The study was published in the Journal of Product & Brand Management.
"There's been a lot of work done on how the public processes and responds to fake news on social media in the context of politics, but very little research has been done on how fake news may affect brand trust," said study author Yang Cheng, Assistant Professor at the North Carolina State University.
"We wanted to see what kind of impact fake news could have for companies," Cheng said.
To explore the issue, researchers drew on a real-life incident from 2016, in which a false news story circulated on Facebook, asserting that Coca-Cola had recalled bottles of its Dasani-brand water due to the presence of aquatic parasites.
For this study, 468 consumers were shown an example of the 2016 Facebook posts, but were not told the information was false.
The consumers then answered a range of survey questions.
At this point, the researchers informed the consumers that the Facebook posts were fake news, after which the study participants answered another series of survey questions.
The researchers found that the more consumers felt they could detect and evaluate misinformation, the more likely they were to feel that the post was intended to manipulate readers.
Similarly, the less consumers trusted Facebook, the more likely they were to feel that the post was intended to manipulate readers and they become more sceptical.
The results also showed that the more likely people were to think the post was manipulative, the less likely they were to find the post helpful or relevant to themselves.
However, consumers' trust of the brand was not affected - due to the consumers' increased skepticism and inference of manipulative intent regarding the fake news.
"One takeaway here is that when communicating the truth in response to a fake news story, such as when Coca-Cola made clear that it was not recalling Dasani, managers need to consider the trustworthiness of each media channel and choose the appropriate media channels to communicate with consumers," Cheng said.
( With inputs from IANS )