Summary
Climate and animal rights activists are increasingly resorting to provocative tactics, such as targeting cultural institutions and disrupting sporting events. Often, these actions don’t seem to make much sense - they have what is known as ‘low action logic’ - meaning there is little or no direct connection between the action and the movement’s goals and demands. This research examines the impact of these often controversial tactics on two outcomes: (1) the media coverage of protests, and (2) the level of active support groups receive, measured through financial donations to them. Our analysis shows that lower action logic and higher disruptiveness are associated both with a greater level of media attention and a higher level of active support. A mediation analysis suggests that the increased active support is largely driven by media coverage - that is, protests which are more illogical and disruptive get more media coverage and this drives more people to donate. While further research is needed to understand potential negative public opinion effects of these tactics, our findings help to balance the narrative that illogical, disruptive tactics are purely detrimental to the cause of social movements.
Cover image by AP, used under CC BY 4.0. Available here.
Contents
Activist groups
Active support
Media hits
Identifying and classifying protests
Action logic and disruptiveness coding
Analysis
Protests lead to more donations
Low action logic/high disruptiveness associated with more media hits/donations
Media attention mediates the effects of action logic and disruptiveness
Additional findings
Introduction
There is wide scientific and public consensus that climate change is a threat to human well-being and planetary health. For several decades, scientists have signalled that global carbon emissions need to dramatically reduce if we are to avoid the worst-case scenarios of our heating planet (IPCC, 2023). So far, however, governments have failed to act with the necessary urgency and commitment to make these changes (Watts et al., 2021) and a wide range of actors have risen up in social movements to address the inaction (De Moor et al., 2020).
But which strategies and tactics of social movements are likely to be most effective? This is a question of acute interest and much debate amongst both movement leaders, researchers and anyone with an interest in addressing the crisis we face (Taylor & Van Dyke, 2004). The evidence base, while growing, remains sparse and answers are complicated by the fact that the success of different approaches depends on the context of time and place (Thomas-Walters et al., 2024). Also, though they are focused on similar end goals, social movements are typically composed of sub-groups with different tactics and demands (Downey & Rohlinger, 2008). This has been particularly true of the climate movement which has deployed a wide variety of protest tactics in recent years, ranging from peaceful marches to motorway roadblocks to performative symbolic acts of civil disobedience, such as throwing soup on celebrated artworks (Damien Gayle [@damiengayle], 2022; Lu, 2022). While some hold firm views as to what works (Hallam, 2019), it is generally not well understood what the advantages and disadvantages of different protest tactics are, nor the balance of their trade-offs (Shuman et al., 2023; Thomas-Walters et al., 2024).
So, what do we know so far? There is strong evidence that protests can lead to significant increases in public knowledge about the groups carrying them out (Brehm & Gruhl, 2024; Kenward & Brick, 2023; Schuster et al., 2023). While climate action can move public opinion in a positive (i.e., synergistic) direction (Brehm & Gruhl, 2024; Bugden, 2020; Kenward & Brick, 2023; Shuman et al., 2021), this is not necessarily the case for all protests. Disruptive protests tend to get considerably more media and public attention (Kenward & Brick, 2023; Lochner et al., 2024) and can improve public attitudes towards more moderate groups (Dasch et al., 2024; Ostarek et al., 2024; Simpson et al., 2022), but are often also deeply unpopular and can have backlash effects on public opinion (Feinberg et al., 2017, 2020; Fuller et al., 2025). By contrast, non-disruptive protests tend to be seen more favourably but typically receive little attention (Feinberg et al., 2017, 2020). This points to an ‘activists’ dilemma’: the actions which tend to offer the greatest benefits in raising awareness of an issue, might simultaneously undermine support for it (Feinberg et al., 2020).
