pith. sign in

arxiv: 2607.00103 · v1 · pith:NAWTSJYOnew · submitted 2026-06-30 · 💻 cs.HC

Comparing the Emotional Impact of Thematic Versus Episodic Framing in Visualization Text

Poorna Talkad Sukumar , Maurizio Porfiri , Oded Nov This is my paper

Pith reviewed 2026-07-02 17:50 UTC · model grok-4.3

classification 💻 cs.HC
keywords visualization textepisodic framingthematic framingemotional valencedata visualizationsgun controlmass shooting datamediation analysis
0
0 comments X

The pith

Episodic framing in visualization text elicits more negative emotions than thematic framing, indirectly boosting support for gun control.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The study compares how text framing around data visualizations affects viewers' emotions using U.S. mass shooting data. Participants saw the same charts but with different titles and annotations: one focusing on a specific event and others on broader trends. Episodic framing produced stronger negative feelings, which in turn predicted more support for gun control policies. Adding an annotation to a thematic title made no difference to emotions. Although framing had no direct impact on policy views, the emotional path mattered.

Core claim

In a preregistered experiment with 800 participants, episodic framing of mass shooting visualization text caused significantly more negative emotional valence than thematic framing conditions. While direct effects on policy attitudes were not significant, the increased negative emotion mediated greater support for gun control. Annotations added to thematic titles did not change the emotional response.

What carries the argument

The textual framing manipulation in data visualizations, specifically episodic versus thematic approaches, and its measurement through self-reported emotional valence scales.

Load-bearing premise

That participants' self-reported emotional valence on the scales accurately reflects their actual emotional responses to the visualizations.

What would settle it

Finding no difference in negative emotional valence between episodic and thematic framing conditions in a direct replication of the experiment would challenge the central result.

Figures

Figures reproduced from arXiv: 2607.00103 by Maurizio Porfiri, Oded Nov, Poorna Talkad Sukumar.

Figure 1
Figure 1. Figure 1: Experimental conditions showing textual framing manipulations. All conditions display identical bar charts of U.S. mass [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Estimated change in emotional valence by framing condi [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
read the original abstract

Although textual framing in data visualizations is known to influence comprehension, recall, and perceptions of bias, its effects on viewers' emotional responses remain underexplored. Drawing on two widely studied framing strategies in political communication, we examine how episodic framing (foregrounding a specific event) versus thematic framing (foregrounding broader trends) affects emotional and attitudinal responses to visualizations. We conducted a preregistered, between-subjects online experiment (N = 800) in which participants viewed identical visualizations of U.S. mass shooting data that varied only in textual framing: a thematic title, a thematic title with annotation, or an episodic title paired with the same annotation. Results show that episodic framing elicited significantly more negative emotional valence than both thematic conditions. In contrast, adding an annotation to a thematic title did not alter emotional impact. While framing did not significantly affect policy attitudes, mediation analysis revealed a significant indirect effect: increased negative emotion under episodic framing predicted greater support for gun control. These findings position emotion as a critical, yet underexamined, dimension of how textual framing shapes responses to data visualizations.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper reports a preregistered between-subjects online experiment (N=800) in which participants viewed identical mass-shooting visualizations that differed only in textual framing (thematic title, thematic title plus annotation, or episodic title plus annotation). It claims that episodic framing produced significantly more negative emotional valence than the two thematic conditions, that adding an annotation to a thematic title had no effect, and that the increased negative valence under episodic framing mediated greater support for gun control even though framing had no direct effect on policy attitudes.

Significance. If the self-report valence measures are valid, the work supplies empirical evidence that episodic versus thematic framing in visualization text can shape affective responses and, through them, policy attitudes. The preregistration, large sample, and mediation analysis are strengths that would allow the result to extend political-communication framing research into visualization contexts.

major comments (2)
  1. [Methods] Methods section: the manuscript provides no information on the specific items, response format, source, prior validation, or reliability of the emotional valence scales. Because the headline claims (framing effect on valence; valence-mediated policy effect) rest entirely on these measures, the absence of scale details, internal-consistency statistics, or convergent-validity evidence is load-bearing for the central interpretation.
  2. [Results] Results / Mediation analysis: the abstract states that mediation was significant, yet the manuscript supplies no details on the statistical model (e.g., PROCESS macro, bootstrapping parameters, covariates), effect sizes, or sensitivity checks. Without these, it is impossible to evaluate whether the indirect effect is robust or artifactual.
minor comments (2)
  1. [Abstract] Abstract: the phrase 'statistically significant main effects plus mediation' is too vague; it should name the exact tests and report effect sizes or confidence intervals.
  2. [Methods] The three conditions are described clearly in the abstract, but the manuscript should include a table or figure that shows the exact wording of the titles and annotations side-by-side for reproducibility.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive and detailed feedback. We address each major comment below and have prepared revisions to improve the transparency and completeness of the manuscript.

read point-by-point responses

  1. Referee: [Methods] Methods section: the manuscript provides no information on the specific items, response format, source, prior validation, or reliability of the emotional valence scales. Because the headline claims (framing effect on valence; valence-mediated policy effect) rest entirely on these measures, the absence of scale details, internal-consistency statistics, or convergent-validity evidence is load-bearing for the central interpretation.

