语义干扰效应发生阶段的ERP研究

采用延时命名任务,运用事件相关电位(ERP)技术考察了图-词干扰范式下语义干扰效应的发生阶段。在屏幕上呈现图片(如"狗")及其干扰词800ms后,要求被试忽略干扰词命名图片。干扰条件有三:语义关联干扰(如"猫")、语义无关干扰(如"树")和无刺激干扰。结果发现,从图片呈现大约180ms-700ms的时窗内,语义关联干扰和语义无关干扰引起的脑电波与无刺激干扰引起的脑电波均有显著差异。表明语义干扰效应发生在言语产生中的多个阶段。     

Simon效应的近红外研究

该研究采用近红外成像技术,以30名大学生为被试,旨在验证近红外技术对于探讨反应层面冲突控制神经机制的有效性.实验过程中,要求被试按键判断计算机屏幕上出现的颜色块的颜色,包含2个实验条件:不一致条件(刺激颜色对应的按键位置和刺激出现的位置不同)和一致条件(刺激颜色对应的按键位置和刺激出现的位置相同).结果发现,不一致条件的反应时显著长于一致条件的反应时,说明被试需要更多的时间抑制冲突.此外,氧合血红蛋白浓度(HBO)的变化在刺激出现后9~12s时达到最高值.在此时间段内对Simon效应和HBO在两条件之间的差异进行相关分析,发现在9~10s时额叶左侧的HBO差值与Simon效应有统计学意义(P0.05);在11~12s时额叶左侧和额中右侧的HBO差值也与Simon效应有统计学意义(P0.05).这说明额叶与个体冲突控制能力有着显著相关,验证了近红外技术对于探讨冲突控制神经机制的有效性.     

语言理解中P600效应的新解读——监控理论及双路线模型的扩展

语义违例句诱发了P600效应对P600是句法加工的标志的说法提出了质疑.监控理论从执行控制参与语言理解的角度阐释了P600,并提出了双路线模型.而后的研究发现在各个语言水平上都存在对错误加工的监控,双路线模型得到扩展,提出P600反映了预期和非预期表征之间出现冲突时引发的监控反应.在对监控理论和模型进行分析和评价的基础上,认为它为P600效应实质的研究提供了新视角,但是还有待完善和深入研究.     

融合文本内容和标签引导文本编码的分类方法

针对当前文本分类算法未能充分利用标签的语义表示导致文本表示学习与分类预测割裂的问题,提出一种融合文本内容编码和标签引导文本编码的文本分类方法.在文本内容编码部分,通过长短时记忆网络获得文本序列累计语义表示,通过自注意力机制捕获和强化文本长距离语义依赖.在标签引导文本编码部分,设计交互注意力机制,通过标签引导得到经由标签的语义表示过滤下文本的新表示.最后将两部分输出融合,得到同时具有本地内容语义信息和全局任务指导信息的文本表示,使得模型在文本表示阶段即对分类任务具有早期感知力.实验结果表明本研究方法在真实数据集上可有效提升文本分类任务性能.     

利手与两半球的听觉词汇语义加工

左右利手被试在双耳分听呈现条件下完成单个汉字具体名词(动物、物体和身体部位)的范畴归类任务, 检查不同利手者大脑两半球加工听觉词汇语义的大脑功能偏侧化与整合效应.结果显示:(1) 右利手被试用左手或右手反应在左耳内的错误率显著低于右耳内,提示存在明显的右半球优势;(2) 左利手被试用左手反应在左耳内的错误率也显著低于右耳内,但用右手反应在右耳内与左耳内的错误率差异不显著,提示两半球可能都有一定的听觉词汇语义加工能力;(3) 右利手被试用左手或右手反应在左-右耳间或右-左耳间的错误率均显著高于左耳内,而与右耳内没有差异,提示缺乏明显的语义加工在半球间的整合优势;(4) 左利手被试用右手反应在左-右耳间和右-左耳间的错误率均明显低于左耳内和右耳内,提示左利手被试的左半球可能具有更强的听觉词汇语义加工在半球间的整合能力.     