Research has often focused on how radical or extreme a tactic is—that is, the extent to which it breaks social conventions or disrupts the public (Feinberg et al., 2020; Giugni, 1999; Orazani et al., 2021). Here, we consider a subtly different topic, which is to do with the perceived ‘logic’ of protest actions. One reason why many disruptive protests might be so unpopular could be that people struggle to see the point: why do activists do what they do? Why would a climate activist throw soup on a painting? Why don’t they protest in front of Parliament or an oil company’s headquarters? This dimension of how intuitively a protest tactic relates to the goals and demands of an activist group is referred to as “action logic”.
Here, we test the idea that it is specifically this element of lacking logic—that jarring quality of certain disruptive protests—that gets the most media attention and encourages people to support a group’s cause. To test this, we assess how different degrees of action logic and disruptiveness impact two important outcomes for social movements, the amount of media coverage they get, and the level of active support they receive, here taking financial donations as a proxy for active support. We look at two major UK protest groups: Just Stop Oil (a climate group) and Animal Rising (an animal rights group). Our main prediction is that protests with lower action logic and higher degrees of disruptiveness get more media attention, which in turn leads to greater active support in the form of direct donations to the groups. The rationale for this prediction is that more disruptive and shocking protests are more newsworthy: the media cover them more, more people hear about them and some of those people feel energised and motivated to support the groups doing those protests.
Method
Activist groups
We drew on data received from the A22 Network (an international community of climate activists engaged in civil disobedience). This included data from several member groups, including on donations and action sign-ups—the number of people expressing willingness to take action in support of a group. However, most groups had incomplete or unreliable data, particularly on action sign-ups. Additionally, the A22 Network noted that they didn’t have the impression action sign-ups reliably predicted actual training attendance.
Given these limitations, we decided to limit this analysis to groups with good data on donations. The groups for which weekly donations data had satisfactory quality and completeness were Just Stop Oil (JSO), Animal Rising (AR), and Direct Action Everywhere (DxE). However, we excluded DxE from the final analysis because nearly all of their protests received the same disruptiveness score (3), leaving little variation to assess potential effects. To ensure our analysis captures the impact of differing levels of protest disruptiveness, we selected Just Stop Oil and Animal Rising for this study. For the purposes of this analysis, Animal Rebellion and Animal Rising are treated as the same group. For Just Stop Oil, we have data for the years of 2022 and 2023, for Animal Rising/Rebellion we have data for 2021-2023.
Active support
To measure active support for activist groups, we use donations as a proxy. Donations serve as a tangible measure of support for a group's growth and organisational capacity. Our analysis focuses on the number of donations made to activist groups, measured as the number of individuals donating per week. We use the sum of the number of donations in a given week plus the following week because the effects of a protest are likely to spill over from one week into the next, especially if protests happened towards the end of the week. While other forms of active support, such as attendance at protests and training sessions, would provide valuable additional insights, the available data on these were too incomplete to include here. Future research could incorporate a broader range of metrics for active support.
Media hits
TThe key mediator variable is media hits, i.e. how many articles are written about the protest group directly following a given protest action. Media hits were obtained from Mediacloud, an open source platform for media analysis, via a Python script. For each week, we take the sum of the media hits in that week plus the following week as the main measure of interest. This is to capture instances where a protest happens during the weekend and the bulk of the media response will spill over into the next week. Future analyses could use additional measures of media attention, such as the reach of a protest (how many people heard about it) or by including social media metrics.
Identifying and classifying protests
To conduct our analysis, we required a time series of protest events, including details on their timing, location, and nature. For this, we relied on the ACLED database, a human-curated dataset that tracks violent and non-violent conflicts, including most forms of protest since 2020. To identify relevant events, we filtered ACLED entries using GPT-4, which we queried via Python code. This process extracted key details such as the date and location of each protest and classified them into predefined protest types, as follows:
● March
● Rally
● March and Rally
● Picketing/gathering
● Sit-in
● Road Blockade/ traffic disruption
● Artwork Targeting
● Oil Terminal/Pipeline Targeting
● Artistic Creation/Performance
● Symbolic Act
● Occupation of Building
● Sport Event Interruption
● Airport Disruption
● Railway Disruption
● Hunger Strike
● School Strike
● Labor Strike
● Legal Action
● Shareholder Activism
● Vandalism
● Property Damage
● Counter Protest
Action logic and disruptiveness coding
We used GPT-4 to code protest actions for action logic and disruptiveness. To refine its responses, we iteratively adjusted our prompts until GPT-4 gave responses that resembled how humans would code the protest events. Initially, we found that the term "action logic" caused inconsistencies in GPT-4's interpretation, tending to assume "logic" referred to rationality or effectiveness rather than the internal coherence between protest methods and stated goals. To resolve this, we replaced the term with "protest transparency," which led to more accurate coding.
