What's a picture worth? Too much, a new study suggests. The mere presence of a picture of a brain, in an article about cognitive processes, makes us more likely to believe what we're reading -- even though most of us have no ability to understand the picture.
The paper is "Seeing Is Believing: The Effect Of Brain Images On Judgments Of Scientific Reasoning," by Colorado State's David McCabe and UCLA's Alan Castel, forthcoming in the journal Cognition but also published on line at Colorado State. There's a press release here.
More credible, better written
In three different experiments, subjects read articles about different aspects of cognition. The articles were identical in everything but the illustrations. In the first experiment, for example, subjects read three fictitious articles, including one called "Watching TV Is Related To Math Ability." As you might guess from that title, the article was flawed. (Or as McCabe and Castel kindly put it, it contained some "claims that were not necessitated by the data," "giving participants some basis for skepticism in their ratings.")
The article purported to compare the brain's responses, as measured by neuroimaging, to watching TV and to math, and that comparison had an illustration. Some of the subjects saw this picture,
but the rest of the subjects saw this one,
and even though the graph is something even I can comprehend and the brain picture is completely inscrutable, the subjects who saw the brain picture found the article more credible and better written. The difference was small, McCabe and Castel said, but it was significant.
Let's try that a different way
So okay, McCabe and Castel said, maybe it's not the brain picture that's making the difference here; maybe it's just that the brain picture is more complicated and impressive. So they did the experiment again, this time with two complicated and impressive pictures. This time half the subjects saw the brain "map" on the left below, and half saw the brain picture on the right:
And it wasn't just the complexity. The subjects who got the brain picture found the article better reasoned than those who got the map, even though both are byzantine. When the researchers repeated the test with a real article, the result was the same.
"It is unlikely that this subtlety is appreciated by lay readers"
McCabe and Castel aren't sure why we're so dazzled by these pictures. They think it may trace back to our naive belief -- "reductionist," they call it -- that the mind is "an extension of the brain," so that we think the brain images are literally pictures of thinking. In fact they're not, as McCabe and Castel explain in the kind of passage that makes these papers so much fun to read and translate:
Indeed, it is important to note that while brain images give the appearance of direct measurement of the physical substrate of cognitive processes, techniques like fMRI measure changes in relative oxygenation of blood in regions of the brain, which is also indirect. Of course, it is unlikely that this subtlety is appreciated by lay readers.
The limits of teaching
At one level, the message for lawyers here is clear: use pictures, and not just any pictures. Pictures may be most powerful when they represent what you're talking about as directly and concretely as possible.
At the next level, though, this research dramatizes a judgment call that lawyers often have to make. When the material is difficult, we can try to teach it to jurors so that they independently get it, or we can simply tell them, with the authority that comes from credibility, that we're right. A large body of research and writing tells us that when we can really teach jurors, they're not only convinced, but they're grateful for the experience.
But is there a limit to teaching? This research suggests (at least to me) that there may be -- that at some level of complexity, jurors are more comfortable with trusting than they are with learning. The amazing thing about McCabe and Castel's research is those first two pictures, the graph and the brain. One picture makes sense to most laypeople, the other doesn't -- and the one that makes no sense was more believable. The article implied that the brain picture demonstrated the author's claims; the subjects had no way of judging whether this was true; and yet they believed more strongly than when they were given a graph they themselves could read.
______________________
The full cite for this article is McCabe, D. P., & Castel, A. D., Seeing is believing: The effect of brain images on judgments of scientific reasoning, Cognition (2007), doi:10.1016/j.cognition.2007.07.01
(Drawing by lucianvenutian, done when he was three, at http://flickr.com/photos/lucianvenutian/477929996/; license details there.)
