Digital Prompting in Education: A Design Framework and Bibliometric Analysis

Digital prompts are a widely used instructional method to support student learning. However, the diversity of research topics and the variety of ways in which prompts are designed and labeled result in terminological inconsistencies (e.g., jingle-jangle fallacies) and make it difficult to integrate similar research. Additionally, the rise of artificial intelligence has introduced new meanings to the term “prompt,” threatening the discoverability of research on digital prompting interventions. An updated analysis of recent records showed that in 2025, 19% of all publications using the term “prompt” referred to AI-related contexts. Therefore, this bibliometric analysis of 1,238 publications provides a systematic overview of how digital prompts are labeled and how well the research field is integrated. We identified 283 distinct prompt labels, either referring to the content of the prompt or its presentation format. Our findings highlight that a few prompt types such as metacognitive prompts, self-explanation prompts, and question prompts dominate the field, while the majority is used inconsistently or idiosyncratically. Network analyses revealed limited integration across the field, with metacognitive prompting emerging as a central research area. We provide concrete recommendations to improve the visibility and integration of this research field, including the use of standardized terminology in metadata, multi-level labeling practices, and engagement with conceptually similar prompting interventions.

Prompting is a prominent type of instructional intervention used to support student learning. Prompts are brief hints, questions, or requests that are integrated into the learning context to help students recall and execute their knowledge, strategies, and skills. An advanced way to deliver prompts is through digital technology. Commentators have noted the great variety in the way prompts are designed and labeled, but a comprehensive overview and understanding of the diversity of the field remains limited. To address this issue, the present study aims to map the research field on digital prompting interventions, with a focus on providing an overview of how the term prompting is used in the context of digital (educational) interventions and how the field is integrated.

Table of Contents: AI in LENS - Search results on artificial intelligence–related papers retrieved from LENS.org. analysis code - Scripts and code used for data cleaning, analysis, and modeling. dataset - Final curated bibliographic dataset used in the analyses. figures - Generated plots and visualizations used in the paper and supplementary materials. networks - Derived network data (e.g., citation or co-occurrence networks) used for analysis. paper identification workflow - Materials documenting the paper identification and screening process. prompt_types - Data and materials related to the classification of different prompt types. search queries - Search strings and query documentation used for database retrieval. topic_modeling - Inputs, outputs, and scripts related to topic modeling analyses. VOSviewer - Citation network and mapping files prepared for use with VOSviewer.