Charles R. Fletcher

Causal reasoning in the comprehension of simple narrative texts

Abstract This research represents an attempt to unify two separate approaches to the study of text comprehension and recall. The first of these approaches, exemplified by the work of Trabasso and his colleagues ( Trabasso & Sperry, 1985 , Journal of Memory and Language , 24 , 595–611; Trabasso & van den Broek, 1985 , Journal of Memory and Language , 24 , 612–630) views comprehension as a problem-solving task in which the reader must discover a series of causal links that connect a texts opening to its final outcome. The second approach, typified by Kintsch and van Dijk (1978 , Psychological Review , 85 , 363–394; van Dijk & Kintsch, 1983 , Strategies in discourse comprehension , Academic Press, New York) emphasizes the importance of short-term memory as a bottleneck in the comprehension process. We combine these two approaches by assuming that the most likely causal antecedent to the next sentence is always held in short-term memory. This allows a reader to discover the causal structure of a text within the constraints of a limited-capacity short-term memory. We show that three variables derived from this hypothesis (time in short-term memory, causal connections allowed, and referential connections allowed) account for 31% of the variance in the free recall of propositions from eight simple narrative texts.

Surface forms, textbases, and situation models: Recognition memory for three types of textual information

It has recently been argued that memory for a text or discourse consists of three separate representations: a surface representation, a propositional textbase, and a situation model. This tri‐partite division is supported by a recognition memory procedure that demonstrates that (a) subjects can reliably differentiate between sentences actually read and meaning preserving paraphrases of those sentences, (b) performance improves when the distractors also differ at the propositional level but are consistent with the situation described by the text, (c) recognition is best when the distractors are inconsistent with all three levels of representation. These data are subject to alternative interpretations because all changes within a level of representation may not be equal and inferences based on one level of representation may be responsible for differences attributed to another. A series of control experiments eliminates these alternatives.

An on-line assessment of causal reasoning during comprehension

Fletcher and Bloom (1988) have argued that as readers read narratives, clause by clause, they repeatedly focus their attention on the last preceding clause that contains antecedents but no consequences in the text. This strategy allows them to discover a causal path linking the texts opening to its final outcome while minimizing the number of times long-term memory must be searchedfor missing antecedents or consequences. In order to test this hypothesis, we examined the reading times of 25 subjects for each clause ofeight simple narrative texts. The results show that: (1) causal links between clauses that co-occur in short-term memory (as predicted by the strategy) increase the time required to read the second clause (2)-poten-tiaicausal links between clauses that never co-occur in short-term memory (again as predicted by the strategy) have no effect on reading time; and (3) reinstatement searches are initiated-at-the-end-of-sentences thatare causally unrelated to the contents of short-term memory or that contain clauses that satisf~, goals no longer in short-term memory. These results support the claim that subjects engage in a form of causal reasoning when they read simple narrative texts.

Investigations of Inferential Processes in Reading: A Theoretical and Methodological Integration.

The study of inferential processes constitutes a major focus of research on text comprehension. Recent research has resulted in interesting and important findings, as exemplified by the contributions in this issue of Discourse Processes. Studies differ substantially, however, in the types of inferences that they consider and in the methodologies that they use. As a result, it is difficult to interpret and integrate the findings. In this article, we present a theoretical and methodological framework for organizing empirical investigations of inferential processes in reading. The theoretical component provides an account of the various types of inferences that have been investigated. The methodological component, based on Jenkins’ (1979) tetrahedral model of psychological experimentation, captures the impact of methodological variations across studies on the meaning of the results. The framework is applied in a discussion of central issues in current research on inference generation in reading.

Markedness and topic continuity in discourse processing

Two experiments are presented in support of the hypothesis that various syntactic constructions hierarchize along a continuum which codes the degree of topic continuity in a discourse. In the first experiment subjects were asked to rewrite two short sentences as a single long sentence. They were allowed to rephrase the second sentence to make their new, combined sentence sound natural. The form of the referent in the second sentence was found to depend on its continuity with the topic of the other sentence. Subjects tended to use unmarked linguistic forms in cases of high topic continuity and marked forms when there was a topic shift. In the second experiment subjects were asked to interpret discourse fragments which contained ambiguous referents. The less marked the form of the referent was, the more likely subjects were to interpret it as coreferential with the preceding topic.

