当代西方语篇阅读信息加工过程理论的分歧与整合

当代西方语篇阅读研究过程中产生了多种理论与模式,它们试图对读者阅读过程的信息加工过程做出解释。文章从发展的角度介绍了建构主义理论、最低限度假设理论和记忆基础的文本加工观的历程、基本假设和实证研究,从中可以看出各种理论派别的分歧与整合的趋势。     

语篇阅读中高级目标推理的机制

建构主义理论和最小假设模型是有关推理研究的两个比较系统的理论。建构主义理论认为在自然阅读过程中,读者会即时产生高级目标推理,而最小假设模型认为在自然阅读过程中,高级目标推理不会即时产生。该研究采用词汇判断和命名的方法,250ms和500ms两种延时程序,以较连贯的自然语篇为实验材料,对语篇阅读中的高级目标推理机制进行了研究。实验结果发现,被试在阅读过程中即时产生了高级目标推理,这与建构主义理论的预测相一致。     

Semantic associations and elaborative inference

In this article, a theoretical framework is proposed for the inference processes that occur during reading. According to the framework, inferences can vary in the degree to which they are encoded. This notion is supported by three experiments in this article that show that degree of encoding can depend on the amount of semantic-associative information available to support the inference processes. In the experiments, test words that express possible inferences from texts are presented for recognition. When testing is delayed, with other texts and test items intervening between a text and its test word, performance depends on the amount of semantic-associative information in the text. If the inferences represented by the test words are not supported by semantic associates in the text, they appear to be only minimally encoded (replicating McKoon & Ratcliff, 1986), but if they are supported by semantic associates, they are strongly encoded. With immediate testing, only 250 ms after the text, performance is shown to depend on semantic-associative information, not on textual information. This suggests that it is the fast availability of semantic information that allows it to support inference processes.     

文本阅读双加工理论与实验证据

文本阅读信息加工过程研究一直是国内外心理学界高度重视和关注的领域, 形成了建构主义理论、最低限度假设理论和记忆基础文本加工理论等派别百家争鸣的局面。这些理论争议的焦点在于自然阅读是主动的、积极的、目标策略驱动的过程, 还是被动的、消极的、自动的过程。在全面总结国内外心理学界有关文本阅读的研究成果的基础上, 文本阅读双加工理论提出并对文本阅读的主要争议进行了整合。该理论的核心观点是文本的自然阅读过程是连贯阅读与焦点阅读的双加工过程。文本阅读中读者所阅读的材料特点不同, 引发的阅读信息加工活动也不同, 而不同性质的阅读过程, 又会引发不同的推理整合, 从而会建构不同类型的文本表征。文本阅读双加工理论已经形成了比较完整的理论框架, 并获得了很多实验证据的支持。当然, 文本阅读双加工理论作为一个新的理论, 其中的有些观点还需要进一步检验。在未来的文本阅读研究领域, 有很多问题还需要研究者进一步关注。     

文本阅读中的信息整合研究

文本阅读中的信息整合研究是当前文本阅读研究的热点问题,在这一问题上,虽然当前研究已经达成了一些基本共识,但仍然存在着建构主义理论、最低限度假设和记忆基础文本加工理论这三种理论争议。本文详细介绍了这三种理论争议的主要观点和实验证据,特别是重点分析了三种理论争议的最新发展,在此基础上对未来的研究趋势进行了展望。     

On some sources of metacomprehension

Metacomprehension is defined as readers' evaluations of their performance on a task, involving inferences derived from a text. In two studies metacomprehension was related to characteristics of the inferences required, in terms of 1) amount of information to be kept in long-term memory (Study 1), 2) amount of information to be kept in short-term memory (Study 2) and 3) existence of negatives (Study 2). In the first study, 41 psychology students read a text and afterwards judged 1) the correctness of a set of pragmatic inference statements, and 2) their confidence in being correct or the difficulty of judging each statement. In the second study, 81 high-school students read a text and simultaneously judged 1) the correctness of a set of logical inference statements and 2) their confidence in being correct and the difficulty of judging each statement. In both studies, metacomprehension was not significantly correlated with actual performance. The results indicate that one important source of metacomprehension consists in information processing load. In Study 1, long-term memory requirements represent this load, in Study 2, the existence of negatives.     

