Search

Text comprehension is an essential skill in today’s information-rich world, and self-explanation practice helps students improve their understanding of complex texts. This study was centered on leveraging open-source Large Language Models (LLMs), specifically FLAN-T5, to automatically assess the comprehension strategies employed by readers while understanding Science, Technology, Engineering, and Mathematics (STEM) texts. The experiments relied on a corpus of three datasets (N = 11,833) with self-explanations annotated on 4 dimensions: 3 comprehension strategies (i.e., bridging, elaboration, and paraphrasing) and overall quality. Besides FLAN-T5, we also considered GPT3.5-turbo to establish a stronger baseline. Our experiments indicated that the performance improved with fine-tuning, having a larger LLM model, and providing examples via the prompt. Our best model considered a pretrained FLAN-T5 XXL model and obtained a weighted F1-score of 0.721, surpassing the 0.699 F1-score previously obtained using smaller models (i.e., RoBERTa).

Academic discourse communities and learning circles are characterized by collaboration, sharing commonalities in terms of social interactions and language. The discourse of these communities is composed of jargon, common terminologies, and similarities in how they construe and communicate meaning. This study examines the extent to which discourse reveals “shared language” among its participants that can promote inclusion or affinity. Shared language is characterized in terms of linguistic features and lexical, syntactical, and semantic similarities. We leverage a multi-method approach, including (1) feature engineering using state-of-the-art natural language processing techniques to select the most appropriate features, (2) the bag-of-words classification model to predict linguistic similarity, (3) explainable AI using the local interpretable model-agnostic explanations to explain the model, and (4) a two-step cluster analysis to extract innate groupings between linguistic similarity and emotion. We found that linguistic similarity within and between the threaded discussions was significantly varied, revealing the dynamic and unconstrained nature of the discourse. Further, word choice moderately predicted linguistic similarity between posts within threaded discussions (accuracy = 0.73; F1-score = 0.67), revealing that discourse participants’ lexical choices effectively discriminate between posts in terms of similarity. Lastly, cluster analysis reveals profiles that are distinctly characterized in terms of linguistic similarity, trust, and affect. Our findings demonstrate the potential role of linguistic similarity in supporting social cohesion and affinity within online discourse communities.

Literacy is a critically important and contemporary issue for educators, scientists, and politicians. Efforts to overcome the challenges associated with illiteracy, and the subsequent development of literate societies, are closely related to those of poverty reduction and sustainable human development. In this paper, the authors examine literacy from the lens of text and discourse theorists who focus on the higher-order comprehension processes involved in literacy. Discourse processing models make the assumption that comprehension emerges from the construction of a mental model of the text, which relies on the reader generating inferences to connect ideas within the text and to what the reader already knows. The article provides a broad overview of the theoretical models that drive research on text comprehension and production, as well as how this research shapes literacy instruction and effective interventions. The authors focus on two interventions with proven success in improving deep comprehension and writing, iSTART and the Writing Pal. Increasing literacy across the world call for a greater focus on theory driven strategy interventions to be integrated within classrooms and community at large.

Limited health literacy is a barrier to optimal healthcare delivery and outcomes. Current measures requiring patients to self-report limitations are time-consuming and may be considered intrusive by some. This makes widespread classification of patient health literacy challenging. The objective of this study was to develop and validate "literacy profiles" as automated indicators of patients' health literacy to facilitate a non-intrusive, economic and more comprehensive characterization of health literacy among a health care delivery system's membership. To this end, three literacy profiles were generated based on natural language processing (combining computational linguistics and machine learning) using a sample of 283,216 secure messages sent from 6,941 patients to their primary care physicians. All patients were participants in Kaiser Permanente Northern California's DISTANCE Study. Performance of the three literacy profiles were compared against a gold standard of patient self-reported health literacy. Associations were analyzed between each literacy profile and patient demographics, health outcomes and healthcare utilization. T-tests were used for numeric data such as A1C, Charlson comorbidity index and healthcare utilization rates, and chi-square tests for categorical data such as sex, race, poor adherence and severe hypoglycemia. Literacy profiles varied in their test characteristics, with C-statistics ranging from 0.61-0.74. Relations between literacy profiles and health outcomes revealed patterns consistent with previous health literacy research: patients identified via literacy profiles indicative of limited health literacy: (a) were older and more likely of minority status; (b) had poorer medication adherence and glycemic control; and (c) exhibited higher rates of hypoglycemia, comorbidities and healthcare utilization. This represents the first successful attempt to employ natural language processing to estimate health literacy. Literacy profiles can offer an automated and economical way to identify patients with limited health literacy and greater vulnerability to poor health outcomes.