Effects of signaling topic structure on text recall.

Two experiments studied effects of signaling devices (headings, overviews, and summaries) on text memory. In Exp 1, Ss read a text with or without signals, then recalled the topics of the text. Signaling produced better memory for the topics and their organization. In Exp 2, Ss recalled the content of the text they read, and recalls were scored for the number of accurately recalled ideas. Signals produced recalls that were better organized by text topics. Signals also influenced the distribution of recall of ideas: Ss remembered more topics but recalled less about each accessed topic if the text they read contained signals than if it did not. The results are interpreted as supporting a model in which signals influence readers' representations of a text's topic structure, which, in turn, is used to guide the recall of text content. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Effects of organizational signals on text-processing strategies.

Two hypotheses about how organizational signals influence text recal1 were tested: (a) that signals cause readers to change their text-processing strategies and (b) that signals facilitate readers' attempts to encode topic structure information but do not cause a shift in strategies. College students read and recalled a text that contained either no signals or contained headings, overviews, or summaries emphasizing the text's topic structure. At recall, students either received no cues or were reminded of the text's topics. Providing cues facilitated recall much more in the 3 conditions involving signaling than in the no-signals condition. The results are consistent with the hypothesis that organizational signals induce readers to change their text-processing strategies. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Context and the recallability of concrete and abstract sentences.

Examined the influence of contextual information on the recall of abstract and concrete sentences in 3 experiments, using 216 undergraduates. In Exp I, concrete and abstract target sentences were presented in either a coherent paragraph context or a random paragraph context. In the random context, Ss recalled more concrete target sentences than abstract ones, but there was no difference between the 2 groups when the sentences were presented in a coherent context. Exp II extended this finding by adding a moderately coherent context that used many of the same nouns as the coherent paragraph, but it was not as thematically coherent. Exp II replicated the results of Exp I and found that the moderately coherent context provided intermediate facilitation for the recall of abstract sentences relative to the random context and the coherent context; context structure had no effect on the recall of concrete sentences. In Exp III, the target sentences were abstract and the concreteness of the context was varied. Abstract context sentences were recalled as well as concrete context sentences if the contexts formed a coherent paragraph; if the context was a randomly ordered list of sentences, concrete context sentences were recalled better than abstract context sentences. Results were interpreted in terms of the differential availability of contextual information for abstract and concrete materials and support the context availability model. (27 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Differences in the use of selective attention by more successful and less successful tenth-grade readers.

Both more successful and less successful readers appear to use the selective attention strategy (SAS) to learn important text information; however, more successful readers tend to learn and recall considerably more important, as well as unimportant, information. The 2 studies reported here investigated the reason(s) for the more successful readers' learning and recall advantage. In Exp 1, 10th graders were asked to read, learn, and recall information from a text on marine biology. Questions were inserted every 4 pages to manipulate text item importance. The results showed that more successful readers learned and recalled more important information than less successful readers because they were more metacognitively aware of how and when to use the SAS. In Exp 2, perceptual and conceptual attention were measured for both more and less successful readers. More successful readers used significantly more conceptual attention while reading than did less successful readers. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Effects of organizational signals on free recall of expository text.

How is text memory influenced by organizational signaling devices? In Experiments 1 and 2, college students read a text with or without headings, overviews, and summaries; then they did a free recall. When the text's topic structure was simple (Experiment 1), signaling had no effect. When the structure was complex (Experiment 2), signaling affected the distribution of recall of text content and recall organization. Experiment 3 compared recall for texts in which all, half, or none of the topics were signaled. Signaled content was recalled equally well for the half-signaled and fully signaled texts. However, unsignaled content was recalled more poorly for the half-signaled text than for the unsignaled text, suggesting that signals direct processing of text content. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Selective recall for personally relevant communications.