On Certain Limitations Accompanying Readers' Interpretations of Corrections in Episodic Text

Where information concerning an ongoing event is acquired piecemeal over time, it may become necessary to correct a particular item of information after it is first encountered. Previous work (Johnson & Seifert, 1994; Wilkes & Leatherbarrow, 1988) has shown that if the content targeted for correction is thematically salient, neither a delayed nor an immediate correction may be sufficient to suppress its influence when subjects are subsequently tested for their comprehension of the message sequence as a whole. The present experiments are aimed at clarifying the conditions under which corrections fail to achieve their intended effects. In Experiment 1, message sequences were prepared in which the correction occurred midway in the text. It was found that the pre- or post-correction status of a message had no bearing on how it was interpreted on a later test of comprehension; both types of statements were equally likely to be associated with inferences based on the discredited message. In Experiment 2 the message targeted for correction could be directly linked to only one other message in the overall sequence. Despite this reduction in the scope of the correction, illegitimate inferences based on the discredited source continued to be made during comprehension testing. These results suggest that the critical inferences were made at the time of comprehension testing and are interpreted as providing support for the minimalist stance on the incidence of on-line inferencing during normal reading.     

文本阅读中情境模型建构和更新的机制

采用移动窗口阅读技术探讨了文本阅读中情境模型建构和更新的机制。实验材料按照特征描述与目标行为 的关系总体上分为可能、不可能和重新可能三个版本。实验1主要探讨了文章开始部分有明显的焦点信息时情境模 型建构和更新的机制,检验了更新追随假设的合理性;实验2主要探讨了文章开始部分没有明显的焦点信息时情境 模型建构和更新的机制,检验了记忆基础文本加工理论的合理性;实验3使用再认探测任务,进一步检验了更新追 随假设和记忆基础文本加工理论的差异。实验结果表明,文本阅读中情境模型的建构和更新是一个双加工过程,在 自然阅读过程中,读者根据所阅读的信息不同会产生不同的信息加工活动,既会有记忆基础文本加工理论所支持的 协调性整合,也会有建构主义的更新追随假设所支持的焦点整合     

Logical processing of set inclusion relations in meaningful text

The present experiments attempted to resolve some recent conflicting findings in cognitive structure research between processing linear ordering and set inclusion relations described in meaningful paragraphs of text. In a self-paced study-test paradigm, college students studied such paragraphs. Definite processing differences were found for the two set-theoretic relations. Set inclusion test results were found to be due to faulty logical processing and not to memory retrieval problems. Subjects made invalid conversions of universally quantified statements and failed to make valid transitive inferences between such statements. This failure was found to be an increasing function of the distance separating the two terms in the set inclusion. No such problems were observed for linear orderings. The usual distance function was found: Accuracy on test questions about the ordering was an increasing function of the distance between the terms in the question. Results similar to those for linear orderings were observed for set inclusions when subjects were given special instructions about the validity and invalidity of symmetric and transitive inferences.     

Testing psychological theories about inference making

Two contrasting views about elaborative inferences in text comprehension are that they occur as the text is read and that they are deferred until they are made necessary by a question about the inferable information. This paper develops these views in greater detail and examines some of their consequences. The model of memory performance underlying the first of these theories has two quite different interpretations: the standard one and one in which inferable information, whether it is implicit or explicit in a text, is not encoded into memory for content. Two experiments are reported. The results are not consistent with the most straightforward interpretation of the claim that inferences are always deferred. However, that theory can be modified so that it makes the same predictions as its rivals. The various interpretations of the formal model underlying all three accounts of inference making can be distinguished only in terms of parsimony, explanatory adequacy, and their ability to handle results from other experimental paradigms, such as sentence verification.     

What readers bring to the processing of fictional texts

Research on text processing has generally focused on the types of inferences that all readers draw in common. Our research examines aspects of processing that depend on the particular relation of the reader to the text. Students read fictional stories that contained weak and unsupported assertions and that were set either at their own school or at another school. We expected that they would be prompted to process the story information thoroughly enough to reject the assertions only if they were familiar with the story setting. Consistent with this expectation, the results showed that the away-school story, but not the home-school story, had a significant impact on students’ beliefs. These results support the view that readers must actively construct disbelief when processing fictional information.     

Cumulative Keyword Index for Volumes 93-95

Individuals with autism frequently show impairments in text reading comprehension. This often is attributed to poor ability to draw inferences during reading and to inadequate access to relevant knowledge. The current study tested this hypothesis by measuring the time taken to read the same question, relating to either physical or social world knowledge, when it was either relevant or irrelevant to the bridging inference evoked by a preceding two-sentence vignette. In the study, 16 normally developing adolescents and 16 adolescents with autism were matched on word reading accuracy, chronological age, and vocabulary but differed significantly in text comprehension. A strong priming effect was found, robust over participants and over items; participants read those questions that were relevant to the inference evoked by the vignette faster than they read those questions that were irrelevant, and no interaction with group membership or type of knowledge was found. This indicates that readers with autism, just like controls, were activating appropriate world knowledge primed by implicit inferences while reading the vignettes. Thus, the comprehension problems in these readers cannot be attributed to an inability to make implicit inferences or to draw on relevant world knowledge. Instead, we suggest that these problems must be sought at a higher level of text processing.     

