In this study, researchers investigated the use of psycholinguistic characteristics to differentiate between social media users who spread fake news and those who debunk it. They developed a model called CheckerOrSpreader, which utilized a Convolution Neural Network to classify users as potential spreaders or checkers of fake news. The results showed that the model was effective in identifying users who tended to spread fake news versus those who debunked it.
The analysis also revealed that users who debunked fake news tended to use more positive language and a higher number of terms that showed causality compared to users who spread fake news. These findings have significant implications for identifying and combating the spread of fake news on social media, highlighting the importance of language use and psycholinguistic characteristics in understanding social dynamics online. The CheckerOrSpreader model could be a useful tool for identifying potential spreaders of fake news and targeting efforts to combat its dissemination.