Used 2 experiments to test whether people display a recall bias for supportive over refutive information. While previous work has used broad social attitudes, the present research investigated recall for communications relevant to Ss' behavior. In Exp I, 15 smokers, 36 nonsmokers, and 24 ex-smokers read and recalled information related to smoking. In Exp II, 28 seatbelt wearers, 18 occasional wearers, and 15 nonwearers read and recalled pro-seatbelt material. As predicted, in both experiments Ss' behavior significantly affected their recall of information: Nonsmokers recalled more anti-smoking material than smokers, and seatbelt wearers recalled more pro-seatbelt information than nonwearers. There were no significant effects on any measures when Ss in Exp I were divided into seatbelt wearers and nonwearers and when Ss in Exp II were divided into smokers and nonsmokers. Thus, Ss were selective only on the communication relative to their own behavior. (French abstract) (10 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Learner-generated organizational aids: Effects on learning from text.

Compared the effects on learning from text of 4 contexts (topic sentences, headings, related sentences, and unrelated sentences) when they were generated by readers or provided in text. 133 female undergraduates read a 522-word passage based on 15 related topics within a conceptual hierarchy on minerals, and performance under each of the treatments was examined on a free recall test, matching test, and test for knowledge of passage structure. The contexts had differential effects on knowledge of passage structure only when they were generated; readers who generated topic sentences recalled more than all other learners. The provided contexts had no effect on knowledge of passage structure but increased recall of subordinate information over generated contexts. (28 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Effects of number signals on reading and recall.

Examined the influence of number signals (i.e., numbers or number words preceding important textual information) on text recall. 120 undergraduates read and recalled 2 texts containing 10 target sentences each. Reading times were recorded for each target sentence. For half of the Ss, the target sentences were preceded by numbers indicating their organization; for the other half, the target sentences were not signaled. Half of the Ss completed a free-recall task, while the remainder completed a cued-recall task. Results indicate that Ss read target sentences more slowly if they were signaled than if they were unsignaled. Ss' recalls of target information followed the text organization more closely if the sentences were signaled. Signaling aided free recall of target sentences, but had no effect on cued recall. Results demonstrate that number signals directed attention to the sentences they marked, led to better encoding of the organization of target information, and influenced the process of recalling the target information. (33 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Integration of topic and subordinate information during reading.

Four experiments examined how readers integrate subordinate information with relevant context as they read. Ss read texts a sentence at a time with occasional interruptions lasting 30 sec. Following a distractor task, they resumed reading after being reminded of the topic sentence of the last paragraph they read (topic cue condition), being reminded of the last sentence they had read (local cue condition), or receiving no reminder of what they had been reading (no cue condition). Reading times on the 1st sentence following interruption were faster in the topic and local cue conditions than in the no cue condition (1) when the topic and local cues supplied missing referents for the target sentences, (2) when the target sentences were written to be understood as independent statements, and (3) whether the target sentences were embedded in short or long texts. Results are interpreted as demonstrating that readers integrate subordinate information with relevant topics, as well as with the immediate local context. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Misleading postevent information and recall of the original event: Further evidence against the memory impairment hypothesis.

Evaluated the hypothesis that misleading postevent information impairs memory for the original event using 174 undergraduates in Exp I and 228 Ss in Exp II. Ss were assigned either to a recall test condition or an original recognition test condition. Ss viewed a sequence of slides depicting an event, read a postevent narrative that presented neutral or misleading information about critical details, and were tested on their ability to recall the critical details. No difference in recall performance between misled and control conditions was found. These results, in conjunction with the finding of M. McCloskey and M. Zaragoza (see record 1986-03053-001) that misleading information did not affect Ss' ability to recognize original information, argue strongly against the memory impairment hypothesis. (15 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Echoic memory processes in good and poor readers.

