ABSTRACT. Written language can be used to measure the Big Five personality traits using computational models of language. The aim of this study is to test the moderating role of different variables of computational models in a meta-analysis of 23 independent estimates. While the results showed significant combined estimates of the correlations for the five traits, these estimates were moderated by the type of information in the texts, the use of prediction mechanisms, and the source of publication of the primary studies. It is concluded that written language analyzed through computational methods could be used to extract relevant information of personality.

ABSTRACT. The discourse of mathematical word problems can give students a plethora of challenges. However, one question may have been overlooked: does the code of the numbers (“4” vs. “four”) affect processing and accuracy? In this study, fifty-three adults completed tasks involving numerical and word problems containing the two codes. Results indicated slightly higher accuracy and significantly quicker response times for problems containing the arabic code compared to their verbal written code counterparts. Though accuracy differences were not significant, cluster analysis revealed that for participants that struggle with math problems, the verbal written code can cause significant difficulties.

ABSTRACT. We investigated how responses by college students to a reading motivation measure (RMM) was related to comprehension performance. 166 college students completed scenario-based and traditional reading comprehension assessments and a measure targeting five constructs: self-efficacy, curiosity, challenge, intrinsic motivation, and value. Surprisingly, regressions showed no significant variance explained with the total RMM, nor subscales. To explore further, we decomposed the RMM into items, and used these in models instead. Full and most subconstructs were now significant for both types of test. In each construct, some items showed unexpected, significant, negative coefficients. We discuss possible interpretations of this unexpected result.

ABSTRACT. By using the same set of demographic questions twice in a reading comprehension (RC) study, we identified students who were disengaged in the study, as indicated by inconsistent responses on the demographic questions. Students with inconsistent responses generally had poorer RC performance, but the effect was stronger in traditional than in scenario-based RC tests. We argue that scenario-based RC tests are more engaging than traditional RC tests, and thus, more robust considering students’ engagement issues.

ABSTRACT. This paper introduces and tests the reliability of a new computer application that facilitates the application of eight different extant readability formulas (i.e., statistically derived readability assessment models that are based on linguistic features and different readability criteria). The reliability is tested by comparing the readability criteria from two separate corpora (the Bormuth corpus and the Newsela corpus) with the formula scores derived from the application. The results show that the formula scores exhibit significant and high correlation to the difficulty of the texts, and that the application produces reliable output suitable for use in research and education.

ABSTRACT. Scenario-Based Assessments (SBAs) are reading comprehension activities that are designed to improve upon the documented shortcomings of traditional tests. However, not much is known about how various psychological measures that correlate with traditional tests may covary with SBA performance. In a large-scale sample of college students, subscales of component reading skills, self-report reading strategies, and metacognitive awareness were used to predict SBA and comprehension test scores within subjects. Many significant predictors were shared, though reading efficiency and problem-solving strategies were significantly associated with SBA scores, but not comprehension test scores. Implications for educational practice are discussed.

ABSTRACT. Word knowledge has long been considered as one of the most important predictors of reading comprehension, academic achievement, and social development. Currently, the examination of word knowledge typically remains limited to vocabulary size, precise understanding of single words, or simple associations between word pairs. No known study has examined understanding of the complex similarities and differences among sets of words (i.e., depth) across a broad range of words (i.e., breadth). This study aims to fill this gap by applying a relational reasoning framework to the assessment of work knowledge by means of a novel measure, the Relational Reasoning with Words.

ABSTRACT. Over the last decades, research into the processes involved in written production has progressed tremendously. While disciplines like (cognitive) psycholinguistics have studied cognitive processes in more and more depth, fostering advances in model building and methodology, text production studies such as so-called genetic criticism (critique génétique) have contributed to a richer understanding of the dynamics of writing and authors’ strategies. My PhD project combines approaches from these fields to look into literary writing in particular. It turns out that existing cognitive models can only partially account for the specifics of literary text production research in genetic criticism has pointed out.

ABSTRACT. The purpose of this work is to offer a better understanding of how okay is used in naturalistic dialogue. We examined the use of okay in a French job interview corpus and in a German research interview corpus. We found that the interviewer was more likely to use okay than the other participant; he or she was also more likely to use okay as a vertical marker. The effect of age as well as changed in the use of okay during the interaction were also examined. The findings contribute to understanding the overarching functions of backchannel utterances in coordinating conversations.

ABSTRACT. In a verbal conversation, processes specific to each interlocutor and contextual features may both influence the memorization of verbal information. The novel aim of this study was to examine how the nature of the information that is exchanged (self- or partner-produced, emotionally charged or neutral) influences conversational memory as a function of the type of interaction (collaborative vs. noncollaborative). Results underlined the importance of hitherto understudied factors, showing the replication of known effects (i.e., (self)-production and emotion effects) and highlighting the modulation of these ordinary effects by the collaborative nature of the interaction between partners.

ABSTRACT. Writing is a fundamental activity present in all disciplinary areas. In Science, a relevant scientific practice is writing scientific explanations. Analytical rubrics have generally been applied to assess students' scientific explanations, but Comparative Judgment (CJ) has emerged as a new, highly reliable, holistic, and less time-consuming assessment tool. This exploratory study aims to understand how CJ relates to analytical assessments specifically designed to evaluate scientific explanations. We discuss the implications of evaluating written scientific explanations through Comparative Judgment against analytical rubrics.

ABSTRACT. Processing of language by humans involves the intertwining of processes of production and comprehension. This paper describes how a cognitively inspired model can be adapted for the modeling of language generation and specifically details how a generated sentence can be parsed within a workspace in order to contribute to the program's self-evaluation.

