Misinformation is a messy problem, and not only because false stories spread far and fast. Even when people encounter questionable claims, they still have to decide whether those claims deserve to be believed and passed along. That decision is influenced by more than just the facts, or lack of facts, at the center of a story. It also depends on how the story is told, including the language that frames its evidence. This is where hedging becomes interesting. Hedging refers to the small linguistic choices that make a claim sound less direct or less certain: vague sources like “some experts say,” imprecise numbers like “a significant number,” or cautious phrases that soften commitment to what is being said. In misinformation, such cues appear all the time. But do they make misleading content more convincing, or do they actually raise red flags? The answer, according to recent research, is more nuanced than a simple yes or no. A study involving Chinese readers looked closely at two specific kinds of ambiguity-based hedging: source ambiguity, when it is unclear where information comes from, and numerical ambiguity, when quantities or probabilities are expressed in fuzzy terms. The researchers wanted to understand how these two cues shape whether people reject misinformation, whether they feel like sharing it, and how quickly they make up their minds. The study is a reminder that misinformation is not just a matter of false content, but also of the subtle stylistic choices that shape how readers judge credibility and decide what to spread.
One reason this question is hard to answer is that previous research has produced conflicting results. In some cases, hedged language seems to make messages seem more careful, transparent, or scientific, which can boost credibility. In other cases, the same kinds of expressions make people more suspicious, because they suggest that the author lacks solid evidence or is trying to hide something. Part of the problem may be that researchers often treated hedging as one broad category, even though ambiguity can creep into a claim in very different places. A claim can be vague about who said something, or it can be vague about how much, how many, or how likely something is. These are different kinds of weakness in the evidential chain. The current study built on a framework called epistemic vigilance, which basically suggests that people are not passive receivers of information. Instead, they watch for cues about whether a source is reliable and whether the content makes sense. Source ambiguity and numerical ambiguity map neatly onto these two concerns. Vague attribution makes it hard to evaluate the source. Vague numbers make it hard to evaluate the precision of the evidence. The researchers therefore hypothesized that these two forms of ambiguity might not work in exactly the same way, and set out to compare them side by side in a controlled experiment.
To test this, the research team recruited 182 university students in China who were native Mandarin speakers and active social media users. Each participant saw five misinformation statements that had already been debunked by authoritative fact-checking sources. The statements were adapted so that they appeared in one of four versions. Some had both a specific source and a precise number. Others had a vague source with a precise number. Still others had a specific source with an imprecise number. And some had both a vague source and an imprecise number. The design allowed the researchers to isolate the effects of each kind of ambiguity. The materials were carefully checked to make sure that the core false claim remained the same across versions, and that only the wording changed. After reading each statement, participants were asked to decide whether the claim was false, and their reaction time was recorded. Then they were asked how likely they would be to share the statement on social media. The researchers also measured how clear and comprehensible the statements felt, to make sure that the ambiguity manipulation did not simply make the materials confusing or unreadable. The use of reaction times was particularly clever, because it offered a window into how quickly people processed the statements, not just what they ultimately decided.
The results told a fairly clear story. Both source ambiguity and numerical ambiguity made people more likely to reject debunked misinformation as false. In other words, when a claim cited a vague source instead of a specific one, participants were more likely to say it was false. When a claim used an imprecise number instead of an exact figure, they were also more likely to reject it. The combination of both cues produced the highest rejection rate. This suggests that, at least in this experimental context, ambiguity acted as a signal of unreliability rather than as a persuasive device. But the story did not end there. Neither kind of ambiguity directly predicted whether someone wanted to share the misinformation. What mattered for sharing was whether the person judged the claim to be false in the moment. When participants correctly rejected a misinformation statement, they were significantly less willing to share it. That might seem obvious, but it is actually an important finding, because previous research has shown that people sometimes share misinformation even when they should know better. The disconnect between judgment and sharing often happens when accuracy is not at the front of their minds. In this study, the explicit truth-judgment task likely made accuracy highly salient, and sharing intentions followed closely from veracity judgments.
The reaction time results added another layer of nuance. Numerical ambiguity was associated with faster rejection decisions, even after controlling for other factors. This means that imprecise numbers were a quick and salient clue for people, prompting them to recognize the claim as false more rapidly. Source ambiguity, by contrast, increased rejection but did not speed up response times. It had a real effect on judgment, but not on processing speed. This suggests that the two kinds of ambiguity may work through somewhat different pathways. Vague numbers might be an early, obvious signal that a claim lacks the kind of precision we expect from trustworthy information. Vague sources might create a more general sense of skepticism that takes a bit more mental work or simply does not change the speed of decision-making. The distinction matters because it shows that ambiguity in misinformation should not be treated as a single, uniform cue. Instead, the specific location of the ambiguity, in the source or in the numbers, shapes how readers react. The findings also suggest practical lessons for media literacy and misinformation prevention. Teaching people to notice vague sourcing and fuzzy numbers could help them evaluate claims more critically. At the same time, the study cautions against treating ambiguity as a simple lie detector. Vague language is not always false, and precise language is not always true. The goal is not to train people to be suspicious of every hedged sentence, but to help them think more carefully about what kind of evidence a claim actually provides.
Of course, the study has limitations. It used only five misinformation statements, all focused on everyday life topics, so the results might not generalize to political misinformation or highly emotional issues. The participants were mostly young female university students studying English, which is not a representative cross-section of the general population. All of the statements were false, so the study could not say how ambiguity affects judgments when people encounter a mix of true and false information, as they do in real social media feeds. Sharing intention was measured after an explicit truth judgment, which may have made people more accuracy-focused than they would be in a natural scrolling context. The researchers also did not directly measure underlying psychological processes like perceived source credibility or perceived evidential strength, so the interpretations remain plausible but not fully proven. Still, the study makes a valuable contribution by showing that the language of misinformation deserves serious attention. It challenges the assumption that hedging always makes misinformation more persuasive, and it highlights the need for a more differentiated view of how message-level cues shape reader judgment. In a world where false claims are abundant and digital attention is short, understanding the small linguistic signals that guide trust is more important than ever. The bottom line is that how a claim is worded can shape whether we believe it, how quickly we judge it, and whether we share it, but not always in the way we might expect. Part of staying informed is paying attention not just to what is said, but to how it is said.