bias, interpretation bias and depression are significantly interrelated among college students. Negative attentional bias not only directly influences depression but also indirectly exacerbates it through interpretation bias. Moreover, interpretation bias moderates the impact of attentional bias on depression, suggesting that adaptive interpretation patterns may mitigate the adverse effects of negative attentional processing.

Keywords: Depression; Negative attention bias; Interpretation bias; College students

Introduction

Depression is a common mood disorder characterized by at least two weeks of depressed mood, loss of pleasure or interest, fatigue or low energy, accompanied by cognitive decline and somatic symptoms [1]. Surveys indicate that as of 2022, there were 95 million known cases of depression in China, with college students accounting for 21.4% of this population. The depression level among college students is higher than the national norm [2]. Excessive depression not only endangers the physical and mental health of college students but also poses a threat to campus safety [3-4]. Therefore, investigating the influencing factors and pathogenesis of depression in college students is of great significance.

Cognitive theories of depression posit that negative cognitive biases in depressed individuals form the basis for the onset and maintenance of depression [5]. Attention bias and interpretation bias are two primary types of cognitive bias [6]. Attention bias refers to the differential allocation of attentional resources toward positive or negative stimuli compared to neutral stimuli. Negative attention bias toward negative stimuli is a crucial factor in the development and maintenance of depression. When individuals exhibit negative attention bias, they predominantly receive negative stimuli, which over time creates adverse effects on cognition and ultimately leads to depression [7]. Furthermore, corrective training targeting attention bias has been shown to effectively reduce depression levels [8-9]. In summary, attention bias plays an important role in both the development and alleviation of depression. Based on this, we hypothesize that attention bias is correlated with depression and can predict depression.

Interpretation bias refers to individuals' tendency to make positive, neutral, or negative interpretations of ambiguous verbal or nonverbal information based on their habitual patterns [10]. Cognitive processing bias theory considers interpretation bias a core cognitive factor in depression [11]. Some researchers suggest that the activation of negative interpretation bias impedes the generation of benign interpretation bias [12]. Interpretation is a crucial component of cognitive processing, and individuals' negative interpretation bias adversely affects their cognition, subsequently triggering mood disorders such as depression [13].

Meanwhile, according to the combined cognitive biases hypothesis, attention bias and interpretation bias are inseparable and mutually influential [14]. Research has indicated a positive correlation between negative attention bias and negative interpretation bias [15]. According to the depression reinforcement feedback process cycle model, attention bias influences interpretation bias, which in turn significantly impacts the onset and development of depression [16]. When individuals exhibit negative attention bias, they tend to notice negative stimuli, and coupled with negative interpretation of received information, this increases the risk of depression onset and development. Therefore, interpretation bias may mediate the relationship between negative attention bias and depression.

Additionally, studies have found that positive interpretation bias can enhance self-regulation ability and increase attention to positive information, thereby alleviating depression [17]. Therefore, this study hypothesizes that interpretation bias may also moderate the relationship between attention bias and depression.

This study examines college students as participants, using eye-tracking experiments to assess individuals' negative attention bias toward emotional face images and interpretation bias materials to explore interpretation bias in ambiguous social situations. The aim is to clarify the relationships among negative attention bias, interpretation bias, and depression in college students, and to further investigate whether interpretation bias plays a mediating and moderating role between attention bias and depression. This research holds important theoretical significance for deeply analyzing the mechanisms through which negative attention bias and interpretation bias affect depression, and for effectively implementing cognitive bias correction training to alleviate depression levels.