ABSTRACT. We examined the cohesion of readers’ constructed responses to multiple documents in relation to source-based essay quality. Participants (n=95) were prompted to think-aloud, self-explain, or evaluate sources while reading a set of texts; they then used the texts to write a source-based essay. Natural language processing techniques were used to analyze the cohesion of the constructed responses at within- and across-documents levels. Results indicated that within-document cohesion was negatively related to essay quality, whereas across-document cohesion was positively related to essay quality. These relations differed by condition; self-explanation and source evaluation instructions were related to greater across-text integration than think-alouds.

ABSTRACT. To investigate the relation between the preparation, execution, and production stages of an ecologically-valid multiple source use (MSU) task, students’ (n = 118) detailed search log data and argumentative essay scores were analyzed using a mediated moderation measured variable path analysis model. Preliminary analyses suggest that the relation between variables in the execution stage directly predict essay quality and that this relation is moderated by variables in the preparation stage. These findings stress the importance of thoroughly researching what unfolds in the preparation stage in the context of a task with high student autonomy.

ABSTRACT. This study implements and expands a differential word use (DWU) measure for automatically assessing the quality of independent and integrated student essays. For independent writing samples, the baseline unigram DWU measure successfully predicted human scores in a training corpus but was less generalizable to a test corpus. For integrated writing, tri-gram measures yielded the best performance, and using n-grams as a unit of feature development improved the performance on both training and testing corpora. The findings provide information about the role of differential word use in explaining writing quality, and implications for providing automatic feedback to student writers.

ABSTRACT. This study aimed to investigate a trade-off relationship between specialized causal connectives and prosody in expressing subjective and objective causality. Several theories predicted that the presence of specialized causal connectives would eliminate the necessity of using prosody to express the distinctions between the two types of causality, and conversely that the absence of specialized causal connectives would make it necessary for speakers to use prosody to distinguish between the two types. These two hypotheses were tested in two production experiments, investigating the prosodic realizations of subjective and objective causality in Mandarin and English, respectively. The results support these hypotheses.

ABSTRACT. We examined the amount of variance attributable to child, text, and assessment in listening comprehension performance, and the roles of language and cognitive skills (child factor), genre (text factor), and comprehension question types (assessment feature) in listening comprehension, using longitudinal data from kindergarten to Grade 4. Children were assessed on listening comprehension, and language and cognitive skills (e.g., vocabulary). Results from explanatory item response models showed that a substantial amount of variance was attributable to items, with lesser amount to child and text. Different patterns and nuances were found across the genres and grades.

ABSTRACT. This study examined the ability of young adults from diverse backgrounds to process past tense counterfactual (PTCF) sentences. An aspect of academic language, PTCF sentences occur in textbooks where the author wishes to express the relative timing of past events with precision (“If women had surrendered the battle over voting rights, the 19th Amendment to the US Constitution would not have been passed”). Results indicated that many young adults had not mastered PTCF sentences and that the past perfect (had surrendered) was more difficult than the present perfect (would not have been passed). Possible reasons for these patterns are discussed.

ABSTRACT. This study examined the ability of poor comprehenders (PC) to process and remember central (main) ideas during and after reading, using eye-tracking. Results indicated that PC identified, (re)read and recognized central ideas to the same extent as good comprehenders (GC), whereas their recall of central ideas was significantly lower. A second experiment explored the origin of this retrieval deficiency, using think-aloud procedure. Results indicated that PC reinstated fewer prior central ideas than GC. This finding suggests that PC form fewer intra-text connections to central ideas during reading and, consequently, construct fewer retrieval cues for their recall after reading.

ABSTRACT. Many second-language reading assessments presume higher-order skills are transferred from first language literacy and may not accurately account for higher-order reading skills in scoring. This study examines the activation of inferencing during realistic L2 reading comprehension tasks. A reading comprehension test with three different tasks was developed and administered to 102 second language English and multilingual undergraduate and graduate students studying at a university in the US. Individual differences data were collected, and inferencing was measured using a sentence verification task. Scores on each task were related to different predictors, and the role of inferencing in successful comprehension is discussed.

ABSTRACT. Second language (L2) readers can maintain coherence as well as first language (L1) readers. Causal questions, which help readers to make connections between causes and effects, may be effective for maintaining causal coherence by L2 readers. This study used think-aloud and post-reading interviews to examine whether causal or comprehension questions maintain causal coherence. Statistical analyses revealed no differences in the coherence break during reading for each question. However, a qualitative analysis of the protocol showed that the readers who read linearly tended to change their reading behavior depending on the question type.

ABSTRACT. Using a sentence recognition task, we investigated whether elementary school children’s (N = 92) memory of the text surface, the textbase, and the situation model differed depending on whether the same information was embedded in an expository or a narrative text. Other than previous research with children that used narrative and expository texts dealing with different topics, our results did not indicate differences between narrative and expository texts for any of the levels of representation. Thus, the effect of text topic on effects of genre on text comprehension might be important to consider and needs to be investigated further.

ABSTRACT. We explored word learning opportunities in children’s novels with text analysis. Forty-seven books (M=67726 words) with an average Lexile level of 990L (SD=127L) from Gutenberg.org were analyzed. We used an age-of-acquisition database to identify words potentially unfamiliar to different age groups and we examined repetitions and spatial distribution of these words within each book. Results showed children’s novels provide amble opportunities for children to encounter and re-encounter unfamiliar words, especially when reading novels above their grade level. When reading a book, children re-encounter the same target word in a few days, creating a natural spaced learning schedule, facilitating long-term retention.

ABSTRACT. The inconsistency paradigm is used to measure metacognitive competences in the domain of reading comprehension. We found that the offline indicator (number of detected inconsistencies) was subject to age-related changes during adolescence but the online indicator (reading time of inconsistent compared to consistent texts) was not. This pattern suggests different demands of offline and online measures on memory capacity and general cognitive abilities. Although both indicators were moderately correlated, only the offline indicator was substantially associated with reading competence. Thus, offline and online indicators obtained with the inconsistency task seem to measure different aspects of metacognitive processing.

