ABSTRACT. Across 4 studies, we investigated children's ability to deliberately and spontaneously construct spatial situation models. Under deliberate instruction, children were more accurate at recalling spatial details from narratives than from non-narrative descriptions. On two different tasks measuring children’s spontaneous spatial representations during narrative processing, we found that children were able to detect spatial inconsistencies in narrative and that the ability to spontaneously infer character movements through space was predictive of narrative comprehension abilities.

ABSTRACT. Instructional influences on productive use of academic vocabulary were investigated among 460 fifth graders. Participants received a 6-week unit on wolf management involving collaborative groups (CG) or direct instruction (DI). In an individual oral transfer problem, whether whaling should be allowed, both CG and DI students used significantly more general and domain-specific academic vocabulary from the Wolf Unit than uninstructed control students. CG students used more domain-specific vocabulary than DI students.

ABSTRACT. We examined reader profiles of good and poor comprehending developing readers using Think-Aloud. In both comprehension groups we found two subgroups of readers: Paraphrasers and Elaborators.Within each Subgroup, the quality of elaborative inferences differentiated between good and poor comprehenders. Good comprehenders produced more valid and fewer invalid elaborative inferences than poor comprehenders. These findings support the existence of subgroups of readers, and emphasize the importance of the correct use of background knowledge during reading.

ABSTRACT. Social cognition shows significant developmental change throughout adolescence, and is likely to influence the way readers of different ages process narratives. We examined the effects of social-cognitive processing demands and importance in narratives on their recall in children, adolescents, and adults. We found age-related increases in correct recall of social story-elements, independent from sensitivity to structural centrality. These findings support the relevance of social-cognitive development in understanding age-related change in narrative comprehension.

ABSTRACT. This research documents the impact of higher level talk and writing about text on students’ growth in reading. Participants included all second and third grade teachers, specialists, and students in twenty-three schools across three years. Students were assessed twice each year using multiple reading measures. Teachers were observed three times each year during reading instruction. Data were analyzed using hierarchical linear modeling to identify instructional practices that were positively related to students’ growth in reading.

ABSTRACT. This study investigated how contradictions between two webpages (from seemingly reliable sources) stimulate university students’ source evaluations. “About us” information on the webpages indicated that one of the two pages actually had commercial interests. Study results show that when having encountered contradicting information across pages, students spent more time reading the “about us” information and more frequently mentioned evaluative judgments about the two sources in a written summary than when having encountered consistent information.

ABSTRACT. This study examined the encoding and later recognition of “sources” (characters holding claims about an event) as a function of the consistency or discrepancy of their statements. Embedded characters that made no claims (non-sources) were also included in the experimental materials. Discrepancies influenced primarily source rather than non-source information (by increasing fixation times on source areas during reading and accuracy and recall confidence during recognition), whereas consistent statements tended to promote an untagged representation.

ABSTRACT. We investigated whether laypeople consider a source’s “commitment to science” when making of their minds about medical online information. Readers evaluated information from conflicting text documents. The scientific commitment of one source was manipulated through variation of their social role (Study 1) and their scientific integrity (Study 2). The results show that readers consider a source’s commitment to science a positive indicator of source credibility and claim validity.

ABSTRACT. The purpose of the present set of experiments was to investigate the extent to which source credibility influences the process of knowledge revision over the course of reading refutation texts addressing common misconceptions. Reading time and post-test findings suggest that readers may not naturally attend to or use source credibility information when reading refutation texts, unless being instructed to do otherwise.

ABSTRACT. Using eye-tracking during reading we compared the comprehension of jokes to that of similarly constructed control stories. Contrary to predictions of the incongruency theory, verbal jokes were read faster than non-funny stories that also involved a linguistic reinterpretation. Effects of instruction (funniness ratings vs. a meta-linguistic revision judgment), gender of the reader (female vs. male), and text type (funny vs. non-funny) suggest a complex interaction between cognitive and social-affective processes.

ABSTRACT. We examined the extent that individual differences in Machiavellianism moderated the effect of lying on event recollection. Subjects watched a video, then either lied or told a truthful recounting to a confederate. One week later, subjects recalled the actual events of the video. After lying, high Machiavellian subjects recalled fewer incorrect details, but more incorrect inferences, compared to low Machiavellians. Results suggest Machiavellianism moderates the influence of lying or telling the truth on later memory.

