What do Digital Humanities Say about visualization? A Bibliometric Exploration

What do Digital Humanities Say about visualization? A Bibliometric Exploration | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article What do Digital Humanities Say about visualization? A Bibliometric Exploration Rongqian Ma, Pei-Ying Chen This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6458217/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Digital humanities, as an interdisciplinary research domain, has been significantly impacted by emerging technological innovations. In recent years, visualizations have played an increasingly prominent role in digital humanities scholarship, gradually driving the field away from its traditionally text-centric orientation. This paper examines how visual topics and themes are addressed in two leading digital humanities journals— Digital Scholarship in the Humanities and Digital Humanities Quarterly —through full-text analysis of their published articles. Using BERTopic and network analysis, we identify key visual topics in each journal and investigate how these are integrated into scholarly narratives. Our findings show that, while visual discussions are diverse and often idiosyncratic, they remain deeply embedded in digital humanities discourse. Digital Scholarship in the Humanities tends to emphasize a more methodologically oriented visual discourse, focusing on visualization as a research method or as part of broader methodological debates. In contrast, Digital Humanities Quarterly offers a more balanced integration of conceptual and methodological perspectives. Building on the notion of visualizations as inscriptions, this study demonstrates how visual elements mobilize humanities ideas and scholarship, offering a foundation for further empirical investigations into visual discourse in digital humanities. Business and commerce/Information systems and information technology Humanities/Cultural and media studies Humanities/Language and linguistics Social science/Science technology and society Digital humanities visualization BERTopic network analysis narrative function Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 INTRODUCTION Since its inception, digital humanities (DH) have evolved as an interdisciplinary research domain shaped by the constant adoption and integration of new technological innovations. Early discussions about DH’s intellectual identities often revolved around how to interpret the “digital” in relation to established humanities paradigms (Svensson, 2012 ). These debates highlighted the extent to which computational methods could—or should—transform the interpretive, critical, and contextual approaches characteristic of the humanities, influencing aspects ranging from research questions and data sources to analytical frameworks and modes of dissemination (Gold, 2012 ; Gold & Klein, 2016 , 2019 , 2023 ). In this paper, we focus on visualization as a prominent example of how these digital technologies have influenced humanities research. While the humanities scholarship has historically emphasized textual analysis and interpretation, recent scholarship underscores a growing reliance on visual methods to examine large, complex, or heterogeneous data sets (Champion, 2016 ; Münster & Terras, 2020 ). This shift from a primarily “text-heavy” orientation toward more visually oriented practices carries the potential to reshape research paradigms, scholarly workflows, and even academic culture in DH. By examining how visualization has been adopted and adapted within DH, we can shed light on how traditional interpretive conventions—such as close reading, hermeneutics, and critical theory—are being rearticulated in dialogue with computational and visual techniques. Within science and technology studies (STS), visualizations are considered as a form of inscriptions, or “immutable mobiles,” which possess optical consistency while the capacity of traveling across diverse contexts to convey and mobilize arguments, ideas, and findings (Latour, 1990 ; Roth & McGinn, 1998 ). The essence and power of inscriptions lie in their embeddedness within knowledge representation systems and the broader context of knowledge production (Caissie et al., 2017 ). Using visuals embedded in scientific discourse and scholarship, for instance, helps constructs facts and shape convincing arguments (Graves, 2014 ). Scholars have also shown that the use of graphs, such as data visualizations and statistical charts, is strongly associated with the public’s collective perception of a field’s “hardness” or “scientificity” (Cleveland, 1984 ; Smith et al., 2000 ). Visual inscriptions can also facilitate the communication of complex, multifaceted ideas and findings to a broad range of audiences, both within and beyond academia (Caissie et al., 2017 ; Myers, 1988 ). Focusing on a case from the geology discipline, Rudwick ( 1976 ) discussed a wide range of visual aids available to engage in geology research and argued for the power of a “visual language” for geological science. More specifically, several mechanisms and ways of how visuals consolidate facts and mobilize arguments are identified in research literature. Lynch ( 1990 ), for example, argued that selection and mathematization are two important mechanisms that help scientists shape their objects of knowledge. Specifically, selection involves the simplification and transformation of objects to fit scientific representation, often through diagrams that abstract and enhance details for clarity. Mathematization further integrates these visualizations into structured, mathematical forms, making the data analyzable and comparable. These STS works demonstrate the powerful yet often invisible ways that visual representations can contribute to the making of knowledge. In this paper, we adopt the STS notion of inscriptions as a theoretical lens to analyze the roles of visuals in DH scholarship. However, we will extend from this inscription perspective to explore the dynamics behind the “embeddedness” within a specific DH context, moving beyond how visuals are constructed and presented in publications to focus on how they are incorporated within scholarly narratives. We will analyze how discussions of visual representations are at interplay with other parts of narrative in the article to effectively mobilize and communicate DH knowledge. Despite numerous case studies documenting the varied applications of visual technologies in DH, relatively little is known about how collectively visualizations are incorporated into DH scholarly narratives to foster research communication. For example, which aspects of visualizations are the most transferable to DH work? How are these visual aspects applied to DH, and do DH scholars primarily use visualizations for certain purposes over others? To address this knowledge gap, we present a quantitative investigation into the representation of visualization-focused topics—or otherwise referred to as visual themes—in DH scholarship. Specifically, we examine a full-text corpus of articles from Digital Scholarship in the Humanities (DSH) and Digital Humanities Quarterly (DHQ) , two leading DH journals with longstanding traditions and broad recognition within the DH research community (Spinaci et al., 2022 ). By focusing on full-text content rather than abstracts or keywords alone, we aim to capture a richer and more nuanced picture of how visualization is discussed in DH scholarly narratives. Using this corpus, we will explore the following research questions: RQ 1: What are the major visual topics represented in DSH and DHQ? RQ 2: How are these visual topics discussed in DSH and DHQ , respectively? RQ 2.1: How are visual topics interwoven with non-visual topics? RQ 2.2: What key functions do visual themes serve to mobilize DH scholarly narratives? Using topic modeling and network analysis, our research aims to illustrate how DH discourse is dynamically constructed through the interplay of visual and non-visual topics within scholarly narratives. LITERATURE REVIEW As visual technologies become increasingly integral to DH, scholars have examined visualization from both conceptual and practical perspectives. Some studies have focused on theoretical models and principles that align visualization methods with the interpretive nature of humanities research. Early efforts in this vein include Drucker's (2011, 2018) concept of “humanistic visualization,” which distinguishes capta from data and advocates a “nonrepresentational approach” that foregrounds the interpretive complexities of humanities materials. Similarly, Manovich ( 2011 ) introduced the concept of “direct visualization,” which aims to preserve the original forms and inherent complexity of humanities data rather than create abstracted or simplified data representations. Hinrichs et al. ( 2019 ) further introduced ideas such as “sandcastling,” emphasizing that visualization should be iterative, fluid, and open to revision—qualities that align well with the evolving, interpretive character of the humanities. This conceptual discourse has diversified over time. Bradley et al. ( 2016 ), for instance, proposed the notion of “slow analytics” in the context of digital literary studies, arguing that effective visualizations should prioritize “slow and methodical interaction with texts [as] part of the workflow and sense-making process,” rather than emphasizing speed and efficiency. Likewise, Kleymann and Stange ( 2021 ) articulated four postulates toward a hermeneutic approach to visualization in literary studies, aiming to reconcile the interpretive traditions of literary scholarship with the scientific and technical processes of visualization. Among these conceptual discussions, one specific issue that has received particular attention is uncertainty. Uncertainty is defined as “the lack of certainty or doubt about the accuracy or reliability of data” (Conroy et al., 2024 ). Distinct from error, which arises from inaccuracies or mistakes in measurement, data entry, or analysis, uncertainty stems from the inherent limitations of data, methods, or contextual ambiguities. Schwandt and Wachter ( 2024 ) further argued that “uncertainty in visualization is not a flaw but a feature that enhances critical analysis.” Translating the uncertainty discourse into a DH context, scholars have discussed various potential sources of uncertainty. For example, Panagiotidou et al. ( 2023 ) argued that uncertainty may emerge from the nature of the source data (e.g., missing, imprecise, ambiguous data), the process of datafication (e.g., complexity reduction, subject manual classification), and modelling (e.g., statistical, model appropriateness). In a specific domain such as text analysis, uncertainty may arise at multiple stages—semantic, comprehension, encoding, transformation, representation, and interpretation—thus posing challenges that extend from data labeling to visualization (Haghighatkhah et al., 2022 ). Collectively, these discussions have also led to the development of visualization approaches for addressing uncertainty, such as the four postulates of “established, dedicated, improvisational, and experiential” proposed by Panagiotidou et al. (2022). Additional issues and concerns related to DH visualizations have been explored in the IEEE VIS4DH workshop—one of the key conference venues for discussions on the topic. Since its inception in 2016, publications from the workshop have revealed a clear evolution in how visualization challenges are conceptualized and addressed. These include themes such as how to enhance trust during DH processes and interpretations (Chen et al., 2019 ; Schetinger et al., 2019 ), how to rethink the notion of “scale” within DH contexts (Chen & Cole, 2021 ; Ciula et al., 2021 ), and how to reimagine visualization approaches when working with multimodality data (Brath, 2022 ; Mayr et al., 2022 ; Zhou & Forbes, 2022 ). Alongside these discussions, much of the DH visualization research also aims to develop specific techniques, applications, and tools that operationalize humanistic ideals. Existing scholarship has shown that visualization can support distant reading of large visual corpora, text materials, and cultural heritage collections (Alharbi & Laramee, 2019 ; Arnold & Tilton, 2019 ; Janicke et al., 2016). Computational techniques such as knowledge graphs have become particularly prominent for examining digital cultural objects and data (Khoo et al., 2024 ; Mayr et al., 2022 ; Zheng et al., 2024 ), owing to their capacity to reveal complex relationships in primary sources. Emerging technologies such as Virtual reality (VR) and 3D visualization tools are also increasingly applied to improve accessibility and interaction with cultural heritage materials (Xia et al., 2021; Meinecke et al., 2022 ). Parallel to these efforts, DH researchers have introduced innovative visualization techniques and tools tailored to humanities research tasks. For example, Valdivia et al. ( 2021 ) presented a Parallel Aggregated Ordered Hypergraph (PAOH) method for depicting dynamic hypergraphs, while Yan and Li ( 2023 ) developed a multi-view framework that integrates semantic aggregation and interactive text exploration to enhance historical knowledge discovery. The various case studies and developments show the great potential of leveraging visualizations to support different stages of humanities research, from exploring large data corpora to conducting interpretive analyses. Although scholars have explored the potential of using DH visualizations as the lens for synergizing theories of traditional humanities and tool-making practices of DH most visualization work in the field has centered on case studies and individual applications. As a result, despite the breadth of these innovations, a systematic investigation into the specific ways in which visualizations are employed in DH scholarship to support the construction of scholarly narratives has yet to be conducted. In this paper, we aim to bridge this gap by analyzing how visualizations are collectively applied and discussed in DH research, drawing on a bibliometric approach. Prior bibliometric analyses have revealed key trends of visual practices within DH (Leydesdorff & Salah, 2010; Münster & Terras, 2020 ). For instance, Weingart and Eichmann-Kalwara ( 2017 ) investigated DH conference abstracts from 2004 to 2015, revealing an increasing emphasis on visual methods. Benito-Santos and Sánchez ( 2020 ) subsequently employed a data-driven citation analysis of 300 publications from IEEE VIS4DH workshops, the ADHO Digital Humanities Conference, and Digital Humanities Quarterly , identifying principal authors, techniques, and data sources central to DH visualization research. Further extending this line of inquiry, Ma and Li (2022) analyzed a corpus of DH publications and highlighted the complexity of DH visualization practices, particularly the differing preferences among humanities and STEM (science, technology, engineering, and mathematics) researchers regarding both the frequency and types of visual forms. Building on these studies, our work applies computational methods to full-text articles published in leading DH journals, offering a more nuanced perspective on how visualization practices and discourses are embedded within the broader DH scholarship. DATA AND METHODS Our data sample comprises all English-language research and review articles, conference and special issues papers, and case studies published in DSH and DHQ between 2006 and 2023 (including those in supplementary issues), with full-text XMLs or HTMLs available. After excluding corrigenda, errata, and retractions, as well as verifying article types and language, 990 DSH and 579 DHQ articles remained in the sample for subsequent analysis. The complete lists of articles from DSH and DHQ analyzed in this study are available in Supplementary Materials 1 and 2. We employed BERTopic to identify topics within DSH and DHQ respectively. This method leverages the power of Bidirectional Encoder Representation from Transformers (BERT) to cluster semantically similar documents and generate coherent topic representations through a class-based variation of TF-IDF (Grootendorst, 2022 ). Compared with other topic modeling techniques (e.g., LDA, NMF, Top2Vec), BERTopic is shown to be capable of producing more distinct topics and encoding contextual information (Egger & Yu, 2022 ). To prepare the corpus for BERTopic, we extracted paragraphs and their associated section headings from the full-text HTMLs and XMLs using the “rvest” and “xml2” packages in R. Since BERT is optimized for texts of fewer than 150 words, paragraphs exceeding this limit were binned into chunks while maintaining sentence-level completeness. To ensure sufficient information was retained, chunks with fewer than 18 words were discarded (i.e., the bottom 5% for DSH and the bottom 2% for DHQ ). In total, the DSH and DHQ corpora contained 55,981 and 40,813 chunks, respectively, for topic modeling. Following BERTopic’s best practices, we precalculated embeddings using the default embedding model, all-MiniLM-L6-v2, and prevented stochastic behaviors in UMAP during dimensionality reduction. Default topic representations were enhanced by removing English stopwords and infrequent words, while keyword diversity within a topic was increased with the diversity parameter set to 0.3. All other parameters were kept at their default values, including the minimum cluster size of 10 for the HDBSCAN clustering process and the extraction of the top 10 words per topic. The retrieved topics were then categorized as either visualization-related or not. We applied regular expressions to identify topics as “visual” if any of their top 10 representation terms matched a term from a pre-defined list: diagram, graph, illustration, image, map, network, photo, simulation, and visual. To further explore the relationships between visual and non-visual topics, we built upon the topic modeling results and constructed weighted co-occurrence networks for the DSH and DHQ topics, respectively. Co-occurrences were defined at the article level: if topic A and topic B appear together in a single article, a link between these two topics is established with a weight of 1. Note that we only consider topics that co-occur with visual topics as defined above. The construction of the networks and their visualization was carried out using the “igraph” package in R. Additionally, we analyzed the headings associated with the visual topics and manually coded them into one of categories of the IMRaD structure—Introduction, Methods, Results, Discussion—to better understand the argumentative functions these visual topics serve