【作者】 杨文辉；

【导师】 姚树桥；

【作者基本信息】 中南大学 ， 应用心理学， 2010， 博士

【摘要】 目的：根据抑郁症负性偏向产生的背-腹侧系统认知情绪交互作用模型,通过行为学和事件相关电位(ERPs)以及选择性五羟色胺再摄取抑制剂(SSRIs)抗抑郁治疗追踪研究,验证(1)抑郁症负性偏向产生机制是否为从上向下注意控制缺陷导致负性信息加工增益；(2)是否与五羟色胺(5-HT)功能低下具有相关性；同时探讨这种负性偏向的方向性和时间进程。方法：从行为学、神经电生理和药物治疗学三个层面的研究逐步验证理论模型和假设。研究1从行为学上初步探讨抑郁症认知控制损害和情绪偏向特征以及两者之间的关系、两者与抑郁症状的关联。17例首次发作抑郁症患者抗抑郁剂治疗前后,以及22名对照者完成连线测验、词语流畅性测验、持续操作测验、威斯康星卡片分类测验和Stroop色词干扰测验,以及情绪词选择任务。研究2从ERPs上探讨抑郁症负性偏向产生的脑电生理机制是否为从上向下注意控制缺陷导致负性信息加工增益。20例首次发作从未治疗抑郁症患者与20名对照者完成视觉三刺激语义oddball任务,并同时记录脑电,在该任务中正性和负性情绪词在不同组块中分别作为任务相关靶刺激和任务无关干扰刺激,中性词作为标准刺激。研究3通过选择性五羟色胺再摄取抑制剂(SSRIs)抗抑郁治疗前后的ERPs追踪研究,探讨抑郁症负性偏向的脑电生理机制与SSRIs抗抑郁治疗效应,即5.HT功能的关系。16例首次发作从未治疗抑郁症患者SSRIs治疗前和治疗9周后完成视觉三刺激语义oddball任务同时记录脑电。结果：(1)研究1行为学结果显示：首次发作抑郁症患者认知控制能力相关的测验成绩受损,情绪偏向以正性情绪加工损害为特征；情绪词选择指标与认知控制能力相关的测验成绩关联；正负性情绪词选择指标的变化对抑郁症状变化有不同预测作用。(2)研究2抑郁症与对照组ERPs比较结果显示：从上向下靶刺激加工条件下,抑郁症患者对正性刺激早期阶段注意调节和认知阶段的策略控制损害,表现为靶正性词诱发的前部P2波幅增加,N2波幅降低；在刺激导向的非靶干扰刺激条件下,抑郁症患者对负性刺激的早期抑制加工损害,表现为非靶负性词诱发的P2波幅增加,对照者则表现为对正性刺激的早期注意偏向。(3)研究3抑郁症SSRIs抗抑郁治疗9周后ERPs自身比较结果显示：在靶刺激加工条件下,SSRIs抗抑郁治疗后抑郁症患者对正性刺激的早期注意和认知控制增加,表现为靶正性词诱发的前部P2波幅降低,前部N2波幅增加；在非靶干扰刺激条件下,SSRIs抗抑郁治疗后抑郁症患者对负性刺激的早期抑制增加,表现为非靶负性词诱发的P2波幅降低,且治疗前靶正性词诱发的前部N2波幅与治疗后抑郁症状和负性思维评分显著负相关。结论：(1)抑郁症情绪偏向与认知控制损害存在关联。(2)在知觉和知觉后加工阶段,正性信息加工缺陷导致负性信息加工相对增益可能是抑郁症负性偏向的潜在机制；抑郁症负性偏向可能源于正性信息加工损害,而非负性信息加工增加；(3)抑郁症负性偏向始于知觉阶段,持续至知觉后认知阶段。(4)抑郁症负性偏向可能与5-HT功能低下相关,慢性应用SSRIs抗抑郁治疗具有提高对正性情绪刺激的皮质电生理反应,抑制对负性情绪刺激皮质反应,转变负性情绪偏向作用。(5)情绪词选择指标和正性靶刺激诱发的前部P2,N2,P3等ERPs成分有可能成为未来抑郁症诊断和疗效评价的客观指标。目的：通过对首次发作抑郁症患者治疗前后认知控制与情绪偏向的行为学研究,探讨(1)抑郁症认知损害与情绪偏向的特征；(2)认知控制损害与情绪偏向的关系,以及两者与抑郁症状的关联。方法：17例首次发作抑郁症患者抗抑郁剂治疗前后,以及22名对照者完成连线测验、词语流畅性测验、持续操作测验、威斯康星卡片分类测验、Stroop色词干扰测验,以及情绪词选择任务。结果：(1)认知测验：抑郁症组治疗前连线测验A和Stroop色词一致完成时间长于对照组,Stroop色词不一致120s内正确数小于对照组,治疗后与对照组差异均无统计学意义；抑郁症组治疗前后差异有统计学意义的认知测验有连线测验A和Stroop色词干扰测验。(2)情绪词选择任务：抑郁症组治疗前后正性词遗漏数均大于负性词遗漏数,治疗后正性词遗漏数小于治疗前,与对照组差异无统计学意义；治疗前后负性词遗漏数以及正负性词错判数与对照组差异均无统计学意义；正性词平均反应时治疗后短于治疗前,负性词平均反应时治疗前后差异无统计学意义。(3)治疗前后情绪词选择指标与涉及认知控制能力的连线测验B、威斯康星卡片分类测验、持续操作测验和Stroop色词测验成绩相关。(4)治疗前正性词遗漏数与抑郁症状和负性思维评分正相关,认知控制相关的认知测验成绩与抑郁症状和负性思维评分相关无统计学意义；治疗前后正、负性词遗漏数变化对抑郁症状评分变化有不同预测作用。结论：(1)抑郁症可能存在认知控制损害和以正性情绪加工损害为特征的情绪偏向；(2)情绪偏向与认知控制损害关联；(3)正负性情绪加工变化对抑郁症状改善可能有不同关联作用。目的：通过首次发作抑郁症对情绪信息加工的事件相关电位(ERPs)研究,验证抑郁症负性偏向产生机制是否为从上向下对于正性信息注意控制缺陷导致负性信息加工增益,以及这种负性偏向发生的时间进程和方向性,为抑郁症情绪认知控制障碍提供电生理证据。方法：20例首次发作从未治疗抑郁症患者与20名对照者完成视觉三刺激语义oddball任务,并同时记录脑电。在该oddball任务中正性和负性情绪词在不同组块中分别作为任务相关靶刺激和任务无关干扰刺激,中性词作为标准刺激。结果：(1)行为学数据：与对照组比较,抑郁症组正性词正确率显著降低,负性词正确率差异无显著意义；抑郁症组正性词正确率显著低于自身负性词正确率；反应时与对照组差异无显著意义。(2)ERPs数据：在靶刺激条件下,与对照组比较,抑郁症患者靶正性词诱发的前部P2(-160-260ms)波幅显著增高,前部N2(-260-430ms)波幅边界降低,后部N2波幅显著降低；靶负性词诱发的前部P2、N2波幅与对照组差异均无显著意义；正负性词之间的组内比较,抑郁症患者靶正性词诱发的前部N2波幅较靶负性词显著降低,靶负性词诱发的P3(-400-600ms)波幅较靶正性词显著降低,而对照者靶正性词诱发的前部P2(-210-260ms)波幅较靶负性词显著降低。在非靶干扰刺激条件下,与对照组比较,抑郁症患者非靶负性词诱发的前部P2(-160-210ms)波幅边界增高；正负性词之间的组内比较显示,对照者非靶正性词诱发的前部P2(-160-210ms)波幅较非靶负性词边界增大。抑郁症患者前部N2潜伏期较对照组显著延迟。此外,抑郁症患者右侧顶枕区正性词诱发的P1(-70-130ms)潜伏期较自身负性词显著延迟。(3)ERP成分与抑郁症状和负性思维的关系：靶正性词诱发的额-中央区P2、N2波幅、非靶负性词诱发的额-中央区P2波幅(-160-210ms)与汉密尔顿抑郁评分(HAMD)和自动思维(ATQ)评分显著正相关。结论：电生理数据证实(1)在知觉和知觉后加工阶段,正性信息加工缺陷导致负性信息相对增益可能是抑郁症负性偏向的潜在机制；抑郁症负性偏向可能源于正性信息加工损害,而非负性信息加工增加；(2)抑郁症负性偏向始于知觉阶段,持续至知觉后阶段。(3)正性目标刺激诱发的前部P2,N2,P3等ERPs成分有可能成为未来抑郁症诊断检测的客观指标。目的：通过对从未治疗首次发作抑郁症患者治疗前和选择性五羟色胺再摄取抑制剂(SSRIs)抗抑郁治疗9周后ERPs追踪研究,探讨抑郁症负性偏向脑电生理机制与SSRIs治疗效应的关系,进一步验证认知情绪交互作用模型,并为抗抑郁剂治疗反应和疗效提供客观指标。方法：16例首次发作从未治疗抑郁症患者治疗前和SSRIs抗抑郁治疗9周后完成视觉三刺激语义oddball任务同时记录脑电。结果：(1)抑郁症患者治疗有效率为87.5%,汉密尔顿抑郁量表(HAMD)总分减分率75%。(2)行为学数据：抑郁症治疗9周后正性词正确率较治疗前显著提高,负性词正确率边界提高；治疗后正性词正确率边界低于负性词正确率；治疗后反应时显著短于治疗前。(3)ERPs数据：在靶刺激条件下,与治疗前相比,治疗后靶正性词诱发的前部P2(-160-260ms)波幅显著降低、前部N2(-260-430ms)波幅显著增大、P3(-400-600ms)波幅边界降低,靶负性词诱发的前部P2、前部N2、P3波幅与治疗前差异均无显著意义；在非靶干扰刺激条件下,与治疗前相比,治疗后非靶负性词诱发的前部P2(-210-260ms)波幅显著降低,非靶正性词诱发的前部P2波幅边界降低。前部N2潜伏期治疗后较治疗前显著缩短,右侧顶枕区正性词P1潜伏期治疗后较治疗前显著缩短,而负性词P1潜伏期治疗后与治疗前差异无显著意义。(4)抑郁症组治疗前靶正性词诱发的前部N2波幅与治疗后汉密尔顿抑郁评分(HAMD)和自动思维(ATQ)评分显著负相关。结论：(1)抑郁症负性偏向可能与5-HT功能低下有关；(2)SSRIs慢性治疗效应具有提高正性情绪刺激的皮质电生理反应,抑制负性情绪刺激的皮质反应,改变负性情绪偏向作用；(3)正性目标刺激诱发的前部N2成分有可能成为未来抗抑郁剂治疗效应的客观评价指标。更多还原

