Verified by 
Psychology seeks to improve our understanding of human minds and behavior in a way that is consistent with the facts. That's the idea, anyway. But what happens when our best answers become, in a sense, unbelievable?
There are various challenges in psychological research. Recent debates over replication and publication practices highlight some. The simple possibility that at some point it all gets too complex for mere mortals to understand is another. Yet we're making tangible progress on these fronts all the time.
But here's another problem that hasn't been generally acknowledged—one that may be fast approaching and that involves centrally the public: What if it turns out that people mostly believe what they want to believe, and that some of the questions we address have answers that almost no one wants to believe?
Rob Kurzban and I have recently run into this problem when studying the public's political views. We took a large database of political and demographic information and sought out the strongest connections. The results provided a pretty clear pattern. It turns out that people are often driven by what can be generally (though not fully) described as self-interest.
Who tends to really like the idea of robust government safety nets? Poorer people with limited access to private social support. Who tends to think that the government ought to stay out of the income redistribution business? Richer, white, Christian men. Who's especially liberal on religious discrimination? People who aren't Christian. The conservatives on these issues? Christians who don't excel at meritocracy (because of less education, etc.). Who's especially liberal on immigration? Immigrants. Who doesn’t like immigration? Native-born whites who don't excel at meritocracy. Who tends to want family planning tools available? People who spend long periods of their lives being sexually active but not wanting to have children. And so on.
But our findings go against people's introspective beliefs. Hardly anyone points to self-interest in accounting for their own views. Instead, people generally claim that they and their political allies are just smart people looking out for what's best for society as a whole (or for some popular subset of society, like women, children, or the middle class). If there's self-interest at work here, most people think it's limited to their political opponents.
Psychologists, though, are often unbothered by disconnects between the data patterns and people's self-descriptions. There’s a long line of research in social psychology and cognitive neuroscience showing that people often aren’t aware of their own motives.
It's worse than that, really. Not only are people often ignorant of their own motives, they're also masters of self-deceptive self-presentation. The introspective "mistakes" are often strategic maneuvers aimed at convincing others that the person is smart, nice, competent, altruistic, principled, and so on. People believe nice-sounding stories about themselves so that they can more effectively sell them to other people.
In explaining the political mind, then, we were ultimately able to arrive at an account that is consistent with psychology and public opinion data. In short, people often prefer political positions that advance their own interests. Yet people also engage in self-deceptive spin—they consciously believe that their own political views stem from high and mighty motives. And it hardly does any good to point this out, because people usually prioritize their preferred policy outcomes and strategic narratives over such bloodless luxuries as empirical coherence.
While the self-deceptive stories appeal to their various constituencies, our account, as far as we can tell, will please no one, including our own political selves (we have political views typical of our researcher demographic). Part of our defense of our efforts, in fact, is that our account is unwelcome to pretty much every conceivable political orientation—right, left, center, libertarian, or whatever.
And so arises a practical problem. We're left telling people something that our own account predicts they probably won't believe. In fact, if our account ends up changing the way lots of people talk about their own political views, then it's probably wrong.
The broader issue is that much of the best social science is trending in this direction, towards explanations that reveal and undermine people's self-presentational biases. When these explanations have no natural constituency—when there's no compelling reason why people ought to believe things they find unpleasant to believe—then we're approaching a different kind of practical limit in social science.
You wouldn't believe us if we told you.