within an article (Sollaci & Pereira, 2004 ). We also manually distinguished and annotated the nature of articles as empirical, conceptual, or methods-oriented to provide more context regarding the likely concentrations of visual topics within specific IMRaD sections. Empirical articles typically follow the standard IMRaD structure, while conceptual articles are composed of Introduction and Discussion, and methods papers include Introduction, Methods, and Discussion sections only. RESULTS 1. Topic modeling results show comparative trends and emphases in visual topics across DSH and DHQ In total, 919 were identified in DSH and 650 in DHQ (see Supplementary Materials 3 and 4). Among these topics, 21 in DSH and 26 in DHQ are visual topics. Figure 1 illustrates the temporal trends of articles containing visual topics, showing that DHQ exhibits a stronger pattern of publishing visual-related articles over time. Proportionally, 11% (n = 112) of DSH articles and 20% (n = 116) of DHQ articles contain at least one visual topic, with the highest yearly shares reaching 17.6% (2021) and 43.8% (2019) and the lowest falling to zero (2007) and 9.1% (2009), respectively. Since half of the topics appear only once or twice in the corpus, we limited our subsequent analyses to topics appearing in at least three articles. This leaves us with 416 and 309 topics, including 10 and 12 visual topics, from DSH and DHQ respectively. Table 1 and Table 2 show these major visual topics from DSH and DHQ. A comparative analysis of these two topic lists reveals convergent commitments to visualization and computational methods in DH scholarship, yet each journal demonstrates slightly different emphases (Table 1 , Table 2 ). DSH shows a more technical orientation, evident in visual topics such as “Classification & Machine Learning,” “Word-Sense Disambiguation & Linguistics,” and “Geometric & Graph-Based Analysis,” all of which underscore a strong focus on algorithmic approaches to digital scholarship. By contrast, DHQ offers a broader conceptual and theoretical scope. Its largest visual topic, “Humanistic Visualization & Graphics,” highlights an engagement with the epistemological and aesthetic dimensions of DH visualization. DHQ also addresses specialized cultural or heritage topics—ranging from “Monastic Architecture & Mapping” to “Folklore and Gameworld & Interactive Visualizations”—illustrating how digital methods can be applied to diverse historical, social, or creative contexts. Meanwhile, “Agent-based Modeling & Simulation,” which does not appear as prominently in DSH , indicates a commitment to computational experiments aimed at simulating social or cultural phenomena. Despite these differences, the two journals share important common ground in areas such as network analysis, geospatial inquiry, temporal visualization, and the study of art. Both journals emphasize the topic of “Network Analysis & Visualization” (i.e., T96 in DSH and T585 in DHQ) , reflecting a keen interest in mapping relationships—whether social, linguistic, or otherwise—through graph-based models. Geospatial research is similarly central: Topics such as “Maps & Geographic Information” are represented in DSH , while DHQ features themes such as “Monastic Architecture & Mapping.” Similar patterns also emerge in temporal analysis and art: DSH includes “Time and Spatial Visualization,” while DHQ discusses “Chronologies & Temporal Diagrams;” DSH showcases themes like “Photography”, whereas DHQ focuses on “Image Atlas & Art”. Finally, both journals consider best practices in visual scholarship, though they prioritize different dimensions. DSH features a topic on “Provenance & Transparency in Visualization,” which raises specific ethical and methodological concerns. DHQ , in comparison, addresses the theme of “Archiving” more explicitly, emphasizing a particular context of use for visualization practices. Figure 2 shows how different topics in DSH and DHQ are distributed across time. As we can observe from Fig. 2 , certain visual topics hold long-term prominence in DH scholarship. For DSH , “Time & Spatial Visualization” is one such topic, the discussion of which appeared in the journal in the 2005–2009 period and has been increasing since then until the peak during the period of 2020 and 2024. By contrast, “Word-Sense Disambiguation & Linguistics” first appeared during the 2010–2014 period, while “Classification & Machine Learning” was a relatively new visual topic adopted in the communities during 2015–2019. For DHQ , “Humanistic Visualization & Graphics” holds a similar trend. The discussion of this topic started in the early stages of the journal, from 2005 to 2009, and gained prominence in the period of 2015–2019. Afterward, there was a slow decline for its representation, but this visual theme remained as the most represented and discussed visual topics from 2020 to 2024. Interestingly, “Archiving” in DHQ also appears to be a longstanding topic among the communities, with peak representations during 2015–2019 and then a decreasing representation during 2020–2024. In summary, our results show that both journals integrate visualization discussions in their scholarship, although the topic emphases change across time, reflecting the collective interests among the DH research communities. More specifically, while DSH provides a technically oriented, methodologically detailed approach to DH visualization, DHQ integrates theoretical, conceptual, and domain-specific explorations into its visualization discourse. Table 1 Major visual topics in DSH (N = Number of articles). Topic Label Keywords N 31 Time and Spatial Visualization ['chronotopic', 'commodities', 'grid', 'wheat', 'spatiotemporal', 'visualizations' , 'geohistorical', 'gir', 'heatmap', 'timelines'] 22 12 Word-sense disambiguation & linguistics ['wsd', 'senses', 'concordancer', 'glosses', 'supervised', 'véronis', 'disambiguation', 'graph' , 'korean', 'relatedness'] 13 347 Illustration and Iconography ['reproductions', 'vision', 'pollock', 'illustrations' , 'artworks', 'motifs', 'iconographic', 'dietrich', 'postures', 'microscopy'] 11 517 Classification & Machine Learning ['burrowsian', 'nsc', 'centroid', 'classify', 'classifications', 'classifiers', 'b5', 'flavours', 'simulations' , 'mf3c'] 11 806 Photography ['cesr', 'cvma', 'slides', 'études', 'digitising', 'courtauld', 'romanesque', 'scanned', 'photographs' , 'renaissance'] 11 234 Infographics [ 'infographics' , 'mastering', 'literacy', 'kandinsky', 'graphics' , 'visualizations' , 'hackneyed', 'charter', 'visual' , 'counterfactual'] 10 11 Maps & Geographic Information ['cartographic', 'historic', 'georeferencing', 'maps' , 'geoportals', 'topographic', 'cartography', 'cadastral', 'mercator', 'geographic'] 9 10 Geometric & Graph-Based Analysis ['bisector', 'counter', 'zeta', 'centroids', 'segments', 'randomized', 'howells', 'marker', 'eyed', 'graphs' ] 8 96 Network Analysis & Visualization ['vertices', 'diameter', 'shells', 'cores', 'networks' , 'nodes', 'edges', 'neighbours', 'connectivity', 'geodesic'] 7 43 Provenance & Transparency in Visualization ['provenance', 'disclosure', 'vancisin', 'visualization' , 'reproducibility', 'artifactual', 'transparency', 'curatorial', 'participants', 'cs4'] 3 Table 2 Major visual topics in DHQ (N = number of articles). Topic Label Keywords N 1 Humanistic Visualization & Graphics ['deleuze', 'infovis', 'folding', 'visualizations' , 'folds', 'graphical', 'charts', 'interactivity', 'transitions', 'humanistic'] 43 284 Archiving ['ellen', 'archival', 'pilers', 'storing', 'photocopying', 'dropbox', 'discloses', 'photos' , 'subfolders', 'folder'] 13 461 Text Visualization ['themeriver', 'tei2html', 'reversible', 'tapor', 'visualizations' , 'textarc', 'computationalstylistics', 'darker', 'tufte', 'shading'] 11 135 Graph Layouts & Structures ['layered', 'graphs' , 'layouts', 'nodes', 'dashboard', 'topology', 'hierarchical', 'graphviz', 'mental', 'visualizations' ] 10 321 Image Atlas & Art ['warburg', 'rembrandt', 'atlas', 'heinrich', 'wooden', 'reproductions', 'artworks', 'renaissance', 'antiquity', 'photographic' ] 8 44 Chronologies & Temporal Diagrams ['chronologies', 'heterochronologies', 'timelines', 'lunar', 'leonid', 'synchronicity', 'solar', 'chronology', 'temporal', 'diagrams' ] 7 585 Network Analysis & Visualization ['gephi', 'poms', 'eigenvector', 'centrality', 'closeness', 'cliques', 'bimodal', 'weingart', 'graph' , 'cosmopolitan'] 7 61 Monastic Architecture & Mapping ['monastic', 'samos', 'polygons', 'monastery', 'monasteries', 'architectural', 'buildings', 'maps' , 'village', 'romanesque'] 4 14 Gameworld & Interactive Visualizations ['verbum', 'visuality' , 'wallpaper', 'rooms', 'npcs', 'plotkin', 'exits', 'gameworld', 'commands', 'puzzles'] 3 150 Folklore ['folktale', 'folktales', 'dutch', 'folklore', 'tales', 'visualises' , 'mock', 'facets', 'visitors', 'fairy'] 3 257 Agent-based Modeling & Simulation ['abm', 'simulations', 'agents', 'ideation', 'artificial', 'modeling', 'behaviors', 'archeological', 'simulated' , 'autopilot'] 3 516 Text Mining & Networks ['bigram', 'skipgrams', 'preprocessing', 'mining', 'edges', 'visualisations' , 'summarises', 'vertices', 'networks' , 'contextually'] 3 2. Topic co-occurrence networks reveal the embeddedness of visual topics in DSH and DHQ The topic co-occurrence networks of DSH and DHQ , as shown in Fig. 3 , further reveal how visual themes are integrated into DH scholarship. The complete DSH network comprises 241 nodes (topics) and 1,272 links (Fig. 3 A), 91.2% of which appear only once. After filtering out the one-off links, 69 nodes (28.6%) and 112 links (8.8%) remain (Fig. 3 B). Similarly, the complete DHQ network comprises 284 nodes and 1,877 inks (Fig. 3 C); after filtering out one-off links (94.1%), 80 nodes (28.2%) and 111 links (5.9%) remain (Fig. 3 D). This result aligns with our previous observation that the visual topic discourse in DH scholarship is diverse and dispersed, with a large proportion of topics being niche, unique, and specialized and rarely discussed in combination with other themes. From a broader perspective, our results also correspond to the commonly shared perception of DH being a highly heterogeneous and “amorphous” field (Deegan, 2014 ; Luhmann & Burghardt, 2022 ). Our analysis also reveals that visual topics are embedded to varying degrees within the broader subject landscapes of DSH and DHQ . A comparison of the co-occurrence network modularity—using the built-in function in the “igraph” package that identifies the optimal community structure based on maximal modularity scores—shows that DSH exhibits a clearer community structure than DHQ. Specifically, Fig. 3 B yields 8 clusters with a modularity of 0.70, whereas Fig. 3 D returns 13 clusters with a modularity of 0.56. In DSH , visual topics tend to connect with non-visual topics and form small, interconnected clusters. For example, the largest visual topic, “Time & Spatial Visualization” (T31), co-occurs with non-visual topics such as “Metadata and Postcolonial Curation” (T425), “Spatial Humanities and Gazetteers” (T55), and “Digital Humanities as a Field” (T41), while co-occurring with only one visual topic, “Infographics” (T234). Similarly, “Word-Sense Disambiguation & Linguistics” (T12) and “Classification & Machine Learning” (T517) co-occur exclusively with a group of non-visual topics. We can also observe from Fig. 3 B that some non-visual topics in the DSH network play a crucial role not only in terms of their degrees but also as bridge nodes connecting clusters. For instance, among non-visual topics, “Gephi” (T145) ranks highest in terms of betweenness in the DSH subnetwork, serving as a bridge node between two communities (Figs. 3 B and Fig. 4 A). The non-visual topic “HCI and usability” (T603) follows T145, connecting three communities (Figs. 3 B and 4 A). A complete list of DSH node centralities is available in Supplementary Material 5. In DHQ , visual topics are more seamlessly integrated into the broader DH discourse, co-occurring with both visual and non-visual topics and forming multiple interconnected groups of nodes. “Humanistic Visualization & Graphics” (T1) appears to be the most central node in Fig. 3 D. It widely connects with a variety of visual and non-visual topics. For example, it co-occurs with the visual themes of “Text Visualization” (T461) and “Archiving” (T284). Among the non-visual topics, themes in the DHQ subnetwork such as “Digital Tool Building & Criticism” (T7) and “Humanities Landscape” (T0) are the most predominantly represented, followed by high-degree topics such as “Digital Practices & Infrastructures” (T66), “Archival Profession” (T171), “LDA” (T11), and “Debates in Digital Humanities” (T87). These topics range from more technical ones (e.g., LDA) to more conceptual ones (e.g., Debates in Digital Humanities), indicating that “Humanistic Visualization & Graphics” has been discussed across diverse contexts. A complete list of DHQ node centralities is available in Supplementary Material 6. Synthesizing the comparisons above, the co-occurrence networks of DSH and DHQ reveal slightly different approaches to incorporating visualization knowledge into DH scholarship across these two leading journals. The visual topics in DSH represent a strong methodological emphasis, and these various visual methods are often applied to niche and specific subjects and research areas. This may explain the more prominent community structure in the DSH co-occurrence network. DHQ , however, places a stronger emphasis on conceptual discussions of visual topics, with “Humanistic Visualization & Graphics” occupying a central position in the network and connecting to a wide array of both visual and non-visual topics. By contrast, most other visual topics—such as Graph Layouts & Structures (T135), Network Analysis & Visualization (T585), and Chronologies & Temporal Diagrams(T44) appear in more peripheral positions within the network. 3. Narrative Functions of Visual Topics We identified 75 DSH articles and 75 DHQ articles that include section headings associated with visual topics. A manual coding of their article types as well as the associated headings reveals the narrative functions of visualization in DH scholarship. Two authors independently mapped the headings to the IMRaD structure and categorized the articles into three types: Conceptual, Methodology, and Empirical. For DSH , the IMRaD classification produced a Cohen’s (1960) kappa of 0.67, indicating substantial agreement, while the DHQ IMRaD classification yielded a slightly higher kappa of 0.69. For article type classification, the kappa was 0.58 for DSH and 0.59 for DHQ , reflecting moderate agreement. Our detailed mapping results are also available in Supplementary Materials 7 and 8. In DSH , most articles related to visual themes (70.7%, n = 53) are classified as Methods, while Conceptual (16%, n = 12) and Empirical (13.3%, n = 10) papers together make up less than one-third. By contrast, DHQ exhibits a more balanced distribution, with Methods (42.7%, n = 32) and Conceptual (40%, n = 30) constituting the majority, and Empirical (17.3%, n = 13) accounting for the remainder. Tables 3 and 4 illustrate the narrative functions of visual topics in both journals. In DSH , 39% of visual-topic headings map to “M,” emphasizing their prominent role in illustrating methods. Such headings appear more frequently in Methods paper (47.9%) than in Empirical ones (36.8%). The second most common function of visual themes is supporting the introduction (29.3%), with most instances in Conceptual (38.5%) and Methods (30.3%) papers. In DHQ , however, the most frequent function of visual topics is supporting the introduction (37%) across all article types, followed by illustrating methods (24%). Overall, these findings suggest that both journals feature relatively few empirical studies on visualization. Comparatively, DSH favors a more methodologically oriented style, with a high proportion of Methods papers and a predominant use of visualizations to illustrate methods, while DHQ is more balanced with methodological detail and conceptual discourse. Table 3 Distribution of IMRaD headings associated with visual topics across Article Type in DSH . IMRaD Conceptual Paper Empirical Paper Methods Paper Total I 10 (38.5%) 2 (10.5%) 36 (30.3%) 48 (29.3%) M – 7 (36.8%) 57 (47.9%) 64 (39.0%) R – 8 (42.1%) – 8 (4.9%) D 16 (61.5%) 2 (10.5%) 26 (21.8%) 44 (26.8%) Total 26 19 119 164 Table 4 Distribution of IMRaD headings associated with visual topics across Article Type in DHQ . IMRaD Conceptual Paper Empirical Paper Methods Paper Total I 30 (48.4%) 7 (25.9%) 17 (29.8%) 54 (37.0%) M – 5 (18.5%) 30 (52.6%) 35 (24.0%) R – 10 (37.0%) – 10 (6.8%) D 32 (51.6%) 5 (18.5%) 10 (17.5%) 47 (32.2%) Total 62 27 57 146 Figure 5 further illustrate how specific visual topics are distributed across the IMRaD structure in DSH and DHQ . As shown in Fig. 5 A, most of the visual topics are concentrated in the Introduction and Methods sections, suggesting their primary narrative function of supporting conceptual and methodological design of research in DSH. This is consistent with our earlier finding about the journal’s methodological orientation and emphasis. In addition, some visual topics are well represented in the Discussion section, such as “Illustration and Iconography” (T347) and “Infographics” (T234). In terms of the distribution of narrative functions, four topics in DSH —"Network Analysis & Visualization” (T96), “Maps & Geographic Information” (T11), “Infographics” (T234), and “Time and Spatial Visualization” (T31)—appear in all four sections. In particular, “Network Analysis & Visualization” shows a balanced presence across the IMRaD structure, suggesting that it is commonly used for empirical research and for highlighting study results. All other visual topics in DSH are also fairly well integrated into the IMRaD structure, except in the Results section, indicating that visual topics are holistically interwoven in the DH scholarly narrative. Compared to DSH, DHQ demonstrates a more balanced distribution of the narrative roles that visual topics serve (Fig. 5 B). “Humanistic Visualization & Graphics” (T1) emerges as the most predominant visual topic in DHQ , deeply embedded across the IMRaD structure. Similarly, “Archiving” (T284), “Graph Layouts & Structures” (T135), and “Text Mining & Networks” (T516) also appear in all four sections. Among the remaining topics, “Images Atlas & Art” (T321) and “Gameworld & Interactive Visualizations” (T14) are the only ones that lack representation in the Methods section, which may be attributed to their strong association with conceptual papers in the journal. DISCUSSION AND FUTURE WORK The comparative analysis