We assigned action logic scores ranging from -2 (very illogical) to 2 (very logical) and disruptiveness scores from 0 (not disruptive at all) to 5 (extremely disruptive). The prompts were designed to classify protests as disruptive if they either (a) concretely disrupted public life or (b) strongly deviated from societal norms—or both. The exact prompt wording can be found in our pre-registration document.
To validate the coding, one of the authors manually coded the protests to see if GPT-4 successfully mimicked the intended human judgements. Human and GPT-derived coding for both variables was relatively strongly correlated (ρ > .6). While some variation in tactics and subjective interpretation contributed to this correlation level, the results indicate that GPT-4 effectively captured the intended distinctions. When more than one protest happened in a given week, the action logic and disruptiveness scores for the individual protests were averaged.
Analysis
We conducted Bayesian regression analysis using the R package brms for all analyses with weakly informative priors, four Markov chains, 10000 iterations, and a warmup of 4000. All analyses used baseline-corrected donation and media hit data. Specifically, we subtracted the sum of the number of donations in the two closest weeks in the past without protests, ensuring a baseline that is temporally close and unaffected by protest activity. This is particularly important for the main analyses looking at different protest tactics because protests using similar tactics often cluster together. Without this approach, a traditional time-series baseline (simply using the previous weeks) would risk masking the distinct effects of different tactics, as similar protests would frequently appear in both the treatment and baseline periods.
Our first analysis assessed whether there was an increase in the outcome variables when protests happened (compared to when no protests happened). This analysis does not model action logic or disruptiveness; it simply evaluates whether and to what extent donations differed when there was a protest vs. there was no protest in a given week. We then conducted a second analysis using the number of protests per week as a predictor to test whether a higher protest frequency led to more media coverage and donations. To test for effects of action logic, disruptiveness, media effects, and mediation effects, we used brms models and the mediation package.
Our analysis addressed four key questions:
1. Do action logic and disruptiveness influence donations?
2. Do action logic and disruptiveness influence media coverage?
3. Does media coverage influence donations?
4. Does media coverage mediate the effect of action logic and disruptiveness on donations?
These analyses are summarised in Table 1.
Table 1 – analysis structure table
The method takes a brms model predicting media hits from the main predictor variables and a model predicting donations as input and then calculates the direct effect of action logic on donations, the effect of the mediator (media hits) on donations, and the indirect effects of action logic on donations via media hits. The summary function mediation calculates the direct effects of action logic and disruptiveness, the effect of the mediator, and the indirect effects of action logic and disruptiveness via media hits.
The models include two covariates to account for time-varying effects that may be correlated with the predictor variables. The first is a numeric variable starting at 1 and increasing by 1 for each week in the dataset, thus capturing any linear increases over time (donations or media hits may increase over time and this could be correlated with increasing/decreasing action logic and/or disruptiveness). The second is a categorical variable whose levels are all the different months in the dataset, thus controlling for potential months in which donations were exceptionally high/low due to factors unrelated to protest tactics–for example spikes due to well-publicised climate-induced weather disasters or global climate summits. We further included a covariate for the two activist groups (JSO and AR) to capture differences in donations/media hits attributable to differences between groups that may be confounded with the predictor variables (JSO tended to get more donations and tended to use protests with lower action logic).
The data and analysis scripts for all the analyses below can be found in this OSF repository.