Methods

1.1 Participants

Using a random sampling method, college students from Xinyang College were selected as research participants between January and February 2023. Inclusion criteria were: (1) normal or corrected-to-normal vision; (2) no sedatives or stimulants taken within 24 hours. Exclusion criteria were: (1) color weakness or color blindness; (2) history of brain organic disease, severe physical illness, or family history of mental illness; (3) presence of anxiety symptoms [Beck Anxiety Inventory (BAI) score >14]. G*Power 3.1 was used to estimate sample size, with the following parameters: effect size f=0.25, α=0.05, 1-β=0.95, resulting in a minimum requirement of 54 participants. A total of 300 questionnaires were distributed, 251 were returned, and 245 were valid. Through text messages, participants were asked about their willingness to participate, and 70 participants were eventually recruited for the eye-tracking experiment. Four participants were excluded due to excessive eye-tracking calibration errors, leaving 66 participants who actually completed the experiment. This study was approved by the Xinyang College Biomedical Ethics Review Committee [approval number: 2023(01)], and all participants signed informed consent forms.

1.2 Research Instruments

1.2.1 General Information Questionnaire

This study collected general demographic information including participants' gender, grade, only-child status, residence, and family structure.

1.2.2 Beck Depression Inventory (2nd Edition) (BDI-Ⅱ)

The BDI-Ⅱ revised by Wang et al. [18] was used. This self-report scale assesses depressive symptoms over the past two weeks. It consists of 21 items, each using a four-point rating scale with symptom descriptions ranging from mild to severe. The sum of all items yields the total score, with a score of 14 or above indicating depression. The Cronbach's α coefficient was 0.94.

1.2.3 Beck Anxiety Inventory (BAI)

The BAI revised by Zheng et al. [19] was used for self-assessment of anxiety symptoms. It contains 21 items, each rated on a four-point scale, with higher scores indicating more severe anxiety symptoms. A total score above 14 suggests the presence of anxiety. The Cronbach's α coefficient was 0.95.

1.2.4 Facial Expression System

Emotional face images were selected from the facial expression system established by Liu et al. [20] for emotion disorder research. The system includes 20 positive emotional face images, 20 negative emotional face images, and 40 neutral face images, all in black and white. The emotional valence of neutral face images (4.31±0.23) differed significantly from both negative face images (2.57±0.24) and positive face images (6.43±0.28) (t=-40.23, 14.05, P<0.01). The emotional valence of positive face images (6.43±0.28) also differed significantly from negative face images (2.57±0.24) (t=21.39, P<0.01). For emotional arousal, neutral face images (4.00±0.54) differed significantly from both negative face images (5.04±0.22) and positive face images (5.11±0.06) (t=4.78, 4.01, P<0.01). However, the arousal levels of positive (5.11±0.06) and negative (5.04±0.22) face images did not differ significantly (t=1.81, P=0.18). The 80 images were paired into four types: positive-neutral, neutral-positive, negative-neutral, and neutral-negative [21], with each type containing 10 pairs. Face images were uniformly sized at 7 cm × 10 cm, with a distance of 2 cm between the two emotional face images in each pair.

1.2.5 Eye Tracker

An Eyelink 1000 Plus eye tracker was used to collect eye movement data to assess participants' attention bias. The display was a 19-inch monitor with a resolution of 1024×768 pixels and a refresh rate of 60 Hz. The sampling frequency was 1000 Hz, using a 9-point calibration method. During use, participants placed their chin on a chinrest positioned approximately 65 cm from the screen, with their gaze centered at the upper quarter of the screen to more accurately capture eye movement trajectories and extract eye movement parameters [21]. The total gaze duration bias score was used as the eye movement index. Total gaze duration was defined as the total fixation time on the area of interest for a single face image. The area of interest was defined as a rectangle matching the size of the face image (7 cm × 10 cm). The total gaze duration bias score was calculated as the total time fixated on negative emotional face images divided by the total time fixated on both images in the trial. A total gaze duration bias score >0.5 indicated overall attention maintenance toward negative emotional face images, a score of 0.5 indicated no attention bias, and a score <0.5 indicated attention avoidance of negative emotional face images [22].