of DSH and DHQ reveals both shared patterns and distinct emphases in how visual discourse is represented in DH scholarship. Both journals demonstrate that visual discourse in DH tends to be diverse yet niche, encompassing a wide range of specific discussions, themes, with each taking up relatively small narrative space within only a few articles. Despite being limited, the visual topic overlap between the two journals—primarily in areas such as network analysis and temporal or spatial analysis—suggests a few popular visual themes in DH scholarship. Empirical studies that employ visualization to present analysis and research findings remain a small portion of the publications in both journals, which suggests the lack of this genre of scholarship in DH research. At the same time, the journals exhibit important differences in their treatment of visual discourse. DSH in general has a stronger methodological orientation to visual discussions, while DHQ presents a more balanced approach incorporating both conceptual and methodological discussions of visual themes. A close examination of the two journals suggests that this practice is aligned with the editorial conventions, histories, and preferences of each journal: DSH evolved from Literary and Linguistic Computing (LLC , 1986–2015), a journal that originally focused on applying computing techniques to literary and linguistic research, particularly in text analysis and corpus linguistics. Over time, LLC expanded its scope to cover broader digital methodologies in the humanities, leading to its rebranding as DSH in 2015. This editorial trajectory helps explain DSH ’s continued emphasis on computational methods and data-driven applications in visual scholarship. By contrast, DHQ , which was launched in 2007 as an open-access, web-native publication, was designed to serve as a flexible, community-driven venue for DH research, theory, and practice. Closely affiliated with the Alliance of Digital Humanities Organizations (ADHO), DHQ embraces a broader range of contributions and has consistently prioritized conceptual and interpretive engagement with digital tools. This orientation is reflected in its more diverse and balanced visual discourse. A further look into the most well-represented visual topic, “Humanistic Visualization & Graphics” (T1), also supports this interpretation. This topic appears in 43 articles published between 2008 and 2023 and consistently engages with questions about what constitutes ideal visualizations for humanities and cultural data. For example, Drucker’s ( 2011 ) essay “Humanities Approaches to Graphical Display” proposed an emphasis on capta and the idea to adopt a more interpretive and constructivist view of humanistic visualizations. This was also followed by discussions such as “generous interfaces” incorporating rich, complex, and critical visual approaches to represent cultural heritage materials digitally (Whitelaw, 2015 ). More recent articles have also extended these discussions by developing more specific and creative conceptual models, such as “the fold” to rethink and critique interactivity in data visualization (Brüggemann et al., 2020 ), a “hermeneutic visualization” for literary studies (Kleymann & Stange, 2021 ), and how to leverage a sociological approach to critical design for humanities visualizations (Forberg, 2022 ). These works showcase a continuous interest in conceptual engagements with humanities visual design and analysis in DHQ and the DH communities involved in it. This stream of discussions and publications have likely contributed to the more balanced visual discourse presented in DHQ. While our findings offer valuable insights, we acknowledge limitations of our study. For example, given the analytical focus on DSH and DHQ , the results may primarily reflect the editorial conventions and priorities of these two journals rather than the broader landscape of DH scholarship. We do not claim that the patterns observed in the present study are fully generalizable to all DH journals; rather, we see these findings as a foundation for future empirical studies of visual rhetoric in DH. Furthermore, by focusing exclusively on English-language publications, this study may not fully capture the multilingual and culturally diverse realities of the global DH community. Finally, we recognize that valuable DH work appears in venues well beyond journal publications. Books, conference proceedings, extended abstracts, and presentations also shape the intellectual landscape of the field and can offer valuable insights. We plan to incorporate such materials in future iterations of this research. In future research, we aim to expand the scope of our analysis in multiple ways. First, we plan to expand the dataset to include a broader array of multilingual DH journals, as well as additional sources such as conference proceedings and books. This effort will enable us to construct a more comprehensive representation of visual discourse and languages in the field. The ongoing development of the Index of DH Conferences ( https://dh-abstracts.library.virginia.edu/ ) presents a valuable opportunity to enhance the coverage of our corpus. Meanwhile, other ongoing efforts—such as the curation of a public dataset on DH visual rhetoric (Ma et al., 2025 ) and the development of comprehensive, reusable book metadata from the Library of Congress and the British Library (Li et al., 2025 )—highlight the feasibility and promise of this broader direction. These initiatives will collectively lay the groundwork for extending and enriching the present study. Additionally, we intend to refine our analysis of co-occurrence networks by further defining and categorizing the nature of connections between topics. Rather than treating all co-occurrences as equivalent, we aim to distinguish between different types of relationships among visual and non-visual topics—for example, whether they serve complementary, contrasting, supporting, or hierarchical functions in the overall articles and discourse. This deeper semantic and structural differentiation will allow us to capture more nuanced patterns of association and provide a richer understanding of how visual and thematic elements interact within DH scholarship. We will also refine our analysis by constructing time-focused networks to examine how visual discourse evolves over time. Furthermore, by incorporating edge-weighting strategies and contextual metadata, we can begin to explore the strength, frequency, and directionality of topic relationships, which may shed light on underlying rhetorical or epistemic structures in the field. Finally, we will enhance our network analysis by examining the structural placement of visual discussions within DH scholarship, analyzing how visual themes function across different sections of research articles. Such extended analysis will offer a more nuanced understanding of how visual elements are embedded in DH research narratives and how they shape scholarly communication in the field. Building upon the visual-focused case study, we hope our work may serve as a model for empirically investigating the broader implications of technological integration in DH scholarship. Declarations Ethical Approval Ethical approval is not required for this study. Informed Consent Informed consent is not required for this study. Artificial Intelligence (AI) usage The authors used ChatGPT for grammar and stylistics checks. All the content was carefully reviewed, and the authors take full responsibility of the content in this paper. Competing Interests One of the authors, Rongqian Ma, is a member of the Editorial Board Member of Humanities and Social Sciences Communications. Author Contribution Author Contributions: Rongqian Ma—Conceptualization, Methodology, Investigation, Resources, Data Curation, Writing - Original Draft, Writing - Review & Editing, Funding Acquisition, Project Administration; Pei-Ying Chen—Conceptualization, Methodology, Investigation, Formal Analysis, Software, Visualization, Data Curation, Writing - Original Draft, Writing - Review & Editing Data Availability The original full-text corpus analyzed during the current study are available from the corresponding author on reasonable request given copyright restrictions. Other datasets generated and/or analyzed during the current study are available in the supplementary files. References Alharbi M, Laramee RS (2019) SoS TextVis: An Extended Survey of Surveys on Text Visualization. Computers 8(1). Article 1. https://doi.org/10.3390/computers8010017 Arnold T, Tilton L (2019) Distant viewing: Analyzing large visual corpora. Digit Scholarsh Humanit 34(Supplement1):i3–i16. https://doi.org/10.1093/llc/fqz013 Benito-Santos A, Sánchez RT (2020) A Data-Driven Introduction to Authors, Readings, and Techniques in Visualization for the Digital Humanities. IEEE Comput Graph Appl 40(3):45–57. https://doi.org/10.1109/MCG.2020.2973945 Bradley AJ, Mehta H, Hancock M, Collins C (2016) Visualization, Digital Humanities, and the Problem of Instrumentalism. IEEE VIS4DH 1st Workshop on Visualization for the Digital Humanities Brath R (2022) Multimodal analogs to infer humanities visualization requirements (arXiv:2209.15029). arXiv. https://doi.org/10.48550/arXiv.2209.15029 Brüggemann V, Bludau M-J, Dörk M (2020) The Fold: Rethinking Interactivity in Data Visualization. Digit Humanit Q, 14 (3). https://digitalhumanities.org/dhq/vol/14/3/000487/000487.html Caissie LT, Goggin C, Best LA (2017) Graphs, Tables, and Scientific Illustrations: Visualisation as the Science of Seeing Gerontology. Can J Aging / La Revue Canadienne Du Vieillissement 36(4):536–548. https://doi.org/10.1017/S0714980817000447 Champion EM (2016) Digital humanities is text heavy, visualization light, and simulation poor. Digit Scholarsh Humanit fqw053. https://doi.org/10.1093/llc/fqw053 Chen AT, Cole CL (2021) Reflexivity in Issues of Scale and Representation in a Digital Humanities Project. IEEE VIS4DH 6th Workshop on Visualization for the Digital Humanities. https://arxiv.org/abs/2109.14184 Chen AT, Raza M, Gazula Y, Huang D, Mendoza J, Andrews WG (2019) Grounding Users in Interpretive Acts: Lessons Learned in the Iterative Design of a Digital Collection. IEEE VIS4DH 4th Workshop on Visualization for the Digital Humanities. https://vis4dh.dbvis.de/2019/papers/2019/VIS4DH2019_paper_6.pdf Ciula A, Vieira M, Ferraro G, Ong T, Perovic S, Mucignat R, Valmori N, Deseure B, Mannucci EJ (2021) Small Data and Process in Data Visualization: The Radical Translations Case Study IEEE VIS4DH 6th Workshop on Visualization for the Digital Humanities . arXiv: https://doi.org/10.48550/arXiv.2110.09349 Cleveland WS (1984) Graphs in Scientific Publications. Am Stat 38(4):261–269. https://doi.org/10.2307/2683400 Conroy M, Gillmann C, Harvey F, Mchedlidze T, Fabrikant SI, Windhager F, Scheuermann G, Tangherlini TR, Warren CN, Weingart SB, Rehbein M, Börner K, Elo K, Jänicke S, Kerren A, Nöllenburg M, Dwyer T, Eide Ø, Kobourov S, Betz G (2024) Uncertainty in humanities network visualization. Frontiers in Communication , 8 . https://doi.org/10.3389/fcomm.2023.1305137 Cuijuan X, Lihua W, Wei L (2021) Shanghai memory as a digital humanities platform to rebuild the history of the city. Digit Scholarsh Humanit 36(4):841–857. https://doi.org/10.1093/llc/fqab023 Deegan M (2014) This ever more amorphous thing called Digital Humanities’: Whither the Humanities Project? Arts Humanit High Educ 13(1–2):24–41. https://doi.org/10.1177/1474022213513180 Drucker J (2011) Humanities Approaches to Graphical Display. Digit Humanit Q, 005 (1) Drucker J (2018) Non-representational approaches to modeling interpretation in a graphical environment. Digit Scholarsh Humanit 33(2):248–263. https://doi.org/10.1093/llc/fqx034 Egger R, Yu J (2022) A topic modeling comparison between LDA, NMF, Top2Vec, and BERTopic to demystify Twitter posts. Front Sociol 7:886498. https://doi.org/10.3389/fsoc.2022.886498 Forberg P (2022) Critical Design as Theory, Experiment, and Data: A Sociologically-Informed Approach to Visualizing Networks of Loss. Digit Humanit Q, 16 (3). https://www.digitalhumanities.org/dhq/vol/16/3/000621/000621.html Gold MK (2012) Debates in the Digital Humanities. University of Minnesota Press Gold M, Klein L (eds) (2016) Debates in the Digital Humanities. University of Minnesota Press Gold M, Klein L (eds) (2019) Debates in the Digital Humanities. Univ Of Minnesota Gold M, Klein L (eds) (2023) Debates in the Digital Humanities. University of Minnesota Press Graves H (2014) The Rhetoric of (Interdisciplinary) Science: Visuals and the Construction of Facts in Nanotechnology. Poroi 10(2). https://doi.org/10.13008/2151-2957.1207 Grootendorst M (2022) BERTopic: Neural topic modeling with a class-based TF-IDF procedure. arXiv preprint arXiv:2203.05794 . https://doi.org/10.48550/arXiv.2203.05794 Haghighatkhah P, El-Assady M, Fekete J-D, Mahyar N, Paradis C, Simaki V, Speckmann B (2022) Characterizing Uncertainty in the Visual Text Analysis Pipeline. IEEE 7th Workshop on Visualization for the Digital Humanities (VIS4DH). https://doi.org/10.1109/VIS4DH57440.2022.00010 Hinrichs U, Forlini S, Moynihan B (2019) In defense of sandcastles: Research thinking through visualization in digital humanities. Digit Scholarsh Humanit 34(Supplement1):i80–i99. https://doi.org/10.1093/llc/fqy051 Jänicke S, Franzini G, Cheema MF, Scheuermann G (2017) Visual Text Analysis in Digital Humanities. Comput Graphics Forum 36:226–250. https://doi.org/10.1111/cgf.12873 Khoo CSG, Tan EAL, Ng S-G, Chan C-F, Stanley-Baker M, Cheng W-N (2024) Knowledge Graph Visualization Interface for Digital Heritage Collections: Design Issues and Recommendations. Inform Technol Libr 43(1). Article 1. https://doi.org/10.5860/ital.v43i1.16719 Kleymann R, Stange J-E (2021) Towards Hermeneutic Visualization in Digital Literary Studies. Digit Humanit Q, 15 (2). https://www.digitalhumanities.org/dhq/vol/15/2/000547/000547.html Latour B (1990) Drawing things together. In: Lynch M, Woolgar S (eds) Representation in Scientific Practice. The MIT Press Leydesdorff L, Hammarfelt B, Salah A (2011) The structure of the Arts & Humanities Citation Index: A mapping on the basis of aggregated citations among 1,157 journals. J Am Soc Inform Sci Technol 62(12):2414–2426. https://doi.org/10.1002/asi.21636 Li K, Dobreski B, Busch M (2025) How are library cataloging metadata used differently over time: A large-scale quantitative analysis of the Library of Congress catalog. J Doc 81(2):385–402. https://doi.org/10.1108/JD-08-2024-0199 Luhmann J, Burghardt M (2022) Digital humanities—A discipline in its own right? An analysis of the role and position of digital humanities in the academic landscape. J Association Inform Sci Technol 73(2):148–171. https://doi.org/10.1002/asi.24533 Lynch M (1990) The externalized retina: Selection and mathematization in the visual documentation of objects in the life sciences. In: Lynch M, Woolgar S (eds) Representation in Scientific Practice. The MIT Press Ma R, Li K (2021) Digital humanities as a cross-disciplinary battleground: An examination of inscriptions in journal publications. J Association Inform Sci Technol 73(2). https://doi.org/10.1002/asi.24534 Ma R, Li K, Virk RS (2025) GraphEidos: A Dataset of Visual Rhetoric in Digital Humanities. https://www.researchgate.net/publication/390764252_GraphEidos_a_Dataset_of_Visual_Rhetoric_in_Digital_Humanities/references Manovich L (2011) What is visualisation? Visual Stud 26(1):36–49. https://doi.org/10.1080/1472586X.2011.548488 Mayr E, Windhager F, Liem J, Beck S, Koch S, Kusnick J, Jänicke S (2022) The Multiple Faces of Cultural Heritage: Towards an Integrated Visualization Platform for Tangible and Intangible Cultural Assets. IEEE 7th Workshop on Visualization for the Digital Humanities (VIS4DH) , 13–18. https://doi.org/10.1109/vis4dh57440.2022.00008 Meinecke C, Hall C, Jänicke S (2022) Towards Enhancing Virtual Museums by Contextualizing Art through Interactive Visualizations. J Comput Cult Herit 15(4):1–26. https://doi.org/10.1145/3527619 Münster S, Terras M (2020) The visual side of digital humanities: A survey on topics, researchers, and epistemic cultures. Digit Scholarsh Humanit 35(2):366–389. https://doi.org/10.1093/llc/fqz022 Myers G (1988) Every picture tells a story: Illustrations in E.O. Wilson’s Sociobiology. Hum Stud 11:2–3. https://doi.org/10.1007/BF00177305 Panagiotidou G, Lamqaddam H, Poblome J, Brosens K, Verbert K, Moere V, A (2023) Communicating Uncertainty in Digital Humanities Visualization Research. IEEE Trans Vis Comput Graph 29(1):635–645. https://doi.org/10.1109/TVCG.2022.3209436 Roth W-M, McGinn MK (1998) Inscriptions: Toward a Theory of Representing as Social Practice. Rev Educ Res 68(1):35–59. https://doi.org/10.2307/1170689 Rudwick MJS (1976) The Emergence of a Visual Language for Geological Science 1760–1840. Hist Sci 14(3):149–195. https://doi.org/10.1177/007327537601400301 Schetinger V, Raminger K, Filipov V, Soursos N, Zapke S, Miksch S (2019) Bridging the Gap between Visual Analytics and Digital Humanities: Beyond the Data-Users-Tasks Design Triangle. IEEE VIS4DH 4th Workshop on Visualization for the Digital Humanities. https://vis4dh.dbvis.de/2019/papers/2019/VIS4DH2019_paper_13.pdf Schwandt S, Wachter C (2024) Visualization as irritation: Producing knowledge about medieval courts through uncertainty. Frontiers in Big Data , 7 . https://doi.org/10.3389/fdata.2024.1188620 Smith LD, Best LA, Stubbs DA, Johnston J, Archibald AB (2000) Scientific Graphs and the Hierarchy of the Sciences: A Latourian Survey of Inscription Practices. Soc Stud Sci 30(1):73–94 Sollaci LB, Pereira MG (2004) The introduction, methods, results, and discussion (IMRAD) structure: A fifty-year survey. J Med Libr Association 92(3):364–371 Spinaci G, Colavizza G, Peroni S (2022) A map of Digital Humanities research across bibliographic data sources. Digit Scholarsh Humanit 37(4):1254–1268. https://doi.org/10.1093/llc/fqac016 Svensson P (2012) Beyond the Big Tent. In Gold (Ed.), Debates in the Digital Humanities (pp. 36–49). University of Minnesota Press. https://doi.org/10.5749/minnesota/9780816677948.003.0004 Valdivia P, Buono P, Plaisant C, Dufournaud N, Fekete J-D (2021) Analyzing Dynamic Hypergraphs with Parallel Aggregated Ordered Hypergraph Visualization. IEEE Trans Vis Comput Graph 27(1):1–13. https://doi.org/10.1109/TVCG.2019.2933196 Weingart SB, Eichmann-Kalwara N (2017) What’s Under the Big Tent? A Study of ADHO Conference Abstracts. Digital Studies/Le Champ Numérique , 7 (1), Article 1. https://doi.org/10.16995/dscn.284 Whitelaw M (2015) Generous Interfaces for Digital Cultural Collections. Digit Humanit Q, 9 (1). https://www.digitalhumanities.org/dhq/vol/9/1/000205/000205.html Yan C, Li S (2023) Exploring Multiview Interactive Visualization for Chinese Historical Texts Using Knowledge Aggregation. 