ABSTRACT. Based on "embodied" theories of comprehension and on studies demonstrating the beneficial impact of mental imagery trainings on children’s comprehension, this study examine the hypothesis of a mental imagery deficit in children with a specific comprehension deficit (S-LSC). Several imagery tasks (implicit/explicit; visual/motor,) have been administered to both skilled readers (N=37) and S-LSC (N=14) in Grade 5. Comparisons of these two groups reveal that S-LSC showed poorer imagery capabilities on explicit imagery tasks but not on implicit ones. These results are discussed in terms of a possible lack of automaticity in mental imagery in S-LSC.

ABSTRACT. This pilot study examines the links between L2 undergraduate writers' (n = 74) fluency features (operationalized using keystroke log measures) and the argument and discourse elements (human annotated based on a modified Toulmin's model of argument) produced in their argumentative writing process. Results suggested that two fluency features (i.e., pause time and P-burst length) can be combined to make relatively accurate classifications of the argument elements constructed in L2 writers' text production. These findings provide implications for links between discourse production and process as well as the development of automated writing evaluation (AWE) and L2 argumentative writing instruction.

ABSTRACT. This study examined the effects of individual differences and genre on the cohesion of readers’ think-aloud protocols. Participants (n=119) were instructed to think-aloud while reading a history and science text and subsequently completed reading skill and working memory assessments. Correlations and mixed-effects models were used to examine relations between think-aloud cohesion, individual differences, and genre. Results revealed that working memory was related to cohesion for history texts whereas reading skill was related to cohesion for both texts. These findings indicate that the interaction between genre and individual differences may be used to model coherence-building processes during reading.

ABSTRACT. The role of Theory of Mind (ToM), which refers to one’s understanding of the thoughts and feelings of others, in reading comprehension is a burgeoning research area. We synthesized the results of 15 studies on ToM and reading comprehension published between January 1983 and May 2018. Our findings indicate the relation between ToM and reading comprehension in the general population is small, potentially stronger for second order and advanced ToM, and mediated by listening comprehension. For those with autism, there may be a small direct relation. We conclude with a critique for the consideration of including ToM in future studies.

ABSTRACT. This presentation reports on a corpus analysis of Agatha Christie’s use of discourse in the person of her fictional Belgian detective, Hercule Poirot. By focusing on a specific linguistic trait – false friends – it presents evidence of how Christie used this trait to develop Poirot’s character across her thirty-three novels in which he appears. Moreover, it shows that this trait is almost completely absent from the four novels featuring Hercule Poirot written by Sophie Hannah, who was chosen by Christie’s estate to continue his adventures.

ABSTRACT. We introduce the Anonymous Ease of Readability (AEAR) corpus, which provides researchers within the reading community with a resource from which to develop and test readability metrics and to model text readability. The AEAR corpus has a number of improvements over previous readability corpora, including size (N = ~5,000 reading excerpts), the breadth of the excerpts available, which cover over 250 years of writing in two different genres, and the readability criterion used (teachers’ ratings of text difficulty for their students). This paper discusses the development of the corpus and presents reliability metrics as well as initial analyses of readability.

ABSTRACT. This presentation introduces the Persuasive Essays for Rating, Selecting, and Understanding Argumentative and Discourse Elements (PERSUADE) corpus which was developed to spur the creation of open-source algorithms that identify and evaluate argumentative and discourse elements in writing. The corpus will not only allow for large-scale analyses related to the categorization of discourse and argumentation elements, but also investigations into relationships among these elements, the quality of these elements and the links between these elements, their relationships, their quality, and holistic scores of essay quality.

ABSTRACT. Using eye tracking, we extended ERP evidence that word-to-text integration at the beginnings and ends of sentences is primarily determined by local text factors (antecedents in a previous sentence) but that global factors (central theme) may make these antecedents more accessible in memory and thus facilitate their integration. Eye-tacking and ERP measures aligned generally but reading times did not consistently correspond to reduced N400s. A distinctive eye-tracking result is that when antecedents were not related to the central theme of the passage there was a greater likelihood of return to the antecedent from the regions beyond the target word.

ABSTRACT. This study examined the role of the causal connectivity of the statements and the modality of presentation of discourse in the generation of emotion inferences. We asked a group of Argentine college students to either listen to or read an excerpt of a radio interview, and to perform an emotions recall task. Statements that promoted the generation of emotion inferences and had a high number of causal connections were more often included in the recall protocols than those that had a low number of them. Also, listeners included a higher number of statements than readers.

ABSTRACT. Teacher education aims to move students from novice to expert level. In this study, we analysed student’s textual peer feedback on video recordings of their teaching practice. First, the impact of the curriculum and literature on students’ feedback by the network analysis of prominent words. Secondly, the lexical richness and the semantic cohesion of students’ feedback and reflections. Our findings show that students created stronger connections between the prominent words from the literature. The lexical richness and semantic cohesion also increased. This means that students incorporated vocabulary from expert sources and maintained semantic consistency while using the expert vocabulary.

ABSTRACT. The current study employed a corpus stylistic approach to analyze James Joyce’s Dubliners and A Portrait of the Artist as a Young Man (hereafter Portrait). Joyce’s two works were selected for their literary significance and distinct nature in terms of themes and literary movements. The analyses demonstrated that the overall linguistic complexity was higher in Portrait than in Dubliners and that cohesion was lower in Portrait. The present findings were interpreted in relation to literary criticisms and literary movements associated with each work. This study broadens the extant research in literary corpus stylistics by analyzing literary texts on multidimensional features.

