Journalist tend to rely on rapid, low-effort cognitive processes when posting content to the social media website Twitter, according to new research that analyzed more than 12 million words produced by campaign reporters during the run-up to the 2016 election.
The study, which has been published in PLOS One, was based on what is known as dual process theory of thought, which has been popularized by psychologist Daniel Kahneman. The theory proposes that there are two distinct cognitive modes in the human mind: System 1, which is quick, intuitive and emotional; and System 2, which is slower, logical and deliberative.
People use System 1 thinking when they are faced with a situation that is familiar or when they don’t have to put much thought into their decision. System 2 thinking is more effortful and is used when people need to focus their attention on a task or when they are facing a new or challenging situation. While both System 1 and System 2 thinking have their advantages, research has shown that people tend to rely too heavily on System 1 thinking, which can lead to errors in judgment.
“This paper applies cognitive theories to examine when and how System 1 thinking can be evident in journalists’ minds through empirical investigations of their word choices across different media,” explained study authors Jihye Lee and James T. Hamilton in their study.
“Theories of the mind predict that journalists on the campaign trail may engage in System 1 thinking when they sift through volatile campaign events. Of particular interest to the current discussion is whether journalists are more likely to engage in System 1 thinking when they navigate Twitter than when they compose texts and scripts for traditional news media such as newspapers and broadcasts.”
For the study, the researchers constructed dataset of news articles, broadcast comments, and tweets published by 73 reporters covering the 2016 presidential campaign between November 7, 2015 and November 7, 2016. The dataset contained 9,745,292 words from papers, 237,583 words from broadcast transcripts, and 2,643,593 words from Twitter.
Lee and Hamilton used an automated text analysis tool called Linguistic Inquiry and Word Count to examine how journalists’ language varied across the different types of media. The program can be used to study a wide variety of topics, including the psychological state of the author, the emotions conveyed in a text, and the underlying themes of a piece of writing.
In line with their hypothesis, the researchers found that the reporters tended to use more emotional language when writing on Twitter, compared to when writing in news articles. The reporters’ tweets also tended to be more focused on the present, contained more certainty and used more informal language, while using fewer analytical words and fewer numerical terms.
When Lee and Hamilton compared tweets to broadcast comments, they found a similar pattern of results. However, there were also some notable differences. For instance, broadcast comments tended to be more focused on the present and contain fewer analytical words than tweets.
“Overall, these findings indicate that journalists routinely engage in System 1 thinking in covering the evolving world of presidential campaigns, and System 1 thinking can be especially amplified in journalists’ minds when they navigate Twitter to engage with their audiences in a fast and personalized manner,” the researchers said.
The study, “Anchoring in the past, tweeting from the present: Cognitive bias in journalists’ word choices“, was published March 2, 2022.