Challenges to political beliefs might engage brain networks tied to self-reflection, new study suggests

A preregistered brain-imaging study led by Małgorzata Kossowska, a professor at Jagiellonian University's Institute of Psychology in Kraków, found that Polish participants with strongly held left-wing views barely changed their political beliefs when confronted with counterarguments, while they were more willing to revise factual beliefs. Challenges to political beliefs produced greater activity in brain networks linked to self-reflection, which the researchers interpret as evidence that political convictions are tied to personal identity, according to findings published in Scientific Reports that largely replicate a 2016 American study. The earlier U.S. study had also found that participants less willing to change their minds showed higher activity in the amygdala and insula.

People with strongly held left-wing views barely budged when confronted with arguments against their political beliefs, but they were more willing to revise factual beliefs, according to a brain imaging study conducted in Poland. Challenges to political beliefs were accompanied by greater activity in brain networks linked to self-reflection, which the researchers interpret as a sign that political convictions are tied to personal identity. The findings, published in Scientific Reports https://doi.org/10.1038/s41598-026-35397-6 , largely replicate an earlier American study. When people encounter facts or arguments that contradict their existing views, they rarely evaluate the new information like impartial judges. Instead, they tend to engage in a psychological process known as motivated reasoning, which involves filtering, reinterpreting, or dismissing evidence to protect their prior conclusions. This resistance to changing one’s mind is especially pronounced for political beliefs, which are normative statements about how society should work rather than simple matters of testable fact. For many individuals, political stances on issues like abortion or immigration are deeply woven into their sense of self and social belonging. Modern neuroscience has begun to map how these psychological defenses play out physically inside the brain. For instance, a study covered by PsyPost in 2024 https://www.psypost.org/neuroscientists-uncover-brain-mechanisms-underlying-political-myside-bias/ found that reading belief-confirming political messages activates the brain’s reward centers, whereas encountering contradictory messages triggers regions associated with detecting errors. Similarly, research covered by PsyPost in 2020 https://www.psypost.org/neuroscience-study-finds-political-attitudes-can-influence-how-the-brain-responds-to-information/ indicated that liberal and conservative viewers exhibit distinct patterns of brain activity in the prefrontal cortex when watching the exact same political videos, and that alignment with a group’s brain activity predicts adopting that group’s stance. The direct blueprint for the new research comes from a 2016 neuroimaging study https://doi.org/10.1038/srep39589 that compared how American liberals responded to counterarguments targeting their political and non-political views. That earlier experiment showed that challenging a person’s political beliefs increased activity in the default mode network, a collection of connected brain structures along the midline and sides of the brain that becomes active when people look inward and think about who they are. The 2016 study also observed that people who were less willing to change their minds showed higher activity in the amygdala and the insula, two deep brain structures involved in processing fear, disgust, and emotional arousal. Because that earlier experiment took place in the United States, which features a rigid two-party system and a distinct history of partisanship, it remained uncertain whether those brain patterns represent a universal human trait or a product of American politics. The new research, led by Małgorzata Kossowska, a professor at the Institute of Psychology at Jagiellonian University in Kraków and head of the Behavior in Crisis Lab https://becrilab.id.uj.edu.pl , tested whether the same neural responses emerge in a European country shaped by a post-communist transformation, different cultural values, and a multi-party political landscape. By examining both everyday factual beliefs and deeply held political convictions, the researchers sought to determine if identity-protective brain activity crosses cultural borders. The study was preregistered, meaning the researchers publicly recorded their plans before collecting data. “One of the most intriguing questions in psychology is why people hold on to their beliefs even when confronted with information that challenges them,” Kossowska told PsyPost. “This is particularly important in the political domain, where presenting facts or counterarguments often does little to change people’s minds. We were interested in understanding why political beliefs are so resistant to change and whether this resistance reflects something fundamental about how our brains process information.” To investigate this question, the researchers recruited 46 politically engaged Polish adults through social media, though a technical error left a final behavioral sample of 43 participants. The participants had an average age of about 25 years, ranging from 18 to 62, and included 34 women, 8 men, and 1 