1.3 Experimental Procedure

Upon arrival at the laboratory, participant information was verified and participants were assigned ID numbers. The experimental program was initiated, and participants were guided to sit at the computer with their chin placed on the chinrest while adjusting the seat height. After preparation, participants read the instructions. Once they confirmed understanding of the experimental operations and procedures, a nine-point calibration was performed, with both calibration errors controlled within 1 degree. The formal experiment then began. First, a fixation point "+" appeared in the center of the screen for 1000 ms. After participants focused their attention, emotional face images were presented, one on each side of the screen. Each pair of emotional face images was presented for 5000 ms before automatically disappearing, after which the "+" reappeared on the screen. This process repeated until all 40 pairs of emotional face images had been presented. The presentation order of the emotional face image pairs was randomized, and participants were required to maintain fixation on the screen throughout the experiment. The experimental procedure is illustrated in Figure 1 [FIGURE:1].

After the eye-tracking experiment, participants were guided to a lounge to listen to relaxing music and rest, reducing any emotional priming effects from the emotional images in the eye-tracking experiment. Ten minutes later, each participant was given a paper version of the interpretation bias materials. Participants were asked to carefully read the materials, imagine themselves in the described social situations, and rate each interpretation based on their own feelings. There was no time limit for the rating process.

1.5 Statistical Analysis

Data statistical analysis was completed using SPSS 23.0 and the PROCESS macro. Data conforming to normal distribution were expressed as (mean ± SD), while non-normally distributed data were expressed as M (Q1, Q3). Between-group comparisons were performed using Mann-Whitney U tests. Harman's single-factor test was used to assess common method bias. Spearman correlation analysis was employed to explore the relationships among attention bias, interpretation bias, and depression. Model 4 in PROCESS was used to test the mediating effect of interpretation bias. Linear regression analysis was conducted to examine the moderating effect of interpretation bias, with simple slope graphs plotted. P<0.05 was considered statistically significant.

Results

2.1 Basic Information of Participants

Comparison of BDI-Ⅱ scores among participants with different only-child status showed statistically significant differences (P<0.05). Comparisons of BDI-Ⅱ scores across different genders, ages, residences, and family structures showed no statistically significant differences (P>0.05). See Table 1 [TABLE:1].

2.2 Common Method Bias Test

Harman's single-factor test was used to assess common method bias. An unrotated exploratory factor analysis of all measurement items revealed that eight common factors with eigenvalues greater than 1 were extracted, and the first common factor explained 26.92% of the total variance, which is lower than the 40% criterion proposed by Podsakoff et al. [26]. Therefore, this study does not have serious common method bias.

2.3 Correlation Analysis

Participants' mean BDI-Ⅱ score was 13.500 (5.000, 21.000), mean total gaze duration bias score was (0.468±0.091), and mean relative interpretation bias score was (15.410±18.423). BDI-Ⅱ scores were positively correlated with total gaze duration bias scores (P<0.01); higher BDI-Ⅱ scores corresponded to greater total gaze duration bias scores, meaning participants with higher BDI-Ⅱ scores spent more time viewing negative images. BDI-Ⅱ scores were negatively correlated with relative interpretation bias scores (P<0.01); higher BDI-Ⅱ scores corresponded to lower relative interpretation bias. Total gaze duration bias scores were negatively correlated with relative interpretation bias scores (P<0.01); greater total gaze duration bias scores corresponded to lower relative interpretation bias.

Table 2 [TABLE:2] shows the correlation analysis of depression, attention bias, and interpretation bias among college students. The correlation matrix indicates: BDI-Ⅱ score with total gaze duration bias (r=0.688, P<0.01), BDI-Ⅱ score with relative interpretation bias (r=-0.731, P<0.01), and total gaze duration bias with relative interpretation bias (r=-0.580, P<0.01). Note: a indicates P<0.05; — indicates duplicate data not shown.

2.4 Mediating Effect of Interpretation Bias Between Attention Bias and Depression

Correlation analysis of BDI-Ⅱ scores, relative interpretation bias scores, and total gaze duration bias scores revealed significant relationships among the three variables. Further analysis of their relationships was conducted following the mediation test procedure proposed by Wen and colleagues [27]. Mediation analysis was performed using the PROCESS macro, selecting Model 4 (simple mediation) with Bootstrap sampling limited to 5,000 iterations. If the 95% confidence interval did not contain zero, the mediation effect was considered significant. Test results are shown in Table 3 [TABLE:3] and Table 4 [TABLE:4].