Results
Protests lead to more donations
Fig. 1. Effects of protests on donations. Left panel: Average baseline-corrected donations and 95% credible intervals for weeks where protests happened vs. no protests happened. Right panel: Estimated linear relationship (blue line) between the number of protests in a given week and the number of donations. The shaded grey area shows the 95% credible interval.
Our analysis shows that protests are associated with an increase in donations (Figure 1). In weeks with protests, activist groups on average received 58 more donations than in weeks without protests (95% CrI [16, 100]). Additionally, the frequency of protests matters—each additional protest in a given week is associated with an additional 20 donations (95% CrI [12, 29]).
Figure 2 illustrates the fluctuations in donation levels over time, offering a more granular view of the weekly variability in donations. Although there is some variability in weeks with no protests, it is much greater in weeks when protests occur.
Fig. 2. Overview of the time series donation data for both groups. Dots show the baseline-corrected number of donations in a given week and the following week, with blue/teal dots indicating that at least one protest happened in that week and red dots indicating that no protests happened.
Low action logic and high disruptiveness are associated with more media hits and donations
We examined the relationship between action logic and disruptiveness, on the one hand, and media coverage and donations on the other. For clarity, in this section we refer to ‘action illogic’. We make this change here as high disruptiveness and low logic (i.e. ‘action illogic’) are correlated; and framing the variable in this way makes it easier to interpret in the graphs that follow.
Our analysis finds that higher perceived action illogic and disruptiveness are associated with more media hits, such that a one-unit increase in disruptiveness is associated with 77 additional media hits (95% CrI [6, 146]) and a one-unit increase in action illogic is associated with 96 additional media hits (95% CrI [42, 150]). Additionally, increased media attention is linked to increased donations, such that each additional media hit was associated with 0.23 additional donations (95% CrI [.17, .29]). This means that a protest generating 1000 media hits on average would be expected to result in approximately 230 additional donations. Similarly, disruptive protests are associated with more donations, such that a one-unit increase in disruptiveness is associated with 64 additional donations (95% CrI [18, 109]). While a similar pattern was observed for action illogic, this effect was not robust when all the relevant covariates were included in the model (estimate = 26, 95% CrI = [-9, 59]).
Fig. 2. Effects of action illogic and disruptiveness on donations and media hits. Top panels: Estimated linear effect (blue line) of action illogic (left) and disruptiveness (right) on the number of donations. Bottom panels: Estimated linear effect (blue line) of action illogic (left) and disruptiveness (right) on the number of media hits. The shaded grey area shows the 95% credible interval (which is missing for extreme values for which the dataset did not have enough data points).
Tables 2 and 3 show how action logic and disruptiveness relate to raw donation numbers, providing more detail on how donations are distributed across different levels of action logic and disruptiveness. While in the previous analyses, we treated these as continuous variables, here we have binned them into categories to give a sense of how each category level is related to donations.
Table 2: Binned action illogic to give an overview of how action logic relates to raw donation numbers and to show the distribution of action logic in the dataset.
Table 3: Binned disruptiveness scores to give an overview of how disruptiveness relates to raw donation numbers and to show the distribution of disruptiveness in the dataset.
Media attention mediates the effects of action logic and disruptiveness
The results so far suggest that higher perceived action illogic and higher disruptiveness trigger more media attention which then causes more people to donate to the groups. To more formally evaluate this causal hypothesis of how action logic, disruptiveness, media hits, and donations are connected, we ran a Bayesian mediation analysis. As predicted, it showed that there was a robust indirect effect of (higher) action illogic (estimate = 24 additional donations; 95% CrI [9, 45]) and (higher) disruptiveness (estimate = 19 additional donations, 95% CrI [1, 43]) on donations via media hits.
Our expectation that media coverage mediates donation levels was based on the assumption that most people become aware of protests through news reports, rather than direct exposure. Compared to the number of people who witness protests firsthand, traditional media coverage likely reaches a far broader audience, amplifying the protest’s impact on donations.