Conducted 4 experiments to determine whether echoic memory plays a role in differences between good and poor readers. In Exp I, with 9 poor (mean age 11.05 yrs) and 9 good (mean age 10.9 yrs) readers, and Exp II, with 12 poor (mean age 10.85 yrs) and 12 good (mean age 10.7 yrs) readers, a suffix procedure was used in which the S was read a list of digits with either a tone control or the word go appended to the list. For lists that exceeded the length of the Ss' memory span by 1 digit (i.e., that avoided ceiling effects), poor readers showed a larger decrement in the suffix condition than did good readers. In Exp III, with 14 poor (mean age 10.64 yrs) and 14 good (mean age 10.83 yrs) readers, Ss shadowed words presented to 1 ear at a rate determined to give 75–85% shadowing accuracy. The item presented to the nonattended ear were words and an occasional digit. At various intervals after the presentation of the digit, a light signaled that the S was to cease shadowing and attempt to recall any digit that had occurred in the nonattended ear recently. Whereas good and poor readers recalled the digit equally if tested immediately after presentation, poor readers showed a faster decline in recall of the digit as retention interval increased. In Exp IV, using Ss from Exp II, bursts of white noise were separated by 9–400 msec of silence, and the S was to say whether there were 1 or 2 sounds presented. There were no differences in detectability functions for good and poor readers. (39 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Advance organizers that compensate for the organization of text.

In Exp I, 56 college students with no computer experience read a 24-frame text on computer programing that was presented in logical or random order. For random organization, Ss given an advance organizer performed better on a posttest than controls, but the opposite pattern obtained for logical organization. In Exp II, 96 students read a 4-paragraph text concerning imaginary countries that was presented in name or attribute organization. Low-ability (determined by questionnaire data on academic background and test scores) Ss given an organizer prior to reading performed better on questions that required integrating across different paragraphs of the presented text, and Ss given the organizer after reading performed relatively better on questions concerning information they had read within the same paragraph. Apparently, advance organizers served as an assimilative context for unfamiliar organizations. (18 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Distinctiveness of encoding and recall of text materials.

Four text-recall experiments, involving 243 undergraduates, examined the distinctiveness-of-encoding hypothesis. Exp I investigated the comparative recall of distinctively (DE) vs nondistinctively (NDE) encoded summary sentences and revealed a significantly higher rate of recall for DE summary sentences and associated text material. Exp II combined an activity that required Ss to interact with the semantic base of the text with the DE operations employed in Exp I. Data indicate that each approach resulted in a greater level of recall than a control condition and that the combination of the 2 produced the highest overall recall. Exp III examined possible mechanisms by which semantic content associated with the summary sentences was encoded and recalled. The most facilitative arrangement of activities involved the physical presence of text materials at the time at which the operations on summary sentences took place. Exp IV investigated the role of inspection time and rereading on DE processes. Although controlled inspection time reduced the amount recalled compared with uncontrolled conditions, DE summary sentences still led to a significantly higher rate of recall than NDE sentences. (43 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Reorganization and the recall of prose.

The hypothesis that reorganizing written material while taking notes would aid free recall was evaluated in 2 experiments with high school juniors. 82 Ss read a passage on anthropology organized either by society or by topic. They were then instructed to take notes either (a) in the same order as the material was presented or (b) by reorganizing the information according to specified categories. Free recall was higher for Ss who reorganized the material. In Exp II, with 146 Ss, the effect persisted when the categories were given to Ss in all conditions. Ss of average verbal aptitude benefited from reorganizing to a greater extent than did Ss of higher verbal aptitude. Results are discussed in terms of the encoding variability hypothesis. (16 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Reading in perspective: What real cops and pretend burglars look for in a story.

Tested 2 explanations of how readers' existing knowledge structures (their schemata) function during reading. According to the selective attention (SA) hypothesis, activated schemata identify certain text elements as important and direct increased processing toward those elements. According to the slot-filling hypothesis, schemata provide ready structures into which relevant information can be assimilated without additional processing. Two experiments were performed. 55 Ss participated in Exp I and 106 in Exp II. In both experiments, Ss chosen to represent 3 naturally occurring perspectives (police, real estate, and education students) were assigned to 1 of 3 perspectives (burglar, homebuyer, control). In Exp I, text elements relevant to readers' assigned perspective were rated more important. In Exp II, text was presented by computer, and reading times for individual sentences were recorded. Results confirm the powerful influence of assigned perspective on recall. Consistent with the SA hypothesis, readers spent more time on sentences containing information important to their assigned perspective. Naturally occurring perspectives had little effect in either experiment. (17 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Reading both high-coherence and low-coherence texts: Effects of text sequence and prior knowledge.