ABSTRACT. This paper demonstrates the need for a new methodological approach to ecolinguistics termed contrastive eco-discourse analysis. While writing about the 2016 movement at Standing Rock, D/Lakota authors used Lakota words and phrases which cannot be translated to English. Contrasting the differences in the ways these Lakota words and phrases express the relationship between human life and water with their closest possible English translations reveals that English discourse does not reflect the relationship between human beings and water at level of etymology, morphology, semantic meaning, and language practices in ways only made clear through contrast to Lakota linguistic alternatives.

ABSTRACT. Small group discussions can lead to elaborated argumentation and deep learning. But they also necessitate socially difficult situations, e.g. when students want to offer criticism or unusual perspectives. Competent groups will use linguistic politeness strategies to mitigate the impact of these face-threatening acts. Because argumentative and interpersonal skills develop over time, this should vary with age. Analyses of 27 small group discussions from German classrooms in two age groups, recorded in the context of the DIALLS project, show how these cognitive and social aspects are intertwined. The results offer conclusions for argumentation research as well as practical implications.

ABSTRACT. This paper applies the methods of critical discourse analysis on the coverage of political protests that happened across the world during the years 2019 and 2020 (such as 2019-2020 Hong Kong protests, 2019 Bolivian protests, or George Floyd protests) in Czech media, focusing on the political statements of Czech politicians commenting discussed protests. It aims to identify how different events are covered and viewed in Czech discourse(s) and how this coverage is connected to Czech foreign policy and neoliberal ideology in general. Selected politicians either officially represent Czech foreign policy or are just a highly influential political figure.

ABSTRACT. We investigated the effects of reading medium and reading time-frame on readers’ attention, metacognitive calibration, and comprehension. We asked 140 undergraduates to read an expository text in print or on a computer under self-paced or constrained reading time. We probed their generation of task-unrelated thoughts (mindwandering) while reading. After reading, they predicted their performance on a subsequent test about the text content and completed the test. Under time pressure, on-screen readers mindwandered more than in-print readers, and they scored lower on the comprehension test. No differences in calibration appeared. Reading on screen hindered readers’ attention and comprehension under time pressure.

ABSTRACT. Task-unrelated thought (TUT) is a common phenomenon with highly variable rates in everyday life, but less is known about the dynamics of TUT in conversations. The current study explores how the Topic constraints of a conversation influence 1) overall rates of TUT, 2) divergence of TUT ratings between partners, and 3) TUT and TUT divergence over time. Topic constraints did not influence overall divergence, but divergence between partners over time was influenced by the Topic constraints of the conversation.

ABSTRACT. Little research has been conducted to examine the phenomenological experiences of reading. This study examined how participants’ ratings of their visual and verbal thoughts during reading related to comprehension and transportation into the text, and whether this was influenced by perspective-taking instructions. Participants (n=147) received either perspective-taking or control instructions and read a narrative while being periodically asked to rate their thought characteristics. Results indicated that visual imagery was positively associated with comprehension and transportation; conversely, verbal-thinking was negatively associated with comprehension but not transportation. No effects of condition were found for either thought probes, comprehension, or transportation.

ABSTRACT. The goal of this project was to determine how textual features give rise to certain types of thoughts during reading. We analyzed five studies that each assessed thinking while reading using experience sampling probes that measured reports of task-unrelated thought (TUT), personally connective thought, visual imagery, and verbal thoughts. Results show that the concreteness of text negatively predicted TUT and verbal thoughts, but positively predicted personal connections and visual imagery. These results provide a nuanced and comprehensive view of how features of language in text predict various types and characteristics of individuals’ thoughts during reading.

ABSTRACT. Seventy participants read excerpts from the novel ”Memories of the Future” by Siri Hustvedt while their eye movements were recorded. Mind-wandering was probed after 30 emotionally neutral, negative and positive paragraphs with a 13-item questionnaire (Turnbull et al., 2019) tapping into the content and form of current thoughts. Principal components analysis of the questionnaire data revealed five thought patterns: immersion, rumination, voluntary verbalization, mind-wandering to the future, and thoughts about past and others. Emotionally positive paragraphs induced more immersion, whereas negative paragraphs reduced mind-wandering to the future. Eye movement data provides insights into the attentional processes underlying these thought patterns.

ABSTRACT. False assertions appear in news and social media discourse. They also routinely appear in fiction, as authors are not bound to truth in their narratives. Experimental materials intended to examine these misleading fictional experiences tend to be lengthy and inflexible, and exclude assessments of attention or comprehension. We developed a set of 39 short stories, each containing a normed assertion, appropriate in length and difficulty for online data collection. Two experiments verify that these materials replicate detrimental effects of exposure to inaccurate information in fiction. The materials can be used to test a variety of to-be-identified issues in future work.

ABSTRACT. With the proliferation of unfiltered information continuing to drive polarization, building critical thinking skills around media use is a pertinent challenge for scholars and practitioners. News media literacy (NML) is an educational intervention that utilizes journalism practices to develop mindful and critical consumption strategies for consumers of news media. The current project presents a case study of secondary Civics teachers as they utilize NML lessons and current events information as part of their pedagogy and practice. Analyses of classroom routines and discourses reveal teachers' uses of a variety of constructivist and instructionist approaches when analyzing news content and NML content.

ABSTRACT. This study examined the effectiveness of using self-explanation prompts to enhance students’ ability to produce bridging inferences, improve their text comprehension, and correct their misconceptions. College students were prompted to self-explain or think-aloud while reading a non-refutational or refutational text. The students were assessed on their text comprehension, conceptual understanding, vocabulary, and prior knowledge. The results demonstrated that students prompted to self-explain produced more bridging inferences than students prompted to think-aloud. In addition, students who generated more bridging inferences better text comprehension had fewer misconceptions after reading. However, students’ conceptual understanding also depended on their prior knowledge and reading skill.