Additional findings
Discussion
Climate and animal rights activists increasingly employ controversial tactics with low perceived action logic, where the chosen methods have no obvious connection to activists' goals. While these tactics have sparked extensive debate and concerns over potential backlash, they also serve strategic functions. Our analysis shows lower action logic and higher disruptiveness are associated both with a greater level of media attention and a higher level of active support, as measured via donations. These findings challenge the dominant narrative that illogical and disruptive protest tactics are inherently harmful to social movements. Instead, they highlight how such actions can increase media attention and stimulate active support, providing a more nuanced understanding of the activists' dilemma.
While these findings offer useful insights, further research is needed to determine the extent to which these patterns generalise to other protest groups, movements, and contexts. Our analysis focused on two UK-based activist groups—Just Stop Oil and Animal Rising—that frequently employ at least somewhat disruptive tactics. It remains an open question whether the same patterns would hold for groups that rely more heavily on traditional protest methods. Additionally, media responses to protest tactics may vary across national contexts, suggesting that similar analyses using data from other countries would be beneficial. Moreover, media attention on illogical, highly disruptive protests is likely to diminish over time. Once a tactic becomes familiar—such as throwing soup on artwork—it loses its novelty, making it far less newsworthy.
Future research could also explore why disruptive and illogical protests attract more media attention. Based on our current data, it is difficult to disentangle the specific factors driving media interest. Disruption, as defined here, includes both direct inconvenience to the public (e.g., roadblocks) and more symbolic, attention-grabbing acts (e.g., targeting artwork). Similarly, protests perceived as illogical may attract media attention because they are sensational and scandalous, because they are perceived as innovative, or due to a combination of these or other factors. Understanding which specific characteristics drive media coverage would provide further clarity on the mechanisms at play.
While we used donations as a proxy for active support, other measures of active support would also have been possible, for example, the number of people who participate in a training or protest. However, these data were not available to us in sufficient quality to allow for a systematic analysis. Donation numbers are more straightforward for protest groups to collect and they constitute a tangible way for people to help a movement grow and increase its organisational capacity. As such, we believe that donation patterns reflect, at least in part, how successful protests are in energising supporters and prompting them to take action.
More research is needed to understand the broader effects of illogical and disruptive tactics beyond media attention and active support. While such tactics often generate high levels of media coverage, prior research suggests that this coverage is frequently negative (Scheuch et al., 2024). If this is indeed the case, this may impact the effects such protests have on public opinion (McLeod & Detenber, 1999; Shanahan et al., 2011). Evidence has shown that disruptive and provocative tactics can reduce public support—both for the activist groups themselves and for the broader cause they represent (Feinberg et al., 2017, 2020; Fuller, 2025). Forthcoming experimental studies which we are aware of similarly indicate that disruptive and illogical tactics by climate activists can reduce public support for climate policies. Further research in this area will help in assessing the overall balance of advantages and disadvantages of disruptive and illogical protest actions.
About Social Change Lab
Social Change Lab conducts empirical research on disruptive protest and people-powered movements. Through research reports, workshops, and trainings, we provide actionable insights to help movements and funders be more effective. You can find all of our research projects and resources on our website. You can contact us at info@socialchange.lab.org
This report would not have been possible without data kindly provided by the A22 network, or the generous support of the Climate Emergency Fund.
References
Andrews, K. T., & Caren, N. (2010). Making the News: Movement Organizations, Media Attention, and the Public Agenda. American Sociological Review, 75(6), 841–866. https://doi.org/10.1177/0003122410386689
Brehm, J., & Gruhl, H. (2024). Increase in concerns about climate change following climate strikes and civil disobedience in Germany. Nature Communications, 15(1), 2916. https://www.nature.com/articles/s41467-024-46477-4
Bugden, D. (2020). Does climate protest work? Partisanship, protest, and sentiment pools. Socius, 6, 2378023120925949.