    Authors: We agree that the Methods section requires expanded reporting on the emotional valence measures to support the central claims. The scales consisted of three 7-point semantic differential items (unpleasant-pleasant, negative-positive, bad-good) drawn from validated valence instruments in the emotion and media effects literature. In the revised manuscript we will add a dedicated paragraph specifying the exact items and response format, citing the source scales and prior validation studies, and reporting internal consistency (Cronbach’s α) and any convergent validity checks performed with our sample. revision: yes

  2. Referee: [Results] Results / Mediation analysis: the abstract states that mediation was significant, yet the manuscript supplies no details on the statistical model (e.g., PROCESS macro, bootstrapping parameters, covariates), effect sizes, or sensitivity checks. Without these, it is impossible to evaluate whether the indirect effect is robust or artifactual.

    Authors: We accept that the mediation results section must be expanded for full evaluability. The analysis followed our preregistration and used the PROCESS macro (Model 4) with 5,000 bootstrap resamples; no covariates were included. The revised Results section will report the complete model specification, unstandardized and standardized indirect effect sizes with 95% CIs, direct and total effects, and sensitivity checks (e.g., outlier robustness and alternative model specifications). These additions will allow readers to assess the robustness of the valence-mediated path to policy support. revision: yes

Circularity Check

0 steps flagged

No circularity: purely empirical experiment with independent data

full rationale

The paper reports a preregistered between-subjects experiment (N=800) collecting new participant responses to visualizations under different framing conditions, followed by standard statistical tests and mediation analysis. No equations, fitted parameters, self-citations, or derivations are present that reduce any result to its own inputs by construction. All load-bearing claims rest on fresh empirical measurements rather than any of the enumerated circular patterns.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim depends on the validity of self-report measures for emotion and attitudes plus the assumption that the online sample and manipulation isolate the framing variable.

axioms (2)
  • domain assumption Self-reported emotional valence and policy attitude scales accurately capture the intended psychological constructs
    The study measures all dependent variables through participant survey responses.
  • domain assumption The online participant sample is appropriate for detecting the framing effects under study
    The experiment relies on an online between-subjects design with N=800.

pith-pipeline@v0.9.1-grok · 5727 in / 1313 out tokens · 41769 ms · 2026-07-02T17:50:30.686309+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

35 extracted references · 25 canonical work pages

  1. [1]

    L. Aarøe. Investigating Frame Strength: The Case of Episodic and Thematic Frames.Political communication, 28(2):207–226, 2011. doi: 10.1080/10584609.2011.568041 1, 2

    work page doi:10.1080/10584609.2011.568041 2011

  2. [2]

    M. A. Borkin, Z. Bylinskii, N. W. Kim, C. M. Bainbridge, C. S. Yeh, D. Borkin, H. Pfister, and A. Oliva. Beyond Memorability: Visualiza- tion Recognition and Recall.IEEE transactions on visualization and computer graphics, 22(1):519–528, 2015. doi: 10.1109/TVCG.2015. 2467732 1, 2, 4

    work page doi:10.1109/tvcg.2015 2015

  3. [3]

    Bylinskii, M

    Z. Bylinskii, M. A. Borkin, N. W. Kim, H. Pfister, and A. Oliva. Eye Fixation Metrics for Large Scale Evaluation and Comparison of Infor- mation Visualizations. InWorkshop on eye tracking and visualization, pp. 235–255. Springer, 2015. doi: 10.1007/978-3-319-47024-5 14 1, 2

    work page doi:10.1007/978-3-319-47024-5 2015

  4. [4]

    Campbell and D

    S. Campbell and D. Offenhuber. Feeling numbers: The emotional impact of proximity techniques in visualization.Information Design Journal, 25(1):71–86, 2019. doi: 10.1075/idj.25.1.06cam 1

    work page doi:10.1075/idj.25.1.06cam 2019

  5. [5]

    C. H. De Vreese. News Framing: Theory and Typology.Information design journal+ document design, 13(1):51–62, 2005. doi: 10.1075/ idjdd.13.1.06vre 1, 4