The results indicated that total gaze duration bias scores significantly predicted BDI-Ⅱ scores (β=0.656, P<0.01). After adding the mediating variable, total gaze duration bias scores still predicted BDI-Ⅱ scores (β=0.378, P<0.01). Additionally, total gaze duration bias scores negatively predicted relative interpretation bias scores (β=-0.561, P<0.01), and relative interpretation bias scores negatively predicted BDI-Ⅱ scores (β=-0.496, P<0.01). The 95% confidence intervals for both the direct effect of total gaze duration bias scores on BDI-Ⅱ scores and the indirect effect through relative interpretation bias scores did not contain zero, confirming the mediation effect. The direct effect was 0.378, accounting for 58% of the total effect (95%CI: 0.185–0.570), while the indirect effect was 0.278, accounting for 42% of the total effect (95%CI: 0.148–0.453). In summary, total gaze duration bias scores can not only directly predict BDI-Ⅱ scores but also indirectly influence BDI-Ⅱ scores through the mediating role of relative interpretation bias scores. Relative interpretation bias scores partially mediated the relationship between total gaze duration bias scores and BDI-Ⅱ scores. The mediation model is illustrated in Figure 2 [FIGURE:2].

2.5 Moderating Effect of Interpretation Bias Between Attention Bias and Depression

To examine the moderating effect of interpretation bias on the relationship between attention bias and depression, regression analysis was conducted with total gaze duration bias scores, relative interpretation bias, and their interaction term as predictor variables, and BDI-Ⅱ scores as the dependent variable. As shown in Table 5 [TABLE:5], total gaze duration bias scores had a positive effect on BDI-Ⅱ scores (P<0.01); higher total gaze duration bias scores corresponded to higher BDI-Ⅱ scores. Relative interpretation bias had a negative effect on BDI-Ⅱ scores (P<0.01); higher relative interpretation bias corresponded to lower BDI-Ⅱ scores. The interaction term between total gaze duration bias scores and relative interpretation bias had a negative effect on BDI-Ⅱ scores (β=-3.479, P<0.05), confirming the moderating effect of interpretation bias. To further verify the moderating effect, a simple slope graph was plotted to visually reflect the moderating role of relative interpretation bias. The slope of the line for low relative interpretation bias was steeper than that for high relative interpretation bias, indicating that when relative interpretation bias was low, total gaze duration bias scores had a stronger impact on BDI-Ⅱ scores. In other words, relative interpretation bias negatively moderated the effect of total gaze duration bias scores on depression, as shown in Figure 3 [FIGURE:3].

Discussion

3.1 Effect of Attention Bias on Depression

This study found that attention bias was positively correlated with depression and could positively predict individuals' depressive symptoms. The attentional narrowing theory suggests that depressed individuals tend to allocate attention to depression-related stimuli [28], and this preference causes individuals to receive and process more negative information. As this negative attention bias is continuously reinforced, attentional focus eventually becomes confined to negative stimuli, and this narrowing of attentional focus undoubtedly further exacerbates individuals' depression levels. Therefore, improving negative attention bias in depressed individuals is an important approach to alleviating depressive emotions. Research has shown that attention bias training is more effective than conventional medication in improving depressive emotions in patients with depression [29], and positive attention training for depressed individuals can improve their negative attention bias and subsequently alleviate depressive emotions [30].

3.2 Mediating Effect of Interpretation Bias

This study found that interpretation bias partially mediated the relationship between total gaze duration bias scores and depression. Total gaze duration bias scores can influence depression both directly and indirectly through interpretation bias. The depression reinforcement feedback process cycle model suggests that individuals with negative attention bias are more likely to develop negative interpretation bias, thereby increasing depressive emotions [16]. According to the information processing analysis model, when individuals exhibit attention bias toward negative information, their understanding of social information also shows a negative tendency. Individuals' negative interpretation of information ultimately activates the depression network [31], and negative interpretation bias not only increases depression risk but also predicts the likelihood of future depression onset.

