In a 2010 blog post, researcher Michael Inzlicht told the story of the finding that wasn’t. The now-unsupported idea of ego depletion suggested that self-control was limited: You could run out of it if you used up too much. This meant that, for example, if you went to the gym and chose a healthy lunch earlier in the day, both of which required self-control, then you might yell at your spouse later—the reason being that your self-control was depleted. Inzlicht was one of the early researchers of ego depletion, and, as an ethical scientist, he had the courage to accept dissent and speak out when numerous excellent studies didn’t support it (including large meta-analyses). In his words, “the idea that a few minutes of self-control can leave you unable to resist temptation later has been thoroughly debunked.”
In its heyday, though, ego depletion was all the rage. It was heavily covered in the news: As Inzlicht pointed out, even then-President Barack Obama cited it while in office. Why? It’s a splashy finding, and it makes some kind of sense, even though no scientist found a biological mechanism by which it would work. And if we believed the idea of ego depletion, we might just think we spot it in our own lives—confirmation bias in action. We now know that academic journal editors historically preferred exciting findings, particularly in comparison to studies that found no differences—what researchers call null results. And journalists, being people, too, fell for the glamorous findings as well.
No Spoilers
How do we overcome human nature and get science journalists to be more objective in their choices of what research to cover? A new study by Alexa Tullett and colleagues (2026) explored whether journalists would make different decisions depending on whether they knew a study’s results. 413 journalists, editors, and journalism faculty and graduate students were randomly assigned to one of two groups. Half read traditional summaries of research that included results. Half read what the researchers called “results-blind” summaries that omitted the findings.
Participants in both groups preferred studies with characteristics of good research, including that the study had been preregistered (posted beforehand so that researchers couldn’t zhuzh the results to find what they wanted) and that it replicated (had been repeated by others). (For more on replication, see our posts about baseball cards and experts.) This is good news because it suggests the media publication bias that amplified the ego-depletion research might be waning.
But the study also found that journalists, like all of us, exhibit confirmation bias, a phenomenon where we pay attention to information that fits with our beliefs and ignore information that does not (which we’ve written about in relation to Wordle, overcoming cynicism, zombie facts, and political disagreement). The researchers found that the lack of results tended to reduce journalists’ confirmation bias: That is, those who saw the full summaries tended to prefer results that fit with their preexisting beliefs, whereas those who saw the “results-blind” summaries could not be influenced by that tendency. It’s important to note that the study’s finding is a small effect, but any reduction of bias could be important in an era where journalists are aiming for more objectivity.
Take-Aways
The study’s results offer some suggestions for journalistic best practices (and probably best practices for academic journal editorial teams, too!). Preregistration is important. This is the practice where researchers document their methods before collecting data. Journalists could make decisions on what to cover based on the methodology outlined in preregistration reports and not on the eventual findings. After all, an excellent study that does not support its hypotheses likely tells us more than a poorly designed study with a wow finding.
The challenge is that news cycles move fast and readers don’t generally take the time to understand the intricacies of journalistic decisions. Reporters would have to help audiences understand exactly why a particular study is given media attention, even if the finding seems to fall flat. Over time, increased transparency about journalist decisions on what stories to report could lead to a shift in what news consumers expect.