2023 ACM/IEEE Joint Conference on Digital Libraries (JCDL) , 165–168. https://doi.org/10.1109/JCDL57899.2023.00031 Zheng X, Li M, Wan Z, Zhang Y (2024) Knowledge mining and graph visualization of ancient Chinese scientific and technological documents bibliographic summaries based on digital humanities. Libr Hi Tech 42(6):1693–1721. https://doi.org/10.1108/LHT-11-2022-0538 Zhou H, Forbes AG (2022) Data feel: Exploring visual effects in video games to support sensemaking tasks. In 2022 IEEE 7th Workshop on Visualization for the Digital Humanities (VIS4DH) (pp. 6–12). IEEE. https://doi.org/10.1109/VIS4DH57440.2022.00007 Additional Declarations Competing interest reported. One of the authors, Rongqian Ma, is a member of the Editorial Board Member of Humanities and Social Sciences Communications. Supplementary Files Supple1dsharticleinfosamplepublic.csv Supple2dhqarticleinfosamplepublic.csv Supple3dshtopicinfopublic.csv Supple4dhqtopicinfopublic.csv Supple5dshnodecentralities.csv Supple6dhqnodecentralities.csv Supple7dshirrpublic.csv Supple8dhqirrpublic.csv Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6458217","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":446548446,"identity":"b1dbfbdc-456c-4cae-bbe5-c96e65a0c13d","order_by":0,"name":"Rongqian Ma","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAj0lEQVRIiWNgGAWjYFAC/ocPPhiAGAlEa+FhNpxBqhY2YR4GUrSYTzt7jNmm4DADP3uOAXFaZG7npT3OMTjMINnzhkgtEtIJ5sYgLQY3iLUFqMVM2gKoxZ4ELTlm0gwgWySI15KWbNhjkM4jceZZAbFakg8++PHHWo6/PXkDcVpggIc05aNgFIyCUTAK8AMAmqwk9GJdWnIAAAAASUVORK5CYII=","orcid":"","institution":"Indiana University Bloomington","correspondingAuthor":true,"prefix":"","firstName":"Rongqian","middleName":"","lastName":"Ma","suffix":""},{"id":446548448,"identity":"2574ceba-1251-4eca-a3c0-5cdc1f6912fe","order_by":1,"name":"Pei-Ying Chen","email":"","orcid":"","institution":"Indiana University Bloomington","correspondingAuthor":false,"prefix":"","firstName":"Pei-Ying","middleName":"","lastName":"Chen","suffix":""}],"badges":[],"createdAt":"2025-04-15 23:38:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6458217/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6458217/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":81280571,"identity":"0c7b88ce-208d-4b5a-88b8-03d8a2e9995e","added_by":"auto","created_at":"2025-04-24 10:08:11","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":10877,"visible":true,"origin":"","legend":"\u003cp\u003eOverview of total number and number of articles containing visual topics: (A) DSH (2006–2023) and (B) DHQ (2007–2023).\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6458217/v1/0cd8ef7f9f6a0222f3856321.png"},{"id":81280617,"identity":"ddbcfa50-bcc9-438c-addc-e3aa65e0e195","added_by":"auto","created_at":"2025-04-24 10:08:13","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":14320,"visible":true,"origin":"","legend":"\u003cp\u003eTemporal trends of visual topics: (A) \u003cem\u003eDSH\u003c/em\u003e and (B) \u003cem\u003eDHQ\u003c/em\u003e. Only topics appearing in 10 articles or more in total are color-coded and labelled.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6458217/v1/5cbdc73ffce8e362df793abb.png"},{"id":81281722,"identity":"703ccba1-1eb9-4a08-b590-580c3cdd84ff","added_by":"auto","created_at":"2025-04-24 10:16:14","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":826420,"visible":true,"origin":"","legend":"\u003cp\u003eTopic co-occurrence networks, where nodes represent topics and links indicate their co-occurrence. Visual topics are colored in blue, while others are in orange. (A) Complete network of DSH. (B) Subnetwork of DSH with one-off links removed. (C) Complete network of DHQ. (D) Subnetwork of DHQ with one-off links removed. Node size reflect degree (number of connections). Communities are highlighted with colored polygons.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6458217/v1/c05bc415722d1e3b0a44334f.png"},{"id":81280612,"identity":"7f33e431-b6cf-47e8-97f3-d99aba539be3","added_by":"auto","created_at":"2025-04-24 10:08:13","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":67647,"visible":true,"origin":"","legend":"\u003cp\u003eCentrality measures of visual topics and the top 10 non-visual topics (including ties) in the subnetworks and full networks, ordered by node degree in the subnetwork: (A) \u003cem\u003eDSH\u003c/em\u003e and (B) \u003cem\u003eDHQ\u003c/em\u003e. Rankings are based solely on nodes within the subnetwork.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6458217/v1/aadb18b75a2d71ace60ed630.png"},{"id":81280645,"identity":"5524e220-6ca6-40d8-bea1-b01d3242f029","added_by":"auto","created_at":"2025-04-24 10:08:14","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":53919,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of visual topics across the IMRaD structure: (A) DSH and (B) DHQ. The numbers in the cells represent the total occurrences of a topic at the article level. Numbers in white indicate a concentration of 50% or higher.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-6458217/v1/818dcd4ac15d00130f86f861.png"},{"id":96912931,"identity":"371b160e-322b-41f0-9e20-241a02f0e1e3","added_by":"auto","created_at":"2025-11-27 13:36:43","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1578348,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6458217/v1/8798285a-d7ef-4546-9117-2a435bd95f5b.pdf"},{"id":81280589,"identity":"09ab0f24-710b-4ae4-9cae-2efe8cf9c5a3","added_by":"auto","created_at":"2025-04-24 10:08:12","extension":"csv","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":251004,"visible":true,"origin":"","legend":"","description":"","filename":"Supple1dsharticleinfosamplepublic.csv","url":"https://assets-eu.researchsquare.com/files/rs-6458217/v1/fdad58c8deff038971d0506c.csv"},{"id":81280635,"identity":"79087041-62f9-4900-91ed-8927ebb3e35f","added_by":"auto","created_at":"2025-04-24 10:08:14","extension":"csv","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":83585,"visible":true,"origin":"","legend":"","description":"","filename":"Supple2dhqarticleinfosamplepublic.csv","url":"https://assets-eu.researchsquare.com/files/rs-6458217/v1/20ca1aa26d896764e7b697d1.csv"},{"id":81281723,"identity":"a0fec85e-5189-4330-ad41-42bfdb5dfd5e","added_by":"auto","created_at":"2025-04-24 10:16:15","extension":"csv","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":163616,"visible":true,"origin":"","legend":"","description":"","filename":"Supple3dshtopicinfopublic.csv","url":"https://assets-eu.researchsquare.com/files/rs-6458217/v1/e802208ce404127989338c9a.csv"},{"id":81280620,"identity":"9036c3d1-fc31-4601-a092-873e17205521","added_by":"auto","created_at":"2025-04-24 10:08:13","extension":"csv","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":116848,"visible":true,"origin":"","legend":"","description":"","filename":"Supple4dhqtopicinfopublic.csv","url":"https://assets-eu.researchsquare.com/files/rs-6458217/v1/e3688933182f716867eb1da8.csv"},{"id":81280639,"identity":"70c2c1be-ed33-42bf-9e0a-d780b9b9e9e5","added_by":"auto","created_at":"2025-04-24 10:08:14","extension":"csv","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":17988,"visible":true,"origin":"","legend":"","description":"","filename":"Supple5dshnodecentralities.csv","url":"https://assets-eu.researchsquare.com/files/rs-6458217/v1/95467222b53b8f278bfd1d76.csv"},{"id":81280616,"identity":"4e6a50e9-a4c7-426a-a766-927ab77f596c","added_by":"auto","created_at":"2025-04-24 10:08:13","extension":"csv","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":21175,"visible":true,"origin":"","legend":"","description":"","filename":"Supple6dhqnodecentralities.csv","url":"https://assets-eu.researchsquare.com/files/rs-6458217/v1/5bf384604fbd9dacd5fb1458.csv"},{"id":81280594,"identity":"b5b17e3b-e952-4862-b695-7f2ffeac7921","added_by":"auto","created_at":"2025-04-24 10:08:12","extension":"csv","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":8415,"visible":true,"origin":"","legend":"","description":"","filename":"Supple7dshirrpublic.csv","url":"https://assets-eu.researchsquare.com/files/rs-6458217/v1/f0177f34c30e21246d820229.csv"},{"id":81280642,"identity":"bb0b2209-1590-4e2f-8014-7de5c08dc930","added_by":"auto","created_at":"2025-04-24 10:08:14","extension":"csv","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":6457,"visible":true,"origin":"","legend":"","description":"","filename":"Supple8dhqirrpublic.csv","url":"https://assets-eu.researchsquare.com/files/rs-6458217/v1/2df0763eb569b28c00b11c89.csv"}],"financialInterests":"Competing interest reported. One of the authors, Rongqian Ma, is a member of the Editorial Board Member of Humanities and Social Sciences Communications.","formattedTitle":"What do Digital Humanities Say about visualization? A Bibliometric Exploration","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eSince its inception, digital humanities (DH) have evolved as an interdisciplinary research domain shaped by the constant adoption and integration of new technological innovations. Early discussions about DH\u0026rsquo;s intellectual identities often revolved around how to interpret the \u0026ldquo;digital\u0026rdquo; in relation to established humanities paradigms (Svensson, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). These debates highlighted the extent to which computational methods could\u0026mdash;or should\u0026mdash;transform the interpretive, critical, and contextual approaches characteristic of the humanities, influencing aspects ranging from research questions and data sources to analytical frameworks and modes of dissemination (Gold, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Gold \u0026amp; Klein, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2016\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2019\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn this paper, we focus on \u003cem\u003evisualization\u003c/em\u003e as a prominent example of how these digital technologies have influenced humanities research. While the humanities scholarship has historically emphasized textual analysis and interpretation, recent scholarship underscores a growing reliance on visual methods to examine large, complex, or heterogeneous data sets (Champion, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; M\u0026uuml;nster \u0026amp; Terras, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). This shift from a primarily \u0026ldquo;text-heavy\u0026rdquo; orientation toward more visually oriented practices carries the potential to reshape research paradigms, scholarly workflows, and even academic culture in DH. By examining how visualization has been adopted and adapted within DH, we can shed light on how traditional interpretive conventions\u0026mdash;such as close reading, hermeneutics, and critical theory\u0026mdash;are being rearticulated in dialogue with computational and visual techniques.\u003c/p\u003e \u003cp\u003eWithin science and technology studies (STS), visualizations are considered as a form of inscriptions, or \u0026ldquo;immutable mobiles,\u0026rdquo; which possess optical consistency while the capacity of traveling across diverse contexts to convey and mobilize arguments, ideas, and findings (Latour, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e1990\u003c/span\u003e; Roth \u0026amp; McGinn, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e1998\u003c/span\u003e). The essence and power of inscriptions lie in their \u003cem\u003eembeddedness\u003c/em\u003e within knowledge representation systems and the broader context of knowledge production (Caissie et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Using visuals embedded in scientific discourse and scholarship, for instance, helps constructs facts and shape convincing arguments (Graves, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Scholars have also shown that the use of graphs, such as data visualizations and statistical charts, is strongly associated with the public\u0026rsquo;s collective perception of a field\u0026rsquo;s \u0026ldquo;hardness\u0026rdquo; or \u0026ldquo;scientificity\u0026rdquo; (Cleveland, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e1984\u003c/span\u003e; Smith et al., \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2000\u003c/span\u003e). Visual inscriptions can also facilitate the communication of complex, multifaceted ideas and findings to a broad range of audiences, both within and beyond academia (Caissie et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Myers, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e1988\u003c/span\u003e). Focusing on a case from the geology discipline, Rudwick (\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e1976\u003c/span\u003e) discussed a wide range of visual aids available to engage in geology research and argued for the power of a \u0026ldquo;visual language\u0026rdquo; for geological science. More specifically, several mechanisms and ways of how visuals consolidate facts and mobilize arguments are identified in research literature. Lynch (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e1990\u003c/span\u003e), for example, argued that selection and mathematization are two important mechanisms that help scientists shape their objects of knowledge. Specifically, selection involves the simplification and transformation of objects to fit scientific representation, often through diagrams that abstract and enhance details for clarity. Mathematization further integrates these visualizations into structured, mathematical forms, making the data analyzable and comparable. These STS works demonstrate the powerful yet often invisible ways that visual representations can contribute to the making of knowledge.\u003c/p\u003e \u003cp\u003eIn this paper, we adopt the STS notion of inscriptions as a theoretical lens to analyze the roles of visuals in DH scholarship. However, we will extend from this \u003cem\u003einscription\u003c/em\u003e perspective to explore the dynamics behind the \u0026ldquo;embeddedness\u0026rdquo; within a specific DH context, moving beyond how visuals are constructed and presented in publications to focus on how they are incorporated within scholarly narratives. We will analyze how discussions of visual representations are at interplay with other parts of narrative in the article to effectively mobilize and communicate DH knowledge. Despite numerous case studies documenting the varied applications of visual technologies in DH, relatively little is known about how \u003cem\u003ecollectively\u003c/em\u003e visualizations are incorporated into DH scholarly narratives to foster research communication. For example, which aspects of visualizations are the most transferable to DH work? How are these visual aspects applied to DH, and do DH scholars primarily use visualizations for certain purposes over others?\u003c/p\u003e \u003cp\u003eTo address this knowledge gap, we present a quantitative investigation into the representation of visualization-focused topics\u0026mdash;or otherwise referred to as visual themes\u0026mdash;in DH scholarship. Specifically, we examine a full-text corpus of articles from \u003cem\u003eDigital Scholarship in the Humanities (DSH)\u003c/em\u003e and \u003cem\u003eDigital Humanities Quarterly (DHQ)\u003c/em\u003e, two leading DH journals with longstanding traditions and broad recognition within the DH research community (Spinaci et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). By focusing on full-text content rather than abstracts or keywords alone, we aim to capture a richer and more nuanced picture of how visualization is discussed in DH scholarly narratives. Using this corpus, we will explore the following research questions:\u003c/p\u003e \u003cp\u003eRQ 1: What are the major visual topics represented in \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ?\u003c/em\u003e\u003c/p\u003e \u003cp\u003eRQ 2: How are these visual topics discussed in \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ\u003c/em\u003e, respectively?\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eRQ 2.1: How are visual topics interwoven with non-visual topics?\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eRQ 2.2: What key functions do visual themes serve to mobilize DH scholarly narratives?\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eUsing topic modeling and network analysis, our research aims to illustrate how DH discourse is dynamically constructed through the interplay of visual and non-visual topics within scholarly narratives.\u003c/p\u003e"},{"header":"LITERATURE REVIEW","content":"\u003cp\u003eAs visual technologies become increasingly integral to DH, scholars have examined visualization from both conceptual and practical perspectives. Some studies have focused on theoretical models and principles that align visualization methods with the interpretive nature of humanities research. Early efforts in this vein include Drucker's (2011, 2018) concept of “humanistic visualization,” which distinguishes \u003cem\u003ecapta\u003c/em\u003e from \u003cem\u003edata\u003c/em\u003e and advocates a “nonrepresentational approach” that foregrounds the interpretive complexities of humanities materials. Similarly, Manovich (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2011\u003c/span\u003e) introduced the concept of “direct visualization,” which aims to preserve the original forms and inherent complexity of humanities data rather than create abstracted or simplified data representations. Hinrichs et al. (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) further introduced ideas such as “sandcastling,” emphasizing that visualization should be iterative, fluid, and open to revision—qualities that align well with the evolving, interpretive character of the humanities.\u003c/p\u003e \u003cp\u003eThis conceptual discourse has diversified over time. Bradley et al. (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), for instance, proposed the notion of “slow analytics” in the context of digital literary studies, arguing that effective visualizations should prioritize “slow and methodical interaction with texts [as] part of the workflow and sense-making process,” rather than emphasizing speed and efficiency. Likewise, Kleymann and Stange (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) articulated four postulates toward a hermeneutic approach to visualization in literary studies, aiming to reconcile the interpretive traditions of literary scholarship with the scientific and technical processes of visualization.