In a recent study published in Psychological Science, a team of scientists revealed that people generally underestimate the positive connections they can form when discussing political and religious disagreements with strangers. Contrary to popular belief, these discussions often turn out more positively than expected, suggesting that our fears about the consequences of such exchanges may be overstated.
People are often advised to avoid discussing contentious topics like politics and religion in conversations, especially with strangers. This advice is rooted in the belief that such discussions will inevitably lead to negative outcomes, such as discomfort, arguments, or even lasting animosity.
However, this perspective may be overly cautious and could contribute to greater social polarization. As political and religious divides widen, especially in societies like the United States, the lack of engagement across differing viewpoints may reinforce echo chambers and diminish empathy between groups.
Researchers were driven to investigate whether the actual outcomes of such conversations matched the negative expectations. Prior studies had hinted that people might naturally avoid those who disagree with them, not just out of discomfort but also due to an innate preference for homophily—associating with similar others. Yet, this might lead to missed opportunities for connection and understanding.
“I’m interested in how people can have better social interactions, including how they can best approach discussing disagreement,” said study author Kristina A. Wald, a postdoctoral researcher at the University of Pennsylvania. “Political disagreement seems like an area people especially want to avoid, out of the fear that discussing it will become hostile and unpleasant, so my collaborators and I wanted to test whether people’s fears are actually warranted.”
In the first experiment, the researchers recruited 471 participants through an online platform, ensuring a diverse demographic by including various age groups, genders, and racial backgrounds. Participants were asked to express their opinions on a set of nine potentially divisive topics, which included sensitive subjects like abortion, climate change, and belief in God.
Participants were then randomly assigned to either an agreement or disagreement condition for one of the topics chosen randomly, except those topics where the participant had neutral feelings (neither agree nor disagree). They were asked to imagine a 10-minute conversation about this topic with a stranger who either shared or opposed their views. Participants subsequently reported their expected enjoyment, awkwardness, learning opportunities, hostility, liking towards the partner, and perceived connection and likability from the conversation partner.
When participants anticipated discussing a topic with someone who agreed with them, their expectations for a positive conversation experience were higher compared to those expecting to talk with someone who disagreed. This expectation significantly influenced their willingness to engage in the conversation: participants were generally more inclined to avoid discussions they expected to be contentious and less rewarding. These findings underscore the idea that people’s reluctance to engage in potentially divisive discussions could stem from negative expectations.
Experiment 2 shifted the focus from hypothetical scenarios to actual interactions to test the accuracy of participants’ expectations against real conversation experiences. This time, 198 participants were recruited to participate in a lab setting. They were paired based on their pre-discussed stances on the same nine topics used in Experiment 1, creating conditions of agreement, disagreement, and uncertainty (where the stance of the conversation partner was not revealed beforehand).
Before the conversation, participants reported their expectations regarding how positive the interaction would be. They then engaged in a 10-minute discussion, following which they were asked about their actual experiences. The conversations were also recorded and later evaluated by researchers to assess the level of agreement or disagreement expressed and the overall tone of the interaction.
The results demonstrated that participants, regardless of whether they agreed or disagreed with their conversation partner, consistently underestimated the positivity of the interactions. This effect was most pronounced in the disagreement condition, where the disparity between expected and actual positive experiences was greatest.
“In our experiments, very few participants reported a negative experience after their conversation (that is, participants’ experiences were not only better than their expectations, but also positive in an absolute sense),” Wald told PsyPost. “And we found that people’s experiences discussing these topics with someone who disagreed with them were actually equally as positive as their experiences discussing with someone who agreed with them (despite the fact that their expectations were far more negative in the former case than the latter).”
The third experiment explored the dynamics of conversation versus monologue in the context of agreement or disagreement. The researchers recruited 240 participants from both an online platform and a university-based online lab. The experiment was structured to allow some participants to have a live dialogue where both individuals could interact freely, while others were placed in a monologue condition where they recorded their views and then listened to their partner’s recorded views without live interaction.
Participants first completed a questionnaire on their political and religious views and were matched with a partner who either agreed or disagreed with them on a selected topic. Expectations were measured before the interaction, which was either a dialogue or a monologue based on their assignment. Following the interaction, participants once again evaluated their experience, allowing researchers to compare expected and actual outcomes across different modes of communication (dialogue vs. monologue) and types of agreement (consensus vs. dissent).
The researchers found that the format of the interaction played a role in shaping both expected and actual outcomes. Participants in dialogue conditions reported more positive experiences than those in monologue conditions, indicating that active engagement and reciprocal exchange offered more satisfaction and connection, even in cases of disagreement.
The results showed that participants expected dialogues to be more contentious when disagreements were involved but found these interactions to be surprisingly positive. This was especially true when comparing their experiences in dialogues versus monologues; dialogues not only provided a platform for more nuanced understanding but also revealed more areas of common ground than participants expected.
“Discussing political and religious disagreement with a stranger is not as bad an experience as people expect it to be!” Wald said. “So people may be avoiding discussing divisive political issues more than they would if they knew what these conversations were really like, and may thus miss out on opportunities to have informative conversations and to create social connection.”
But the study, like all research, includes some caveats.
“Our experiments were only conducted among strangers, so we do not know how these findings would translate to conversations among existing friends or family members,” Wald explained. “We also did not examine typed interactions (e.g., engaging via social media posts), and in fact, one of our experiments suggests that the dynamic nature of real conversations is a key driver of our effects (that is, when participants simply recorded videos of themselves stating their opinions in a monologue and then listened to the other participant’s recording—rather than having a back-and-forth conversation—we did not observe our effects as strongly).”
Future research could explore more naturalistic settings and examine the long-term effects of such interactions on social relationships. Additionally, further studies might investigate whether similar miscalibrations occur in other interpersonal situations beyond politics and religion.
“Generally, I’m interested in studying a variety of ways people can improve their interpersonal interactions,” Wald said.
The study, “Misplaced Divides? Discussing Political Disagreement With Strangers Can Be Unexpectedly Positive,” was authored by Kristina A. Wald, Michael Kardas, and Nicholas Epley.