nonbinary individual. All participants were right-handed, had no history of neurological or psychiatric conditions, and identified as strongly left-wing on a screening questionnaire. Focusing on a single, strongly committed ideological group allowed the team to match the design of the 2016 American study and control for baseline differences in political interest and personal investment. Add PsyPost to your preferred sources https://www.google.com/preferences/source?q=psypost.org Before visiting the laboratory, participants rated their agreement with various claims to verify that they strongly endorsed eight political statements and eight non-political statements adapted for the Polish cultural context. Rather than asking about American topics like gun control, the political statements addressed locally relevant social issues, such as “Abortion should be legal,” “Poland should agree to accept refugees,” and “Sex education should be mandatory for every student.” The non-political statements consisted of factual or widely held empirical claims, such as “A college education generally improves a person’s economic prospects,” “Thomas Edison invented the light bulb,” and “Reading ability in early childhood is a sign of intelligence.” During the laboratory session, participants lay inside a functional magnetic resonance imaging fMRI scanner, a machine that indirectly measures neural activity by tracking changes in blood oxygen. Across four scanning runs lasting about seven minutes each, participants viewed each of the 16 statements they had previously endorsed for 10 seconds at a time. Each statement was immediately followed by five separate counterarguments or pieces of opposing evidence, with each challenge displayed on the screen for 10 seconds. For political items, these challenges consisted of opposing social or moral arguments drawn from public discourse in Poland, whereas non-political items were paired with contradictory factual evidence. After reading all five counterarguments for a given statement, the original claim reappeared, and participants had 12 seconds to rate how strongly they still believed it on a scale from 1 strongly disbelieve to 7 strongly believe . When analyzing the brain scans, the scientists statistically controlled for six head-motion parameters so that physical movement inside the scanner would not distort the neural data. After the session, participants were told that not all of the counterarguments they had read were accurate. Roughly one week after the scanning session, 32 of the participants completed an online follow-up survey in which they rated their belief in the exact same 16 statements once again. The behavioral ratings showed a sharp contrast in how participants responded to factual versus political pushback. When confronted with contrary evidence for non-political statements, participants’ average belief ratings dropped from 6.13 out of 7 before the challenges to 5.37 afterward. When presented with counterarguments against their political beliefs, their ratings held nearly steady, shifting from 5.72 to 5.68, a tiny change that was not statistically significant. That drop for non-political beliefs corresponded to a large effect a Cohen’s d of 1.13, where values of 0.8 or higher are considered large , compared with a negligible one for political beliefs 0.06 . This gap between factual openness and political rigidity persisted one week later among the 32 participants who completed the follow-up survey. Their average belief rating for non-political statements remained lower at 5.65 compared to their initial baseline of 6.07, while their political belief ratings showed no meaningful change, sitting at 5.83 compared to a baseline of 5.79. Inside the scanner, participants also took less time to evaluate political statements after reading the counterarguments, submitting their ratings in an average of 5.16 seconds compared to 5.70 seconds for non-political statements. The authors suggest this half-second difference may reflect quicker, less deliberate judgments about political statements, though other explanations are possible. The brain imaging data showed different patterns for the two kinds of challenges. When participants read counterarguments targeting their political beliefs, they showed heightened activity in the default mode network compared to when they read non-political challenges. Specifically, political challenges triggered greater activation in the medial prefrontal cortex at the front midline of the brain, the posterior cingulate cortex near the back midline, and the right supramarginal gyrus above the ear. The authors interpret this activation as a sign that participants processed political counterarguments as challenges to their personal identity, since these areas are often active during introspection and thinking about the self. On the other hand, reading counterarguments against non-political statements sparked greater activity in brain networks responsible for external attention and flexible thinking. These regions included the dorsolateral prefrontal cortex and the orbitofrontal cortex, which sit behind the forehead and above the eyes and are thought to help people weigh new evidence and revise their expectations. When the researchers examined how strongly the group clung to specific statements across both categories, they found that greater belief resistance corresponded with higher activity in the dorsomedial prefrontal cortex, an