ABSTRACT. Participants completed either a simple task (control), or a cognitively taxing task (resource depletion) and then participants read six passages written to be either vague or concrete. For each passage, participants answered questions that assessed reading comprehension at the surface form, textbase, and situation model levels. The results provide evidence that depletion of cognitive resources negatively impacts readers’ ability to form a coherent situation model, particularly when reading vague passages that are difficult to understand. These findings suggest that successful reading comprehension outcomes are dependent on the attentional and working memory resources available to the reader.

ABSTRACT. The effect of different learner variables on two multiple source-based tasks was examined. One-hundred and fourteen students read two sets of multiple texts and completed two integration tasks; A sentence verification task and essay writing. Students were scored on three levels: main ideas in each text, intertext links and generating inferences. The following learner variables were assessed: reading comprehension skill, reading strategies, prior knowledge, interest, vocabulary, and scores of answers to questions in-between texts. Results showed that the two tasks and the three integration levels were predicted by different variables, supporting a multifaceted nature of multiple text integration tasks.

ABSTRACT. In this study, I examine the online processing of satirical news as it unfolds within a text. Reading times were compared between authentic satirical and non-satirical texts using within- and between-subjects designs. Reading times were measured for six text segments, three of which did (satirical) or did not (non-satirical) contain a lexical incongruity. Multilevel modelling results indicated the initial headline incongruity was associated with significantly slower reading times for satirical news, but only for participants in the within-subjects design. These results suggest readers were able to adjust their coherence thresholds when encountering initial and subsequent incongruities in authentic satirical texts.

ABSTRACT. What are the relative roles of visual and verbal modalities when constructing mental models during the processing of short visual narratives? The Independent Contribution Hypothesis states that each modality independently contributes to the mental model. The Integrated Impact Hypothesis states that the modalities are highly interactive and contribute to a single representation. In two experiments participants were presented four-panel comic strips. We manipulated the order of the text and images for panels 1-3. We analyzed viewing time data. Interestingly, viewing times were most supportive of the Integrated Impact Hypothesis.

ABSTRACT. Research has shown that comprehension of narrative and expository texts differs, presumably due to differences in text characteristics across the two text genres. Although other studies suggest that task-unrelated thought may also be sensitive to some of these same text characteristics, these effects have not been systematically examined. Two experiments thus aimed to assess how text genre impacts the frequency of task-unrelated thought and whether those thoughts are personally-connected to the text. Findings suggest that readers are more likely to have task-unrelated thought in the absence of a personal connection when reading expository texts compared to when reading narrative texts.

ABSTRACT. The present study measured the impact of working memory (WM) and prior knowledge (PK) on text comprehension. Participants read either a high-coherence or low-coherence text about mitosis, or an unrelated control text. They then answered comprehension questions about mitosis. Finally, they were tested for their WM, vocabulary, and biology PK. Using hierarchical regression, results demonstrated that PK was the best predictor of reading comprehension across conditions. However, a three-way interaction demonstrated an increased role of PK specifically for low-WM individuals who read the low-coherence text. This suggests a complex relationship for the roles of WM and PK during reading.

ABSTRACT. Understanding a text is based on its mental representation. Detailed incoming information are first organized at a local level of discourse but then restructured and connected to the reader’s world knowledge, so that finally a more global representation emerges (e.g., O’Brien & Cook, 2015). The current study aims to test whether text comprehension empirically reflects these different levels. Participants read short stories and subsequently wrote summaries and answered comprehension items. Confirmatory factor analysis (CFA) was used to implementing comprehension as uni-dimensional or multi-dimensional construct resulting in better fit for the latter.

ABSTRACT. Journalists are typically taught for breaking-news stories to keep paragraphs to one to three sentences, whereas for feature-length stories longer paragraphs are used. We examined the impact of these conventions on readers by having them read either a breaking-news or feature-length news article with short- or medium-length paragraphs or no paragraph markings. Enjoyment, ease of reading, reading time, and memory was measured. There was tentative evidence that short paragraphs led to greater ease of reading and faster reading.

ABSTRACT. Natural language quantifiers -- words such as few, a few, and many -- have a powerful impact on readers’ focus. The polarity of a quantifier, positive (a few) or negative (few), largely determines focus effects. In the current series of experiments, we demonstrate that a naturalistic story context can influence the interpretation of positive quantifiers. Although positive quantifiers almost always lead to focus on the reference set, this is not always the case, as the polarity of a quantifier interacts with the pragmatics of the discourse.

ABSTRACT. In two experiments, we examined the effects of context on the comprehension of semantic anomalies that were encountered twice. Narratives contained two encounter sentences that were either correct (e.g., Noah brought two animals of each kind on the ark), or anomalous (e.g., Moses brought two animals of each kind on the ark). In both experiments, processing difficulty occurred on the anomalous first encounter. Processing difficulty on the second encounter depended on the contextual differences between the first and second encounter and the availability of the first encounter. Results are discussed within the RI-Val framework.

ABSTRACT. We report one offline experiment examining the relationship between the processing of referential form and the so-called Voice, a literary phenomenon occurring in writers’ manipulation of distance between character and reader. Participants read either narratives that were filtered through the main character’s voice solely or narratives of which the interpretation could be ambiguous; their preferences concerning the eventual linguistic realisation of the character were recorded. Participants were more likely to choose the pronoun to refer to the target character under single voice conditions. Our results demonstrate that preferences regarding referential form are indeed sensitive to a subtle distinction in Voice.

ABSTRACT. Traditional Humanities have explained the creative human abilities from a metaphysical paradigm of knowledge that describes art as a hedonist artifact. However, observing the communicative dimension of artistic-narrative discourses from the conclusions of Cognitive Sciences distances us from that vision. We propose to utilize the results of neuroscience to delve into the neurocognitive processes that take place in the producers and receivers of a literary text when it comes to arrange or transmit mental image narratives. This new perspective of studying the creation and the impact of artistic discourses leads us to redefine the role we traditionally assigned to art.