\u003c/p\u003e \u003cp\u003eAmong these conceptual discussions, one specific issue that has received particular attention is uncertainty. Uncertainty is defined as “the lack of certainty or doubt about the accuracy or reliability of data” (Conroy et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Distinct from error, which arises from inaccuracies or mistakes in measurement, data entry, or analysis, uncertainty stems from the inherent limitations of data, methods, or contextual ambiguities. Schwandt and Wachter (\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) further argued that “uncertainty in visualization is not a flaw but a feature that enhances critical analysis.” Translating the uncertainty discourse into a DH context, scholars have discussed various potential sources of uncertainty. For example, Panagiotidou et al. (\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) argued that uncertainty may emerge from the nature of the source data (e.g., missing, imprecise, ambiguous data), the process of datafication (e.g., complexity reduction, subject manual classification), and modelling (e.g., statistical, model appropriateness). In a specific domain such as text analysis, uncertainty may arise at multiple stages—semantic, comprehension, encoding, transformation, representation, and interpretation—thus posing challenges that extend from data labeling to visualization (Haghighatkhah et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Collectively, these discussions have also led to the development of visualization approaches for addressing uncertainty, such as the four postulates of “established, dedicated, improvisational, and experiential” proposed by Panagiotidou et al. (2022).\u003c/p\u003e \u003cp\u003eAdditional issues and concerns related to DH visualizations have been explored in the IEEE VIS4DH workshop—one of the key conference venues for discussions on the topic. Since its inception in 2016, publications from the workshop have revealed a clear evolution in how visualization challenges are conceptualized and addressed. These include themes such as how to enhance trust during DH processes and interpretations (Chen et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Schetinger et al., \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), how to rethink the notion of “scale” within DH contexts (Chen \u0026amp; Cole, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Ciula et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), and how to reimagine visualization approaches when working with multimodality data (Brath, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Mayr et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Zhou \u0026amp; Forbes, \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAlongside these discussions, much of the DH visualization research also aims to develop specific techniques, applications, and tools that operationalize humanistic ideals. Existing scholarship has shown that visualization can support distant reading of large visual corpora, text materials, and cultural heritage collections (Alharbi \u0026amp; Laramee, \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Arnold \u0026amp; Tilton, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Janicke et al., 2016). Computational techniques such as knowledge graphs have become particularly prominent for examining digital cultural objects and data (Khoo et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Mayr et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Zheng et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), owing to their capacity to reveal complex relationships in primary sources. Emerging technologies such as Virtual reality (VR) and 3D visualization tools are also increasingly applied to improve accessibility and interaction with cultural heritage materials (Xia et al., 2021; Meinecke et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Parallel to these efforts, DH researchers have introduced innovative visualization techniques and tools tailored to humanities research tasks. For example, Valdivia et al. (\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) presented a Parallel Aggregated Ordered Hypergraph (PAOH) method for depicting dynamic hypergraphs, while Yan and Li (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) developed a multi-view framework that integrates semantic aggregation and interactive text exploration to enhance historical knowledge discovery. The various case studies and developments show the great potential of leveraging visualizations to support different stages of humanities research, from exploring large data corpora to conducting interpretive analyses. Although scholars have explored the potential of using DH visualizations as the lens for synergizing theories of traditional humanities and tool-making practices of DH most visualization work in the field has centered on case studies and individual applications. As a result, despite the breadth of these innovations, a systematic investigation into the specific ways in which visualizations are employed in DH scholarship to support the construction of scholarly narratives has yet to be conducted.\u003c/p\u003e \u003cp\u003eIn this paper, we aim to bridge this gap by analyzing how visualizations are collectively applied and discussed in DH research, drawing on a bibliometric approach. Prior bibliometric analyses have revealed key trends of visual practices within DH (Leydesdorff \u0026amp; Salah, 2010; Münster \u0026amp; Terras, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). For instance, Weingart and Eichmann-Kalwara (\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) investigated DH conference abstracts from 2004 to 2015, revealing an increasing emphasis on visual methods. Benito-Santos and Sánchez (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) subsequently employed a data-driven citation analysis of 300 publications from IEEE VIS4DH workshops, the ADHO Digital Humanities Conference, and \u003cem\u003eDigital Humanities Quarterly\u003c/em\u003e, identifying principal authors, techniques, and data sources central to DH visualization research. Further extending this line of inquiry, Ma and Li (2022) analyzed a corpus of DH publications and highlighted the complexity of DH visualization practices, particularly the differing preferences among humanities and STEM (science, technology, engineering, and mathematics) researchers regarding both the frequency and types of visual forms. Building on these studies, our work applies computational methods to full-text articles published in leading DH journals, offering a more nuanced perspective on how visualization practices and discourses are embedded within the broader DH scholarship.\u003c/p\u003e "},{"header":"DATA AND METHODS","content":"\u003cp\u003eOur data sample comprises all English-language research and review articles, conference and special issues papers, and case studies published in \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ\u003c/em\u003e between 2006 and 2023 (including those in supplementary issues), with full-text XMLs or HTMLs available. After excluding corrigenda, errata, and retractions, as well as verifying article types and language, 990 \u003cem\u003eDSH\u003c/em\u003e and 579 \u003cem\u003eDHQ\u003c/em\u003e articles remained in the sample for subsequent analysis. The complete lists of articles from \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ\u003c/em\u003e analyzed in this study are available in Supplementary Materials 1 and 2.\u003c/p\u003e\u003cp\u003eWe employed BERTopic to identify topics within \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ\u003c/em\u003e respectively. This method leverages the power of Bidirectional Encoder Representation from Transformers (BERT) to cluster semantically similar documents and generate coherent topic representations through a class-based variation of TF-IDF (Grootendorst, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Compared with other topic modeling techniques (e.g., LDA, NMF, Top2Vec), BERTopic is shown to be capable of producing more distinct topics and encoding contextual information (Egger \u0026amp; Yu, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eTo prepare the corpus for BERTopic, we extracted paragraphs and their associated section headings from the full-text HTMLs and XMLs using the “rvest” and “xml2” packages in R. Since BERT is optimized for texts of fewer than 150 words, paragraphs exceeding this limit were binned into chunks while maintaining sentence-level completeness. To ensure sufficient information was retained, chunks with fewer than 18 words were discarded (i.e., the bottom 5% for \u003cem\u003eDSH\u003c/em\u003e and the bottom 2% for \u003cem\u003eDHQ\u003c/em\u003e). In total, the \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ\u003c/em\u003e corpora contained 55,981 and 40,813 chunks, respectively, for topic modeling.\u003c/p\u003e\u003cp\u003eFollowing BERTopic’s best practices, we precalculated embeddings using the default embedding model, all-MiniLM-L6-v2, and prevented stochastic behaviors in UMAP during dimensionality reduction. Default topic representations were enhanced by removing English stopwords and infrequent words, while keyword diversity within a topic was increased with the diversity parameter set to 0.3. All other parameters were kept at their default values, including the minimum cluster size of 10 for the HDBSCAN clustering process and the extraction of the top 10 words per topic.\u003c/p\u003e\u003cp\u003eThe retrieved topics were then categorized as either visualization-related or not. We applied regular expressions to identify topics as “visual” if any of their top 10 representation terms matched a term from a pre-defined list: diagram, graph, illustration, image, map, network, photo, simulation, and visual.\u003c/p\u003e\u003cp\u003eTo further explore the relationships between visual and non-visual topics, we built upon the topic modeling results and constructed weighted co-occurrence networks for the \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ\u003c/em\u003e topics, respectively. Co-occurrences were defined at the article level: if topic A and topic B appear together in a single article, a link between these two topics is established with a weight of 1. Note that we only consider topics that co-occur with visual topics as defined above. The construction of the networks and their visualization was carried out using the “igraph” package in R.\u003c/p\u003e\u003cp\u003eAdditionally, we analyzed the headings associated with the visual topics and manually coded them into one of categories of the IMRaD structure—Introduction, Methods, Results, Discussion—to better understand the argumentative functions these visual topics serve within an article (Sollaci \u0026amp; Pereira, \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). We also manually distinguished and annotated the nature of articles as empirical, conceptual, or methods-oriented to provide more context regarding the likely concentrations of visual topics within specific IMRaD sections. Empirical articles typically follow the standard IMRaD structure, while conceptual articles are composed of Introduction and Discussion, and methods papers include Introduction, Methods, and Discussion sections only.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cp\u003e1. Topic modeling results show comparative trends and emphases in visual topics across \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ\u003c/em\u003e\u003c/p\u003e \u003cp\u003eIn total, 919 were identified in \u003cem\u003eDSH\u003c/em\u003e and 650 in \u003cem\u003eDHQ\u003c/em\u003e (see Supplementary Materials 3 and 4). Among these topics, 21 in \u003cem\u003eDSH\u003c/em\u003e and 26 in \u003cem\u003eDHQ\u003c/em\u003e are visual topics. Figure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e illustrates the temporal trends of articles containing visual topics, showing that \u003cem\u003eDHQ\u003c/em\u003e exhibits a stronger pattern of publishing visual-related articles over time. Proportionally, 11% (n\u0026thinsp;=\u0026thinsp;112) of \u003cem\u003eDSH\u003c/em\u003e articles and 20% (n\u0026thinsp;=\u0026thinsp;116) of \u003cem\u003eDHQ\u003c/em\u003e articles contain at least one visual topic, with the highest yearly shares reaching 17.6% (2021) and 43.8% (2019) and the lowest falling to zero (2007) and 9.1% (2009), respectively. Since half of the topics appear only once or twice in the corpus, we limited our subsequent analyses to topics appearing in at least three articles. This leaves us with 416 and 309 topics, including 10 and 12 visual topics, from \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ\u003c/em\u003e respectively. Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e show these major visual topics from \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ.\u003c/em\u003e\u003c/p\u003e \u003cp\u003eA comparative analysis of these two topic lists reveals convergent commitments to visualization and computational methods in DH scholarship, yet each journal demonstrates slightly different emphases (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). \u003cem\u003eDSH\u003c/em\u003e shows a more technical orientation, evident in visual topics such as \u0026ldquo;Classification \u0026amp; Machine Learning,\u0026rdquo; \u0026ldquo;Word-Sense Disambiguation \u0026amp; Linguistics,\u0026rdquo; and \u0026ldquo;Geometric \u0026amp; Graph-Based Analysis,\u0026rdquo; all of which underscore a strong focus on algorithmic approaches to digital scholarship. By contrast, \u003cem\u003eDHQ\u003c/em\u003e offers a broader conceptual and theoretical scope. Its largest visual topic, \u0026ldquo;Humanistic Visualization \u0026amp; Graphics,\u0026rdquo; highlights an engagement with the epistemological and aesthetic dimensions of DH visualization. \u003cem\u003eDHQ\u003c/em\u003e also addresses specialized cultural or heritage topics\u0026mdash;ranging from \u0026ldquo;Monastic Architecture \u0026amp; Mapping\u0026rdquo; to \u0026ldquo;Folklore and Gameworld \u0026amp; Interactive Visualizations\u0026rdquo;\u0026mdash;illustrating how digital methods can be applied to diverse historical, social, or creative contexts. Meanwhile, \u0026ldquo;Agent-based Modeling \u0026amp; Simulation,\u0026rdquo; which does not appear as prominently in \u003cem\u003eDSH\u003c/em\u003e, indicates a commitment to computational experiments aimed at simulating social or cultural phenomena.\u003c/p\u003e \u003cp\u003eDespite these differences, the two journals share important common ground in areas such as network analysis, geospatial inquiry, temporal visualization, and the study of art. Both journals emphasize the topic of \u0026ldquo;Network Analysis \u0026amp; Visualization\u0026rdquo; (i.e., T96 in \u003cem\u003eDSH\u003c/em\u003e and T585 in \u003cem\u003eDHQ)\u003c/em\u003e, reflecting a keen interest in mapping relationships\u0026mdash;whether social, linguistic, or otherwise\u0026mdash;through graph-based models. Geospatial research is similarly central: Topics such as \u0026ldquo;Maps \u0026amp; Geographic Information\u0026rdquo; are represented in \u003cem\u003eDSH\u003c/em\u003e, while \u003cem\u003eDHQ\u003c/em\u003e features themes such as \u0026ldquo;Monastic Architecture \u0026amp; Mapping.\u0026rdquo; Similar patterns also emerge in temporal analysis and art: \u003cem\u003eDSH\u003c/em\u003e includes \u0026ldquo;Time and Spatial Visualization,\u0026rdquo; while \u003cem\u003eDHQ\u003c/em\u003e discusses \u0026ldquo;Chronologies \u0026amp; Temporal Diagrams;\u0026rdquo; \u003cem\u003eDSH\u003c/em\u003e showcases themes like \u0026ldquo;Photography\u0026rdquo;, whereas \u003cem\u003eDHQ\u003c/em\u003e focuses on \u0026ldquo;Image Atlas \u0026amp; Art\u0026rdquo;. Finally, both journals consider best practices in visual scholarship, though they prioritize different dimensions. \u003cem\u003eDSH\u003c/em\u003e features a topic on \u0026ldquo;Provenance \u0026amp; Transparency in Visualization,\u0026rdquo; which raises specific ethical and methodological concerns. \u003cem\u003eDHQ\u003c/em\u003e, in comparison, addresses the theme of \u0026ldquo;Archiving\u0026rdquo; more explicitly, emphasizing a particular context of use for visualization practices.\u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows how different topics in \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ\u003c/em\u003e are distributed across time. As we can observe from Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, certain visual topics hold long-term prominence in DH scholarship. For \u003cem\u003eDSH\u003c/em\u003e, \u0026ldquo;Time \u0026amp; Spatial Visualization\u0026rdquo; is one such topic, the discussion of which appeared in the journal in the 2005\u0026ndash;2009 period and has been increasing since then until the peak during the period of 2020 and 2024. By contrast, \u0026ldquo;Word-Sense Disambiguation \u0026amp; Linguistics\u0026rdquo; first appeared during the 2010\u0026ndash;2014 period, while \u0026ldquo;Classification \u0026amp; Machine Learning\u0026rdquo; was a relatively new visual topic adopted in the communities during 2015\u0026ndash;2019. For \u003cem\u003eDHQ\u003c/em\u003e, \u0026ldquo;Humanistic Visualization \u0026amp; Graphics\u0026rdquo; holds a similar trend. The discussion of this topic started in the early stages of the journal, from 2005 to 2009, and gained prominence in the period of 2015\u0026ndash;2019. Afterward, there was a slow decline for its representation, but this visual theme remained as the most represented and discussed visual topics from 2020 to 2024. Interestingly, \u0026ldquo;Archiving\u0026rdquo; in \u003cem\u003eDHQ\u003c/em\u003e also appears to be a longstanding topic among the communities, with peak representations during 2015\u0026ndash;2019 and then a decreasing representation during 2020\u0026ndash;2024.\u003c/p\u003e \u003cp\u003eIn summary, our results show that both journals integrate visualization discussions in their scholarship, although the topic emphases change across time, reflecting the collective interests among the DH research communities. More specifically, while \u003cem\u003eDSH\u003c/em\u003e provides a technically oriented, methodologically detailed approach to DH visualization, \u003cem\u003eDHQ\u003c/em\u003e integrates theoretical, conceptual, and domain-specific explorations into its visualization discourse.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eMajor visual topics in \u003cem\u003eDSH\u003c/em\u003e (N\u0026thinsp;=\u0026thinsp;Number of articles).