A pair of experiments conducted in San Francisco indicate that exposure to videos depicting local crime can nudge liberal voters toward tougher, more conservative positions on crime policy at the municipal level, without shifting their broader political identities. The research focused on California’s Proposition 47 and was conducted by a team led by University of San Francisco psychologist Violet Cheung-Blunden.
The researchers frame their work within terror management theory, which suggests that reminders of death or existential threat provoke anxiety. To cope, individuals may seek comfort in structured, predictable and historically conservative worldviews.
Across decades of study, terror management theory has produced mixed findings on how fear or threat affects voting. Some experiments have pointed to a shift toward the political right, while others have observed polarisation that pushes liberals and conservatives further apart.
The study’s California setting provided a real‑world test bed. Proponents and critics have debated Proposition 47, a policy designed to reduce prison populations by classifying minor thefts as misdemeanours; opponents argue that a $950 threshold for felony charges invites organised retail theft with limited consequences.
In the first experiment, 206 adults from the San Francisco area were recruited in the week before an election. Participants answered questions about their beliefs on authority, immigration and economics, then were randomly assigned to view one of two short videos: a baseline tour of local sites such as the Golden Gate Bridge, or a compilation of news clips showing organised looting and retail theft.
After viewing the videos, participants completed a mock exercise to revise Proposition 47, proposing an ideal dollar threshold for classifying retail theft as a felony, and setting minimum and maximum ballot thresholds. They also rated their current political orientation on a liberal–conservative spectrum.
Those who watched the crime footage proposed lower minimum and ideal thresholds for felony classification than those who watched the tourist video, indicating a move toward a tougher, more punitive stance on retail theft. Importantly, this shift on the policy question did not coincide with a broader shift in participants’ self‑reported political identities.
To determine whether exposure to crime media also altered overall ideology, researchers assessed whether liberal participants became more liberal or conservative participants more conservative. The data did not support this polarization effect; general ideological footing remained largely unchanged.
A second experiment, conducted with 98 San Francisco college students in a controlled laboratory setting, replicated the first using a nearly identical procedure. Before viewing the assigned footage, participants watched a five‑minute breathing and relaxation video to stabilise mood and reduce baseline variation.
The results again showed a shift toward stricter crime policy thresholds among those who viewed the crime video, with lower proposed felony thresholds across all categories. As in the first study, there was no broad realignment of overall political orientation or evidence of polarisation.
Limitations acknowledged by the researchers include the highly liberal demographic of the city and a sample with a high proportion of female participants, which may affect generalisability to more conservative or politically divided regions. The experiments relied on a mock policy threshold and a general ideology scale, suggesting future work could incorporate a wider array of political choices, including actual electoral matters.
The researchers note that it remains unclear whether the crime videos triggered a specific fear of mortality as posited by terror management theory, or simply heightened general anxiety about safety and social order. They call for further studies to measure the exact emotional drivers behind such policy shifts.
The study, Threat-Induced Conservatism in a Liberal Electorate: Issue-Based Shifts without Ideological Transformation, is authored by Violet Cheung-Blunden, Serena Rajkumar, Yuka Hirata, Caelyn A. McDonough and Chonhei Liao.