Accessing distant premise information: how memory feeds reasoning

According to current psychological models of deduction, people can draw inferences on the basis of information that they receive from different sources at different times. In 3 reading-comprehension experiments, the authors demonstrated that premises that appear far apart in a text (distant) are not accessed and are therefore not used as a basis for logical inferences (Experiment 1), unless the premises are reinstated by a contextual cue (Experiment 2). In Experiment 3, the authors investigated whether these deductions are then integrated into the reader's situation model of the text. The results are interpreted in terms of a collaboration between memory-based text processing and higher level schema-driven logical reasoning.     

An investigation of the ability to process inferences in language following right hemisphere brain damage

Twenty-two right hemisphere brain-damaged and 22 non-brain-damaged patients were given a multiple-choice recognition task which contained true statements, statements which were inferentially true but not actually heard before, and false statements. It was hypothesized that if right hemisphere brain damage disturbs the ability to comprehend inferences, these subjects, unlike their normal counterparts, would not falsely recognize true inferences as heard before. This hypothesis was not confirmed. However, the right hemisphere group was poorer than controls at rejecting false statements. This behavior was speculated to be a retrieval difficulty, which was exacerbated if the information contained spatial or semantically similar material.     

Improving argumentative writing skills: Effect of two types of aids

Young children have difficulties writing argumentative texts which contain well-linked arguments and counterarguments even though they are capable of arguing by oral. Two main explanations have been provided to account for those difficulties: a) The writer has to manage alone two different points of view, whereas each of the two (or more) speakers can take charge of one of the points of view. b) The inability of young children to attribute an argumentative valence to statements.In order to improve the ability of 8-year-old writers (skilled or less skilled) to manage the dialogical dimension of the argumentative text, two types of aids were tested. 1) A collaborative writing in which the children worked in twos to recompose an experimental argumentation. 2) A classification task was presented before the text recomposition task. The results show that 8-year-old children are capable of processing the argumentative valence of statements. However, only skilled writers take advantage of the two types of aids to compose a text comparable to the standard argumentative schema.     

Effects of context-induced processing operations on semantic information acquired from discourse

Effects of context on processes involved in understanding discourse were investigated. An experimental context designed to directly affect discourse processing by inducing subjects to generate inferences involving text content was compared to a context in which subjects simply listened to and recalled the content of a text. The extent of inferential discourse processing was estimated from derived semantic information in subjects' text recalls. The experimental context produced effects on inferred and other derived information in subjects' text recalls which were predicted from a constructive model of discourse processing. Also consistent with this model, the amount of derived information increased with repeated exposures to the text. In a third condition, subjects repeatedly listened to the text in the experimental context, but only recalled it once on the last trial. The context produced very substantial effects on proportions of derived information under this condition. Context effects were still present one week later.     

(Implicitly) judging a book by its cover: the power of pride and shame expressions in shaping judgments of social status

How do we decide who merits social status? According to functionalist theories of emotion, the nonverbal expressions of pride and shame play a key role, functioning as automatically perceived status signals. In this view, observers automatically make status inferences about expressers on the basis of these expressions, even when contradictory contextual information about the expressers' status is available. In four studies, the authors tested whether implicit and explicit status perceptions are influenced by pride and shame expressions even when these expressions' status-related messages are contradicted by contextual information. Results indicate that emotion expressions powerfully influence implicit and explicit status inferences, at times neutralizing or even overriding situational knowledge. These findings demonstrate the irrepressible communicative power of emotion displays and indicate that status judgments can be informed as much (and often more) by automatic responses to nonverbal expressions of emotion as by rational, contextually bound knowledge.     

Retrieving text inferences: Controlled and automatic influences

Bridging inferences contribute to text coherence by identifying the connections among ideas, whereas elaborative inferences simply specify sensible extrapolations from text. Bridging inferences have been indistinguishable from explicit text ideas on numerous measures, suggesting similar longterm memory (LTM) representations for the two, whereas elaborative inferences are inferior. To evaluate the LTM representations of text ideas, we used the extended process-dissociation procedure (Buchner, Erdfelder, & Vaterrodt-Plunnecke, 1995; Jacoby, 1991) to partition the controlled, recollective contributions to text retrieval from the automatic, familiarity-based contributions. The automatic contribution to the recognition of implied concepts was consistently negligible, an outcome consistent with the absence of perceptual processing of those concepts during the original reading. In addition, the controlled basis of recognition was consistently higher for explicit than for implicit concepts, which suggests a more robust conceptual representation for explicit text ideas (Yonelinas, 2002). These results were interpreted to reflect the asymmetric representation of explicit ideas and inferences (elaborative and even bridging inferences) in the surface, propositional textbase, and situational levels of text representation.