【Abstract】 Objective:Based on dorsal-ventral model of cognitive emotional interaction in major depression, we aimed to test (1) whether a negativity bias resultes from deficient top-down attentional control causing gain control of negative emotional information, and (2) the correlation between low serotonin (5-HT) function and negativity bias, and to explore the directionality and temporal specificity of such bias in major depression by behavioral performance and event-related potentials (ERPs) in a longitudinal study.Methods:The model and hypothesis were tested step by step from three domains of behavior, neurophysiology and pharmatherapeutics. Study 1 investigated the characteristics of impaired cognitive control and emotional bias, and their relationships, as well as both impaired cognitive control and emotional bias related with depression symptoms in major depressive disorder (MDD).17 first-episode MDD patients and 22 healthy controls performed cognitive tests of Trial Making, Verbal Fluency, Continuous Performance Test, the Wisconsin Card Sorting Test (WCST), and Stroop Color Word Test, as well as the visual choice tasks of Chinese emotional words. After 9-week antidepressants treatment,17 MDD patients were re-administered these tasks. Study 2 explored whether neurophysiological mechanisms of a negativity bias occur as a result of deficient top-down attentional control causing gain control of negative emotional information in MDD.20 nonmedicated patients with a first episode of unipolar MDD and 20 matched controls complete a visual three-stimulus semantic oddball task while electroencephalogram (EEG) activity was recorded. In the tasks, each of the positive and negative categories served in separate runs as task-relevant target or task-irrelevant nontarget, and neutral category as standard. Study 3 investigated the relationship of electroencephaphysiological mechanisms of the negativity bias and selective serotonin re-uptake inhibitors(SSRIstreatment effects. 16 nonmedicated patients with a first episode of unipolar MDD at initial time and after 9-week treatment with SSRIs, and 20 matched controls complete a visual three-stimulus semantic oddball task while electroencephalogram (EEG) activity was recorded.Results:(1) Study 1 showed that first episode MDD patients had impaired performance associated with cognitive control, and the emotional bias was characteristics of impaired processing of positive emotion, and there was a correlation of the choice index of emotion words and the performance associated with cognitive controls. The change of the choice index of positive and negative emotion words predicted the change of depressive symptoms differently. (2) Study 2 showed in the target condition, MDD patients had impaired attentional modulation of positive stimuli at early stage and impaired strategic controls of positive stimuli during later cognitive processing, reflected by enhanced anterior P2 and reduced N2 amplitude elicited by target positive stimuli. In nontarget condition, MDD patients showed impaired inhibition of negative stimuli at early stage, reflected by enhanced P2 amplitude to nontarget negative stimuli. (3) Study 3 showed that in target condition, there was enhanced attention modulation to