Attention not necessary for visual awareness, large study suggests

The famous “invisible gorilla” experiment has captivated the psychology community since it was published in 1999. About half of the participants—who had to count the number of basketball passes in a video—failed to notice a 5-second cameo in which a person in a gorilla suit danced her way among the players. Some scientists interpreted that failure, called inattentional blindness, as evidence that visual awareness requires attention.

But attention is not essential after all, a new study suggests. People can accurately describe some details in a scene even if they focused their attention elsewhere and reported not noticing them, the study shows.

Inattentional blindness “is a side effect of something that we do really, really well … we have to be able to focus our attention on things we care about and not be distracted,” says Daniel Simons, professor of psychology at the University of Illinois Urbana-Champaign and co-investigator of the gorilla experiment. Simons was not involved in the new study. But awareness is not all-or-nothing, as the new study demonstrates, he adds.

The results, published last month in eLife, challenge some scholars’ interpretation of inattentional blindness, says Ian Phillips, the study’s principal investigator and professor of philosophy and psychological and brain sciences at Johns Hopkins University. Several theories of consciousness, such as the global neuronal workspace, attended intermediate representation, and attention schema theories, “have seized on this kind of data as showing that attention is the crucial ingredient. It’s kind of the gate or the magic that needs to turn information coming in through the eyes into conscious experience,” Phillips says. He believes those views are incorrect, he says.

Most studies of inattentional blindness, including the gorilla experiment, ask participants a yes-or-no question about whether they noticed something unusual in a visual prompt. “These questions are famously problematic,” Phillips says. Researchers typically assume that a “no” means someone did not see or retain traces of information about the unexpected stimulus, he says. But what counts as unusual? And how sure does someone need to be before they would respond yes? Phillips asks. These uncertainties could lead the participants to report not noticing a stimulus when they did, he says.

So Phillips and his team came up with ways to quantify biases in participants’ responses and gauge their visual sensitivities. More than 25,000 people participated in one of the team’s five-part online experiments, which required participants to focus their attention on simple shapes, such as moving black or white squares, or a cross with varying arm lengths in the center of a screen.

Participants had to answer questions about how these objects changed or moved from trial to trial. After several trials, a new object—such as a red or blue line, or a circle or triangle—appeared on one side of the screen. Afterward, the participants were asked whether they had noticed anything unusual and, regardless of their answers, to guess the color and location of the unusual item on the screen.

In one iteration of the experiment, participants were asked to rate how confident they were in their answers. In another iteration, the scientists deliberately did not introduce anything unusual—an “absent trial” that gauged how often people would still say they had noticed something unusual even in the absence of a stimulus.

Across these experiments, about 35 percent of the participants reported that they had not seen anything unusual. And up to 12 percent had a “false alarm” during an absent trial—meaning they mistakenly reported seeing something unusual.

The team then analyzed their data using a mathematical framework called the signal detection theory to account for such false alarms and other responses driven by people’s tendency to respond a certain way regardless of the stimuli.

This is an important step because measures such as the proportion of correct answers “can be confounded by people’s response bias,” says Samuel Robson, a psychology lecturer and postdoctoral fellow at the University of New South Wales, who was not involved in the study.

More than half of the participants who said they did not notice anything unusual during an experiment were still able to correctly guess the color or location of the stimuli, the study shows. “There’s clearly evidence that they are, in fact, picking up on [the stimuli] to some extent,” Phillips says.

Even the group of more than 200 participants who were most confident that they did not notice anything unusual performed better on their guesses than would be expected to occur by chance. “We thought that was like the hardest test, if there was still any residual sensitivity to the stimuli,” Phillips says. It was “a nice surprise.”

“Looking at confidence is particularly important,” says Anna Kosovicheva, assistant professor and co-director of the Applied Perception and Psychophysics Laboratory at the University of Toronto, who was not involved in the study. The results suggest that failing to report the existence of a stimulus does not necessarily mean that someone failed to perceive it, and they raise questions about what other kinds of information are available to people beyond what they pay attention to, Kosovicheva adds.

It remains unclear if some people are more susceptible to inattentional blindness than others. “There might be people who are noticers and people who are missers, right?” Simons says. Or it could be “just chance who notices and who doesn’t,” he adds. “I think it’s more likely that it affects all of us.”