ABSTRACT. Answers to questions in conversations are often delayed. In job interviews, candidates respond to recruiters’ questions. In doing so, they trade off between (1) delaying responding to search for an appropriate response at the risk of appearing inept and (2) responding quicker but less appropriately. In a corpus of job interviews, response delays increased the probability of inappropriate responses and decreased hireability ratings, illustrating how response delays can entail social consequences beyond the conversation itself.

ABSTRACT. This study explores the extent to which a large set of listeners (n=239) share understanding with a performing duo, and with each other, of what happened in three improvisations on a jazz standard. Although listeners endorsed statements the performers had agreed upon much more than statements the performers had disagreed upon, almost no listeners' judgment patterns overlapped with performers' at a level greater than chance. Listeners with similar musical backgrounds could interpret improvisations radically differently.

ABSTRACT. Reading hardly ever occurs out of any context. Thus, recent models of reading (e.g., Goal-focusing model, McCrudden & Schraw, 2007; MD-TRACE model, Rouet & Britt, 2011) have emphasized on the necessity to consider the reading context, i.e. the activity or purpose for reading (e.g., read to write a summary, read for entertainment, or read to search for a specific information). The reading context determines readers' objectives for reading. To reach these objectives, the reader might have to use different reading strategies. For example, to answer a specific question (e.g., to find an author’s date of birth in his complete biography), it may not be useful to read the text entirely and linearly from top to bottom. Other questions may require the reader to read several portions of a text, or even several texts. Thus, the demands of the task modulate readers' access to text information as well as the depth of processing of that information. Within this framework, the present study aims at examining the strategies adult readers use when they have to work with a document in order to answer a specific question. In this specific situation, different properties of the document could promote more or less efficient search strategies and rapid access to information. This is for instance the role played by textual organizers such as headings, subheadings and paragraphs. Information search in documents has indeed proved to be easier when readers are able to make use of these organizers (Lorch et al., 2001; Rouet et al., 2001). The type of information the readers have to find might also influence their search strategies. For example, Rouet and Vidal-Abarca (2002) have shown that reading times were very different depending on the type of questions students had to answer. In addition to task demands and text features such as paragraphs and headers, readers’ characteristics could also explain inter-individual differences in searching efficiently information in written documents. Previous research conducted with children has shown that less skilled comprehenders were less able than skill ones to adapt their reading strategies as a function of the task they have to complete (Cain, 1999) or to search for a specific piece of information in a text (Cataldo & Oakhill, 2000). Efficient search strategies may also be conditional upon readers’ knowledge of the structural properties of texts, including the role of text organizers (Rouet & Coutelet, 2008). Finally, based on the observation that executive capabilities are involved in text comprehension (especially in high-level inferential comprehension tasks; Potocki et al., in press), one might hypothesize that executive skills would be particularly recruited in information search activities. The aims of this research is to investigate the role of these different textual, contextual and individual factors in adult readers in a situation in which they have to read a document in order to answer a specific question. More specifically, we expect the inclusion of headers to trigger top-down selective strategies, as opposed to linear scanning strategies. Header effects may be stronger for more demanding tasks such as questions requiring the location and comparison of several pieces of information. Finally, reading skills and executive control were expected to affect the flexibility of readers' strategies.

Method

Sixty undergraduate students from a large public French university participated in two sessions. In the first session (collective, with 4 to 8 participants per group) the participants completed a series of tasks assessing decoding; narrative and informative text comprehension; inhibition in working memory and metatextual knowledge. In the second session, the participants individually performed a series of information search tasks with the TOBII 1750 eye-tracker. The tasks involved six texts; each text comprised a main theme (e.g., African diseases) divided in three different subtopics (e.g., Malaria, HIV and Ebola). Each subtopic included three paragraphs. For each text, we constructed a “signaled” version (with headings and subheadings for each paragraph) and an “unsignaled” version (paragraphs only). For each text, the participants were asked to answer a practice question and four critical questions by using the information found in the text. The experimenter first read the question twice, while the participant looked at a fixation cross on the computer screen. Then the participant pressed the space bar in order to display the text, searched the text until they could articulate an answer. When they felt ready to answer, the participant pressed the space bar again which caused the text to be replaced with a blank screen, and gave their answer orally. The experimenter transcribed the answer on a sheet of paper. No feedback was provided. The four critical questions included two Locate questions, which required the participant to find a single piece of information in the text, and two Integrate questions, which required the participants to compare two or three pieces information located in different paragraphs.