positive stimuli at early stage and enhanced cognitive controls of positive stimuli at later cognitive stages after 9-week treatment with SSRIs, reflected by decreased anterior P2 amplitude and enhanced anterior N2 amplitude to target positive stimuli across time. In nontarget condition, the ablility of inhibition of negative stimuli was enhanced after treatment with SSRIs, reflected by reduced anterior P2 to nontarget negative stimuli across time. Anterior N2 amplitude to target positive stimuli at initial time was negatively associated with depressive symptoms and negative thoughts after treatment.Conclusions:(1) There is a correlation of emotional bias and impaired cognitive controls in MDD; and (2) the potential mechanism of negativity bias in major depression may be deficits of attentional control on positive stimuli causing relative gain control of negative stimuli at perceptual and post-perceptual stages, and the negativity bias may stem from impaired processing of positive emotional information rather than an enhanced processing of negative emotional information; and (3) the negativity bias onsets at perceptual stage and lasts until post-perceptual stages; and (4)negativity bias might be associated with low 5-HT, and SSRIs treatment effects have the role of improvement of positive emotional processing and the ablility of inhibition of negative stimuli, causing the shift of negativity bias; (5) The choice index of emotion word and the ERPs index of anterior P2 and the N2 elicited by target positive stimuli might provide an objective index for MDD in the diagnosis, the treatment response of antidepressant and prognosis in the future. Objective:We aimed to test (1) the characteristics of impaired cognitive control and emotional bias; and (2) the relationships of impaired cognitive control and emotional bias, and their correlation with depressive symptoms using the peformance of cognitive tests and the choice tases of emotional words in first-episode major depressive disorder (MDD) at initial time and after treatment.Methods:17 first-episode MDD patients and 22 healthy controls performed cognitive tests of Trial Making, Verbal Fluency, Continuous Performance Test (CPT), the Wisconsin Card Sorting Test (WCST), and Stroop Colour Word Test as well as the visual choice tasks of Chinese emotional words. After 9-week antidepressants treatment,17 patients were re-administered the tasks.Results:(1) Cognitive tests:At initial assessment, MDD patients required more time on Trial Making A and Stroop Test-congruent than controls, and performed fewer correct numbers on Stroop Test-incongruent than controls. At follow-up, MDD patients did not differ from controls on cognitive tests. In MDD patients, cognitive tests of Trial Making A and Stroop Test had statistical significance in the comparson across initial time and follow-up. (2) Chinese emotional words choice tasks:MDD patients made more errors of omission during positive than negative block before and after treatment. At follow-up, fewer errors of omission were found during positive block relative to the initial time in MDD patients, and no difference was found on errors of omission during positive block compared to controls. No difference was found on errors of omission during negative block across time in depressed patients and across groups. Nor was on distracter errors. Depressed patients required less time to respond to positive words at follow-up than at initial time, no significant difference was observed for negative across time. (3) The choice index of emotion