upper-front brain area involved in thinking about oneself and others, which the authors link to protecting one’s identity. Meanwhile, greater willingness to change one’s mind on a specific item was linked to stronger activity in the right inferior temporal gyrus and the left temporal pole, areas positioned along the sides of the brain that assist in processing meaning and conceptual knowledge. “The main takeaway is that changing someone’s political beliefs is not simply a matter of presenting better arguments or more information,” Kossowska said. “Our findings suggest that political beliefs are often closely connected to our sense of who we are.” “This suggests that political disagreement is not always just about facts,” she continued. “It can also involve identity, values, and the need to maintain a coherent understanding of ourselves. The practical implication is that if we want to encourage openness to different perspectives, we need to consider not only what information we communicate, but also how that information relates to people’s identities and values.” To see whether emotion-related brain activity tracked belief change, as it had in the 2016 American study, the researchers closely examined the amygdala and three subregions of the insular cortex. In contrast to the earlier American findings, the Polish study found no statistically significant correlation between activity in any of these emotion-related brain structures and the degree to which individual participants changed their non-political beliefs. Additional statistical tests provided moderate evidence that activity in the amygdala and insula was unrelated to belief updating in this group. The authors suggest that holding onto existing views may be driven primarily by cognitive, identity-focused processes rather than emotional arousal. “One particularly interesting finding concerned the role of emotion,” Kossowska explained. “We did not replicate these associations. Instead, our results pointed more strongly toward brain systems involved in self-referential processing and maintaining a coherent sense of identity. This suggests that resistance to changing one’s mind may not necessarily be driven by strong emotional reactions.” The faster reaction times for political items also stood out to the research team. “One might expect politically sensitive issues to require more deliberation, but our findings suggest that strongly held political convictions may sometimes lead people to reach conclusions relatively quickly, perhaps because their positions are already well established,” Kossowska noted. “These findings reinforce the idea that resistance to belief change is not necessarily a matter of intense emotional reactions or extensive deliberation. Sometimes, people may simply rely on deeply established convictions without seriously reconsidering them.” Taken together, the Polish results reproduced the American study’s main behavioral pattern and its default mode network findings, but not its link between emotion-related brain regions and belief change. “By reproducing several key findings of an earlier American study in Poland, we provide additional evidence that these patterns are not restricted to a single national political context,” Kossowska said. “At the same time, the differences we observed, particularly concerning emotion-related brain regions, remind us that replication is also an opportunity to refine existing theories.” There are several methodological caveats to keep in mind when interpreting these results, and one independent researcher views the neuroimaging evidence with skepticism. “This paper provides unsurprising evidence that political beliefs are hard to change,” Paul J. Zak https://scholar.google.com/citations?user=psG2HSUAAAAJ&hl=en , a professor of economic sciences, psychology, and management at Claremont Graduate University and director of its Center for Neuroeconomics Studies https://neuroeconomicstudies.org/ who was not involved in the study, told PsyPost. “The functional imaging results are thin and over-generous in their interpretations. Just observing human beings shows that political beliefs, especially among strongly committed individuals, are unlikely to change.” A primary consideration is that the political statements in the experiment were value-based claims about what society “should” do, whereas the non-political statements were factual claims that can be tested against objective evidence. This distinction aligns with a 2015 neuroimaging study https://doi.org/10.1371/journal.pone.0124596 of 14 adults, which found that evaluating testable factual claims engages different brain regions than evaluating untestable propositions. As a result, the differences in default mode network activity could reflect how the brain processes value judgments versus verifiable facts in general, rather than politics alone. The authors acknowledge this difference but argue it is meaningful rather than incidental, since political beliefs are bound up with values and identity in a way that factual beliefs typically are not. Zak’s broader objection concerns how the brain activity was interpreted. Inferring a specific mental state from activity in a brain area, such as reading activity in the medial prefrontal cortex PFC and posterior cingulate cortex PCC of the default mode network DMN as a sign of self-reflection, is a practice neuroscientists call reverse inference, which Zak referred to as reverse causation. “The