Results and discussion

Analyses conducted on response time revealed a significant effect of Question (F(1, 59) = 318.7, p < .001) with slower response time for Integrate questions (mean = 37511) than Locate questions (mean = 15634.2); as well as a significant effect of document Organization (F(1, 59) = 11.6, p = .001) with signaled documents leading to faster response time (mean = 24849.7) than unsignaled documents (mean = 28295.6). The interaction between these two factors was also significant (F(1, 59) = 8.4, p = .005). The difference between signaled and unsignaled documents was only significant for the Integrate questions. The data on fixation time on areas of interest (target headings and subheadings for one specific question) are being processed at the time of submission. We also assumed the effects of signals to be mediated by the decoding, comprehension, working memory ability, and metatextual knowledge. The mediation would be due to the use of different search/scan strategies as evidence in the visual exploration patterns. We expected the results to support the view that goal-based reading is a complex and resource demanding process even for young adults. The data were also meant to provide a baseline for further research on text search with children and teenagers.

References

Cain, K. (1999). Ways of reading: How knowledge and use of strategies are related to reading comprehension. British Journal of Developmental Psychology, 17(2), 295-312. Cataldo, M.G., & Oakhill, J. (2000). Why are poor comprehenders inefficient searchers? An investigation into the effects of text representation and spatial memory on the ability to locate information in text. Journal of Educational Psychology, 92, 791-799. McCrudden, M.T., & Schraw, G. (2007). Relevance and goal-focusing in text processing. Educational Psychology Review, 19, 113-139. Lorch, R.F., Jr., Lorch, E.P., Ritchey, K., McGovern, L., & Coleman, D. (2001). Effects of headings on text summarization. Contemporary Educational Psychology, 26, 171-191. Potocki, A., Sanchez, M., Ecalle, J., & Magnan, A. (in press). Linguistic and cognitive profiles of 8- to 15-year-old children with specific reading comprehension difficulties: The role of executive functions. Journal of Learning Disabilities. Rouet, J.-F. & Britt, M.A. (2011). Relevance processes in multiple document comprehension. In M.T. McCrudden, J. P. Magliano, & G. Schraw (Eds.), Text Relevance and Learning from Text (pp. 19-52). Greenwich, CT: Information Age Publishing. Rouet, J.-F., & Coutelet, B. (2008). The acquisition of document search strategies in grade school students. Applied Cognitive Psychology, 22, 389-406. Rouet, J.F., & Vidal-Abarca, E. (2002). "Mining for meaning": A cognitive examination of inserted questions in learning from scientific text. In J. Otero, J.A. Leon, & A.C. Graesser (Eds.), The Psychology of Science Text Comprehension (pp. 417-436). Mahwah, NJ: Lawrence Erlbaum Associates. Rouet, J.-F., Vidal-Abarca, E., Bert-Erboul, A. & Millogo, V. (2001). Effects of information search tasks on the comprehension of instructional text. Discourse Processes, 31(2), 163-186.

ABSTRACT. The present study was aimed at reconciling the seemingly contradictory findings that readers routinely reject false information (epistemic validation) yet are susceptible to inaccurate information embedded in stories (misinformation effect). We investigated the online processing of clearly inaccurate information in a story via eyetracking while also testing for potential misinformation effects in a subsequent verification task. The findings suggest that when the inaccurate information contradicts commonly held knowledge, readers successfully detect and reject this information.

ABSTRACT. We investigated the differential effects of self-explaining a refutational text, over thinking aloud or rereading. Students who self-explained the text subsequently outperformed their peers on a test of natural selection knowledge. Additionally, both instructional and performance differences were significantly linked to the degree of causal cohesion present within students’ responses to the text. We interpret these results to indicate that self-explanation promotes specific coherence-building processes that are more conducive to conceptual change than other processes.

ABSTRACT. In this study, we investigated how early semantic anomalies induced by inconsistent connectors are detected using eye-tracking and whether there are interindividual differences in the repair of such inconsistencies. Results show that inconsistencies are detected early in canonical but not in non-canonical structures. Late repair processes were observed in both types of sentences, but were moderated by cognitive resources. Together, results support the distinction between the detection and repair of semantic inconsistencies during comprehension monitoring.

ABSTRACT. People frequently rely on inaccurate information, but what happens when their processing of that information is interrupted? Participants were repeatedly probed while reading a narrative text containing potentially inaccurate assertions about general knowledge topics. Subsequent use of that misinformation was assessed with a questionnaire. Participants produced more incorrect responses after reading inaccurate than accurate information. While interruptions might disrupt encoding of text contents, the disruptions here were insufficient to reduce participants’ reliance on misinformation.

