ABSTRACT. This presentation examines whether using an Artificial Intelligence tool to complete an assigned university task (writing an essay after reading a text), harms the comprehension of the text information provided to the students and the attention paid to the source information included in the text. Data were collected from university students’ spontaneous use of these tools. Results suggest that students employed these tools for writing the essay rather than extracting information from the text.

ABSTRACT. We compared how 147 adults engaged in multiple text reading and writing tasks with versus without ChatGPT. Essays that were written with ChatGPT assistance were longer and had higher levels of integrative argumentation, source use, and source citations. However, ChatGPT users read less, had lower single- and multiple-text comprehension, and found the task less interesting. ChatGPT use also influenced the essays' conclusions. Thus, ChatGPT use appears to involve substantial trade-offs that raise critical instructional challenges.

ABSTRACT. This study compares ChatGPT’s performance in sourcing and corroboration tasks with that of lower secondary, upper secondary, and teacher students. While ChatGPT outperformed all human groups in sourcing, it struggled with corroboration, performing similarly to lower secondary students. Findings indicate a developmental trend in evaluation skills, with improvements through adolescence but a plateau in corroboration abilities. Results suggest that while ChatGPT can assist with source evaluation, it cannot replace human judgment in online information assessments.

ABSTRACT. Large Language Models (LLMs) can rate texts in writing instruction, thereby reducing subjective bias and the workload of teachers. Benchmark ratings provide a global judgment by comparing texts to benchmarks. In this study, texts (grade 4/5) were assessed by human raters and LLMs. Preliminary results show high correlations between human and LLM scores, with humans rating slightly higher. This suggests that LLMs have the potential to augment text assessment in education and research.

ABSTRACT. Secondary and tertiary students often do not possess sufficient Information Literacy (IL) skills in the online context, creating the need to identify particularly effective interventions. Therefore, we have conducted a systematic review to examine the availability and effectiveness of IL interventions targeting online IL skills of secondary and tertiary students. This contribution will present the first results, and the coding used to identify relevant literature.

ABSTRACT. Many college students are underprepared for discipline-specific reading. To better understand students’ development of this skill, we randomly assigned two groups to read a textbook excerpt for understanding (spontaneous) or with a task model prompt. We expected that providing this prompt would increase use of discipline-specific goals and strategies. We found that both groups learned theory (initiating, primary path and outcome) but those given the prompt learned more overall and about the studies in particular.

ABSTRACT. We conducted a systematic literature review, adhering to PRISMA guidelines, and analyzed 34 studies to synthesize empirical evidence on disciplinary literacy and effective strategies for teaching reading to learn within disciplines. The results underscore the effectiveness of interventions aimed at fostering disciplinary reading, writing, reasoning, and vocabulary acquisition, with differences across school grades. The critical role of teachers’ professional development is discussed. Future research should explore the potential of digital resources in enhancing disciplinary literacy.

ABSTRACT. This systematic review and meta-analysis synthesizes findings from 20 studies, focusing on ten that quantitatively assess reading comprehension interventions for poor comprehenders. Results reveal moderate positive effects among poor comprehenders in pre-post and pre-follow-up comparisons. Test type and delivery agent factors influence effects: standardized tests reduce effect sizes while teacher-led interventions increase them. These findings highlight how teacher involvement and assessment methods affect intervention effectiveness, offering insights for evidence-based educational practices.

ABSTRACT. Generalizations are a prerequisite for cognition and communication in the context of academic language functions. By focusing on significant linguistic generalization processes and defining three linguistic dimensions as the basis for assessing generalization competence, this study aimed to answer the question of how the comprehension of generalizations can be measured in primary school children and how to use the outcomes for data-based decision-making. Six task formats were evaluated for internal consistency and construct validity (n=261).

ABSTRACT. In this study, we applied Item Response Theory (IRT) to examine the psychometric properties of an Italian battery of reading comprehension tasks for primary and middle school students. Each grade is assessed with separate tasks, and IRT enables comparisons across these tasks. After identifying the best-fitting IRT model, we examined the Test Information Function and item discrimination parameters to highlight the additional insights these metrics offer into the tasks’ functioning.

