Why Experts Get It Wrong

New research explains why those we rely upon for advice so frequently lead us astray

Ravi Mehta is not vaguely surprised that most high-profile basketball "experts" screwed up their Final Four predictions. Overall, of 5.9 million brackets submitted to ESPN.com's Tournament Challenge by those following the games, only 192 had Butler meeting Virginia Commonwealth in Saturday's semi-finals.

"Yes, I think it's an example of the fallibilities of expertise people can bring to their work," says Mehta, a business administration professor at the University of Illinois. With most basketball experts congregating in Houston, the matter is especially relevant to him because he's just dissected why those we often rely upon for advice can be so wrong.

"Knowing Too Much: Expertise Induced False Recall Effects in Product Comparison," to be published shortly in the Journal of Consumer Research, is specifically about consumer products and why we err in buying into expert opinion. But when I tracked down Mehta, he suggested that the Final Four blunders are of a piece with research co-authored with Joandrea Hoegg and Amitav Chakravarti, associate professors of marketing in the graduate business schools at the University of British Columbia and New York University, respectively.

"There is a blanket assumption that knowledge and expertise are always good," Mehta says. "What we show is that it's not always true. Expertise is a double-edge sword."

There is no shortage of popular literature from various fields, including foreign affairs and business, about smart people making dumb mistakes. David Halberstam's "The Best and the Brightest" remains a primer on botched government decision-making by experts; namely, how key aides to Presidents John Kennedy and Lyndon Johnson, most of them products of an American academic and social elite, got us into the mess of the Vietnam War.

And, for sure, there is previous academic literature suggesting that experts can fall prey to false memories, and thus make false comparisons and inferences, precisely because of their greater-than-average memories of specific subjects.

Mehta and his business school colleagues expand on that literature via four related studies, each concluding that consumer products experts mistakenly, if inadvertently, tend to suggest apples-to-apples comparisons when such may simply not exist.

They open with a far less weighty matter than whether we should have escalated our involvement in Vietnam, namely what experts might erroneously tell you if you were wondering whether to get an Xbox 360 or a PlayStation 3. For sure, there are attributes of each product which are directly comparable, such as a 60 GB hard drive for the Xbox versus a 120 GB hard drive for the PlayStation. But there are many elements which describe just one of the products, such as the BluRay video playback of the PlayStation.

The core problem with an expert dissecting the differences, the researcher found, involves an impulse to "maximize comparability," or stretching to make comparisons and falsely recalling features that simply aren't there for one product. This consistent "false recall" was, they concluded, partly fueled by an expert's sense of accountability and resulting pressure to be, well, an expert.

Such an impulse was not found in non-experts and, interestingly, in experts when the researchers told them not to be worried about consequences as they answered a study's questionnaire. "When we told them they were off the hook, they made better decisions. Their false recall call rate went down!" said Mehta.

They tested their basic hypotheses in four experiments. For example, one examined if reducing an expert's sense of accountability for his judgments could "debias" them and improve their memories. Another inspected whether experts have a greater frequency of false recalls than novices when involved in memory-driven product comparisons.

"One reason for this tendency to provide more thorough and detailed information is that experts naturally feel a heightened sense of accountability for their judgments," the authors write. "It is this greater effort to compare individual features across options that, we believe, leads experts to commit more recall errors than novices."

Expanding on that notion of greater smarts somehow leading to more mistakes, they raise the question of "What is it about [the] feeling of accountability that underlies the recall difficulty?"

Their suspicion is that "feelings of accountability cause experts to focus more on the steps involved in making their judgments, rather than the final decision itself, and ironically this focus leads to memory errors."

In addition, they found that "consumer experts have a far better developed schema than novices; thus, consistent with prior research on false memories, their more complex schemas increase the likelihood of falsely recalling as associative link from memory."

Ultimately, theirs is not a call to rely on amateurs. They agree that expertise has its many benefits. And since they all teach in big-time business schools, I wondered what their own counsel would be to Fortune 500 chief executives who read their work.

"When experts start making decisions based on memory, they start filling in," Mehta told me. "If you have information in front of you, there should not be that problem of false recalls. But if the information is not available, we say that rather than depending on your memory, go back and look for the facts. The chief executive shouldn't just depend on experts' memory or their vast experiences. Get the facts before making a decision."

And, with that said, and having not closely followed the past basketball season, I go unequivocally with Butler and Kentucky in the final. As for the winner, give me until halftime on Monday.