ABSTRACT. Narrative comprehension is believed to be a general skill that can transfer across media. We examined differences in comprehension of narratives, by presenting children (ages 8-12) with three stories in different formats (text, audio, video). The results indicate that the medium in which the story was presented affected offline comprehension. More specifically, skilled comprehenders performed equally well across media, whereas less-skilled comprehenders performed significantly better on video understanding compared to text understanding. The impact of media affordances on narrative comprehension will be explored further through analysis of the think-aloud and recall data that were collected as well.

ABSTRACT. We investigated the role perspective-taking has on preference for character outcomes. In all experiments, participants were instructed to read for comprehension or adopt the perspective of the main character. When character descriptions were neutral and participants adopted perspective, successful outcomes were preferred over failed outcomes (Exp 1b); no difference for outcome type occurred under comprehension instructions (Exp 1a). For heroic characters, participants preferred successful outcomes over failed outcomes independent of instruction (Exp 2a & 3a). Participants preferred successful outcomes for villains when they adopted the villain’s perspective (Exp 2b & 3b).

ABSTRACT. Engaging in explanation while studying expository science texts can improve comprehension and comprehension monitoring. Readers often make judgments of comprehension (JOCs) so that metacomprehension accuracy can be assessed, but making JOCs may have its own benefits. The present study used a fully-crossed design to explore the independent effects of explanation activities and making JOCs on comprehension outcomes. In addition, the timing of restudy opportunities was varied. When students had the opportunity to restudy in the same session as initial study, then comprehension benefited from explanation activities. When there was a delay before restudy opportunities, then comprehension benefited from making JOCs.

ABSTRACT. Testing students on to-be-learned materials can be an effective learning activity. It has been demonstrated that practice tests with open-ended questions can promote understanding from text. The present study explored whether practice tests with closed-ended questions can also improve understanding from text and whether differences could be observed between different types of closed-ended practice test formats: multiple-choice and true-false tests. While no significant differences were seen between the two formats, results showed closed-ended practice testing can lead to benefits in both comprehension monitoring and comprehension outcomes compared to a re-reading condition.

ABSTRACT. This study tested how the presence of instructional visuals affects students’ judgments of learning. Undergraduates studied four biology texts with or without instructional visualizations. For each text, students rated how well they believed they would perform on a test of the material, how well they could explain the material, and how well they could draw the material. Then students completed comprehension tests on all the texts. The presence of visuals did not affect comprehension performance; however, it did significantly hinder relative monitoring accuracy, specifically for judgments of drawing. This finding suggests cues from provided visuals uniquely influence students’ drawing judgments.

ABSTRACT. This study examined the manner that time constraints and domain knowledge modulate the underlying processing and relative effectiveness of two strategies employed in reading comprehension tests: Reading the text before (TB) versus while (TW) answering the questions. Results demonstrated higher performance accuracy, albeit longer performance time, and inferior search efficiency for TW, when test was timed. Domain knowledge enhanced performance accuracy regardless of strategy and time constraints, although performance time and search efficiency were similar. These findings suggest that examinees should apply TW strategy in timed tests. The theoretical implications on goal-oriented models and RC tests’ validity will be discussed.

ABSTRACT. The study investigates which reading strategies university students use on the fly, how they perceive their processing retrospectively and which strategies they generally judge as appropriate for understanding narrative and expository texts. Participants read an expository and a narrative text. Reading strategies were assessed by text traces, note-taking techniques, subsequent interviews and a standardized questionnaire to reflect on strategies for both genres. Whereas text tracings and note-taking techniques differed substantially with respect to the types of texts, the interviews revealed differences in terms of memorization, organization and elaboration, but also similarities. In contrast, the questionnaire demonstrated more communalities than differences.

ABSTRACT. This study explored, whether learning from a scientific text on plate tectonics can be enhanced by an interactive drawing strategy. 8th-grade students (N = 94, M = 13.34) of a Swiss- secondary school read a scientific text and answered questions on recall, transfer, and drawing accuracy and on cognitive load and interest. Instructions varied according to a 2 x 2-factorial design with “learner-generated-drawings” (yes, no) and “interactive drawings” (yes, no) as two factors. Results across outcome measures (transfer, recall, drawing) indicate that interactive drawing groups performed better than the groups who learned without interactive drawings.

ABSTRACT. The newest version of the Medical College Admission Test (MCAT) includes a Critical Analysis and Reasoning Skills (CARS) section which tests students’ ability to understand, infer, and reason from text. Since its inception, it has generally become the lowest scoring section on the exam. The current study explored whether training to identify question types and the strategies useful for each question type would aid pre-medical students in preparing for the CARS section of the MCAT.

ABSTRACT. In this study, we compared case-based learning and learning by design pedagogies on learning how to evaluate “blog” like research summaries. Working in pairs, students in the case-based learning conditions critiqued research summaries on the presence of “flawed” variables (e.g., small sample size) whereas students in the learning by design condition created and evaluated summaries that contained flaws. Both groups learned equally well but students in the learning by design condition were more engaged in the activities and gave higher scores of perceived learning. The findings indicate that students learn and value the experience of designing materials.

ABSTRACT. We examined the effect of students’ ability to state task-specific strategies in solving analytical reasoning problems on exam performance. To solve these reasoning problems, the reader must represent the goals and strategies necessary to solve them. We predicted that students who stated specific strategies after solving the problems would perform better on an analytical reasoning exam compared to those who stated general strategies. We found that more task-specific strategies explicitly mentioned positively predicted exam performance.

