Social Science Bites

One of the promises of artificial intelligence is that it will mimic, and perhaps even improve, on human thinking. One of those hoped-for improvements was that AI would not exhibit human biases. Turns out that in one area, AI can indeed mimic human thinking, and it's in that field of bias. As Harvard psychologist Mahzarin Banaji -- one of the creators of the widely used implicit bias test -- explains in this Social Science Bites podcast, AI platforms both mimic human bias and even amplify it.

In her second appearance on the podcast series, Banaji tells interviewer David Edmonds that even she was surprised how overtly bias shows up in AI results. She recalls her jaw dropping after she queried a large language model about what biases it might have, and it replied "I am a white male," and then how, a month later when queried the same thing it came back with a lengthy 'correct' answer about how it could be biased.

"[W]hat stunned me, and why I began to work on these LLMs, is because it became clear that the creators of these models were actually doing us a massive disservice by creating in these machines two kinds of thought: what the machine knows that it's learned, and now what the machine is going to say, which I'll just call LLM hypocrisy." 

Banaji is the Richard Clarke Cabot Professor of Social Ethics in the Department of Psychology at Harvard, a position she has held since 2002. She is also the first Carol K. Pforzheimer Professor at the Radcliffe Institute for Advanced Study, and the George A. and Helen Dunham Cowan Chair in Human Dynamics at the Santa Fe Institute. A former president of the Association of Psychology Science (2010-11), she was named William James Fellow by the APS and is also a fellow of the American Academy of Political and Social Science, the Society for Experimental Psychologists, Society for Experimental Social Psychology, and the American Academy of Arts and Sciences.

The human brain works very hard behind the scenes even in the most mundane aspects of daily life, like enjoying a nice day or determining the meaning of chit-chat with a friend. Ferreting out the basis and structures of our brain's labor is the domain of Daniel Yon, a psychologist and neuroscientists and director of the Uncertainty Lab at Birkbeck, University of London.

In this Social Science Bites podcast, Yon - author of the 2025 book A Trick of the Mind: How the Brain Invents Your Reality -- details for interviewer David Edmonds why he feels that just as science itself represents a solid - but not "bullet-proof" way of interpreting the natural world, science also well describes how the brain itself does the same.

"I think that at the heart of what I think science and the brain share is this preoccupation with building theories and models based on the data that you've gathered and using those theories to make sense of the world around you. That's a very powerful way to make sense of things," he explains, before adding the caveat, "but it also means that once you start to build your theories and paradigms, they can become the filter and the lens through which everything else gets seen."

Yon's scholarship has earned him a number of honors, such as the Experimental Psychology Society's EPS Prize and the Janet Taylor Spence Award from the Association for Psychological Science. He has also been named a Rising Star by the Cognitive Neuroscience Society and received a mid-career fellowship by the British Academy.

Tom Gilovich finds it fun to study the whys and wherefores of how human beings make sense of the information delivered by the world around them. And why not, he explains to interviewer David Edmonds in this Social Science Bites podcast. "We're dynamic, very complicated creatures who do all sorts of things and sometimes make you go, 'Huh?' That's interesting."

He adds, "At the same time, some of the things that people do have great consequences," which means understanding how understandings come about also has great import.

"A lot of the research on judgment and decision making is that there's a schism between the rational choice and the psychologically compelling choice," Gilovich continues, "and that has provided fertile ground for psychologists like me to explore it: "OK, this is what the rational analysis suggests. Why don't we do that?" And there's often some interesting psychological answers to that. Doesn't make logical sense, but it makes lots of psychological sense."

In that spirit, Edmonds and Gilovich, the Irene Blecker Rosenfeld Professor of Psychology in the Department of Psychology at Cornell University, run through what Edmonds calls "the greatest hits" of Gilovich's research findings. These include the "spotlight effect," which posits that individuals often assume others pay more attention to them than they are, and its cousin, "the illusion of transparency," in which people assume others recognize their feelings and emotions accurately. They also look at regret, bias blind spots, and why third-place finishers are happier than second-place ones.

Gilovich is the co-director of the Cornell Center for Behavioral Economics and Decision Research. He's written or co-written several books, ranging from the academic (the textbook Social Psychology written with Dacher Keitner, Serena Chen and Richard Nisbett), titles that bridge academia and the general public (2002's The psychology of intuitive judgment: Heuristic and biases written alongside Dale Griffin and Daniel Kahneman), and books that bring psychological insights directly to the public (such as 1999's Why smart people make big money mistakes—and how to correct them: Lessons from the new science of behavioral economics with Gary Belsky and 2015's The wisest in the room: How you can benefit from social psychology's most powerful insights with Lee Ross).

This Social Science Bites podcast offers a dollop of good news and heaping helping of bad. The good news is that since the end of American Civil War the economic condition of Back Americans has improved, using as a comparison the presumed status quo population of white Americans. According to Princeton University economist Ellora Derenoncourt, this "wealth gap" has fallen from 60-to-one to six-to-one in the intervening 160 years.

While that's heartening, as Derenoncourt details for interviewer David Edmonds, that six-to-one gap hasn't budged since the 1950s. The academic, the founder and faculty director for Princeton's Program for Research on Inequality, breaks down that stall using historical data, parsing out differences between classes and also discussing the difference between income and assets.

"Income," she notes, "has its own growth process, and income between the two groups has been converging over the last 150 years, and savings from income helped Black Americans accumulate some wealth, driving the racial wealth gap down." But as incomes came closer, accumulated assets and the wealth derived from that have only inched closer, driven in part by generational wealth, especially in housing.

"[F]or most Americans, housing is their wealth," she explains. "And we can keep going down the distribution to ask, '"'When is it the case that white Americans at this point in the distribution are mostly renters versus homeowners?'"' That's where we're going to start to see these dynamics of the wealth gap shift.

Derenoncourt closes with some policy ideas that could accelerate closing the gap, including the politically hot topic of slavery reparations.

There is a value to shared knowledge that tends to go unrecognized because it's so ubiquitous. Nonetheless, experimental psychologist Steven Pinker explains in this Social Science Bites podcast, common knowledge underlies things like paper money, governance, and even coral reefs.
And common knowledge, he makes clear to host David Edmonds, "does not have its ordinary sense of conventional wisdom or an open secret or something that everyone knows, but rather something that everyone knows that everyone knows, and everyone knows that, and everyone knows that, and so on, ad infinitum."
Possing that shared knowledge – and the knowledge that others share that knowledge – creates the conditions for coordination, and thus action beyond what an individual could achieve. That's the reason, he says, "that autocrats fear common knowledge of the regime's shortcomings is that no regime has the firepower to intimidate every last citizen."
Pinker, the Johnstone Professor of Psychology at Harvard University, details his understanding of the virtues and vices of common knowledge in his most recent book, When Everyone Knows That Everyone Knows...: Common Knowledge and the Mysteries of Money, Power, and Everyday Life. The book, his 13th, continues his streak as one of the most publicly recognized of public intellectuals, including recognition as one of Foreign Policy's "World's Top 100 Public Intellectuals" and Time's "100 Most Influential People in the World Today." He is also only the second (so far) returning guest to Social Science Bites, having addressed violence and human nature in a 2012 podcast.