ABSTRACT. There are no previous studies on how readers resolve white lies presented in text context. In the present experiment, 60 participants read white lies and true utterances embedded in short texts while their eye-movements were recorded. The role of individual differences and context in resolving the meaning of the lie were examined. Results showed that white lies were slower to process and more difficult to comprehend than true utterances.

ABSTRACT. Emotion-provoking information can have negative effects on visual perception and cognition, but little is known about how emotion affects memory for what we read. Memory was measured for short narrative texts, some of which included emotion-provoking sentences. Implicit memory for the neutral sentences was poorer when they followed emotional sentences than when they preceded them. In contrast, explicit memory for the neutral sentences was poorer when they preceded emotional sentences than when they followed them.

ABSTRACT. Spoilers have been shown to enhance enjoyment of stories (Leavitt & Christenfeld, 2011, PsychScience). However, this effect has been established only for stories that readers had no prior interest in. To test the beneficial-spoiler effect's generality, we spoiled stories either before reading or in the middle of reading. Before-spoiled stories were enjoyed less than unspoiled stories, and middle-spoiled stories were enjoyed slightly more. Findings are discussed with respect to a fluency-based explanation and reader characteristics.

ABSTRACT. We assessed mind wandering (MW) during film comprehension. We tested the possibility that having prior-knowledge of the plot of the film would either increase or decrease MW. A prior-knowledge condition read a text rendition of the film prior to viewing and MW less than a control condition. Prior-knowledge also suppressed MW throughout the duration of the film. Interest moderated the effect of condition: higher interest resulted in less MW when the plot was known.

ABSTRACT. Mood can influence attention to and comprehension of text content. We examined whether particular moods encourage or discourage readers’ encoding of and reliance upon information contained in narrative stories. Participants in happy and neutral moods utilized information presented in previously presented texts, including obviously inaccurate statements, to answer subsequent questions. This is consistent with a variety of projects documenting the general effects of misinformation exposure. But participants in anxious moods showed substantially reduced misinformation use.

ABSTRACT. Forty-five participants read immigration-related arguments while their eye movements were recorded. After reading each argument, participants rated the credibility of the argument and the valence and arousal induced by the sentence. First fixation duration was not influenced by these factors. Valence and arousal had significant effects on gaze duration. Credibility only affected regression path duration. The results demonstrate that the reader’s subjective emotional experience plays a crucial role in eye movement guidance during reading.

ABSTRACT. Three experiments manipulated the degree of semantic overlap amongst messages provided by various information sources. Readers’ source recognition was consistently poorer when texts' sources presented semantically-congruent compared to semantically-distinct messages. Participants did, however, display longer reading times and better recall memory for the claims and evidence statements from the semantically-congruent compared to distinct texts. We discuss implications for contemporary accounts of multiple text comprehension and implications for future research.

ABSTRACT. Task instructions can affect reading processes and products. We investigated readers’ comprehension processes when they were asked to focus on particular types of information (i.e., standards of relevance), but for different purposes (i.e., standards of coherence) while reading a dual-position text. The results suggest that standards of coherence and relevance uniquely and interactively affect reading, but that pre-existing attitudes play an important role in both processing and memory.

ABSTRACT. Inference generation is central to reading skill, and the ability to do so has been linked to characteristics of brain functioning. The current experiment used fMRI to investigate neural differences during optional (predictive) or required (bridging) inferences. Results suggested that higher-working-memory-capacity readers generated inferences in all conditions, recruiting highly overlapping brain areas under varying text constraints. Lower-capacity readers only reliably generated inferences under required conditions and showed smaller inference-related neural changes across conditions.

ABSTRACT. This research studied the manner that working memory supports inference generation. Specifically, we tested whether working memory contributes to predictive and bridging inferencing by enhancing retention, (re)activation, and/or suppression of textual and/or inferential information. Naming latencies of inferential and textual probes by high- and low-span readers were compared. Findings suggest that high-span readers activate and suppress predictive inferences faster than low-span readers, and activate more bridging inferences due to enhanced retention of prior text.

ABSTRACT. Subjects read passages and then applied either verbatim, gist, or relatedness criteria to judgements about explicit, paraphrased, inference, or lure test items. Multinomial tree processing analysis revealed that readers phantomly remembered the gist and even perceptual details of previously unencountered paraphrases. Bridging but not elaborative probes were also phantomly recollected, replicating prior findings. The results support theoretical claims that discourse paraphrases are consistent with both the textbase representation of a message and its situation model.