认知与发展心理学研究室知识库

在对文本进行阅读的过程中，读者的眼动行为受到视觉感知、词汇识别、句法加工、语义整合和语用加工等各种内在的认知过程的影响。同时，这些认知过程会受到工作记忆等认知能力个体差异的制约，也受到动机的影响。在本论文中，我们研究阅读能力、工作记忆和自我增强的动机因素如何影响阅读时的眼动行为。该研究一共包含如下三个方面的研究内容。
在研究一中， 利用支持向量机分析 61个被试阅读科技说明文时的眼动数据。采用的眼动指标是位于文章的不同部分的向前注视时间、首次阅读回视时间，第二次阅读时间和回视路径时间。根据支持向量机建立的模型，可以成功预测被试的阅读能力，预测正确率为 80.3%，敏感性为 84%，特异性为 77.8%。这一结果说明将眼动数据与机器学习的算法相结合可以很好的区分被试的阅读能力。
研究二由两个实验组成，探讨了工作记忆容量是如何影响眼动的。在第一个实验中，我们操纵的是关键区的词频，在第二个实验中，我们在关键区设置暂时性切分歧义的字段。对平均注视时间和眼跳长度的分析都发现了工作记忆的影响。工作记忆高的个体平均注视时间较短，眼跳长度较长。在第一个实验中，我们还分析了在关键区的注视时间和离开关键区的眼跳长度，发现在高频词的首次注视时间较短，离开高频词的眼跳长度较长，但是这些效应只存在于高工作记忆的被试。在第二个实验中，我们分析了在关键区的注视时间和离开关键区的眼跳长度，发现关键区右侧区域语义信息对于注视时间，眼跳长度的影响也只存在于高工作记忆的被试。这些结果说明中央凹、旁中央凹信息对于眼动的影响受到工作记忆的调控。同时，这些研究也再次证实了在中文阅读中基于加工的眼跳落点选择策略。
在第三个研究中，我们考察了自我增强对眼动的影响。被试阅读了描述自己或者他人积极或者消极特征的句子，并同时监控他们的眼动数据。相较于他人，在描述自己积极特征的词上首次注视时间和凝视时间都较长。这一结果说明除了文本特征及读者认知能力的个体差异外，眼动还受到自我增强的影响。这一结果不仅加深了对眼动控制的理解，同时也表明了利用眼动研究动机的可能性。
综合这三个研究，读者的眼动行为受到阅读能力、知觉广度的大小及自我增强的影响。因此，只有理解了这些因素如何影响眼动，才能够完整地理解阅读中的眼动控制机理。这也提示我们，未来的眼动控制模型也必须考虑这些因素。

Readers’ eye movements are influenced by many cognitive processes such as visual perception,  word processing, syntax processing, semantic processing, and pragmatics  processing. These cognitive processes are in turn influced  by working memory capacity, and  also by  motivation. This dissertation describe three studies investing how reading ability, working memory and  motivation such as  self-In the first study, the Support Vector Machines  (SVM) was used to analyze eye movements. data from 61 undergraduate students who read a multiple-paragraph, multiple-topic expository text. Forward fixation time, first-pass rereading time, second-pass fixation time, and regression path reading time on different regions of the text were provided as features. The SVM classification algorithm assisted in distinguishing high-literacy-skilled readers from low-literacy-skilled readers with 80.3% accuracy, 84% sensitivity, and 77.8% specificity. Results demonstrate the effectiveness of combining eye tracking and machine learning techniques to detect readers with low literacy skills, and suggest that such approaches can be potentially used in predicting other cognitive abilities.
The second study included two experiments,  which  investigated how eye movement behavior are mediated by working memory. In Experiment 1, the critical region was two-character word with different frequencies. In Experiment 2, the critical region was a 3-characters overlapping ambiguous string. For this kind of overlapping ambiguous string, the first two characters could constitute a word, and the last two charactes could constitute another word. Readers have to decide whether the middle character belongs to the first word or the last word during reading. Average eye fixation, average saccade length were all influenced by working memory. Fixations were shorter and saccades were longer for readers with a higher working memory capacity than readers with lower working memory capacity. What’s more, in the first experiment, we found that fixation durations were shorter in the high-frequency condition than in the low-frequency, and outgoing saccade length was longer in the high-frequency condition than  in the low-frequency condition. Interestingly, this effect only appeared in the people with high working memory capacity, but did not appear for readers with low working memory capacity. In Experiment 2, for the people with high working memory, fixation duration and saccade length all influenced by  the semantic information from the region to the right of the critical region. For the readers with low working memory capacity, we failed to find such effect. These results suggest that the effect foveal and parafoveal information influence the eye movement behavior were mediated by working memory. What’s more, our study provides additional evidence that support the processing-based strategy for saccade target selection in Chinese reading.
In the third study, we further showed that readers' eye movements are influenced by a social motivation of self-enhancement. We asked participants to silently read sentences that describe self or others with positive or negative traits while their eyes were monitored. First fixation duration and gaze duration were longer when positive words were used to describe self than to describe others. These results suggested that eye movements can be influenced by the motivation of self-enhancement beyond various stimuli features and  cognitive factors. This finding not only deepen the understanding the implication of eye movements but also highlight the potential of eye movements in studying (implicit) social cognition.
Taken together, these three studies showed that reading ability, working memory capacity, and the motivation of self-enhancement all affect eye movements. Therefore, all of these factors have to be taken into account when we understand the mechanism of eye movement control during reading, and all of eye movement models should also be able to explain the findings from these studies.