ABSTRACT. In the current study, we examine introductory psychology students’ reported use of discipline-specific reading strategies as a measure of the extent to which they hold discipline-appropriate task models for reading in psychology (e.g., focusing on causal mechanisms, data from studies, and understanding how evidence supports theory). Student’s reported use of discipline-specific reading strategies predicted comprehension performance over and above their reported use of general college reading strategies when answering both text-based and inference questions. These results suggest that helping students develop discipline-appropriate task models and reading strategies in introductory college courses may be an effective way to support students.

ABSTRACT. Studies suggest that children’s understanding of how questions about a text are answered and whether they are important to main ideas may facilitate improvement in reading comprehension. However, whether adults seem to have this understanding is unknown. In the ongoing study, we seek to fill this gap in the literature. Our findings may add to our understanding of question asking while reading. We hope to inform the development of reading interventions. Data collection is ongoing.

ABSTRACT. Making scientific evaluations and plausibility judgments involve domain-specific and domain-general factors. In the current study, we explored the contribution of certain domain-general processes--executive functions--in making plausibility judgments. Seventy-one adults completed tasks that measured their executive functioning skills (inhibitory control and shifting), as well as a scaffolding activity in which they made plausibility judgments and evaluated explanatory statements about climate change. The results showed that adults’ performance on executive function tasks was not related to their evaluations or their plausibility judgments. However, we found a relation between inhibitory control, but not shifting, on their reading time.

ABSTRACT. Scientific knowledge can only provide a provisional orientation for political decision making. Past discussions about mandatory mask-wearing illustrate this clearly. In this context, we investigated how the communication of scientific uncertainty affects source trustworthiness. Participants (N = 398) read an evidence-based text supporting mandatory mask-wearing. The text included lexical hedges (e. g. “maybe”) (or not), that is epistemic uncertainty was communicated (or not). Second, the communicator was a scientist or a politician. Including hedging did not affect trustworthiness judgements. However, the scientist was perceived as having more integrity and as more competent but not as more benevolent than the politician.

ABSTRACT. Today, online videos are a popular resource for learning about scientific topics (List & Ballenger, 2019). In this study, we examined whether the use of simplified language pervasive in online videos inclines laypeople to overestimate their own capabilities of making informed validity judgments. Forty-six medical laypeople watched videos about health topics that were either easy or difficult to comprehend. Participants agreed more strongly and confidently with contained knowledge claims and decreased their desire to consult an expert for support after watching easy videos. This suggests that easy-to-comprehend videos can tempt laypeople to overlook the limits of their own evaluative capabilities.

ABSTRACT. This study explores academic language development trajectories in written scientific explanations of fourth-grade students. The researchers analyzed 136 explanations produced by 35 students who participated in a learning sequence aimed at fostering scientific explanation skills, with explicitly academic language teaching. Four explanations per student --initial, draft, revision, and final-- were coded according to genre mastery, voice consistency, organization of information, and textual connection. Differences and similarities were found in the changes over time in academic language use principles. The results shed light on the diverse trajectories by language principle and writing task type throughout the learning sequence.

ABSTRACT. It can be challenging for students to evaluate the connections between scientific evidence and alternative models of natural phenomena in classroom learning situations. Adolescent students also seek to be more autonomous in their science learning. Our team has hypothesized and subsequently compared two instructional scaffolds---one which is more autonomy supportive--for facilitating more scientific evaluations, judgments, and understanding, about complex geological topics: hydraulic fracturing (fracking) and fossils as indicators. Both MEL activities led to increases in knowledge, about 5%, which is an appreciable gain for a 90 minute activity. However, the more autonomy supportive scaffold resulted in more scientific evaluations.

ABSTRACT. Given recent disappointing PISA results, the Dutch reading curriculum could benefit from evidence-based recommendations to teach children about text structures (Pyle et al., 2017) and combine reading with Science education (Bradbury, 2014). In this materials analysis, we therefore examined to what extent current Dutch Science learning materials for grades 3-6 are appropriate for text structure instruction. We discuss whether science texts provide opportunities to teach children about text structure at the paragraph and the text level, whether assignments direct children’s attention to coherence relations, and whether visualizations can be related to text structures such as cause-effect and chronology.

ABSTRACT. Comprehension can be measured by asking readers to reason beyond the text, such as applying concepts to new situations, or manipulating or integrating new elements into a preexisting mental model. The current study tested whether poor performance on hypothetical inference questions might be due to difficulty accessing the textbase or from reasoning with the information. Results suggested that readers may be able to reason with the information provided it is active in working memory.

ABSTRACT. This study examines learning from a simulation as a multimedia comprehension task. Biology students completed a computer-based simulation about photosynthesis. They were randomly assigned to a 2(instruction: control-of-variable instruction, control) x 2(agency: passive, active) between-subjects design. After working through the simulation, participants completed a comprehension test. We examine how the manipulations influence comprehension as well as how these effects are influenced by individual differences in prior knowledge.

ABSTRACT. The current experiments investigate memory for source information when reading multiple texts reflecting gossip. Results demonstrate that discrepancies promote memory for links between sources and their content, but not links between the sources themselves, relative to when texts are consistent. Discussion focuses on the potential for cognitive conflict brought about by discrepant information in texts to aide in constructing more complete intertext models, which could likely result in more accurate mental representations of the texts.

ABSTRACT. Corrections to readers’ misconceptions should elicit higher belief when information sources are high-credibility. However, evaluations of credibility are malleable, and we do not yet fully understand how changes to a source’s credibility influence knowledge revision. Thus, we examined how updating a source character’s credibility (high-credibility vs. low-credibility vs. neutral/no change) influenced knowledge revision and source evaluations after readers engaged with refutation and non-refutation texts. Results indicated that revision was superior when refutations came from sources that were later updated to be high-credibility. Sources were also rated as higher in credibility in the refutation condition than in the non-refutation condition.