words were associated with the performance of Trial Making B, WCST, CPT, and Stroop Color Word Test related with cognitive controls. (4) Errors of omission during positive block were positively associated with the scores of depression and negative thoughts at initial time. The performance of cognitive tests associated with cognitive controls was not correlated with the scores of depression and negative thoughts. The change of errors of omission during positive and negative block predicted the change of the Hamilton Depression Rating Scale (HAMD) scores.Conculusions:(1) There were impaired cognitive controls and emotional bias which is characteristic of impaired processing of positive emotion in MDD; and (2) there is a correlation of emotional bias and impaired cognitive controls; and (3) the change of the processing of positive and negative emotional information might correlate with the change of symptoms differently. Objective:we aimed to test whether a negativity bias occurs as a result of deficient top-down attentional control causing gain control of negative emotional information, and to explore the directionality and temporal specificity of such bias in major depression by event-related potentials (ERPs) in first-episode major depressive disorder (MDD). Methods:20 nonmedicated patients with a first episode of unipolar MDD and 20 matched controls complete a visual three-stimulus semantic oddball task while electroencephalogram (EEG) activity was recorded. In the tasks, each of the positive and negative categories served in separate runs as task-relevant target or task-irrelevant nontarget, and neutral category as standard.Results:(1) Behavior data showed that MDD patients had decreased hit rates for positive words, but no difference for negative words compared with controls; and there were significantly lower hit rates for positive words than for negative words in MDD patients; and there was no group difference in reaction time. (2) ERPs data showed that in target condition, MDD patients showed enlarged anterior P2 (～160-260ms) and marginally reduced anterior N2(～260-430ms) elicited by positive target stimuli compared with controls, while no difference was found for anterior P2 and N2 elicited by negative target stimuli between groups. Within group, controls showed significantly decreased anterior P2 (-210-260ms) elicited by target positive stimuli relative to target negative; MDD patients showed significantly reduced N2 elicited by target positive stimuli relative to target negative, and significantly reduced P3 (-400-600ms) elicited by target negative stimuli relative to target positive. In nontarget condition, MDD patients exhibited a marginal enhancement P2 for nontarget negative words compared with controls, while no difference was found to the P2 elicited by nontarget positive words and nor to the N2 elicited by nontarget stimuli including positive and negative words, compared with controls. Within group, controls showed marginally increased P2 (～160-210ms) to nontarget positive words relative to nontarget negative words. Additively, MDD patients showed delayed latency of anterior N2 compared with controls, and MDD patients presented delayed P1 latency (～70-130 ms) for positive words relative to negative over right posterior regions. (3) The relationship of ERP data with the scores of HAMD and automatic thoughts questionnaire (ATQ) showed that the amplitude of the P2 and the N2 elicited by target positive words over frontal-central regions was positively associated with the scores of HAMD and ATQ, so was