In the online world, news stories easily crawl into social media feeds, where real and fake, fun and serious, sourced and fabricated content pile up on top of each other. In this environment, how do people remember what they see?

Younger and older adults both better remember content of tweets, but when trying to recall the source information, older adults rely a lot more on content to determine where something might have come from, according to a new study, which compared tweets and news headlines, by Kimberly Bourne ’16, Sarah Boland ’17, Grace Arnold ’17, and Associate Professor of Psychology Jennifer Coane recently published in Cognitive Research: Principles and Implications.

“What it suggests is that we can be misled, and we all can misremember where we came across a piece of news or a piece of information,” said Coane, whose research focuses on cognitive psychology. “If you fail to remember the source and you say, ‘Oh, I saw this thing and I saw it on social media,’ you might think it’s less reliable, whereas if you see a piece of news and it looks like it comes from a reputable news source, you might give it more credibility because we rely on both the content and the source—and remembering the source is hard.”

“This was really to suss out: is the content driving the memory for these items? When [the] source is manipulated, how does that impact memory?” —Sarah Boland ’17

This study sprouted in the spring of 2015 when Bourne, Boland, and Arnold were enrolled in “Research Methods and Statistics I and II,” a prerequisite for psychology majors, where students learn to design and execute follow-up studies on recent research. The three, along with Shanna Grant ’17, teamed up with Coane to further explore a study that found content of social media was better remembered.

“Our first question, based on the paper that we had read, was, ‘Oh, this is interesting that we have this better memory for Facebook posts than news headlines, but … there are so many things about Facebook posts that are different from a news headline,” said Bourne, the paper’s lead author and a psychology and Spanish double major now pursuing a Ph.D. in social psychology at the University of Washington. “We were interested in [finding], why is that? Is it the way that this content is being talked about? Is it the way that it’s framed, it’s so visual? That’s when we decided to explore both the content and the format of social media and news headlines.”

They combed through the internet to select 80 real CNN headlines and 80 publicly posted tweets. The researchers said they didn’t want the tweets and headlines to be really memorable but wanted tweet and headline content to be fairly similar in terms of topic, whether it be about sports, food, or politics. From CNN, they picked headlines that read: “Jury finds Eddie Ray Routh guilty in ‘American Sniper’ case,” or “Hush! There’s a secret bar inside this bar.” The tweets they found included: “It’s official. Katy Perry is magic,” or “Getting new tools to the people fighting Ebola is harder than it needs to be.”

Next, they photoshopped these to create four different representations: tweets on a Twitter background, CNN headlines on a CNN background, tweets formatted as CNN headlines, and CNN headlines dressed up as Twitter posts.

This is an original CNN headline that researchers photoshopped to create two versions of the same content.

This is an original CNN headline that researchers photoshopped to create two versions of the same content.

Why?

“We were adding this additional layer of source memory to the research questions,” explained Boland, a psychology major with a minor in anthropology and currently a first-year student at Washington State University’s clinical psychology Ph.D. program. “This was really to suss out: is the content driving the memory for these items? When [the] source is manipulated, how does that impact memory, etcetera?”

This was also how they built onto the previous study and created an even more challenging memory test.

In the study, participants were first shown 80 items—20 tweets on Twitter background, 20 tweets on CNN background, 20 CNN headlines on CNN background, and 20 CNN headlines on Twitter background. Then, they were tested on 160 items, all formatted in black Times New Roman font and presented on plain white background, with two questions: had you seen it before, yes or no; and if they answered yes, had you seen it on a CNN background or a Twitter background?

This is one of the 80 tweets, all of which were publicly available, that researchers pulled out from the Internet and modified for their study.

This is one of the 80 tweets, all of which were publicly available, that researchers pulled out from the Internet and modified for their study.

Testing 42 Colby students (mean age of 20) and 32 healthy older adults (mean age of 71), the findings were revealing.

“When comparing the results that we found for younger adults [and] older adults, … if the content was a tweet they had a better time remembering it, which is pretty interesting considering we thought possibly the exposure to social media may impact that,” she said. In terms of identifying the source, both groups had an easier time when the content and background matched.

Looking deeper, they noticed something interesting in older adults when they couldn’t recall the source. Relying on content and their intuition, they were more accurate at guessing what the item was. “If it sounded like a news headline, it had that tone of being a headline, they made the judgment that it came from a CNN item,” said Boland. But this, they concluded, might make them more susceptible to source errors.

Contributing to the literature with these findings, Arnold, a psychology major who is a third-year student in an industrial-organizational psychology Ph.D. program at George Washington University, felt accomplished.

“What it suggests is that we can be misled, and we all can misremember where we came across a piece of news or a piece of information.” —Associate Professor of Psychology Jennifer Coane

“It’s exciting to not only see all that hard work you did for class had amounted to something really significant and important, but it’s always exciting to see how research findings you find can be shared to the general public,” said Arnold, who enrolled in graduate school immediately after Colby and felt prepared because of all the research experience she had, including this one. “Being able to get involved in a lab, really getting to know professors who guide you through this process, give you advice, and they still let you be in charge of it when needed, which is really great. I’d say the psychology program at Colby has really done a fantastic job of preparing me.”

Coane, who supervised the research and witnessed the students’ growth, was also proud of the outcome.

“I will say it is probably one of the projects I’ve done with students that I’m proudest of, not just because it was published in a really good journal, but because the students started this,” said Coane, who stressed that when they began the study, Bourne was a junior and the other three students were all sophomores. “We have a lot of students doing amazing research in their senior year, like honors theses—these are second-year students doing this project. And it just reminds me of what phenomenal students we have … that they could do a project of such skill and sophistication.”

 


Listen to a podcast where Laura Mickes, whose research inspired this Colby study, interviewed Professor Coane for Psychonomic Society April 17.