ABSTRACT. The study examined local coherence processing in German elementary school children using a timed and untimed auditory task. Results supported the cumulative complexity approach: positive-causal coherent relations were the easiest, while negative-causal incoherent items were the most difficult. In the timed condition, all items were more difficult, and the difference in difficulty between items was larger. Findings suggest that timed testing may enhance the differentiation between ability levels, improving the assessment of listening component skills.

ABSTRACT. Standards of coherence are criteria of comprehension that readers implicitly or explicitly adopt. Despite its importance for reading comprehension, almost no measure for standards of coherence has been available. Thus, we developed and tested a measure for reader-based dimensions of standards of coherence . The measure includes five factors: (1) reading attitudes and habits, (2) goal setting and monitoring, (3) coherence building strategies, (4) metacognition strategies, and (5) reading self-efficacy.

ABSTRACT. With the advent of large language models (LLMs), a major concern has become how students can be taught artificial intelligence (AI) literacy. Arguably, a necessary starting point for teaching AI literacy is knowledge about students’ pre-existing understanding of how LLMs work and the textual output they generate. In this study, we therefore asked high-school students (n = 267) broad open-ended questions about these issues and used thematic analysis in qualitatively exploring their conceptions and misconceptions.

ABSTRACT. This study examines how ChatGPT instruction affects reading comprehension, knowledge application, and user experience in struggling students. An intervention with 180 students compares guided instruction on AI use versus self-directed learning. The impact on comprehension, critical prompting, and user experience is assessed through pre-, post-, and follow-up tests. Findings will provide insights into AI-assisted learning, its benefits for weaker students, and how educators can integrate ChatGPT effectively into teaching practices.

ABSTRACT. The rapid advancement of AI in modern society underscores the urgent need to foster AI literacy among students and educators. Developing this literacy is essential to ensuring that individuals can critically, responsibly, and ethically engage with AI technologies. Despite ongoing efforts, misconceptions, uncertainties, and vague fears about AI persist among both educators and students. This study leverages a Large Language Model (LLM) chatbot as an educational tool to identify and debunk common misconceptions about AI.

ABSTRACT. STEM teams drive innovation, yet the underrepresentation of women and underrepresented racial minorities (URMs) limits equitable participation. This study proposes AI-driven interventions to foster inclusivity in STEM collaboration. AI teammates will provide real-time interventions to mitigate exclusionary behaviors and enhance engagement by analyzing dynamic communication patterns and developing inclusion analytics. This research advances AI in education, team dynamics, and equity in STEM, offering data-driven insights to create more inclusive collaborative learning environments.

ABSTRACT. Comprehending texts requires readers to build a coherent representation of the text by making global inferences and integrating textual information with one’s prior knowledge. The Deep Cloze test, validated among adults, is developed to measure these skills. The aim of the ongoing study is to assess convergent, concurrent, and discrimination validity of the test adapted for Finnish sixth graders.

ABSTRACT. This study examined predictors of reading comprehension (RC) gains in bilingual minority language children (BLMC) learning Italian and monolingual peers (7-10 years). BLMC showed lower vocabulary, decoding, and RC, but similar developmental patterns to monolinguals. Hierarchical regression revealed vocabulary and working memory as key predictors of RC gains, while decoding was not significant. Findings support the Simple View of Reading, emphasizing the importance of linguistic and cognitive skills in literacy interventions for diverse learners.

ABSTRACT. Text-belief consistency effects, reflecting readers' better understanding of information aligned with their beliefs, have been established in both L1 and L2 reading contexts. Two experiments extended this effect to bilingual reading settings by examining Iranian and German students reading conflicting texts in their L1s and their L2 English. Across both samples, the effect emerged consistently, regardless of the language. These findings highlight the robustness of belief-driven validation when reading in bilingual settings like the Internet.

ABSTRACT. In the present study, recent upper secondary graduates read four texts on a controversial topic to examine their ability to overcome the text-belief consistency effect. Eye tracking assessed immediate and delayed text processing, while an essay task measured their mental representation. Results showed no significant differences in processing based on belief consistency, and essays reflected a balanced approach. While findings suggest they may overcome this effect, study limitations must be addressed before drawing firm conclusions.

ABSTRACT. The purpose of this study is to establish a baseline understanding for how an adult reader navigates a parallel version of a narrative text. In Study 1, the original and simplified versions that make up the parallel text will be separated and tested. In Study 2, eye-tracking technology and a self-report measure will be used to determine the strategies adult readers use when reading a parallel text.