authors, like in many neuroimaging studies, misinterpret their findings using reverse causation,” Zak said. “That is, region or network X was more active in one task and then claim this means psychological process Y has occurred. Such a 1:1 mapping in the brain simply does not occur for higher-level cognitive functions. For example, greater political-condition activity in medial PFC/PCC is described as DMN activity reflecting ‘self-referential processing,’ which becomes evidence for the ‘identity-protective function’ of political beliefs.” Kossowska also pointed out that neuroimaging data must be interpreted with caution. “Brain activation does not provide direct evidence of a specific psychological process,” she noted. “Our interpretation in terms of identity protection is consistent with the observed neural patterns and previous research, but it is not the only possible explanation.” The study also relied on a narrow slice of the public, consisting exclusively of strongly left-wing individuals, nearly 80 percent of whom were women. Although recruiting a uniform group helped the researchers isolate specific brain signals, it leaves open the question of whether conservatives or political moderates exhibit the exact same neural responses. In fact, a 2026 behavioral study https://doi.org/10.1002/ejsp.70110 of more than 1,000 Polish adults by several of the same authors indicated that political moderates revise their beliefs more readily than people on either the far left or the far right, suggesting that resistance is tied more closely to ideological extremity than to political direction. Future neuroimaging studies will need to recruit gender-balanced samples across the entire political spectrum to test whether these brain mechanisms operate symmetrically on the right and the left. “First, our findings should not be interpreted as suggesting that left-wing individuals are particularly resistant to changing their political beliefs,” Kossowska emphasized. “We deliberately recruited participants with strong left-wing convictions to replicate an earlier American study using a comparable sample. We did not compare left-wing and right-wing participants, so our results cannot establish whether one ideological group is more resistant to belief change than another.” “Second, our findings do not mean that political beliefs are impossible to change or that people’s responses are neurologically predetermined,” she added. “We examined how participants reacted to particular counterarguments in a controlled experimental setting. Different forms of communication, social contexts, or experiences might produce different outcomes.” Additionally, although Poland has a very different history and party structure than the United States, it shares many institutional and media features with other Western democracies, pointing to the need for future research in non-Western societies. The experiment also measured short-term belief ratings on a 1-to-7 scale inside a laboratory scanner rather than tracking real-world political behaviors such as voting, donating, or activism. Finally, the link between brain activity and belief change for specific statements was based on just 16 statements, with each statement’s results averaged across participants. “This line of analysis needs a better experimental design to compare strongly vs. weakly held political beliefs and see if they can be changed using observable actions, e.g. donating to a political party,” Zak added. “Beliefs are most effectively changed not with facts, but with stories and social proof, that is more interesting and useful research program.” The Polish study also did not measure individual personality traits, such as a person’s need for mental closure, which could help explain why some people are more flexible than others when facing counterarguments. The authors argue that when beliefs are tied to personal identity, simply presenting counterarguments or evidence may not be enough to change their minds. “We are also interested in understanding why some individuals are more willing than others to reconsider their convictions,” Kossowska said. “Factors such as the need for certainty, cognitive rigidity, and the personal importance of political identities may play an important role. Ultimately, we would like to move beyond identifying the mechanisms of belief resistance toward understanding the conditions under which people become more open to alternative perspectives without experiencing those perspectives as threats to their identity. This has implications not only for theories of belief formation and change, but also for political communication, misinformation, and efforts to reduce polarization.” “I think the broader message of this research is that our beliefs are not simply collections of propositions that we evaluate independently of who we are,” she concluded. “They are often embedded in our identities, values, and broader understanding of the social world. This is especially true of political beliefs. When people defend a political position, they may be defending something much more personally meaningful than a particular opinion.” The study, “ Neural correlates of belief change in political and non-political domains among left-wing individuals confronted with counterarguments https://doi.org/10.1038/s41598-026-35397-6 ,” was authored by Małgorzata Kossowska, Paulina Szwed, Gabriela Czarnek, Sofiia Honcharowa, Aleksandra Domagalik, and Łucja Doradzińska.