3.3 Moderating Effect of Interpretation Bias

The innovation of this study lies not only in analyzing the mediating role of interpretation bias between attention bias and depression but also in exploring its moderating effect. This study found that when relative interpretation bias was high, it weakened the effect of total gaze duration bias scores on depression; when relative interpretation bias was low, it strengthened the effect of total gaze duration bias scores on depression scores. Positive interpretation bias can mitigate the impact of negative attention bias on negative emotions to some extent [32-33]. Attention bias and interpretation bias both belong to cognitive processes. When individuals tend to interpret negative stimuli with a positive and optimistic attitude, it increases the positivity of their cognitive processes to a certain degree, thereby alleviating depressive emotions.

In conclusion, individuals' depressive emotions are simultaneously influenced by negative attention bias and interpretation bias. The combined cognitive biases hypothesis suggests that cognitive biases do not exist in isolation but rather influence each other, and the interaction among cognitive biases likely has a greater impact on disorders than single biases alone [34]. Beck's negative cognitive schema theory posits that the negative schemas existing in depressed patients' minds are stable and enduring. If only one component of the cognitive system is improved, it may not change the overall nature of the cognitive schema. Therefore, to significantly improve individuals' depressive emotions, attention should be paid to the impact of the overall cognitive process on depression, particularly by correcting negative interpretation bias to reduce the influence of negative attention bias on depression. This study not only provides theoretical support for the combined cognitive biases hypothesis but also offers reference for cognitive therapy of depression.

Based on this research, the following conclusions can be drawn: (1) Attention bias, interpretation bias, and depression are significantly interrelated. Individuals with higher depression scores exhibit more negative attention bias and negative interpretation bias. (2) Interpretation bias mediates the relationship between attention bias and depression. Attention bias can influence depression both directly and indirectly through interpretation bias. (3) Interpretation bias moderates the relationship between attention bias and depression. When participants have positive interpretation bias, it weakens the effect of attention bias on depression; when participants have negative interpretation bias, it strengthens the effect of attention bias on depression.

Study Limitations: First, this study had a relatively small sample size, and participants did not undergo professional clinical diagnosis; therefore, caution is needed in clinical generalization. Future research could include clinically diagnosed depressed patients to increase generalizability. Second, this study only examined participants' delayed interpretation bias and did not measure immediate interpretation bias. Future studies could simultaneously include both immediate and delayed interpretation bias in the same experiment to more comprehensively examine the mediating or moderating role of interpretation bias. Finally, this study only explored the effects of attention bias and interpretation bias on depression and did not include other cognitive processes such as memory bias. Therefore, future research could incorporate memory bias and other cognitive processes to more comprehensively analyze the impact of cognition on depression.

Author Contributions: XU Xiliang proposed the main research objectives, was responsible for research conception and design, implementation, and manuscript writing. CHENG Guo conducted experimental operations, data collection and organization, statistical analysis, and figure and table preparation. LIU Mingfan was responsible for quality control and review of the article and manuscript revision. YANG Lihao was responsible for experimental operations, statistical analysis, and manuscript writing and revision.

Conflicts of Interest: This article has no conflicts of interest.

References

[1] Huang Y. Research on the Correlation Between TCM Syndrome Differentiation of Depression and Objective Indicators Based on Literature and Clinical Evidence[D]. Tianjin: Tianjin University of Traditional Chinese Medicine, 2021. DOI: 10.27368/d.cnki.gtzyy.2021.000504.

[2] Zhang R, Yang C, Zhang Y. Meta-Analysis of Influencing Factors of Depression Among Chinese College Students[J]. Chinese General Practice, 2020, 23(35): 4497-4502.

[3] Suwinyattichaiporn T, Johnson Z D. The impact of family and friends social support on Latino/a first-generation college students' perceived stress, depression, and social isolation[J]. J Hispanic High Educ, 2022, 21(3): 297-314. DOI: 10.1177/1538192720964922.