    2005

  6. [6]

    R. M. Entman. Framing Bias: Media in the Distribution of Power. Journal of communication, 57(1):163–173, 2007. doi: 10.1111/j.1460 -2466.2006.00336.x 1

    work page doi:10.1111/j.1460 2007

  7. [7]

    K. Gross. Framing Persuasive Appeals: Episodic and Thematic Fram- ing, Emotional Response, and Policy Opinion.Political Psychology, 29(2):169–192, 2008. doi: 10.1111/j.1467-9221.2008.00622.x 2

    work page doi:10.1111/j.1467-9221.2008.00622.x 2008

  8. [8]

    ISBN 9781450367080

    J. Heyer, N. K. Raveendranath, and K. Reda. Pushing the (Visual) Narrative: The Effects of Prior Knowledge Elicitation in Provocative Topics. InProceedings of the 2020 CHI Conference on Human Fac- tors in Computing Systems, pp. 1–14, 2020. doi: 10.1145/3313831. 3376887 4

    work page doi:10.1145/3313831 2020

  9. [9]

    Holder and C

    E. Holder and C. Xiong. Dispersion vs Disparity: Hiding Vari- ability Can Encourage Stereotyping When Visualizing Social Out- comes.IEEE Transactions on Visualization and Computer Graphics, 29(1):624–634, 2022. doi: 10.1109/TVCG.2022.3209377 4

    work page doi:10.1109/tvcg.2022.3209377 2022

  10. [10]

    Iyengar.Is Anyone Responsible?: How Television Frames Political Issues

    S. Iyengar.Is Anyone Responsible?: How Television Frames Political Issues. University of Chicago Press, 1994. doi: 10.7208/CHICAGO/ 9780226388533.001.0001 1, 2, 4

    work page doi:10.7208/chicago/ 1994

  11. [11]

    D. H. Kim, V . Setlur, and M. Agrawala. Towards Understanding How Readers Integrate Charts and Captions: A Case Study With Line Charts. InProceedings of the 2021 CHI Conference on Human Fac- tors in Computing Systems, pp. 1–11, 2021. doi: 10.1145/3411764. 3445443 1

    work page doi:10.1145/3411764 2021

  12. [12]

    H.-K. Kong, Z. Liu, and K. Karahalios. Frames and Slants in Titles of Visualizations on Controversial Topics. InProceedings of the 2018 CHI conference on human factors in computing systems, pp. 1–12,

    2018

  13. [13]

    doi: 10.1145/3173574.3174012 1, 2, 4

    work page doi:10.1145/3173574.3174012

  14. [14]

    H.-K. Kong, Z. Liu, and K. Karahalios. Trust and Recall of Informa- tion Across Varying Degrees of Title-Visualization Misalignment. In Proceedings of the 2019 CHI conference on human factors in comput- ing systems, pp. 1–13, 2019. doi: doi/10.1145/3290605.3300576 1, 4

    work page doi:10.1145/3290605.3300576 2019

  15. [15]

    P. J. Lang, M. M. Bradley, B. N. Cuthbert, et al.International Affec- tive Picture System (Iaps): Affective Ratings of Pictures and Instruc- tion Manual. NIMH, Center for the Study of Emotion & Attention Gainesville, FL, 2005. doi: 10.1007/978-3-319-28099-8 42-1 3

    work page doi:10.1007/978-3-319-28099-8 2005

  16. [16]

    Lundgard and A

    A. Lundgard and A. Satyanarayan. Accessible Visualization via Nat- ural Language Descriptions: A Four-Level Model of Semantic Con- tent.IEEE transactions on visualization and computer graphics, 28(1):1073–1083, 2021. doi: doi/10.1109/TVCG.2021.3114770 2

    work page doi:10.1109/tvcg.2021.3114770 2021

  17. [17]

    A. V . Pandey, A. Manivannan, O. Nov, M. Satterthwaite, and E. Bertini. The Persuasive Power of Data Visualization.IEEE trans- actions on visualization and computer graphics, 20(12):2211–2220,

  18. [18]

    doi: 10.1109/TVCG.2014.2346419 4

    work page doi:10.1109/tvcg.2014.2346419 2014

  19. [19]

    Periscopic

    Periscopic. Periscopic. US Gun Deaths, 2013.https://guns. periscopic.com/, 2023. [Online; accessed June 2026]. 4

    2013

  20. [20]

    Peterson and J

    J. Peterson and J. Densley. The Violence Project Mass Shooter Database.https://www.theviolenceproject.org. [Online; ac- cessed 1-Feb-2024]. 2