Short-term memory processes in text comprehension

Because of the limited capacity of the short-term, working memory system texts must be processed one sentences or major clause at a time in cycles. This cyclical processing strategy eliminates the limited capacity problem but creates a new one; namely: How does a reader maintain the coherence of a text which must be processed in fragments? Kintsch and van Dijk (1978) have suggested that part of short-term memory is set aside as a buffer in which propositions from earlier processing cycles are held over the guide the integration of new propositions into the long-term representation of the text as a whole. Two experiments are presented which provide empirical support for this claim. Using the kintsch and van Dijk (Psychological Review, 1978 , 85, 363–394) model to predict which propositions would be selected for inclusion in the short-term buffer, it was demonstrated that selected propositions are more available to subjects than unselected propositions from the same part of the text, and that the latter are no more availabe than unselected propositions from earlier in the text.

Causality and the Allocation of Attention During Comprehension

Recent research has suggested that each statement in a narrative text is understood by relating it to its causal antecedents and consequences and that the text as a whole is understood by finding a causal path linking its opening to its final outcome. Fletcher and Bloom (1988) have proposed that in order to accomplish this goal, while minimizing the number of times long-term memory has to be searched, readers focus their attention on the last clause of a narrative that has causal antecedents but no consequences in the preceding text. As a result, a statement that is followed by a causal antecedent should remain the focus of attention, while the same statement followed by a consequence should not. This prediction was tested and confirmed in three experiments which show that when a target statement is followed by a sentence that includes only causal antecedents, (a) continuation sentences related to it are read more quickly, (b) target words drawn from it are easier to recognize, and (c) subject-generated continuations are more likely to be causally related to it.

The role of causal discourse structure in narrative writing

All writers produce text content and ideally connect it together according to discourse conventions. We investigate whether a particularly strong discourse convention, the need for causal coherence in narratives, can predict the kind of text writers will produce. Causality has been found to be a significant discourse factor in reading comprehension and hence can be expected to determine also what writers produce during composition. In Experiment 1, writers composed short continuations at various points throughout a simple narrative, whereas in Experiment 2, writers composed continuations to complete several narratives. The results indicate that causality indeed plays a major role in composition. Writers tend to produce new text in such a way that it is causally connected to the prior text. Furthermore, writers favored causal relations of necessity or of necessity and sufficiency while largely avoiding relations of sufficiency alone, which suggests a general discourse constraint to be maximally informative (e.g., Grice, 1975).

Assessing recognition memory for surface forms in discourse : a methodological note

Two different experimental procedures have been advocated for testing recognition memory for surface forms in discourse. One involves using a reversed-text control group that reverse the roles of recognition targets and distractors. Experiments using this procedure had led some researchers to conclude that surface memory is fairly a robust phenomenon. The alternative procedure, which makes use of no-text (or guessing) control group, had produced inconsistent results, leading other researchers to question the status of surface memory. The goal of this inquiry are to (a) explore the strengths and weakness of each approach, (b) assess alternative explanations for the inconsistency between them, and (c) evaluate the available evidence for better-than-chance recognition of surface forms in discourse

Using Story-Based Causal Diagrams to Analyze Disagreements about Complex Events.

People often disagree about the causes of complex events. Previous research has shown that disagreement often results from cognitive and pragmatic constraints that govern peoples tendency to attribute outcomes to partial rather than complete causes. This article applies the Trabasso, van den Broek, and Suh (1989) story‐based procedures to a narrative of the 1987 stock market crash. The content and structure of the resulting causal diagrams are then used to analyze some opposing causal explanations that were offered by different investigative commissions. The diagrams reveal that some major disagreements involved conflicting interpretations of the crashs multiple indirect and psychological causes. The diagrams also show that some causal explanations were incomplete and implausible because they ignored or contradicted relevant events and causal relations. Implications for discourse comprehension research are discussed, and the advantages of using causal diagrams to mitigate constraints on causal comprehen...