the amplitude of the P2 (～160-210ms) to nontarget negative words.Conclusions:The electrophysiological data demonstrated that (1) the potential mechanism of negativity bias in major depression may be deficits of attentional control on positive stimuli causing relative gain control of negative stimuli at perceptual and post-perceptual stages; and (2) the negativity bias may stem from impared processing of positive emotional information, rather than an enhanced processing of negative emotional information. (3) The onset of the negativity bias is at perceptual stage and such negativity bias lasts until post-perceptual stages. (4) The ERPs components of anterior P2, N2 and the P3 elicited by target positive stimuli might provide objective index for early detection of MDD in the future. Objective:We aimed to explore the relationship of electroencephaphysiological mechanisms of the negativity bias and selective serotonin re-uptake inhibitors (SSRIs) treatment effects to further test the model of cognitive emotional interaction and to provide objective index for the response of antidepressant treatment and prognosis by SSRIs treatment for 9 weeks in first episode major depression.Methods:16 nonmedicated patients with a first episode of unipolar major depressive disorder (MDD) at initial time and after 9-week SSRIs treatment complete a visual three-stimulus semantic oddball task while electroencephalogram (EEG) activity was recorded.Results:(1) The response rate was 87.5% with 75% total reduction rate of the Hamilton Depression Rating Scale (HAMD-17 item) scores after 9-week treatment in MDD patients.(2) Behavior data showed that after 9-week SSRIs treatment, the hit rates was significantly increased for positive words and marginally for negative words across time; but the hit rateswas still marginally decreased for positive words relative to negative words in MDD patients at follow-up. The reaction time was significantly shortened after 9-week treatment compared with initial time.(3) The ERPs data across time showed that in target condition, there were significantly reduced anterior P2, enhanced anterior N2 and marginally reduced P3 amplitude elicited by target positive words across time, while no difference was found to amplitude of the P2, the N2 and the P3 eilicted by target negative words across time. In nontarget condition, there was significantly reduced anterior P2 elicited by nontarget negative words, and marginally reduced anterior P2 elicited by nontarget positive words across time, while no difference was found to both the N2 and the P3 elicited by nontarget stimuli. There was a significantly fastened N2 latency across time. A shortened P1 latency to positive stimuli over right posterior regions was found across time, while no difference was found to P1 latency to negative stimuli over right posterior regions across time.(4) The amplitude of anterior N2 over frontal-central regions elicited by target positive stimuli at initial time was negatively associated with the HAMD and ATQ scores after 9-week treatment.Conclusions:The electrophysiological data demonstrated that (1) negativity bias might be associated with low 5-HT; and (2) SSRIs treatment effects have the role of enhancement of cortical electrophysiological responses to positive emotional stimuli and suppression brain responses to negative stimuli, resulting in the shift of negativity bias; (3) The ERPs components of anterior N2 elicited by target positive stimuli provide objective index for the response of antidepressant treatment and prognosis of MDD.更多还原