Tracking of spatial information in narratives.

Three experiments were conducted to investigate the process by which location information in narratives is represented in memory and the nature of the resulting memory representation. In Experiments 1 and 2, the results of a recognition task demonstrated that location shifts led to an immediate decrease in the accessibility in memory of protagonists' former locations. In Experiment 3, regardless of the amount of backgrounding after the last mention of the critical location ("the forest"), reference to an implied, location-typical entity ("the trees") was read equally fast as long as the protagonist remained in that location. In contrast to previous findings, we conclude that when location information is salient in a narrative it is included in readers' situation models, being updated immediately and remaining highly accessible even several sentences after it was last mentioned.

Maintaining global coherence in reading: the role of sentence boundaries.

In four experiments, we examined the reinstatement of backgrounded information in locally coherent passages, investigating the influence of syntactic boundaries, such as periods, on the time course of this process. In Experiment 1, using a line-by-line reading paradigm, readers were delayed in noticing a contradiction on a target line when the sentence continued onto a posttarget line. Consistent with this, in Experiments 2A and 2B, the results of a recognition probe demonstrated that relevant concepts from long-term memory were not integrated immediately unless the contradiction was followed by a period. Experiment 3 demonstrated that sentence boundaries are sufficient to facilitate the integration of related background information but are not necessary; additional time served the same purpose. Consistent with a memory-based text-processing framework, it appears that the reactivation of related information is initiated automatically but that an integration stage is influenced by such factors as syntactic boundaries.

Knowing that you don't know: metamemory and discourse processing.

In 6 experiments, the authors used a speeded question-answering task and a recognition task to examine how people know what they don't know. Extending work by S. Glucksberg and M. McCloskey (1981) to examine metamemory judgments about narratives, the authors asked participants to respond to 2 types of "don't know" questions. In certain conditions, readers were faster to respond "don't know" to implicit "don't know" questions (i.e., no information regarding the answers was provided) than to explicit "don't know" questions (i.e., narratives explicitly stated that something was unknown). The speed of responding to the implicit "don't know" questions was related to the familiarity of the question, which is consistent with claims that fast metacognitive judgments are based on a preliminary evaluation of the familiarity of a cue. This is a first step in integrating theories of metacognition and discourse processing.

Causal inferences in reading: from immediate activation to long-term memory.

Three different tasks were used to investigate the time course of drawing causal inferences. Participants read passages that contained a causal coherence break that could be resolved by reactivating a concept presented earlier in the passage. In Experiment 1, participants named a probe word that represented the earlier mentioned cause more quickly after encountering the causal coherence break, suggesting that the causal concept had quickly been reactivated. In Experiment 2, participants were slow to read a sentence after the causal coherence break that contradicted the intended inference, indicating that the inference had been encoded and retained in working memory. In Experiment 3, the results of a recall task indicated that the causal link was also included in the long-term memory representation of the text.

Connecting goals and actions during reading.

Three experiments showed that reading about a character's actions can reactivate a goal of the character stated earlier in the passage and backgrounded by intervening material. Subjects were slower to read a line describing an action that was inconsistent with a goal of the protagonist than they were to read about an action that was consistent with the goal, even though both lines were locally coherent. Goals were reactivated even when the intervening material did not describe attempts to achieve the goal (Experiment 2) and when the intervening material described another goal of the protagonist (Experiment 3). The results suggest that reading a sentence can reactivate relevant information from earlier in the text, even when the sentence is coherent with its immediate context and the reactivated information has been backgrounded by several lines of unrelated text.

Reinstatement of causal information during reading.

An on-line word naming probe was used to test whether information presented earlier in a text, and then backgrounded by several sentences, would be reinstated when Ss were required to understand the cause of a currently processed action or event. In Experiment 1, Ss named a probe word that represented an earlier-mentioned cause more quickly when it followed a causal coherence break than when it followed a neutral sentence. Experiments 2 and 3 replicated this effect and examined 2 conditions that may affect the process of reinstating a cause: (a) Inclusion of part of the context in which the cause was originally presented was not necessary to obtain reinstatement of the cause and (b) reinstatement of the cause was not evidenced when it had been disconfirmed earlier in the text.

Accessibility of information about goals during the processing of narrative texts.

We used a probe procedure to show that a goal established earlier in a text is active in memory at the point of its achievement. An initial experiment demonstrated that a goal category (began an investigation to nab the THIEF) is accessible, relative to a control condition, following the processing of a goal-achievement sentence (had the PURSER brought to his office). The remaining experiments provided evidence against several explanations of this result: (a) that the goal category's accessibility is due to an advantage in the strength of its initial encoding; (b) that the goal category is maintained in memory from the point at which the goal is established; or (c) that the goal category is reinstated at the point of goal achievement as the result of a high-level inference. The results suggest that the goal category is reinstated as the result of a low-level inference similar to the type that links an anaphor and its antecedent.