    2024

  21. [21]

    Mapping what I feel

    V . Prantl, T. Moeller, and L. Koesten. Untangling Rhetoric, Pathos, and Aesthetics in Data Visualization.IEEE Transactions on Visu- alization and Computer Graphics, 2025. doi: 10.1109/TVCG.2025. 3628181 1

    work page doi:10.1109/tvcg.2025 2025

  22. [22]

    M. D. Rahman, B. Doppalapudi, G. J. Quadri, and P. Rosen. A Sur- vey on Annotations in Information Visualization: Empirical Studies, Applications and Challenges.IEEE Transactions on Visualization and Computer Graphics, 2025. doi: 10.1109/VIS60296.2025.00053 2, 4

    work page doi:10.1109/vis60296.2025.00053 2025

  23. [23]

    M. S. Schmidt. F.B.I. Confirms a Sharp Rise in Mass Shoot- ings Since 2000.https://www.nytimes.com/2014/09/25/us/ 25shooters.html, 2014. [Online; accessed 24-Apr-2026]. 2, 4

    2000

  24. [24]

    J. R. Silva and J. A. Capellan. The Media’s Coverage of Mass Public Shootings in America: Fifty Years of Newsworthiness.International Journal of Comparative and Applied Criminal Justice, 43(1):77–97,

  25. [25]

    doi: 10.1080/01924036.2018.1437458 2, 4

    work page doi:10.1080/01924036.2018.1437458 2018

  26. [26]

    W. P. Staff. There have been 7 mass killings with guns in 2026. https://www.washingtonpost.com/nation/interactive/ mass-shootings/. [Online; accessed 24-Apr-2026]. 3

    2026

  27. [27]

    Stokes, A

    C. Stokes, A. Arunkumar, M. A. Hearst, and L. Padilla. An Analysis of Text Functions in Information Visualization.IEEE Transactions on Visualization and Computer Graphics, 2025. doi: 10.1109/TVCG. 2025.3634632 2

    work page doi:10.1109/tvcg 2025

  28. [28]

    Stokes, C

    C. Stokes, C. X. Bearfield, and M. A. Hearst. The Role of Text in Visualizations: How Annotations Shape Perceptions of Bias and In- fluence Predictions.IEEE Transactions on Visualization and Com- puter Graphics, 30(10):6787–6800, 2023. doi: 10.1109/TVCG.2023. 3338451 1, 4

    work page doi:10.1109/tvcg.2023 2023

  29. [29]

    Stokes, V

    C. Stokes, V . Setlur, B. Cogley, A. Satyanarayan, and M. A. Hearst. Striking a balance: Reader takeaways and preferences when integrat- ing text and charts.IEEE Transactions on Visualization and Com- puter Graphics, 29(1):1233–1243, 2022. doi: 10.1109/TVCG.2022. 3209383 1, 2

    work page doi:10.1109/tvcg.2022 2022

  30. [30]

    P. T. Sukumar, M. Porfiri, and O. Nov. Are Mass Shootings in the U.S. Increasing? Understanding How Differing Definitions of Politically Charged Events Impact People’s Perceptions of Expected Trends in Visualizations.IEEE Computer Graphics and Applications, 44(4),

  31. [31]

    doi: 10.1109/MCG.2024.3402790 2

    work page doi:10.1109/mcg.2024.3402790 2024

  32. [32]

    P. T. Sukumar, M. Porfiri, and O. Nov. Connections Beyond Data: Exploring Homophily With Visualizations. In2024 IEEE Visualiza- tion and Visual Analytics (VIS), pp. 231–235. IEEE, 2024. doi: 10. 1109/VIS55277.2024.00054 1, 4

    work page arXiv 2024

  33. [33]

    P. T. Sukumar, M. Porfiri, and O. Nov. Seeing identity in data: Can anthropographics uncover racial homophily in emotional responses? EuroVis 2025 - Short Papers, 2025. doi: 10.2312/EVS.20251078 1

    work page doi:10.2312/evs.20251078 2025

  34. [34]

    Bradley, G

    C. Xiong, J. Shapiro, J. Hullman, and S. Franconeri. Illusion of Causality in Visualized Data.IEEE transactions on visualization and computer graphics, 26(1):853–862, 2019. doi: 10.1109/TVCG.2024. 3456381 4

    work page doi:10.1109/tvcg.2024 2019

  35. [35]

    H. Zhu, S. S. Cheng, and E. Wu. How Do Captions Affect Visu- alization Reading?Workshop on Visualization for Communication (VisComm), IEEE VIS, 2022. 1

    2022