Determined if (1) the advantage for the low-coherence text is due to inferences made while reading, or alternatively, due to inferences generated during testing as a result of less information being available from the low-coherence text; (2) the inferences must rely on prior knowledge, or if inferences based on the text (or recently presented information) are sufficient; and (3) reading 2 different text versions is advantageous for readers. Ss were 80 university students who were assigned to 1 of 4 conditions representing if the Ss read the high-coherence text followed by either the high- or low-coherence text, or the low-coherence text followed by either the high-or the low-coherence text. Methodology involved reading the texts, answering questions about the text, and answering prior knowledge questions. The results indicate that high-knowledge readers benefited from the low-coherence only text when it was read first. Further, low-knowledge readers benefited from the high-coherence text, regardless of whether it was read first, second, or twice. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The next-in-line effect: Encoding or retrieval deficit?

The phenomenon that people cannot remember what happens when they are next-in-line to perform was investigated in 2 experiments with 144 undergraduates to determine whether this memory deficit reflects a failure to encode or an inability to retrieve preperformance events. In Exp I, 96 Ss participated in 4 memory trials; in each trial, half the Ss were called on to read words, and the other half merely listened. Before each trial, Ss were told whether they would be readers or listeners and the order in which numbers would be called. Thus, readers could anticipate precisely when they would be called on to perform. After hearing the 28 words in random order, all Ss wrote down the recalled words on cued or noncued sheets. Although the semantic cues strongly facilitated access to memories, they did not moderate the next-in-line effect. In Exp II, Ss were told—either before or after performing—to make a special effort to remember preperformance events. If instructed afterward, Ss displayed the usual memory deficit. If instructed beforehand, they reversed the deficit and showed a superior preperformance recall. It is concluded that the next-in-line effect is a failure at encoding, not at retrieval. (27 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Role of expectations in sentence integration.

Examined the role of causal expectations in sentence integration in 3 experiments with a total of 115 Ss from a university community. Ss read narratives that did or did not elicit strong expectations about upcoming events (high-expectation vs low-expectation texts). Results from Exp I indicate that Ss were faster to judge a target sentence to be unrelated when it followed a high-expectation text than when it followed a low-expectation text. Exp II results demonstrate that Ss were slower to read an unimportant, expectation-violating sentence embedded in a high-expectation text. Furthermore, Ss were faster both to read and to judge an expectation-fulfilling target sentence when it followed a high-expectation text. Three models of the generation and use of expectations are considered. The results support a model in which expectations are generated selectively. Once generated, expectations become the focus of upcoming sentence integration processes such that correct expectations help and incorrect expectations interfere. Exp III results fail to demonstrate that expectations took the form of highly specific predictions. (35 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Heuristic versus systematic information processing and the use of source versus message cues in persuasion.

In Exp I, 183 undergraduates read a persuasive message from a likable or unlikable communicator who presented 6 or 2 arguments on 1 of 2 topics. High involvement (HI) Ss anticipated discussing the message topic at a future experimental session, whereas low-involvement (LI) Ss anticipated discussing a different topic. For HI Ss, opinion change was significantly greater given 6 arguments but was unaffected by communicator likability. For LI Ss, opinion change was significantly greater given a likable communicator but was unaffected by the argument's manipulation. In Exp II with 80 similar Ss, HI Ss showed slightly greater opinion change when exposed to 5 arguments from an unlikable (vs 1 argument from a likable) communicator, whereas LI Ss exhibited significantly greater persuasion in response to 1 argument from a likable (vs 5 arguments from an unlikable) communicator. Findings support the idea that HI leads message recipients to employ a systematic information processing strategy in which message-based cognitions mediate persuasion, whereas LI leads recipients to use a heuristic processing strategy in which simple decision rules mediate persuasion. Support was also obtained for the hypothesis that content- vs source-mediated opinion change would result in greater persistence. (37 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)