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTopic\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLabel\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eKeywords\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTime and Spatial Visualization\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['chronotopic', 'commodities', 'grid', 'wheat', 'spatiotemporal', \u003cb\u003e'visualizations'\u003c/b\u003e, 'geohistorical', 'gir', 'heatmap', 'timelines']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWord-sense disambiguation \u0026amp; linguistics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['wsd', 'senses', 'concordancer', 'glosses', 'supervised', 'v\u0026eacute;ronis', 'disambiguation', \u003cb\u003e'graph'\u003c/b\u003e, 'korean', 'relatedness']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e347\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIllustration and Iconography\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['reproductions', 'vision', 'pollock', \u003cb\u003e'illustrations'\u003c/b\u003e, 'artworks', 'motifs', 'iconographic', 'dietrich', 'postures', 'microscopy']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e517\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eClassification \u0026amp; Machine Learning\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['burrowsian', 'nsc', 'centroid', 'classify', 'classifications', 'classifiers', 'b5', 'flavours', \u003cb\u003e'simulations'\u003c/b\u003e, 'mf3c']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e806\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePhotography\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['cesr', 'cvma', 'slides', '\u0026eacute;tudes', 'digitising', 'courtauld', 'romanesque', 'scanned', \u003cb\u003e'photographs'\u003c/b\u003e, 'renaissance']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e234\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eInfographics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e[\u003cb\u003e'infographics'\u003c/b\u003e, 'mastering', 'literacy', 'kandinsky', \u003cb\u003e'graphics'\u003c/b\u003e, \u003cb\u003e'visualizations'\u003c/b\u003e, 'hackneyed', 'charter', \u003cb\u003e'visual'\u003c/b\u003e, 'counterfactual']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMaps \u0026amp; Geographic Information\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['cartographic', 'historic', 'georeferencing', \u003cb\u003e'maps'\u003c/b\u003e, 'geoportals', 'topographic', 'cartography', 'cadastral', 'mercator', 'geographic']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGeometric \u0026amp; Graph-Based Analysis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['bisector', 'counter', 'zeta', 'centroids', 'segments', 'randomized', 'howells', 'marker', 'eyed', \u003cb\u003e'graphs'\u003c/b\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e96\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNetwork Analysis \u0026amp; Visualization\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['vertices', 'diameter', 'shells', 'cores', \u003cb\u003e'networks'\u003c/b\u003e, 'nodes', 'edges', 'neighbours', 'connectivity', 'geodesic']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eProvenance \u0026amp; Transparency in Visualization\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['provenance', 'disclosure', 'vancisin', \u003cb\u003e'visualization'\u003c/b\u003e, 'reproducibility', 'artifactual', 'transparency', 'curatorial', 'participants', 'cs4']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eMajor visual topics in \u003cem\u003eDHQ\u003c/em\u003e (N\u0026thinsp;=\u0026thinsp;number of articles).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTopic\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLabel\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eKeywords\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHumanistic Visualization \u0026amp; Graphics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['deleuze', 'infovis', 'folding', \u003cb\u003e'visualizations'\u003c/b\u003e, 'folds', 'graphical', 'charts', 'interactivity', 'transitions', 'humanistic']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e43\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e284\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eArchiving\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['ellen', 'archival', 'pilers', 'storing', 'photocopying', 'dropbox', 'discloses', \u003cb\u003e'photos'\u003c/b\u003e, 'subfolders', 'folder']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e461\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eText Visualization\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['themeriver', 'tei2html', 'reversible', 'tapor', \u003cb\u003e'visualizations'\u003c/b\u003e, 'textarc', 'computationalstylistics', 'darker', 'tufte', 'shading']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e135\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGraph Layouts \u0026amp; Structures\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['layered', \u003cb\u003e'graphs'\u003c/b\u003e, 'layouts', 'nodes', 'dashboard', 'topology', 'hierarchical', 'graphviz', 'mental', \u003cb\u003e'visualizations'\u003c/b\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e321\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eImage Atlas \u0026amp; Art\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['warburg', 'rembrandt', 'atlas', 'heinrich', 'wooden', 'reproductions', 'artworks', 'renaissance', 'antiquity', \u003cb\u003e'photographic'\u003c/b\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eChronologies \u0026amp; Temporal Diagrams\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['chronologies', 'heterochronologies', 'timelines', 'lunar', 'leonid', 'synchronicity', 'solar', 'chronology', 'temporal', \u003cb\u003e'diagrams'\u003c/b\u003e]\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e585\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNetwork Analysis \u0026amp; Visualization\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['gephi', 'poms', 'eigenvector', 'centrality', 'closeness', 'cliques', 'bimodal', 'weingart', \u003cb\u003e'graph'\u003c/b\u003e, 'cosmopolitan']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMonastic Architecture \u0026amp; Mapping\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['monastic', 'samos', 'polygons', 'monastery', 'monasteries', 'architectural', 'buildings', \u003cb\u003e'maps'\u003c/b\u003e, 'village', 'romanesque']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGameworld \u0026amp; Interactive Visualizations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['verbum', \u003cb\u003e'visuality'\u003c/b\u003e, 'wallpaper', 'rooms', 'npcs', 'plotkin', 'exits', 'gameworld', 'commands', 'puzzles']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e150\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFolklore\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['folktale', 'folktales', 'dutch', 'folklore', 'tales', \u003cb\u003e'visualises'\u003c/b\u003e, 'mock', 'facets', 'visitors', 'fairy']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e257\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAgent-based Modeling \u0026amp; Simulation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['abm', 'simulations', 'agents', 'ideation', 'artificial', 'modeling', 'behaviors', 'archeological', \u003cb\u003e'simulated'\u003c/b\u003e, 'autopilot']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e516\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eText Mining \u0026amp; Networks\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e['bigram', 'skipgrams', 'preprocessing', 'mining', 'edges', \u003cb\u003e'visualisations'\u003c/b\u003e, 'summarises', 'vertices', \u003cb\u003e'networks'\u003c/b\u003e, 'contextually']\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e2. Topic co-occurrence networks reveal the embeddedness of visual topics in \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ\u003c/em\u003e\u003c/p\u003e \u003cp\u003eThe topic co-occurrence networks of \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ\u003c/em\u003e, as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, further reveal how visual themes are integrated into DH scholarship. The complete \u003cem\u003eDSH\u003c/em\u003e network comprises 241 nodes (topics) and 1,272 links (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA), 91.2% of which appear only once. After filtering out the one-off links, 69 nodes (28.6%) and 112 links (8.8%) remain (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). Similarly, the complete \u003cem\u003eDHQ\u003c/em\u003e network comprises 284 nodes and 1,877 inks (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC); after filtering out one-off links (94.1%), 80 nodes (28.2%) and 111 links (5.9%) remain (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD).\u003c/p\u003e \u003cp\u003eThis result aligns with our previous observation that the visual topic discourse in DH scholarship is diverse and dispersed, with a large proportion of topics being niche, unique, and specialized and rarely discussed in combination with other themes. From a broader perspective, our results also correspond to the commonly shared perception of DH being a highly heterogeneous and \u0026ldquo;amorphous\u0026rdquo; field (Deegan, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Luhmann \u0026amp; Burghardt, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOur analysis also reveals that visual topics are embedded to varying degrees within the broader subject landscapes of \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ\u003c/em\u003e. A comparison of the co-occurrence network modularity\u0026mdash;using the built-in function in the \u0026ldquo;igraph\u0026rdquo; package that identifies the optimal community structure based on maximal modularity scores\u0026mdash;shows that \u003cem\u003eDSH\u003c/em\u003e exhibits a clearer community structure than \u003cem\u003eDHQ.\u003c/em\u003e Specifically, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB yields 8 clusters with a modularity of 0.70, whereas Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD returns 13 clusters with a modularity of 0.56.\u003c/p\u003e \u003cp\u003eIn \u003cem\u003eDSH\u003c/em\u003e, visual topics tend to connect with non-visual topics and form small, interconnected clusters. For example, the largest visual topic, \u0026ldquo;Time \u0026amp; Spatial Visualization\u0026rdquo; (T31), co-occurs with non-visual topics such as \u0026ldquo;Metadata and Postcolonial Curation\u0026rdquo; (T425), \u0026ldquo;Spatial Humanities and Gazetteers\u0026rdquo; (T55), and \u0026ldquo;Digital Humanities as a Field\u0026rdquo; (T41), while co-occurring with only one visual topic, \u0026ldquo;Infographics\u0026rdquo; (T234). Similarly, \u0026ldquo;Word-Sense Disambiguation \u0026amp; Linguistics\u0026rdquo; (T12) and \u0026ldquo;Classification \u0026amp; Machine Learning\u0026rdquo; (T517) co-occur exclusively with a group of non-visual topics. We can also observe from Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB that some non-visual topics in the \u003cem\u003eDSH\u003c/em\u003e network play a crucial role not only in terms of their degrees but also as bridge nodes connecting clusters. For instance, among non-visual topics, \u0026ldquo;Gephi\u0026rdquo; (T145) ranks highest in terms of betweenness in the \u003cem\u003eDSH\u003c/em\u003e subnetwork, serving as a bridge node between two communities (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB and Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). The non-visual topic \u0026ldquo;HCI and usability\u0026rdquo; (T603) follows T145, connecting three communities (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB and \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). A complete list of \u003cem\u003eDSH\u003c/em\u003e node centralities is available in Supplementary Material 5.\u003c/p\u003e \u003cp\u003eIn \u003cem\u003eDHQ\u003c/em\u003e, visual topics are more seamlessly integrated into the broader DH discourse, co-occurring with both visual and non-visual topics and forming multiple interconnected groups of nodes. \u0026ldquo;Humanistic Visualization \u0026amp; Graphics\u0026rdquo; (T1) appears to be the most central node in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD. It widely connects with a variety of visual and non-visual topics. For example, it co-occurs with the visual themes of \u0026ldquo;Text Visualization\u0026rdquo; (T461) and \u0026ldquo;Archiving\u0026rdquo; (T284). Among the non-visual topics, themes in the \u003cem\u003eDHQ\u003c/em\u003e subnetwork such as \u0026ldquo;Digital Tool Building \u0026amp; Criticism\u0026rdquo; (T7) and \u0026ldquo;Humanities Landscape\u0026rdquo; (T0) are the most predominantly represented, followed by high-degree topics such as \u0026ldquo;Digital Practices \u0026amp; Infrastructures\u0026rdquo; (T66), \u0026ldquo;Archival Profession\u0026rdquo; (T171), \u0026ldquo;LDA\u0026rdquo; (T11), and \u0026ldquo;Debates in Digital Humanities\u0026rdquo; (T87). These topics range from more technical ones (e.g., LDA) to more conceptual ones (e.g., Debates in Digital Humanities), indicating that \u0026ldquo;Humanistic Visualization \u0026amp; Graphics\u0026rdquo; has been discussed across diverse contexts. A complete list of \u003cem\u003eDHQ\u003c/em\u003e node centralities is available in Supplementary Material 6.\u003c/p\u003e \u003cp\u003eSynthesizing the comparisons above, the co-occurrence networks of \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ\u003c/em\u003e reveal slightly different approaches to incorporating visualization knowledge into DH scholarship across these two leading journals. The visual topics in \u003cem\u003eDSH\u003c/em\u003e represent a strong methodological emphasis, and these various visual methods are often applied to niche and specific subjects and research areas. This may explain the more prominent community structure in the \u003cem\u003eDSH\u003c/em\u003e co-occurrence network. \u003cem\u003eDHQ\u003c/em\u003e, however, places a stronger emphasis on conceptual discussions of visual topics, with \u0026ldquo;Humanistic Visualization \u0026amp; Graphics\u0026rdquo; occupying a central position in the network and connecting to a wide array of both visual and non-visual topics. By contrast, most other visual topics\u0026mdash;such as Graph Layouts \u0026amp; Structures (T135), Network Analysis \u0026amp; Visualization (T585), and Chronologies \u0026amp; Temporal Diagrams(T44) appear in more peripheral positions within the network.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e3. Narrative Functions of Visual Topics\u003c/p\u003e \u003cp\u003eWe identified 75 \u003cem\u003eDSH\u003c/em\u003e articles and 75 \u003cem\u003eDHQ\u003c/em\u003e articles that include section headings associated with visual topics. A manual coding of their article types as well as the associated headings reveals the narrative functions of visualization in DH scholarship. Two authors independently mapped the headings to the IMRaD structure and categorized the articles into three types: Conceptual, Methodology, and Empirical. For \u003cem\u003eDSH\u003c/em\u003e, the IMRaD classification produced a Cohen\u0026rsquo;s (1960) kappa of 0.67, indicating substantial agreement, while the \u003cem\u003eDHQ\u003c/em\u003e IMRaD classification yielded a slightly higher kappa of 0.69. For article type classification, the kappa was 0.58 for DSH and 0.59 for \u003cem\u003eDHQ\u003c/em\u003e, reflecting moderate agreement. Our detailed mapping results are also available in Supplementary Materials 7 and 8.\u003c/p\u003e \u003cp\u003eIn \u003cem\u003eDSH\u003c/em\u003e, most articles related to visual themes (70.7%, n\u0026thinsp;=\u0026thinsp;53) are classified as Methods, while Conceptual (16%, n\u0026thinsp;=\u0026thinsp;12) and Empirical (13.3%, n\u0026thinsp;=\u0026thinsp;10) papers together make up less than one-third. By contrast, \u003cem\u003eDHQ\u003c/em\u003e exhibits a more balanced distribution, with Methods (42.7%, n\u0026thinsp;=\u0026thinsp;32) and Conceptual (40%, n\u0026thinsp;=\u0026thinsp;30) constituting the majority, and Empirical (17.3%, n\u0026thinsp;=\u0026thinsp;13) accounting for the remainder. Tables\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e and \u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e illustrate the narrative functions of visual topics in both journals. In \u003cem\u003eDSH\u003c/em\u003e, 39% of visual-topic headings map to \u0026ldquo;M,\u0026rdquo; emphasizing their prominent role in illustrating methods. Such headings appear more frequently in Methods paper (47.9%) than in Empirical ones (36.8%). The second most common function of visual themes is supporting the introduction (29.3%), with most instances in Conceptual (38.5%) and Methods (30.3%) papers. In \u003cem\u003eDHQ\u003c/em\u003e, however, the most frequent function of visual topics is supporting the introduction (37%) across all article types, followed by illustrating methods (24%). Overall, these findings suggest that both journals feature relatively few empirical studies on visualization. Comparatively, \u003cem\u003eDSH\u003c/em\u003e favors a more methodologically oriented style, with a high proportion of Methods papers and a predominant use of visualizations to illustrate methods, while \u003cem\u003eDHQ\u003c/em\u003e is more balanced with methodological detail and conceptual discourse.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDistribution of IMRaD headings associated with visual topics across Article Type in \u003cem\u003eDSH\u003c/em\u003e.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIMRaD\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eConceptual Paper\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eEmpirical Paper\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMethods Paper\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eTotal\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e10 (38.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2 (10.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e36 (30.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e48\u0026nbsp;(29.3%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7 (36.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e57 (47.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e64\u0026nbsp;(39.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8 (42.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8\u0026nbsp;(4.9%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e16\u0026nbsp;(61.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2 (10.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e26\u0026nbsp;(21.