不同工作载荷下的警示性图标认知机理

采用工作载荷范式研究警示性图标在不同工作载荷下的认知机理,以及视觉突显性与图形语义信息对警示性图标认知的作用.首先进行了有语义图标实验,以检验具有语义信息的警示性图标在高低知觉负载下的注意捕获效应,结果表明,在高、低工作载荷下警示性图标都比普通图标具有加工优先性,但与任务无关的警示性图标在高、低工作载荷下的注意捕获能力没有显著差异,说明警示性图标能够自动获取注意系统的加工.为分离突显性视觉特征与语义信息2个因素对实验结果的影响,进行了无语义图标实验,将警示性图标转换为具有突显性但无语义信息的转置图标,发现警示性图标在高工作载荷下的注意捕获效应显著小于低工作载荷,即工作载荷能够显著调节其注意捕获能力,说明警示性图标在有语义图标实验中的注意捕获效应应归因于其本身携带的语义信息.本研究证实语义明确的警示性图标能够被注意系统自动捕获,图形语义信息在其中起到关键性作用,突显性的视觉特征也会增加警示性图标的注意捕获效应.     

不同熟练水平中-英双语者的抑制控制加工

采用中文和英文Stroop测验,检查了不同熟练水平大学生中英双语者在语言内(中中和英英)干扰和语言间(中英和英中)干扰条件下的抑制控制加工.结果发现:(1)熟练双语者在2种语言内条件下的干扰效应均明显小于不熟练双语者,提示双语学习可以明显提高双语者的语言抑制控制能力;(2)在中英语言间条件下没有出现不同熟练水平双语者的干扰效应差异,但在英中语言间条件下熟练双语者的干扰效应明显强于不熟练双语者,提示双语者在用目标语言表达时会受到来自非目标语言的竞争性抑制.本研究结果为双语词汇产生的抑制控制(IC)模型提供了更为直接的证据.     

不同引导任务对图-字Stroop效应的影响——对注意力敏感模型的验证

将引导任务与图-字Stroop任务相结合,通过考察不同引导任务对Stroop效应的影响来探讨注意是否影响自动化加工,以及如何影响的问题,以进一步验证注意力敏感模型.两个实验的结果发现:引导任务与Stroop任务中的条件匹配之间存在显著的交互作用;不同引导任务对Stroop效应量的影响不同,语义引导任务下的Stroop效应量明显高于知觉引导任务;引导任务下的反应时间显著高于无引导任务,正确率恰好相反.结论:实验结果与注意力敏感模型相符.注意系统对自动化加工具有调节作用.这种调节作用是通过增加对相关通路的敏感性,减低无关通路的敏感性来实现的.     

延迟干扰对空间工作记忆信息再认的影响

通过延迟匹配的实验方式,要求被试完成空间工作记忆任务和延迟干扰任务,主要对空间工作记忆再认阶段的行为和ERP数据进行分析.实验发现：延迟干扰任务对空间工作记忆的再认速度产生影响,正探测条件下,位置不同条件显著快于位置相同条件,而负探测条件则正好相反.正探测条件在枕区诱发的P1和N1的波幅大于负探测条件.此外,还发现不同实验条件下,P300的潜伏期存在差异,正探测条件下,位置相同条件的P300潜伏期显著长于位置不同条件;而在负探测条件下,位置相同条件的P300潜伏期显著短于位置不同条件.实验结果说明延迟干扰任务对空间工作记忆再认的影响主要出现在刺激评估和反应决策阶段.     

中国英语学习者的英语词素加工和汉语偏旁加工关联研究

以中国学习者为研究对象,对比英汉词素加工和偏旁加工,探究二语学习者的英汉复杂形态词的加工状况.本文通过三个实验来分别研究英语语义相关的词素、汉字语义相关的偏旁和语义相关的英汉词汇在词汇加工中的作用和异同.实验一研究结果显示,英语语义相关的英语词素在二语者英语词汇加工处理的过程中起着很重要的促进作用.实验二表明,语义相关的汉语偏旁对汉字的加工虽然没有取得具有统计意义的显著水平,但是也有一定的促进加工作用.在实验三中,语义相关的汉语偏旁对相关的英语词汇的启动效应也没达到统计意义;同时,语义相关的英语单词对相关的汉字的启动效应也很微小.然而无一例外的是语义促进效应虽然微弱,但都普遍存在.     