ABSTRACT. We examined how readers' topic interest influences text comprehension. Fifty-three elementary-school children rated their interest in twelve topics, then read and answered questions about texts on their three most-interesting and three least-interesting topics while eye movements were recorded during reading. Topic interest improved comprehension, particularly for deeper questions across all readers and superficial questions among less-skilled comprehenders. Less-skilled comprehenders spent more time on most-interesting than least-interesting texts, whereas skilled comprehenders showed no differences during reading.

ABSTRACT. This contribution investigates how presentation format (paging vs. scrolling) in the digital medium influences reading comprehension, reading time, and eye movements in young readers. Additionally, general reading abilities were assessed and included in the analyses. While no differences in reading comprehension were found between conditions, reading times were longer in the scrolling condition. Implications for reading and learning from screens will be discussed.

ABSTRACT. We investigated the transfer effects of a general rational thinking training on university students’ ability to evaluate arguments using a pre-post between-subjects design (N = 149). Students showed comparable improvement in assessing the (im)plausibility of scientific arguments following the rational thinking training and a general reading strategy training. These findings extend both research on the trainability of argument evaluation and on the relationship between general rational thinking abilities and epistemic components of scientific literacy.

ABSTRACT. This study examines how, according to Relational Frame Theory (RFT), secondary school students’ reading comprehension can be improved using Derived Relational Responding (DRR)-based interventions. We present preliminary data on text comprehension of two groups of Italian students who took part, respectively, in SMART Training (Strengthening Mental Abilities with Relational Training) compared with the Scratch active control group. Text comprehension was assessed pre- and post-intervention using INVALSI text excerpts.

ABSTRACT. The aim of this study is to validate the effectiveness of a digital intervention to improve sourcing and intertextual integration skills in a group of 144 lower secondary school students. The results of linear mixed models examined the effects of fixed factors (experimental condition) and random factors (class group) on the argumentative quality of an essay and the mention of reliable sources, confirming the effectiveness of the intervention.

ABSTRACT. This secondary analysis investigated the role of six cognitive, motivational, and social factors on students’ learning gains from sourcing and corroboration training. Reanalysis of two intervention studies indicated that pretest skills predicted sourcing improvements, reading proficiency predicted success in researcher-led workshops, and SES predicted enhancements in teacher-led lesson series. Findings highlight the importance of understanding how individual differences shape training success. Such insights may guide the development of training programs tailored for diverse students’ needs.

ABSTRACT. The present research presents a systematic literature review on the understanding of verbal irony in typically developing children. It resulted that the two most studied issues concern the cognitive ability to detect an incongruity between what is said and the referred situation, and the sensitivity to pragmatic aspects of ironic communication. Little or no attention has been paid to bilingualism and the emotional attribution involved in children’s understanding of irony, which therefore deserve further investigation.

ABSTRACT. Little is known about semantic lexical and grammatical aspect contents understood in sentences during processing and then integrated into the mental model. Apparently, inherent boundedness plays a particularly important role. Using a truth value judgment task, we show that participants appear to truly understand the nonprototypical semantic meanings of lexical Accomplishments and grammatical Imperfectives rather than simply following distributional patterns. Semantic results will also be compared with a second experiment into related pragmatic aspectual differences.

ABSTRACT. The present study addressed the effects of negative and positive decorative images and the role of participants’ working memory in online video comprehension, in an experimental e-learning setting. Online expository comprehension was similar in videos without images and with positive images, against both the seductive and the emotional design effects. However, negatively valenced images led to worse comprehension than positive ones. Working memory influenced comprehension, but did not moderate the effect of images’ valence.

ABSTRACT. This study explored how the medium (paper vs. tablet) and map presence affected children’s performance, mental effort, and feeling of disorientation during an information search task. Self-efficacy towards literacy was also measured. The map improved performance in the tablet condition but increased mental effort and hindered performance in the paper condition. Self-efficacy positively impacted performance and mental effort. These findings highlight the importance of considering medium and maps in educational task design for young children.

ABSTRACT. In this ongoing study, we examine the extent to which different features of competing explanations affect readers’ evaluations of each explanation. In this study, we present readers with a series of disagreements in explanations for some data related to cancer and treatment. The explanations vary in their simplicity and their sources’ expertise. Participants judge the probability of each explanation, the likely cause of disagreement between sources, and their own curiosity about the topic.