Egalitarians are more aware of inequality

Image of an urban scene.
Enlarge / Do you notice the homeless person’s tent or the city’s skyscrapers?

The COVID-19 pandemic has helped draw attention to the persistent disparities in health care in the US, with minorities and the poor suffering disproportionately worse outcomes from the disease. When it comes to inequality, however, the pandemic is competing for attention with other indications of the US’ problems, with issues ranging from income inequality to police brutality to minority participation in science making headlines.

But perhaps the most striking aspect is that, when polled, a substantial portion of the US public doesn’t seem to recognize that disparities exist, or they think that dominant groups are victims of discrimination. This gap in perception makes it difficult to even discuss inequality, much less decide on policies to address it.

This week, a team of researchers published what may be a partial explanation for why people don’t see things the same way: they don’t actually see the same things. More specifically, people who are sensitive to inequality are simply more likely to notice instances of it.

What you see vs. what you think

When it comes to these sorts of perceptions, every-day language is a bit imprecise. We technically “see” much more than we’re aware of; our brain simply filters out a lot of things that it considers unimportant and keeps us from being consciously aware of them. So there’s really a difference between seeing and noticing. If we give any item further attention, however, we’ll often try to fit it within our existing frameworks of belief and understanding.

That latter point raises the possibility of motivated reasoning being involved. We all tend to filter information in a way that helps reinforce our existing beliefs, questioning the items that seem to contradict them while happily accepting items that agree. There’s obviously a lot of space for our beliefs on inequality to shape our motivated reasoning, and this undoubtedly occurs.

The researchers, however, are actually interested in the earlier steps of this process: how things that are in our visual field rise to the level of our conscious attention. Since we’re not aware of the filtering that takes place, it’s tempting to expect that our beliefs don’t affect this process. But that’s actually not the case. Our visual systems tend to focus attention on things that interest us, allowing our concerns to influence how we perceive the world.

Studying that idea, however, is a challenge. Mostly, you understand what people see by asking them about it. But by that point, motivated reasoning has already played a large role in shaping how people interpret their visual information, as well as what they choose to tell anyone who asks about it. To work around this problem, the research team devised a series of tasks that didn’t directly ask about inequality but make it clear whether or not the test subjects noticed inequality regardless.

Indirect tests

We’ll get to the results in a bit because they’re largely consistent across all the experiments. But we’ll first describe all the different ways the researchers attempted to keep motivated reasoning from interfering with the results. If you don’t care about the experiments, you can just skip down to the next section.

For example, in the first of their studies, the researchers simply showed people images of urban scenes and asked them to report on what they saw. Half the scenes were neutral, but the other half contained indications of inequality, like a high-end sports car or a homeless person. Because there was nothing to overtly trigger motivated reasoning, whether or not someone mentioned the indicators of inequality should provide a better indication of where their attention was directed.

In two other tests, the researchers devised ways of noticing inequality that didn’t leave room for motivated reasoning. For one, they showed drawings of groups of men and women with numerous bags of money in front of them and gave people six seconds to figure out whether either of the groups had more money. In another, they took similar urban images but edited a detail out of the image—in some cases it was a neutral detail, but in others it was one that showed a sign of inequality. They then switched back and forth between the two images quickly and tracked how long it took for people to notice the difference, which is an indication of what attracts their attention.

In yet another experiment, participants were asked to listen to an audio recording in which men and women spoke, but the amount of time given to each varied thanks to editing. Afterward, they were asked about several aspects of the recording, including the division of time between the speakers. But the participants were told there would be a $50 reward for answering the questions accurately, which might be enough to overcome any casual bias in recall.

Finally, the researchers gave participants a set of job applications and employment records for a fictitious company. Each applicant had a different grade point average, college major, race, hometown, and hobby. Whether or not they were “hired” by the company (again, this is all fiction) depended only on the GPA and race, with some people getting hiring decisions that favored whites and others getting decisions that favored minorities. Participants were asked if they noticed anything unusual about the hiring patterns and whether they thought the company should be investigated.

To varying degrees, all of these experiments try to get at what the participants are paying attention to rather than what they think about their perceptions.


If you want to draw conclusions about the connections between beliefs and attention, you have to understand people’s beliefs. To do that, the researchers turned to a set of questions that measured what’s called social dominance orientation. That’s the tendency to accept or actively desire that some groups (generally white males) belong in a dominant position in society. People who score highly on social dominance orientation are more likely to be indifferent to inequalities that conform to their beliefs, since they view that hierarchy as the way things should be.

Although there was some difference between individual experiments, there was a clear trend. Those who were more egalitarian in their outlook were more likely to pick up on indications of inequality that disadvantaged minorities and women than those who scored high on social dominance orientation. There was no indication that egalitarians saw inequalities that didn’t exist—which might have shown up in the experiment involving pictures of money—just that they were more likely to pick up on it when it was there.

In a number of experiments, however, the inequality went against traditionally privileged groups, like males. Here, things were a bit more complicated. Here, the gap between egalitarians and those who favored social dominance was much smaller. On the whole, egalitarians were somewhat less likely to notice these inequalities—and sometimes missed them entirely. By contrast, those who favor social dominance were somewhat more likely to notice these inequalities than they were those that disfavored less powerful groups. But in many cases, these differences weren’t statistically significant.

Overall, if you’re looking for a single explanation for why different groups in the US see the same society through different lenses, this isn’t it. The effects, while consistent, were relatively small and probably can’t explain the large divergence in perspectives within the US population. Of course, motivated reasoning still exists and will be layered on top of this effect outside of these experimental conditions, and the two combined could probably explain a lot.

In any case, the results do make clear that we can view the same thing but come away from it having seen different things, in part based on our ideological commitments. So even if someone is aware of our tendency toward motivated reasoning and tries to avoid it, there’s no guarantee they’ll come away with the same impression as someone with different ideological tendencies.

PNAS, 2021. DOI: 10.1073/pnas.2023985118  (About DOIs).