月经周期中情绪冲突Stroop效应研究

探讨正常育龄妇女月经周期对情绪冲突的影响,以及不同时期的情绪注意偏向.研究采用重复测量设计的词-面孔Stroop任务,分析比较了21名被试在卵泡期和黄体期完成情绪判断任务和性别判断任务的正确率和反应时.结果发现:被试在两个任务中均表现出明显的情绪Stroop效应;在情绪判断任务中,月经周期对情绪冲突Stroop任务的影响有统计学意义(p0.01),在黄体期被试的情绪Stroop效应更加明显,对负性和正性情绪面孔识别的正确率和反应时均存在统计学意义(p0.01);在性别判断任务中发现,与黄体期相比,女性在卵泡期对男性面孔的识别正确率较高,反应时较短.研究结果表明:女性在黄体期的认知控制能力较卵泡期下降,同时对负性情绪表现出明显的注意偏向;卵泡期女性对男性面孔的偏好大于黄体期女性.     

Right prefrontal laterality of Chinese boys during a Chinese high conflict Stroop task measured by multi-channel near-infrared spectroscopy imaging

Some studies have shown that native Chinese speakers have different laterality in matched Stroop tasks from native English speakers. Recently, many imaging data, which show left laterality of English-matched Stroop interference, have been reported. And a few functional imaging studies have been conducted to investigate the phenomenon of the Chinese version of Stroop task. In this study, functional activity in the lateral prefrontal cortex of a group of normal Chinese boys with functional near-infrared imaging during a Stroop color-word task was measured to show different Stroop interferences in the prefrontal cortex. The results show obvious fluctuation of the cerebral blood volume in the right prefrontal cortex in all boys, which agrees with the finding of previous studies, that is, Chinese native boys have right laterality in their brain when the Chinese version of Stroop color-word task is applied.     

Right prefrontal laterality of Chinese boys during a Chinese high conflict Stroop task measured by multi-channel near-infrared spectroscopy imaging

Some studies have shown that native Chinese speakers have different laterality in matched Stroop tasks from native English speakers.Recently,many imaging data,which show left laterality of English-matched Stroop interference,have been reported.And a few functional imaging studies have been conducted to investigate the phenomenon of the Chinese version of Stroop task.In this study,functional activity in the lateral prefrontal cortex of a group of normal Chinese boys with functional near-infrared imaging during a Stroop color-word task was measured to show different Stroop interferences in the prefrontal cortex.The results show obvious fluctuation of the cerebral blood volume in the right prefrontal cortex in all boys,which agrees with the finding of previous studies,that is,Chinese native boys have right laterality in their brain when the Chinese version of Stroop color-word task is applied.     

Conflict adaptation in the perception of emotional valence in a Stroop task

Interference effects are commonly observed when conflicts arise between task-relevant and task-irrelevant dimensions of stimuli. Compared with responses to congruent trials in which the two dimensions are compatible, responses to incongruent trials are usually slower and less accurate. Moreover, the performance in a given trial is affected by the nature of the preceding trial, in that the interference effect for the current incongruent trial is reduced when it follows an incongruent trial than when it follows a congruent one. This conflict adaptation effect has been generally ascribed to the functioning of the cognitive control system, in which higher controls are recruited after conflicts to further constrain the processing of the task-irrelevant dimension. However, due to its uniqueness in neural and cognitive processes, whether the perception of emotional valence shows the same conflict adaptation effect in the conflicting context is an open question. This study adopted a variant of the Stroop task in which the task-relevant dimension was the affective facial expression while the task-irrelevant dimension was the meaning of an emotional word superimposed on the face. Results suggest that the perception of emotional valence in the conflicting context can have aspects that are both similar and dissimilar to the control of conflict for non-emotional information. The capturing of attention by emotional information can alter the functioning of the cognitive control system.     