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e44\u0026nbsp;(26.8%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTotal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e119\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e164\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDistribution of IMRaD headings associated with visual topics across Article Type in \u003cem\u003eDHQ\u003c/em\u003e.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIMRaD\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eConceptual Paper\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eEmpirical Paper\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMethods Paper\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eTotal\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e30 (48.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7\u0026nbsp;(25.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e17 (29.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e54 (37.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5\u0026nbsp;(18.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e30\u0026nbsp;(52.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e35 (24.0%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10 (37.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026ndash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e10 (6.8%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e32\u0026nbsp;(51.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5\u0026nbsp;(18.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e10 (17.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e47 (32.2%)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTotal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e146\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e further illustrate how specific visual topics are distributed across the IMRaD structure in \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ\u003c/em\u003e. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA, most of the visual topics are concentrated in the Introduction and Methods sections, suggesting their primary narrative function of supporting conceptual and methodological design of research in \u003cem\u003eDSH.\u003c/em\u003e This is consistent with our earlier finding about the journal\u0026rsquo;s methodological orientation and emphasis. In addition, some visual topics are well represented in the Discussion section, such as \u0026ldquo;Illustration and Iconography\u0026rdquo; (T347) and \u0026ldquo;Infographics\u0026rdquo; (T234). In terms of the distribution of narrative functions, four topics in \u003cem\u003eDSH\u003c/em\u003e\u0026mdash;\"Network Analysis \u0026amp; Visualization\u0026rdquo; (T96), \u0026ldquo;Maps \u0026amp; Geographic Information\u0026rdquo; (T11), \u0026ldquo;Infographics\u0026rdquo; (T234), and \u0026ldquo;Time and Spatial Visualization\u0026rdquo; (T31)\u0026mdash;appear in all four sections. In particular, \u0026ldquo;Network Analysis \u0026amp; Visualization\u0026rdquo; shows a balanced presence across the IMRaD structure, suggesting that it is commonly used for empirical research and for highlighting study results. All other visual topics in \u003cem\u003eDSH\u003c/em\u003e are also fairly well integrated into the IMRaD structure, except in the Results section, indicating that visual topics are holistically interwoven in the DH scholarly narrative.\u003c/p\u003e \u003cp\u003eCompared to \u003cem\u003eDSH, DHQ\u003c/em\u003e demonstrates a more balanced distribution of the narrative roles that visual topics serve (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). \u0026ldquo;Humanistic Visualization \u0026amp; Graphics\u0026rdquo; (T1) emerges as the most predominant visual topic in \u003cem\u003eDHQ\u003c/em\u003e, deeply embedded across the IMRaD structure. Similarly, \u0026ldquo;Archiving\u0026rdquo; (T284), \u0026ldquo;Graph Layouts \u0026amp; Structures\u0026rdquo; (T135), and \u0026ldquo;Text Mining \u0026amp; Networks\u0026rdquo; (T516) also appear in all four sections. Among the remaining topics, \u0026ldquo;Images Atlas \u0026amp; Art\u0026rdquo; (T321) and \u0026ldquo;Gameworld \u0026amp; Interactive Visualizations\u0026rdquo; (T14) are the only ones that lack representation in the Methods section, which may be attributed to their strong association with conceptual papers in the journal.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"DISCUSSION AND FUTURE WORK","content":"\u003cp\u003eThe comparative analysis of \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ\u003c/em\u003e reveals both shared patterns and distinct emphases in how visual discourse is represented in DH scholarship. Both journals demonstrate that visual discourse in DH tends to be diverse yet niche, encompassing a wide range of specific discussions, themes, with each taking up relatively small narrative space within only a few articles. Despite being limited, the visual topic overlap between the two journals\u0026mdash;primarily in areas such as network analysis and temporal or spatial analysis\u0026mdash;suggests a few popular visual themes in DH scholarship. Empirical studies that employ visualization to present analysis and research findings remain a small portion of the publications in both journals, which suggests the lack of this genre of scholarship in DH research.\u003c/p\u003e \u003cp\u003eAt the same time, the journals exhibit important differences in their treatment of visual discourse. \u003cem\u003eDSH\u003c/em\u003e in general has a stronger methodological orientation to visual discussions, while \u003cem\u003eDHQ\u003c/em\u003e presents a more balanced approach incorporating both conceptual and methodological discussions of visual themes. A close examination of the two journals suggests that this practice is aligned with the editorial conventions, histories, and preferences of each journal: \u003cem\u003eDSH\u003c/em\u003e evolved from \u003cem\u003eLiterary and Linguistic Computing (LLC\u003c/em\u003e, 1986\u0026ndash;2015), a journal that originally focused on applying computing techniques to literary and linguistic research, particularly in text analysis and corpus linguistics. Over time, \u003cem\u003eLLC\u003c/em\u003e expanded its scope to cover broader digital methodologies in the humanities, leading to its rebranding as \u003cem\u003eDSH\u003c/em\u003e in 2015. This editorial trajectory helps explain \u003cem\u003eDSH\u003c/em\u003e\u0026rsquo;s continued emphasis on computational methods and data-driven applications in visual scholarship.\u003c/p\u003e \u003cp\u003eBy contrast, \u003cem\u003eDHQ\u003c/em\u003e, which was launched in 2007 as an open-access, web-native publication, was designed to serve as a flexible, community-driven venue for DH research, theory, and practice. Closely affiliated with the Alliance of Digital Humanities Organizations (ADHO), \u003cem\u003eDHQ\u003c/em\u003e embraces a broader range of contributions and has consistently prioritized conceptual and interpretive engagement with digital tools. This orientation is reflected in its more diverse and balanced visual discourse. A further look into the most well-represented visual topic, \u0026ldquo;Humanistic Visualization \u0026amp; Graphics\u0026rdquo; (T1), also supports this interpretation. This topic appears in 43 articles published between 2008 and 2023 and consistently engages with questions about what constitutes ideal visualizations for humanities and cultural data. For example, Drucker\u0026rsquo;s (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2011\u003c/span\u003e) essay \u0026ldquo;Humanities Approaches to Graphical Display\u0026rdquo; proposed an emphasis on \u003cem\u003ecapta\u003c/em\u003e and the idea to adopt a more interpretive and constructivist view of humanistic visualizations. This was also followed by discussions such as \u0026ldquo;generous interfaces\u0026rdquo; incorporating rich, complex, and critical visual approaches to represent cultural heritage materials digitally (Whitelaw, \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). More recent articles have also extended these discussions by developing more specific and creative conceptual models, such as \u0026ldquo;the fold\u0026rdquo; to rethink and critique interactivity in data visualization (Br\u0026uuml;ggemann et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), a \u0026ldquo;hermeneutic visualization\u0026rdquo; for literary studies (Kleymann \u0026amp; Stange, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), and how to leverage a sociological approach to critical design for humanities visualizations (Forberg, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). These works showcase a continuous interest in conceptual engagements with humanities visual design and analysis in \u003cem\u003eDHQ\u003c/em\u003e and the DH communities involved in it. This stream of discussions and publications have likely contributed to the more balanced visual discourse presented in \u003cem\u003eDHQ.\u003c/em\u003e\u003c/p\u003e \u003cp\u003eWhile our findings offer valuable insights, we acknowledge limitations of our study. For example, given the analytical focus on \u003cem\u003eDSH\u003c/em\u003e and \u003cem\u003eDHQ\u003c/em\u003e, the results may primarily reflect the editorial conventions and priorities of these two journals rather than the broader landscape of DH scholarship. We do not claim that the patterns observed in the present study are fully generalizable to all DH journals; rather, we see these findings as a foundation for future empirical studies of visual rhetoric in DH. Furthermore, by focusing exclusively on English-language publications, this study may not fully capture the multilingual and culturally diverse realities of the global DH community. Finally, we recognize that valuable DH work appears in venues well beyond journal publications. Books, conference proceedings, extended abstracts, and presentations also shape the intellectual landscape of the field and can offer valuable insights. We plan to incorporate such materials in future iterations of this research.\u003c/p\u003e \u003cp\u003eIn future research, we aim to expand the scope of our analysis in multiple ways. First, we plan to expand the dataset to include a broader array of multilingual DH journals, as well as additional sources such as conference proceedings and books. This effort will enable us to construct a more comprehensive representation of visual discourse and languages in the field. The ongoing development of the \u003cem\u003eIndex of DH Conferences\u003c/em\u003e (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://dh-abstracts.library.virginia.edu/\u003c/span\u003e\u003cspan address=\"https://dh-abstracts.library.virginia.edu/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) presents a valuable opportunity to enhance the coverage of our corpus. Meanwhile, other ongoing efforts\u0026mdash;such as the curation of a public dataset on DH visual rhetoric (Ma et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) and the development of comprehensive, reusable book metadata from the Library of Congress and the British Library (Li et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2025\u003c/span\u003e)\u0026mdash;highlight the feasibility and promise of this broader direction. These initiatives will collectively lay the groundwork for extending and enriching the present study.\u003c/p\u003e \u003cp\u003eAdditionally, we intend to refine our analysis of co-occurrence networks by further defining and categorizing the nature of connections between topics. Rather than treating all co-occurrences as equivalent, we aim to distinguish between different types of relationships among visual and non-visual topics\u0026mdash;for example, whether they serve complementary, contrasting, supporting, or hierarchical functions in the overall articles and discourse. This deeper semantic and structural differentiation will allow us to capture more nuanced patterns of association and provide a richer understanding of how visual and thematic elements interact within DH scholarship. We will also refine our analysis by constructing time-focused networks to examine how visual discourse evolves over time. Furthermore, by incorporating edge-weighting strategies and contextual metadata, we can begin to explore the strength, frequency, and directionality of topic relationships, which may shed light on underlying rhetorical or epistemic structures in the field.\u003c/p\u003e \u003cp\u003eFinally, we will enhance our network analysis by examining the structural placement of visual discussions within DH scholarship, analyzing how visual themes function across different sections of research articles. Such extended analysis will offer a more nuanced understanding of how visual elements are embedded in DH research narratives and how they shape scholarly communication in the field. Building upon the visual-focused case study, we hope our work may serve as a model for empirically investigating the broader implications of technological integration in DH scholarship.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eEthical Approval\u003c/h2\u003e\n\u003cp\u003eEthical approval is not required for this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInformed Consent\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eInformed consent is not required for this study.\u003c/p\u003e\n\u003ch2\u003eArtificial Intelligence (AI) usage\u003c/h2\u003e\n\u003cp\u003eThe authors used ChatGPT for grammar and stylistics checks. All the content was carefully reviewed, and the authors take full responsibility of the content in this paper.\u003c/p\u003e\n\u003ch2\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/h2\u003e\n\u003cp\u003eOne of the authors, Rongqian Ma, is a member of the Editorial Board Member of Humanities and Social Sciences Communications.\u003c/p\u003e\n\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\n\u003cp\u003eAuthor Contributions: Rongqian Ma\u0026mdash;Conceptualization, Methodology, Investigation, Resources, Data Curation, Writing - Original Draft, Writing - Review \u0026amp; Editing, Funding Acquisition, Project Administration; Pei-Ying Chen\u0026mdash;Conceptualization, Methodology, Investigation, Formal Analysis, Software, Visualization, Data Curation, Writing - Original Draft, Writing - Review \u0026amp; Editing\u003c/p\u003e\n\u003ch2\u003eData Availability\u003c/h2\u003e\n\u003cp\u003eThe original full-text corpus analyzed during the current study are available from the corresponding author on reasonable request given copyright restrictions. Other datasets generated and/or analyzed during the current study are available in the supplementary files.\u003c/p\u003e\n\u003ch3\u003e\u0026nbsp;\u003c/h3\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAlharbi M, Laramee RS (2019) SoS TextVis: An Extended Survey of Surveys on Text Visualization. Computers 8(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003eArticle 1. https://doi.org/10.3390/computers8010017\u003c/span\u003e\u003cspan address=\"Article 1. 10.3390/computers8010017\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArnold T, Tilton L (2019) Distant viewing: Analyzing large visual corpora. Digit Scholarsh Humanit 34(Supplement1):i3\u0026ndash;i16. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/llc/fqz013\u003c/span\u003e\u003cspan address=\"10.1093/llc/fqz013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBenito-Santos A, S\u0026aacute;nchez RT (2020) A Data-Driven Introduction to Authors, Readings, and Techniques in Visualization for the Digital Humanities. IEEE Comput Graph Appl 40(3):45\u0026ndash;57. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/MCG.2020.2973945\u003c/span\u003e\u003cspan address=\"10.1109/MCG.2020.2973945\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBradley AJ, Mehta H, Hancock M, Collins C (2016) Visualization, Digital Humanities, and the Problem of Instrumentalism. \u003cem\u003eIEEE VIS4DH 1st Workshop on Visualization for the Digital Humanities\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrath R (2022) \u003cem\u003eMultimodal analogs to infer humanities visualization requirements\u003c/em\u003e (arXiv:2209.15029). arXiv. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.48550/arXiv.2209.15029\u003c/span\u003e\u003cspan address=\"10.48550/arXiv.2209.15029\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBr\u0026uuml;ggemann V, Bludau M-J, D\u0026ouml;rk M (2020) The Fold: Rethinking Interactivity in Data Visualization. Digit Humanit Q, \u003cem\u003e14\u003c/em\u003e(3). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://digitalhumanities.org/dhq/vol/14/3/000487/000487.html\u003c/span\u003e\u003cspan address=\"https://digitalhumanities.org/dhq/vol/14/3/000487/000487.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCaissie LT, Goggin C, Best LA (2017) Graphs, Tables, and Scientific Illustrations: Visualisation as the Science of Seeing Gerontology. Can J Aging / La Revue Canadienne Du Vieillissement 36(4):536\u0026ndash;548. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/S0714980817000447\u003c/span\u003e\u003cspan address=\"10.1017/S0714980817000447\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChampion EM (2016) Digital humanities is text heavy, visualization light, and simulation poor. Digit Scholarsh Humanit fqw053. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/llc/fqw053\u003c/span\u003e\u003cspan address=\"10.1093/llc/fqw053\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen AT, Cole CL (2021) Reflexivity in Issues of Scale and Representation in a Digital Humanities Project. \u003cem\u003eIEEE VIS4DH 6th Workshop on Visualization for the Digital Humanities.\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://arxiv.org/abs/2109.14184\u003c/span\u003e\u003cspan address=\"https://arxiv.org/abs/2109.14184\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen AT, Raza M, Gazula Y, Huang D, Mendoza J, Andrews WG (2019) Grounding Users in Interpretive Acts: Lessons Learned in the Iterative Design of a Digital Collection. \u003cem\u003eIEEE VIS4DH 4th Workshop on Visualization for the Digital Humanities.\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://vis4dh.dbvis.de/2019/papers/2019/VIS4DH2019_paper_6.pdf\u003c/span\u003e\u003cspan address=\"https://vis4dh.dbvis.de/2019/papers/2019/VIS4DH2019_paper_6.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCiula A, Vieira M, Ferraro G, Ong T, Perovic S, Mucignat R, Valmori N, Deseure B, Mannucci EJ (2021) Small Data and Process in Data Visualization: The Radical Translations Case Study \u003cem\u003eIEEE VIS4DH 6th Workshop on Visualization for the Digital Humanities\u003c/em\u003e. arXiv: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.48550/arXiv.2110.09349\u003c/span\u003e\u003cspan address=\"10.48550/arXiv.2110.09349\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCleveland WS (1984) Graphs in Scientific Publications. Am Stat 38(4):261\u0026ndash;269. