The author examined memory for text in terms of the independent influences of semantic knowledge associations and text organization. Semantic associations were operationalized as the semantic relatedness between individual text concepts and the text as a whole and assessed with latent semantic analysis. The author assessed text organization by simulating comprehension with the construction integration model. Text organization consistently accounted for unique variance in recall. Semantic associations strongly predicted expository recall and predicted narrative recall significantly but to a lesser extent, even when the familiarity of the narrative content was manipulated. Results suggest that prior semantic associations and novel associations in the text structure influence memory independently, and that these influences can be affected by text genre.

This research examines adolescents' learning about a historical issue from multiple information sources. Adolescents read 2 contradictory texts explaining the Fall of Rome and thought out loud after each sentence. After reading, a series of questions probed their understanding and ability to reason with the information. Think-aloud protocols were coded for the type of processing they reflected, as well as the content that was utilized (e.g., prior knowledge, other text information). Paraphrases and elaborations were the most common types of processing activity. Elaborations involved connections to prior knowledge as well as other text information (both within and across texts) and often-represented self-explanations. The complexity of reasoning about the historical event was predicted by think-aloud comments that increased the coherence of the texts: self-explanations that used prior knowledge or previously processed text information as well as surface text connections. Results are discussed in terms of theories of text processing from single and multiple texts, adolescents' competencies when processing them, and implications of the research for providing students with opportunities to learn to think in discipline-based ways.

Processing time and memory for sentences were examined as a function of the degree of semantic and causal relatedness between sentences in short narratives. In Experiments 1-2B, semantic and causal relatedness between sentence pairs was independently manipulated. Causal relatedness was assessed through pretesting and semantic relatedness was assessed with Latent Semantic Analysis. Causal relatedness influenced processing time and memory. Semantic relatedness influenced memory, and influenced processing time when causality was not manipulated within an experiment and the situation described by the sentence pairs was somewhat difficult to construct. Experiment 3 utilized naturalistic texts. Semantic and causal relatedness between sentences influenced online judgments of fit and free recall. Results are discussed in terms of bottom-up and top-down theories of text processing.

This study examines the hypothesis that the ability of a reader to learn from text depends on the match between the background knowledge of the reader and the difficulty of the text information. Latent Semantic Analysis (LSA), a statistical technique that represents the content of a document as a vector in high-dimensional semantic space based on a large text corpus, is used to predict how much readers will learn from texts based on the estimated conceptual match between their topic knowledge and the text information. Participants completed tests to assess their knowledge of the human heart and circulatory system, then read one of four texts that ranged in difficulty from elementary to medical school level, then completed the tests again. Results show a nonmonotonic relation in which learning was greatest for texts that were neither too easy nor too difficult. LSA proved as effective at predicting learning from these texts as traditional knowledge assessment measures. For these texts, optimal assignment of text on the basis of either prereading measure would have increased the amount learned significantly.

In another article (Wolfe et al., 1998/this issue) we showed how Latent Semantic Analysis (LSA) can be used to assess student knowledge—how essays can be graded by LSA and how LSA can match students with appropriate instructional texts. We did this by comparing an essay written by a student with one or more target instructional texts in terms of the cosine between the vector representation of the student's essay and the instructional text in question. This simple method was effective for the purpose, but questions remain about how LSA achieves its results and how the results might be improved. Here, we address four such questions: (a) What role does the use of technical vocabulary play? (b) how long should the student essays be? (c) is the cosine the optimal measure of semantic relatedness? and (d) how does one deal with the directionality of knowledge in the high‐dimensional space?