Conflict adaptation in the perception of emotional valence in a Stroop task

Interference effects are commonly observed when conflicts arise between task-relevant and task-irrelevant dimensions of stimuli. Compared with responses to congruent trials in which the two dimensions are compatible, responses to incongruent trials are usually slower and less accurate. Moreover, the performance in a given trial is affected by the nature of the preceding trial, in that the interference effect for the current incongruent trial is reduced when it follows an incongruent trial than when it follows a congruent one. This conflict adaptation effect has been generally ascribed to the functioning of the cognitive control system, in which higher controls are recruited after conflicts to further constrain the processing of the task-irrelevant dimension. However, due to its uniqueness in neural and cognitive processes, whether the perception of emotional valence shows the same conflict adaptation effect in the conflicting context is an open question. This study adopted a variant of the Stroop task in which the task-relevant dimension was the affective facial expression while the task-irrelevant dimension was the meaning of an emotional word superimposed on the face. Results suggest that the perception of emotional valence in the conflicting context can have aspects that are both similar and dissimilar to the control of conflict for non-emotional information. The capturing of attention by emotional information can alter the functioning of the cognitive control system.     

儿童Stroop效应加工阶段特点的事件相关电位研究

选用小学六年级儿童为被试者,采用事件相关电位研究了Stroop效应的时间进程。实验中要求被试者判断在计算机屏幕中央出现的汉字颜色,并尽可能又快又准地按下相应的键,同时使用32导电极帽记录被试者的脑电信号。结果除了发现反应时的Stroop效应;还发现在汉字出现后大约300 ms左右,颜色与语义不一致的汉字诱发的P300比其它两种条件的波幅要大、潜伏期要晚;在汉字出现大约400 ms之后引发了一个慢波,不一致条件和控制条件引起的慢波波幅在头皮顶部比一致条件大。结果说明儿童Stroop效应的产生是一个复杂过程,干扰效应是早期颜色识别和后期反应选择两个阶段共同作用的结果,而促进效应则发生在后期的反应选择阶段。     

识别关键冲突任务的定量分析

建立关键链（CC）计划,关键在于识别关键冲突任务.讨论了项目进度计划中与冲突资源相关的关键任务、关键任务组以及关键任务群,提出了统一的定量表示方法.针对一种冲突资源,相应地提出了识别关键冲突任务的一个启发式算法,找出了相关冲突任务,并得出相应的优先顺序,建立了有效的CC计划.通过实例进行了实证研究,为CC技术的应用提供了定量支撑.     

开放域对话系统的抗噪回复生成模型

为缓解输入语句中噪声对回复生成模型的干扰, 提出一个基于编码–解码框架的抗噪模型。首先,在训练集输入序列中随机加入模拟噪声字符; 然后,在编码端输出层训练噪声字符识别, 提升模型对噪声特征的提取能力; 最后, 在编码端输出层融合预训练语言模型, 扩大模型对噪声的覆盖面。为验证该模型的抗噪效果, 构建首个带真实噪声的单轮开放域闲聊系统抗噪能力测试集。在该测试集上的实验结果表明, 所提出的抗噪模型自动评测和人工评测结果均优于基准模型。     

Correspondence patterns: mechanisms and models of human dynamics

A stochastic queue model of human behaviour developed on the basis of the distribution of timing of tasks, as studied in a sample of e-mail messages at a university and in the written correspondence of Albert Einstein, may not be as simple as it seems. Although this model reproduces the apparently non-Poisson distribution of correspondence delays, its interpretation is more complicated than suggested by Barabási, who claims that humans execute their tasks based on some perceived priority, setting up queues that generate very uneven waiting-time distributions for different tasks. Such an explanation is intuitively appealing in the context of the queue metaphor, but does not exclude other mechanisms. By attributing delays in the correspondence to task priorities, this explanation ignores two important classes of mechanism that also contribute to the apparent distributions of task timings: the semantic content of an individual's correspondence and the social context in which this correspondence occurs.