Multimodal Cinemetrics

Why It Matters

Multimodal cinemetrics equips digital humanities with rigorous tools to decode evolving visual media, reshaping scholarship and industry practices around content creation, attribution, and cultural impact.

Key Takeaways

• Define multimodal cinemetrics as computational study of time-based media.
• Identify units of analysis—shots, scenes, flow, and segments.
• Highlight challenges of scaling computational methods to amateur documentary archives.
• Emphasize theoretical foundations—from Eisenstein to McLuhan—guiding film metrics.
• Note gendered editing history reveals women’s pivotal role in TV style.

Summary

Professor Lauren Tilton opens the session by proposing a new framework called multimodal cinemetrics, which seeks to apply computational methods to the study of time‑based media such as film, television, and emerging digital formats. Drawing on her background in American studies and digital humanities, she frames the effort as a provocation that bridges traditional visual‑culture analysis with large‑scale data techniques.

Tilton outlines the core challenges: defining appropriate units of analysis, handling the heterogeneity of amateur documentary archives, and moving from uncontrolled variables to more tractable forms like sitcoms. She surveys a lineage of film theory—from Eisenstein’s metric and rhythmic editing to Pudovkin’s continuity, Griffith’s pacing, and Metz’s semiotics—while also invoking television scholars such as Raymond Williams, John Ellis, and Marshall McLuhan to argue that media form, not just content, shapes cultural meaning.

Notable moments include Tilton’s critique of the misquoted “the medium is the message,” her citation of Barry Salt’s early multimodality work, and the revelation that women editors historically shaped TV style more than credited. She also references contemporary scholars—Jeremy Butler, Jason Mittell, and the upcoming computational results from her colleague Taylor—to illustrate how genre, flow, and editing decisions can be quantified.

The discussion signals a turning point for cultural analytics: by reconciling theory with algorithmic measurement, researchers can uncover hidden patterns in visual storytelling, reassess canonical narratives, and foreground overlooked contributors such as women editors. This interdisciplinary push promises richer, more nuanced insights for scholars, media producers, and platforms navigating the convergence of film and television.

Original Description

This presentation introduces “multimodal cinemetrics,” tracing the evolution from traditional cinemetrics to an expanded conception of cinema. We explore how AI-enabled analysis of time-based media challenges conventional units of analysis in media studies.

Co-sponsored by the Berkeley Institute for Data Science, the School of Information, and the Department of Scandinavian.

Speaker

Lauren Tilton

Lauren Tilton is the E. Claiborne Robins Professor of Liberal Arts and Digital Humanities at the University of Richmond, Virginia.

Her research focuses on analyzing, developing, and applying digital and computational methods to the study of 20th and 21st century documentary expression and visual culture. Tilton’s first book, Humanities Data in R: Exploring Networks, Geospatial Data, Images, and Texts (2015, second edition forthcoming), built off work applying digital humanities to the study of photography for the digital, public humanities project, Photogrammar (photogrammar.org), which she directs. Her scholarship has appeared in journals such as American Quarterly, Digital Humanities Quarterly, Digital Scholarship in the Humanities, and Journal of Cultural Analytics.

Her work has received support from ACLS, CLIR, NEH, and Mellon Foundation, and she recently finished a stint as a researcher with the Library of Congress as a part of the Computing Cultural Heritage in the Cloud Initiative. Her co-authored scholarship also includes Layered Lives: Rhetoric and Representation in the Southern Life History Project (layeredLives.org), which was released with Stanford University Press in 2022, and Distant Viewing: Computational Exploration of Digital Images (distantviewing.org/book), which was released with The MIT Press in 2023. She is the co-editor of the forthcoming “Debates in the Digital Humanities: Computational Humanities” (University of Minnesota Press). She received her Ph.D. in American studies from Yale University.

More information:

https://www.ischool.berkeley.edu/events/2026/multimodal-cinemetrics