Eye-tracking study suggests that negative comments on social media are more attention-grabbing than positive comments

When it comes to reading news stories on social media, new psychology findings suggest that people pay more attention to negative comments than positive ones. This evidence comes from an eye-tracking study published in the journal Computers in Human Behavior.

News posts shared on social media often attract emotionally-charged comments. Moreover, these comments are often negative, which can have harmful consequences. For example, emotional comments under news stories can inspire distrust in news sources and attitude extremity among users. However, given that social media is inundated with content, the extent that users pay attention to these emotional comments is unclear.

Study author Susann Kohout and her colleagues designed an eye-tracking study to investigate the extent that people pay attention to and remember emotional content on social media. They also explored the extent that people pay attention to negative, positive, angry, and fearful content.

In a Dutch university laboratory, 169 students sat in front of an eye tracker while they were shown three social media news posts. The posts were artificial news stories designed to resemble Facebook posts. The posts were each accompanied by four comments which varied in the extent that they were emotional, non-emotional, positive, negative, angry, or fearful.

The students were divided into two groups — the heuristic processing group and the systematic processing group. Participants in the heuristic processing group were given only 30 seconds to read the posts. This condition was meant to mirror the low-effort processing that people usually engage in when using social media. Participants in the systematic processing group were allowed to read the posts carefully and with no time limit.

The eye-tracker measured the students’ eye movements, which the researchers used to assess their visual attention. Specifically, the researchers calculated participants’ dwell time for each comment and news story by adding up all fixations, saccades, and revisits to each area of interest. Later, they measured participants’ recognition of the posts with a survey.

Kohout and her team found that the students showed longer dwell times for negative compared to positive comments, but only in the heuristic processing condition. This suggests that when the students were forced to read the comments quickly, they read the negative comments more often than the positive ones.

The authors say that these findings fall in line with the negativity bias, which is the notion that people place more importance on negative information than positive information. However, participants were not more likely to recognize information about the negative posts compared to the positive ones. This suggests that participants may have been avoiding this negative information or suppressing it so they do not remember it later.

Students in the systematic condition showed longer dwell times and greater recognition for the angry comments compared to the fearful comments. This suggests that when the students were given ample time to read the comments, they were more likely to read and remember the story details of the angry comments over the fearful ones.

The study authors note that their study should be considered a precursor to future research given several limitations. For one, they were not able to consider a wider array of emotions or different social media interfaces other than Facebook. They were also unable to manipulate the ordering of the social media comments. Nonetheless, the findings reveal important insights.

“First, we have shown that it is important to distinguish discrete negative emotions (e.g., anger versus fear), as they can affect readers in significantly different ways,” Kohout and her colleagues write. “Future research can build on our study by testing the effects of different emotions, emotional cues, and processing strategies as well as different news providers, formats, and topics. Second, due to effects as information, future research should consider how emotionally invested people might get when reading comments, and how this might affect their information processing.”

The study, “May I have your Attention, please? An eye tracking study on emotional social media comments”, was authored by Susann Kohout, Sanne Kruikemeier, and Bert N. Bakker.