[4] Wang Q, Qi Y, Hu W, et al. The Relationship Between Social Rejection and Depressive Emotions in College Students: The Mediating Role of Emotional Awareness and the Moderating Role of WeChat Parent-Child Communication[J]. Psychological Science, 2022, 45(5): 1159-1165. DOI: 10.16719/j.cnki.1671-6981.20220517.

[5] Beck A T. Cognitive Therapy and the Emotional Disorders[M]. Penguin USA, 1979.

[6] Everaert J, Duyck W, Koster E H W. Attention, interpretation, and memory biases in subclinical depression: a proof-of-principle test of the combined cognitive biases hypothesis[J]. Emotion, 2014, 14(2): 331-340. DOI: 10.1037/a0035250.

[7] Beck A T, Bredemeier K. A unified model of depression[J]. Psychol Med, 2016, 46(3): 551-562. DOI: 10.1017/s0033291715002044.

[8] Zhao F. The Influence of Mindfulness and Emotion Regulation Strategies on Negative Attention Bias: From the Perspective of Depression Cognitive Model Theory[D]. Qingdao: Qingdao University, 2022. DOI: 10.27262/d.cnki.gqdau.2022.001686.

[9] Xia H, Liu X, Gou X, et al. Meta-Analysis of the Intervention Effect of Attention Bias Modification Therapy on Depressed Patients[J]. Chinese General Practice, 2022, 25(17): 2095-2102.

[10] Goyer M S, McKee L G, Parent J. Positive and negative interpretation biases in the relationship between trait mindfulness and depressive symptoms in primarily white emerging adults[J]. Mindfulness, 2022, 13(5): 1258-1270. DOI: 10.1007/s12671-022-01873-z.

[11] Fodor L A, Georgescu R, Cuijpers P, et al. Efficacy of cognitive bias modification interventions in anxiety and depressive disorders: a systematic review and network meta-analysis[J]. Lancet Psychiatry, 2020, 7(6): 506-514. DOI: 10.1016/S2215-0366(20)30130-9.

[12] Clarke S C, Cooper N R, Rana M, et al. Cognitive interpretation bias: the effect of a single session moderate exercise protocol on anxiety and depression[J]. Front Psychol, 2018, 9: 1363. DOI: 10.3389/fpsyg.2018.01363.

[13] Sugita A, Yoshimura S. Impact of interpretation bias on depression in ambiguous situations: a panel survey with a three-month interval[J]. J Exp Psychopathol, 2022, 13(3): 20438087221123242. DOI: 10.1177/20438087221123242.

[14] Sanchez A, Duque A, Romero N, et al. Disentangling the interplay among cognitive biases: evidence of combined effects of attention, interpretation and autobiographical memory in depression[J]. Cogn Ther Res, 2017, 41(6): 829-841. DOI: 10.1007/s10608-017-9858-5.

[15] Fang X, Liu J. Interpretation Bias of College Students with Different Levels of Self-Support in Positive and Negative Events[J]. Chinese Journal of Clinical Psychology, 2017, 25(4): 709-713. DOI: 10.16128/j.cnki.1005-3611.2017.04.025.

[16] Wittenborn A K, Rahmandad H, Rick J, et al. Depression as a systemic syndrome: mapping the feedback loops of major depressive disorder[J]. Psychol Med, 2016, 46(3): 551-562. DOI: 10.1017/s0033291715002044.

[17] Jung E S, Ha J H, Lee S W. The relationships between self-regulation and depression on university students: the moderating effect of interpretation biases of positive social events[J]. Korean J Youth Stud, 2020, 27(2): 101-121. DOI: 10.21509/kjys.2020.02.27.02.101.

[18] Wang Z, Yuan C, Huang J, et al. Reliability and Validity of the Chinese Version of the Beck Depression Inventory-II in Depressed Patients[J]. Chinese Mental Health Journal, 2011, 25(6): 476-480. DOI: 10.3969/j.issn.1000-6729.2011.06.014.