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2307/2683400\u003c/span\u003e\u003cspan address=\"10.2307/2683400\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eConroy M, Gillmann C, Harvey F, Mchedlidze T, Fabrikant SI, Windhager F, Scheuermann G, Tangherlini TR, Warren CN, Weingart SB, Rehbein M, B\u0026ouml;rner K, Elo K, J\u0026auml;nicke S, Kerren A, N\u0026ouml;llenburg M, Dwyer T, Eide \u0026Oslash;, Kobourov S, Betz G (2024) Uncertainty in humanities network visualization. \u003cem\u003eFrontiers in Communication\u003c/em\u003e, \u003cem\u003e8\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fcomm.2023.1305137\u003c/span\u003e\u003cspan address=\"10.3389/fcomm.2023.1305137\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCuijuan X, Lihua W, Wei L (2021) Shanghai memory as a digital humanities platform to rebuild the history of the city. Digit Scholarsh Humanit 36(4):841\u0026ndash;857. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/llc/fqab023\u003c/span\u003e\u003cspan address=\"10.1093/llc/fqab023\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeegan M (2014) This ever more amorphous thing called Digital Humanities\u0026rsquo;: Whither the Humanities Project? Arts Humanit High Educ 13(1\u0026ndash;2):24\u0026ndash;41. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/1474022213513180\u003c/span\u003e\u003cspan address=\"10.1177/1474022213513180\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDrucker J (2011) Humanities Approaches to Graphical Display. Digit Humanit Q, \u003cem\u003e005\u003c/em\u003e(1)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDrucker J (2018) Non-representational approaches to modeling interpretation in a graphical environment. Digit Scholarsh Humanit 33(2):248\u0026ndash;263. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/llc/fqx034\u003c/span\u003e\u003cspan address=\"10.1093/llc/fqx034\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEgger R, Yu J (2022) A topic modeling comparison between LDA, NMF, Top2Vec, and BERTopic to demystify Twitter posts. Front Sociol 7:886498. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fsoc.2022.886498\u003c/span\u003e\u003cspan address=\"10.3389/fsoc.2022.886498\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eForberg P (2022) Critical Design as Theory, Experiment, and Data: A Sociologically-Informed Approach to Visualizing Networks of Loss. Digit Humanit Q, \u003cem\u003e16\u003c/em\u003e(3). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.digitalhumanities.org/dhq/vol/16/3/000621/000621.html\u003c/span\u003e\u003cspan address=\"https://www.digitalhumanities.org/dhq/vol/16/3/000621/000621.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGold MK (2012) Debates in the Digital Humanities. University of Minnesota Press\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGold M, Klein L (eds) (2016) Debates in the Digital Humanities. University of Minnesota Press\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGold M, Klein L (eds) (2019) Debates in the Digital Humanities. Univ Of Minnesota\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGold M, Klein L (eds) (2023) Debates in the Digital Humanities. University of Minnesota Press\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGraves H (2014) The Rhetoric of (Interdisciplinary) Science: Visuals and the Construction of Facts in Nanotechnology. Poroi 10(2). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.13008/2151-2957.1207\u003c/span\u003e\u003cspan address=\"10.13008/2151-2957.1207\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrootendorst M (2022) BERTopic: Neural topic modeling with a class-based TF-IDF procedure. \u003cem\u003earXiv preprint arXiv:2203.05794\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.48550/arXiv.2203.05794\u003c/span\u003e\u003cspan address=\"10.48550/arXiv.2203.05794\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHaghighatkhah P, El-Assady M, Fekete J-D, Mahyar N, Paradis C, Simaki V, Speckmann B (2022) Characterizing Uncertainty in the Visual Text Analysis Pipeline. \u003cem\u003eIEEE 7th Workshop on Visualization for the Digital Humanities (VIS4DH).\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/VIS4DH57440.2022.00010\u003c/span\u003e\u003cspan address=\"10.1109/VIS4DH57440.2022.00010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHinrichs U, Forlini S, Moynihan B (2019) In defense of sandcastles: Research thinking through visualization in digital humanities. Digit Scholarsh Humanit 34(Supplement1):i80\u0026ndash;i99. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/llc/fqy051\u003c/span\u003e\u003cspan address=\"10.1093/llc/fqy051\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJ\u0026auml;nicke S, Franzini G, Cheema MF, Scheuermann G (2017) Visual Text Analysis in Digital Humanities. Comput Graphics Forum 36:226\u0026ndash;250. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/cgf.12873\u003c/span\u003e\u003cspan address=\"10.1111/cgf.12873\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKhoo CSG, Tan EAL, Ng S-G, Chan C-F, Stanley-Baker M, Cheng W-N (2024) Knowledge Graph Visualization Interface for Digital Heritage Collections: Design Issues and Recommendations. Inform Technol Libr 43(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003eArticle 1. https://doi.org/10.5860/ital.v43i1.16719\u003c/span\u003e\u003cspan address=\"Article 1. 10.5860/ital.v43i1.16719\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKleymann R, Stange J-E (2021) Towards Hermeneutic Visualization in Digital Literary Studies. Digit Humanit Q, \u003cem\u003e15\u003c/em\u003e(2). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.digitalhumanities.org/dhq/vol/15/2/000547/000547.html\u003c/span\u003e\u003cspan address=\"https://www.digitalhumanities.org/dhq/vol/15/2/000547/000547.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLatour B (1990) Drawing things together. In: Lynch M, Woolgar S (eds) Representation in Scientific Practice. The MIT Press\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLeydesdorff L, Hammarfelt B, Salah A (2011) The structure of the Arts \u0026amp; Humanities Citation Index: A mapping on the basis of aggregated citations among 1,157 journals. J Am Soc Inform Sci Technol 62(12):2414\u0026ndash;2426. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/asi.21636\u003c/span\u003e\u003cspan address=\"10.1002/asi.21636\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi K, Dobreski B, Busch M (2025) How are library cataloging metadata used differently over time: A large-scale quantitative analysis of the Library of Congress catalog. J Doc 81(2):385\u0026ndash;402. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1108/JD-08-2024-0199\u003c/span\u003e\u003cspan address=\"10.1108/JD-08-2024-0199\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLuhmann J, Burghardt M (2022) Digital humanities\u0026mdash;A discipline in its own right? An analysis of the role and position of digital humanities in the academic landscape. J Association Inform Sci Technol 73(2):148\u0026ndash;171. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/asi.24533\u003c/span\u003e\u003cspan address=\"10.1002/asi.24533\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLynch M (1990) The externalized retina: Selection and mathematization in the visual documentation of objects in the life sciences. In: Lynch M, Woolgar S (eds) Representation in Scientific Practice. The MIT Press\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMa R, Li K (2021) Digital humanities as a cross-disciplinary battleground: An examination of inscriptions in journal publications. J Association Inform Sci Technol 73(2). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/asi.24534\u003c/span\u003e\u003cspan address=\"10.1002/asi.24534\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMa R, Li K, Virk RS (2025) GraphEidos: A Dataset of Visual Rhetoric in Digital Humanities. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.researchgate.net/publication/390764252_GraphEidos_a_Dataset_of_Visual_Rhetoric_in_Digital_Humanities/references\u003c/span\u003e\u003cspan address=\"https://www.researchgate.net/publication/390764252_GraphEidos_a_Dataset_of_Visual_Rhetoric_in_Digital_Humanities/references\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eManovich L (2011) What is visualisation? Visual Stud 26(1):36\u0026ndash;49. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/1472586X.2011.548488\u003c/span\u003e\u003cspan address=\"10.1080/1472586X.2011.548488\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMayr E, Windhager F, Liem J, Beck S, Koch S, Kusnick J, J\u0026auml;nicke S (2022) The Multiple Faces of Cultural Heritage: Towards an Integrated Visualization Platform for Tangible and Intangible Cultural Assets. \u003cem\u003eIEEE 7th Workshop on Visualization for the Digital Humanities (VIS4DH)\u003c/em\u003e, 13\u0026ndash;18. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/vis4dh57440.2022.00008\u003c/span\u003e\u003cspan address=\"10.1109/vis4dh57440.2022.00008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeinecke C, Hall C, J\u0026auml;nicke S (2022) Towards Enhancing Virtual Museums by Contextualizing Art through Interactive Visualizations. J Comput Cult Herit 15(4):1\u0026ndash;26. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1145/3527619\u003c/span\u003e\u003cspan address=\"10.1145/3527619\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eM\u0026uuml;nster S, Terras M (2020) The visual side of digital humanities: A survey on topics, researchers, and epistemic cultures. Digit Scholarsh Humanit 35(2):366\u0026ndash;389. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/llc/fqz022\u003c/span\u003e\u003cspan address=\"10.1093/llc/fqz022\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMyers G (1988) Every picture tells a story: Illustrations in E.O. Wilson\u0026rsquo;s Sociobiology. Hum Stud 11:2\u0026ndash;3. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/BF00177305\u003c/span\u003e\u003cspan address=\"10.1007/BF00177305\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePanagiotidou G, Lamqaddam H, Poblome J, Brosens K, Verbert K, Moere V, A (2023) Communicating Uncertainty in Digital Humanities Visualization Research. IEEE Trans Vis Comput Graph 29(1):635\u0026ndash;645. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/TVCG.2022.3209436\u003c/span\u003e\u003cspan address=\"10.1109/TVCG.2022.3209436\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoth W-M, McGinn MK (1998) Inscriptions: Toward a Theory of Representing as Social Practice. Rev Educ Res 68(1):35\u0026ndash;59. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2307/1170689\u003c/span\u003e\u003cspan address=\"10.2307/1170689\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRudwick MJS (1976) The Emergence of a Visual Language for Geological Science 1760\u0026ndash;1840. Hist Sci 14(3):149\u0026ndash;195. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/007327537601400301\u003c/span\u003e\u003cspan address=\"10.1177/007327537601400301\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchetinger V, Raminger K, Filipov V, Soursos N, Zapke S, Miksch S (2019) Bridging the Gap between Visual Analytics and Digital Humanities: Beyond the Data-Users-Tasks Design Triangle. \u003cem\u003eIEEE VIS4DH 4th Workshop on Visualization for the Digital Humanities.\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://vis4dh.dbvis.de/2019/papers/2019/VIS4DH2019_paper_13.pdf\u003c/span\u003e\u003cspan address=\"https://vis4dh.dbvis.de/2019/papers/2019/VIS4DH2019_paper_13.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchwandt S, Wachter C (2024) Visualization as irritation: Producing knowledge about medieval courts through uncertainty. \u003cem\u003eFrontiers in Big Data\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fdata.2024.1188620\u003c/span\u003e\u003cspan address=\"10.3389/fdata.2024.1188620\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSmith LD, Best LA, Stubbs DA, Johnston J, Archibald AB (2000) Scientific Graphs and the Hierarchy of the Sciences: A Latourian Survey of Inscription Practices. Soc Stud Sci 30(1):73\u0026ndash;94\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSollaci LB, Pereira MG (2004) The introduction, methods, results, and discussion (IMRAD) structure: A fifty-year survey. J Med Libr Association 92(3):364\u0026ndash;371\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSpinaci G, Colavizza G, Peroni S (2022) A map of Digital Humanities research across bibliographic data sources. Digit Scholarsh Humanit 37(4):1254\u0026ndash;1268. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/llc/fqac016\u003c/span\u003e\u003cspan address=\"10.1093/llc/fqac016\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSvensson P (2012) Beyond the Big Tent. In Gold (Ed.), \u003cem\u003eDebates in the Digital Humanities\u003c/em\u003e (pp. 36\u0026ndash;49). University of Minnesota Press. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5749/minnesota/9780816677948.003.0004\u003c/span\u003e\u003cspan address=\"10.5749/minnesota/9780816677948.003.0004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eValdivia P, Buono P, Plaisant C, Dufournaud N, Fekete J-D (2021) Analyzing Dynamic Hypergraphs with Parallel Aggregated Ordered Hypergraph Visualization. IEEE Trans Vis Comput Graph 27(1):1\u0026ndash;13. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/TVCG.2019.2933196\u003c/span\u003e\u003cspan address=\"10.1109/TVCG.2019.2933196\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWeingart SB, Eichmann-Kalwara N (2017) What\u0026rsquo;s Under the Big Tent? A Study of ADHO Conference Abstracts. \u003cem\u003eDigital Studies/Le Champ Num\u0026eacute;rique\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(1), Article 1. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.16995/dscn.284\u003c/span\u003e\u003cspan address=\"10.16995/dscn.284\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWhitelaw M (2015) Generous Interfaces for Digital Cultural Collections. Digit Humanit Q, \u003cem\u003e9\u003c/em\u003e(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.digitalhumanities.org/dhq/vol/9/1/000205/000205.html\u003c/span\u003e\u003cspan address=\"https://www.digitalhumanities.org/dhq/vol/9/1/000205/000205.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYan C, Li S (2023) Exploring Multiview Interactive Visualization for Chinese Historical Texts Using Knowledge Aggregation. \u003cem\u003e2023 ACM/IEEE Joint Conference on Digital Libraries (JCDL)\u003c/em\u003e, 165\u0026ndash;168. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/JCDL57899.2023.00031\u003c/span\u003e\u003cspan address=\"10.1109/JCDL57899.2023.00031\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZheng X, Li M, Wan Z, Zhang Y (2024) Knowledge mining and graph visualization of ancient Chinese scientific and technological documents bibliographic summaries based on digital humanities. Libr Hi Tech 42(6):1693\u0026ndash;1721. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1108/LHT-11-2022-0538\u003c/span\u003e\u003cspan address=\"10.1108/LHT-11-2022-0538\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou H, Forbes AG (2022) Data feel: Exploring visual effects in video games to support sensemaking tasks. In \u003cem\u003e2022 IEEE 7th Workshop on Visualization for the Digital Humanities (VIS4DH)\u003c/em\u003e (pp. 6\u0026ndash;12). IEEE. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1109/VIS4DH57440.2022.00007\u003c/span\u003e\u003cspan address=\"10.1109/VIS4DH57440.2022.00007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Digital humanities, visualization, BERTopic, network analysis, narrative function","lastPublishedDoi":"10.21203/rs.3.rs-6458217/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6458217/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDigital humanities, as an interdisciplinary research domain, has been significantly impacted by emerging technological innovations. In recent years, visualizations have played an increasingly prominent role in digital humanities scholarship, gradually driving the field away from its traditionally text-centric orientation. This paper examines how visual topics and themes are addressed in two leading digital humanities journals\u0026mdash;\u003cem\u003eDigital Scholarship in the Humanities\u003c/em\u003e and \u003cem\u003eDigital Humanities Quarterly\u003c/em\u003e \u0026mdash;through full-text analysis of their published articles. Using BERTopic and network analysis, we identify key visual topics in each journal and investigate how these are integrated into scholarly narratives. Our findings show that, while visual discussions are diverse and often idiosyncratic, they remain deeply embedded in digital humanities discourse. \u003cem\u003eDigital Scholarship in the Humanities\u003c/em\u003e tends to emphasize a more methodologically oriented visual discourse, focusing on visualization as a research method or as part of broader methodological debates. In contrast, \u003cem\u003eDigital Humanities Quarterly\u003c/em\u003e offers a more balanced integration of conceptual and methodological perspectives. Building on the notion of visualizations as inscriptions, this study demonstrates how visual elements mobilize humanities ideas and scholarship, offering a foundation for further empirical investigations into visual discourse in digital humanities.\u003c/p\u003e","manuscriptTitle":"What do Digital Humanities Say about visualization? A Bibliometric Exploration","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-24 10:08:04","doi":"10.21203/rs.3.rs-6458217/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"16f45bc2-d1e6-4bd3-b5da-1ab12d90a43b","owner":[],"postedDate":"April 24th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":47532673,"name":"Business and commerce/Information systems and information technology"},{"id":47532674,"name":"Humanities/Cultural and media studies"},{"id":47532675,"name":"Humanities/Language and linguistics"},{"id":47532676,"name":"Social science/Science technology and society"}],"tags":[],"updatedAt":"2025-11-25T09:39:12+00:00","versionOfRecord":[],"versionCreatedAt":"2025-04-24 10:08:04","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6458217","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6458217","identity":"rs-6458217","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}