ABSTRACT. Two experiments examined which features of a scientific source laypersons consider to be useful evaluation criteria. Extending previous conceptualizations of source credibility, the studies differentiate between two central aspects of a source’s ability (expertise and pertinence) and motivation (benevolence and scientific integrity). Medical laypersons evaluated scientific sources and their health-related claims. The sources varied in terms of their expertise and pertinence (Experiment 1) or their benevolence and scientific integrity (Experiment 2). Results show that laypersons consider both ability and motivational factors to evaluate source credibility and claims, but they do not clearly distinguish between these source characteristics.

ABSTRACT. Researchers have long examined how individuals determine source trustworthiness, in part by considering author benevolence, or freedom from bias or commercial intent. We examine how pre-service teachers (PSTs) evaluate author benevolence within one specific context, when choosing technology applications for classroom use. PSTs were presented with four app reviews attributed to a commercial website, a commercial website with a “real” testimonial, an objective technology blog, and a technology blog featuring a sponsored post. Among other findings, PSTs considered the objective technology blog to be most trustworthy and the sponsored post least trustworthy, even relative to the two commercial sites provided.

ABSTRACT. This study used eye tracking and cued-retrospective thinking aloud to examine how laypersons as compared to experts in the domain of nanosafety read an online article containing conflicting scientific information and considered source information provided within the article. A sample of 21 laypersons and 20 experts was presented with a mock-up online article discussing whether nanoparticles emitted from laser-printers are a potential health risk. Results showed that experts allocated more visual attention to and reflected more on source information provided in the article and also attributed the scientific conflict to a higher extent to differences in sources’ competence than laypersons.

ABSTRACT. The influence of inadmissible evidence on juror verdicts provides a scenario for exploring the impact of irrelevant information on multiple-source comprehension. Participants read a trial summary presenting target evidence in one of four conditions: Admissible (evidence deemed admissible), Due Process (evidence deemed inadmissible because of fairness), Unreliable (evidence deemed inadmissible because of unreliability), Control (evidence excluded). Afterwards, participants gave a verdict (guilty or not guilty), rated the importance of the target evidence, and generated essays. Results indicated that readers rated inadmissible evidence highly and refer to it in their essays, providing preliminary evidence that readers integrate and use inadmissible evidence.

ABSTRACT. Ability to critically evaluate the contents found online is a critical skill. One way to evaluate contents is by sourcing (i.e., assessment of source dimensions. Sourcing is a challenge for teenagers (e.g., Potocki et al., 2019). This may be related to the development of a theory of mind, which extends throughout childhood and into adolescence (Miller, 2009). The present study examined the role of advanced theory of mind (TOM²) in teenagers' source evaluation skills. We hypothesized that after controlling for basic reading abilities, TOM² would predict the evaluation of sources' dimensions. We found moderate relationships in the expected direction.

ABSTRACT. Texting, like speech, relies on fast-paced, reciprocal exchanges. But texting lacks the nonverbal cues found in speech that are critical for accurately interpreting meaning. Texters use textisms, like emoji, to bridge that gap. Here, we ask whether pairing neutral texts with positive-valence emoji or negative-valence emoji will impact the interpretation of those texts. We found that when participants saw neutral texts with positive emoji, they rated the messages as more positive than when they saw those texts with negative emoji. The results indicate that texters use emoji to convey meaning that would be expressed by non-verbal cues during speech.

ABSTRACT. In this project, we examined how including face-like emojis would influence the interpretation of ambiguous text messages presented as sent between friends or between acquaintances. Participants saw screenshots of brief iPhone conversations and selected from four options the most likely interpretation of ambiguous final messages. Without an emoji, participants preferred literal interpretations, and this did not differ by sender-recipient relationship. With the emoji, participants preferred interpretations congruent with the specific sentiment conveyed by the emoji, especially for conversations between acquaintances. People are sensitive to the ways that emojis can convey more specific meanings, and this may vary across communicative contexts.

ABSTRACT. In this study, we examined teachers’ remote use of a science dashboard, Inq-Blotter, through discourse analyses and students’ corresponding inquiry performance in Inq-ITS. Specifically, Epistemic Network Analyses were applied to compare patterns of support elicited by Inq-Blotter when students improved versus did not improve on inquiry in Inq-ITS. Analyses revealed significant differences in teacher support patterns in relation to student improvement. These results demonstrate how dashboards can support science discourse and learning during remote instruction.

ABSTRACT. We describe efforts to design and test an online, collaborative environment to facilitate the development of middle school students’ argumentation skills. Cognitive lab studies were conducted to examine the patterns of group interaction and individual students’ contribution to the critical discussion (n = 34). We found that more students used text-based communications than audio, but the students who used the audio function collaborated with each other more frequently. The survey and interview results suggested an overall positive experience with this computer-supported critical discussion activity. In this presentation, we will discuss the activity design, study results, and implications.

ABSTRACT. To fully realize the vision of the Next Generation Science Standards, students must be able to make sense of the scientific text that guides their inquiry and understanding of science phenomena. In this study, we analyze the text features of several investigative questions in the intelligent tutoring system [ITS] to understand the ways in which the difficulty of the investigative questions relates to students’ competencies with science inquiry practices. Results show that certain word-level features (e.g., familiarity, concreteness) of the investigative questions can explain students’ performances on science inquiry, but that these features differ across science topics.

ABSTRACT. The digital ecosystem of the 21st century has provided many people with the opportunity to engage in discourse with one another. Still, there is little understanding about the processes that are involved in collaborative extraction of meaning during reading comprehension. The purpose of this poster presentation is to provide an overview of the potential representations and processes involved in collaborative reading given extant theoretical comprehension research and spark further discussion on collaborative reading comprehension research.