Damien Gayle [@damiengayle]. (2022, October 14). Activists with @JustStop_Oil have thrown tomato soup on Van Gogh’s Sunflowers at the national Gallery and glued themselves to the wall. Https://t.co/M8YP1LPTOU [Tweet]. Twitter. https://x.com/damiengayle/status/1580864210741133312
Dasch, S. T., Bellm, M., Shuman, E., & van Zomeren, M. (2024). The radical flank: Curse or blessing of a social movement? Global Environmental Psychology, 2, 1–33. https://gep.psychopen.eu/index.php/gep/article/view/11121
De Moor, J., Uba, K., Wahlström, M., Wennerhag, M., & De Vydt, M. (2020). Protest for a future II: Composition, mobilization and motives of the participants in Fridays For Future climate protests on 20-27 September, 2019, in 19 cities around the world. https://www.diva-portal.org/smash/record.jsf?pid=diva2:1397070
Downey, D. J., & Rohlinger, D. A. (2008). Linking strategic choice with macro-organizational dynamics: Strategy and social movement articulation. In Research in social movements, conflicts and change (pp. 3–38). Emerald Group Publishing Limited. https://www.emerald.com/insight/content/doi/10.1016/S0163-786X(08)28001-8/full/html
Feinberg, M., Willer, R., & Kovacheff, C. (2017). Extreme protest tactics reduce popular support for social movements. Rotman School of Management Working Paper, 2911177.
Feinberg, M., Willer, R., & Kovacheff, C. (2020). The activist’s dilemma: Extreme protest actions reduce popular support for social movements. Journal of Personality and Social Psychology, 119(5), 1086.
Fuller, K., Ferrali, R., & Damesin, L., Gainsburg, I., Simpson, B., and Willer, R. (2025). Extreme Protest Tactics Reduce Support for the Climate Movement and Climate Mitigation Policies. 10.31235/osf.io/n6aw2.
Gifford, R. (2011). The dragons of inaction: Psychological barriers that limit climate change mitigation and adaptation. American Psychologist, 66(4), 290. https://psycnet.apa.org/journals/amp/66/4/290.html?uid=2011-09485-005
Giugni, M. (1999). How social movements matter: Past research, present problems, future developments. How Social Movements Matter, xiii–xxxiii. https://access.archive-ouverte.unige.ch/access/metadata/929d4ac6-72f9-4e70-9c34-e15a3c992edd/download
Gounaridis, D., & Newell, J. P. (2024). The social anatomy of climate change denial in the United States. Scientific Reports, 14(1), 2097. https://www.nature.com/articles/s41598-023-50591-6.
IPCC (2023). CLIMATE CHANGE 2023: Synthesis Report - Summary for Policymakers. https://www.ipcc.ch/report/ar6/syr/downloads/report/IPCC_AR6_SYR_SPM.pdf
Kenward, B., & Brick, C. (2023). Large-scale disruptive activism strengthened environmental attitudes in the United Kingdom.
Lochner, J. H., Stechemesser, A., & Wenz, L. (2024). Climate summits and protests have a strong impact on climate change media coverage in Germany. Communications Earth & Environment, 5(1), 279. https://www.nature.com/articles/s43247-024-01434-3
Lu, D. (2022, November 12). Throwing soup at the problem: Are radical climate protests helping or hurting the cause? The Guardian. https://www.theguardian.com/world/2022/nov/13/throwing-soup-at-the-problem-are-radical-climate-protests-helping-or-hurting-the-cause
McLeod, D. M., & Detenber, B. H. (1999). Framing effects of television news coverage of social protest. Journal of Communication, 49(3), 3–23.