[19] Zheng J, Huang C, Huang J, et al. A Study on the Psychometric Properties, Norm Scores, and Factor Structure of the Beck Anxiety Inventory[J]. Chinese Journal of Clinical Psychology, 2002, 10(1): 4-6.

[20] Liu M, Wu H, Cui Y. Preliminary Establishment of a Facial Expression System for Emotion Disorder Research[J]. Psychological Exploration, 2012, 32(4): 340-346.

[21] Xu X, Liu M, Li N, et al. An Eye Movement Study on Attentional Bias Toward Emotional Faces in Depressed College Students with Alexithymia[J]. Chinese Journal of Behavioral Medicine and Brain Science, 2023(1): 56-61. DOI: 10.3760/cma.j.cn371468-20220517-00269.

[22] Wang M. Analysis of Eye Movement Indicators of Attentional Bias Toward Negative Emotional Stimuli in Depressed Patients[J]. Modern Practical Medicine, 2021, 33(2): 189, 207. DOI: 10.3969/j.issn.1671-0800.2021.02.022.

[23] Jiang D. The Interpretation Bias of Ambiguous Social Information by College Students' Rejection Sensitivity[D]. Yantai: Ludong University, 2017.

[24] Hirsch C R, Meeten F, Krahé C, et al. Resolving ambiguity in emotional disorders: the nature and role of interpretation biases[J]. Annu Rev Clin Psychol, 2016, 12: 281-305. DOI: 10.1146/annurev-clinpsy-021815-093436.

[25] Zheng F. Development and Application of an Assessment Tool for Depressive Interpretation Bias[D]. Chongqing: Army Medical University of the Chinese People's Liberation Army, 2019. DOI: 10.27001/d.cnki.gtjyu.2019.000375.

[26] Zhou H, Long L. Statistical Tests and Control Methods for Common Method Bias[J]. Advances in Psychological Science, 2004, 12(6): 942-950.

[27] Wen Z, Ye B. Mediation Effect Analysis: Method and Model Development[J]. Advances in Psychological Science, 2014, 22(5): 731-745. DOI: 10.3724/SP.J.1042.2014.00731.

[28] Liu L. Study on the Relationship Between Emotional Attention Bias and Cognitive Function in First-Episode Depressed Patients[D]. Nanchang: Nanchang University, 2021. DOI: 10.27232/d.cnki.gnchu.2021.002634.

[29] Zhu W, Zhang E, Yin L, et al. The Effect of Attention Bias Modification Training on Depressed Patients[J]. Journal of Clinical Psychiatry, 2022, 32(6): 462-464. DOI: 10.3969/j.issn.1005-3220.2022.06.012.

[30] Xu X. An Experimental Study on Alleviating Depressive Emotions in College Students Through Positive Attention Training[J]. Journal of Xinyang Agriculture and Forestry University, 2015, 25(3): 155-157. DOI: 10.16593/j.cnki.41-1433/s.2015.03.050.

[31] Lau J Y F, Watkins-Muleba R, Lee I, et al. Promoting helpful attention and interpretation patterns to reduce anxiety and depression in young people: weaving scientific data with young peoples' lived experiences[J]. BMC Psychiatry, 2021, 21(1): 403. DOI: 10.1186/s12888-021-03320-0.

[32] Broadbent P, Schoth D E, Liossi C. Association between attentional bias to experimentally induced pain and to pain-related words in healthy individuals: the moderating role of interpretation bias[J]. Pain, 2022, 163(2): 319-333. DOI: 10.1097/j.pain.0000000000002318.

[33] Yu H, Qian M, Zhou P, et al. Attention Bias and Social Anxiety: The Moderating Role of Interpretation Bias[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2016, 52(3): 574-580. DOI: 10.13209/j.0479-8023.2016.030.

[34] Zhu W, Si F, Deng X, et al. Research Progress on Combined Cognitive Biases in Depression[J]. Journal of Neuroscience and Mental Health, 2022, 22(8): 591-595.