Accuracy Inattention

What it is

The starting point is a puzzle. People judge false headlines as less accurate than true ones, and most say it is important to share only accurate news. Yet the authors found that whether a headline is true had little effect on whether people said they would share it. Truth strongly shaped judgments of accuracy and had much less pull on sharing.

The authors read this as a gap in attention. A person can hold a value and still not bring it to mind when a choice arrives. The authors conclude that people often share misinformation because their attention is on other factors, not on accuracy. The finding separates sharing from belief: sharing does not necessarily reflect what a person thinks is true.

The label used on this site, accuracy inattention, names that gap. It is a description of the result, and it does not claim that people are indifferent to truth. The same participants who shared inaccurate headlines said that accuracy matters.

In effect

The field experiment tested the attention reading outside the survey setting. A subtle prompt toward accuracy improved the quality of the news people subsequently shared. The prompt was light, which fits the idea that the value was already present and only needed to be brought to mind.

This is the most practical part of the finding, and it is also the part to hold at arm's length. It shows that an attention-based nudge moved sharing in the study. It does not show how long the effect lasts, how large it would be elsewhere, or whether it reaches the people whose sharing is most automatic.

That last question connects to habitual sharing. Ceylan, Anderson and Wood found that habitual sharers stayed insensitive to accuracy even when asked to judge it first. Taken together, the studies suggest two routes by which accuracy loses out at the moment of sharing: attention that is elsewhere, and responses that have become automatic. Repetition adds a third, since Effron and Raj found that repeated fake-news headlines seemed less unethical to share even when people knew they were labelled false.

What it does not say

The finding does not say that people believe what they share. It says the opposite: that sharing and belief come apart. It also does not say that every shared falsehood is shared without thought, or that accuracy is the only thing people attend to. Other motives, including the social ones studied by Duffy, Tandoc and Ling, sit beside it.

The paper reports survey experiments and one field experiment. Effects on stated intentions in surveys are one kind of evidence and effects on observed sharing are another, and the claim here is limited to what those experiments showed.

Who developed it

Gordon Pennycook is the first author, working with Ziv Epstein, Mohsen Mosleh, Antonio Arechar, Dean Eckles and David Rand. The paper is the source of the attention account of sharing described here.

---