Mildenberger, M., & Tingley, D. (2019). Beliefs about climate beliefs: The importance of second-order opinions for climate politics. British Journal of Political Science, 49(4), 1279–1307. https://www.cambridge.org/core/journals/british-journal-of-political-science/article/beliefs-about-climate-beliefs-the-importance-of-secondorder-opinions-for-climate-politics/E35B49C0DD4A9F814B4281A00CC42450
Mochon, D., & Schwartz, J. (2024). The confrontation effect: When users engage more with ideology-inconsistent content online. Organizational Behavior and Human Decision Processes, 185, 104366. https://www.sciencedirect.com/science/article/pii/S074959782400058X
Munck Af Rosenschöld, J., Rozema, J. G., & Frye‐Levine, L. A. (2014). Institutional inertia and climate change: A review of the new institutionalist literature. WIREs Climate Change, 5(5), 639–648. https://doi.org/10.1002/wcc.292
Norgaard, K. M. (2006). “We Don’t Really Want to Know”: Environmental Justice and Socially Organized Denial of Global Warming in Norway. Organization & Environment, 19(3), 347–370. https://doi.org/10.1177/1086026606292571
Orazani, N., Tabri, N., Wohl, M. J., & Leidner, B. (2021). Social movement strategy (nonviolent vs. violent) and the garnering of third-party support: A meta-analysis. European Journal of Social Psychology, 51(4–5), 645–658.
Ostarek, M., Simpson, B., Rogers, C., & Ozden, J. (2024). Radical climate protests linked to increases in public support for moderate organizations. Nature Sustainability, 1–7. https://www.nature.com/articles/s41893-024-01444-1
Scheuch, E. G., Ortiz, M., Shreedhar, G., & Thomas-Walters, L. (2024). The power of protest in the media: Examining portrayals of climate activism in UK news. Humanities and Social Sciences Communications, 11(1), 1–12. https://doi.org/10.1057/s41599-024-02688-0
Schuster, M., Bornhöft, S. C., Lueg, R., & Bouzzine, Y. D. (2023). Stock price reactions to the climate activism by Fridays for Future: The roles of public attention and environmental performance. Journal of Environmental Management, 344, 118608. https://www.sciencedirect.com/science/article/pii/S0301479723013968
Shanahan, E. A., McBeth, M. K., & Hathaway, P. L. (2011). Narrative policy framework: The influence of media policy narratives on public opinion. Politics & Policy, 39(3), 373–400.
Shuman, E., Goldenberg, A., Saguy, T., Halperin, E., & van Zomeren, M. (2023). When Are Social Protests Effective? Trends in Cognitive Sciences. https://www.cell.com/trends/cognitive-sciences/fulltext/S1364-6613(23)00261-9?dgcid=raven_jbs_aip_email
Shuman, E., Saguy, T., van Zomeren, M., & Halperin, E. (2021). Disrupting the system constructively: Testing the effectiveness of nonnormative nonviolent collective action. Journal of Personality and Social Psychology, 121(4), 819. https://psycnet.apa.org/journals/psp/121/4/819/
Simpson, B., Willer, R., & Feinberg, M. (2022). Radical flanks of social movements can increase support for moderate factions. PNAS Nexus, 1(3), pgac110.
Taylor, V., & Van Dyke, N. (2004). “Get up, stand up”: Tactical repertoires of social movements. In The Blackwell companion to social movements (pp. 262–293).
Thomas-Walters, L., Scheuch, E., Ong, A., & Goldberg, M. H. (2024). The Impacts of Climate Activism. OSF. https://doi.org/10.31235/osf.io/a7umt
Watts, N., Amann, M., Arnell, N., Ayeb-Karlsson, S., Beagley, J., Belesova, K., Boykoff, M., Byass, P., Cai, W., & Campbell-Lendrum, D. (2021). The 2020 report of The Lancet Countdown on health and climate change: Responding to converging crises. The Lancet, 397(10269), 129–170. https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)32290-X/fulltext?faodatalab=2020-12-03-1
Weber, E. U. (2006). Experience-Based and Description-Based Perceptions of Long-Term Risk: Why Global Warming does not Scare us (Yet). Climatic Change, 77(1–2), 103–120. https://doi.org/10.1007/s10584-006-9060-3
Wien, C., & Elmelund-Præstekær, C. (2009). An Anatomy of Media Hypes: Developing a Model for the Dynamics and Structure of Intense Media Coverage of Single Issues. European Journal of Communication, 24(2), 183–201. https://doi.org/10.1177/0267323108101831