A study headed by a professor at UT Austin's McCombs School of Business asked 80 undergraduates to identify which social media news headlines were real and which ones were fake. The person who did the best in detecting fake news got 66% of the headlines correct. That's a D.
Tricia Moravec is an assistant professor of information management at McCombs. She and researchers from the University of Hawaii at Manoa and Indiana University set out to answer three questions about fake news and social media:
- Can people detect fake news on social media?
- The Disputed Flag that Facebook designed to help people identify false news was debuted in 2017, the same time the study was conducted. Did the flag make people more skeptical of articles?
- What's going on cognitively as people are presented with fake news and then also presented with this warning flag?
The results — released in November — showed that these undergraduates were not very good at all at detecting fake news on Facebook. Moravec says 75% of study participants got less than half of the headlines correct.
Moravec calls the results of this study "particularly concerning" and points to several reasons for the poor performance in detecting fake news headlines.
For many of the headlines, "it's not obvious whether it's true or false," Moravec says. "And you have to be very aware of what's going on in the world to actually be able to assess this based on a variety of different sources."
She also says people cling to their established beliefs tightly, even when faced when contradictory information.
"If something aligns with our beliefs, we believe it to be true, and if it doesn't align, then we automatically are inclined to believe it to be false," Moravec says. "And so it means that the real truth has very little influence on what we actually believe."
Moravec says the context of the information being consumed also impacts how that information is processed.
"If we're going to get news at an actual news site, we are more in a mindset of intentionally consuming news," Moravec says. "Whereas when we're on Facebook, we see memes; we see videos of food and kittens; and then you see a political article. And so how are you supposed to just rapidly switch mindsets?"
Listen to the interviews below or read the transcript to learn more about why people are not always good at detecting fake news and to find out how to separate real news from fake.
Listen to the interview with UT Austin McCombs School of Business Assistant Professor Tricia Moravec:
Most of us like to think we could easily discern fake news headlines from the real thing. But we might not be as skilled at it as we think. Take this quiz using some of the headlines from Moravec's study to find out:
Transcript
This transcript has been edited lightly for clarity.
Tricia Moravec: The first question was: can people detect fake news on social media? The second question was: does the Disputed Flag that Facebook had debuted at that time — did that work to help people be more skeptical of the articles that had that Disputed Flag on it? [The study was conducted before Facebook removed its Disputed Flag warning late in 2017.] And the third question was: what's going on cognitively as people are presented with fake news and then also presented with this warning flag?
We studied that using electroencephalography or EEG, which is where we put a headset on their head and measured their cognition.
KUT: When you're measuring people while they're figuring out whether they think this is a true headline or a false headline — what are you looking at going on in their brains?
Moravec: We actually timed how long they looked at each of these headlines. And then what we did was look at the last four seconds before they moved on. So what we're doing here is looking at whether they think about these headlines differently within subjects. So whether when they see this headline that’s flagged … given their confirmation bias, do they actually believe this more or less based on their a priori beliefs?
KUT: What is confirmation bias and how does that intersect with all this?
Moravec: Confirmation bias is our tendency to believe information that aligns with what we already believe. We already believe something to be true, so if we see something that fits within that world, that schema, then we believe it more.
KUT: So, what did you find?
Moravec: We were studying undergraduates, so that is a limitation of our study. But what we did find is that they were not able to detect fake news. So this first question that we asked was can they detect fake news, and they couldn't. And the way we measured this was looking at how believable, credible or truthful they thought these headlines were. Of all of our 80 subjects, the person who did the best in detecting fake news got 66% of the headlines correct, which is a D.
KUT: And that's only one person out of 80 got a D.
Moravec: Seventy five percent of people did worse than chance. So they did worse than 50 percent.
KUT: Why do you think that was the case?
Moravec: A lot of these headlines — it's not clear. It's not obvious whether it's true or false. And you have to be very aware of what's going on in the world to actually be able to assess this based on a variety of different sources. You know, it's not just these students that aren't good at this. All of us aren't good at it. And it's because if something aligns with our beliefs, we believe it to be true. And if it doesn't align, then we automatically are inclined to believe it to be false. So it means that the real truth has very little influence on what we actually believe.
KUT: Well, that's frightening.
Moravec: It is frightening. And sometimes doing this research, I feel very hopeless about how we can deal with this and what we can do with this issue, fake news. But then sometimes — because we have a lot of different research projects that we're working on — some of the results will give me some hope. This study was particularly concerning.
KUT: The fake news designations — you all used those on some of the headlines that you had. Did that have any impact on people's processing of the headlines and how they were able to perform on assessing them?
Moravec: We did use Facebook’s “disputed by third-party fact-checkers.” That was their little warning message. When this warning message was put on headlines, it did not have a statistically significant influence on people's belief in those headlines. The flag didn't work as intended. It didn't change people's level of belief in the headlines.
What it did do, though: people thought about those headlines longer. And so it seemed like there is some ability for that warning sign to create cognitive dissonance, which means that we're presented with two alternate facts that are contrary and we have to decide which one's true. This looks like people did have some cognitive dissonance. They thought about it longer, but it didn't change their actual belief in the headline.
KUT: What then is the intersection of the fact that it's social media? That obviously has an impact on how we're processing all of this?
Moravec: If we're reading print news or we're reading on an actual news site, depending on how reputable that news site is, we probably won't come across very much fake news. Social media doesn't have that journalistic oversight. And so we see a lot more fake news than we would if we were on a reputable news site. And then the other problem is that mindset.
If we're going to get news at an actual news site, we are more in a mindset of intentionally consuming news, whereas when we're on Facebook, we see memes. We see videos of food and kittens, and then you see a political article. And so how are you supposed to just rapidly switch mindsets to be able to deal with the complexity of the news you're seeing? You can't.
KUT: Your research is showing that we are not very good at detecting fake news headlines and sorting out fake news or real news headlines on Facebook. Our confirmation bias is heavily at play. What do we do with this now?
Moravec: Do not use social media as your primary news source. But if you are seeing news on social media, there are some things you should do. The first thing is when you see news, look at the source. If the source is a source that you know to be reputable, that's a step in the right direction. And then read the headline. If it seems like it's important or influential at all, look it up. Look, search for it and make sure that other news sources are presenting it in a similar way.
And then ideally, get your news outside of social media from a variety of different news sites. And understand the bias of the sites that you're on. Even though it's uncomfortable for us to try to read news that might come from a different bias from our own world view, doing that is going to help you be less susceptible to fake news.
