December 2012 Link
I am sitting in a darkened, closet-size lab at Tufts University, my scalp covered by a blue cloth cap studded with electrodes that detect electric signals from my brain. Data flow from the electrodes down rainbow- colored wires to an electroencephalography (eeg) machine, which records the activity so a scientist can study it later on.
Wearing this elaborate setup, I gaze at a television in front of me, focusing on a tiny cross at the center of the screen. The cross disappears, and a still image appears of Snoopy chasing a leaf. Then Charlie Brown takes Snoopy’s place, pitching a baseball. Lucy, Linus, and Woodstock visit as well. For the next half hour I stare at Peanuts comic strips, one frame at a time. The panels are without words, and while sometimes the action makes sense from frame to frame, at other times the Peanuts gang seems to be engaging in a series of unconnected shenanigans.
At the same time, a freshly minted Ph.D. named Neil Cohn is watching the readout from my brain, an exercise he has repeated with some 100 subjects to date. Many people would consider tracking Peanuts or Calvin and Hobbes comic strips unworthy of scientific inquiry, but Cohn begs to differ. His evidence suggests that we use the same cognitive process to make sense of comics as we do to read a sentence. They seem to tap the deepest recesses of our minds, where we bring meaning to the world.
Comics have been part of Cohn’s life as long as he can remember. He was drawing them by age 8. As a teenager he began to sell his own comics--a graphic novel about two people falling in love in one case, a dreamy series of meditations about philosophy in another. By the time he entered the University of California, Berkeley, as an undergraduate, he was a regular at Comic-Con, the gigantic annual comic book and fantasy convention in San Diego. “If you had asked my friends if I would end up as a scientist,” he says, “they would have all said, ‘No way.’ ”
That changed at Berkeley, where Cohn discovered linguistics. He was fascinated by how our brains find meaning in strings of words. Individual languages differ in terms of particular words and grammar, but all are just systems for building sentences. You could split any sentence into smaller units--the subject and the predicate, for example--and these could be broken down into smaller units still.
This hierarchy of big units and small ones helps make language both versatile and easy. Even though each of us has only a finite number of words in our brain (100,000 or so), we can use the rules of language to combine them into a practically endless number of sentences, conveying an infinite set of meanings.
“I started making connections between what was going on in language and what was going on in comics,” Cohn says. A comic strip is a string of panels, just as a sentence is a string of words. From Cohn’s own experience designing comic strips, he was convinced that both followed a similar set of rules. “Sequential images have a grammar like sequential words do,” he says.
It further seemed to Cohn that comic strips are made up of smaller units, just as sentences are. The narrative arc of a comic strip is made up of an initial group of panels that set up the story, followed by those that convey a narrative peak. These units, in turn, are made up of smaller units that accomplish other tasks, like establishing new characters and resolving conflicts.
Cohn suspected this was no coincidence: Comic strip artists were unwittingly exploiting the brain’s grammar- generating function. “The brain is processing these different kinds of grammars in a common way,” Cohn says.
After this epiphany, Cohn decided to go to graduate school and become a cognitive scientist who studied comics. And he knew whom he wanted to study with: Ray Jackendoff, a linguist then at Brandeis University who had done some of the most important research into language’s hierarchy.
Cohn wrote to Jackendoff, asking if he would take on a student who wanted to study how people’s brains make sense of comics. But Jackendoff couldn’t do it at the time. Crestfallen, Cohn applied to other graduate schools instead, supporting himself in the meantime by drawing comics.
As the years passed, his pile of grad school rejection letters grew. In a typical one, Cohn was informed that he was a great candidate, but there was no way a comic book artist could fit in to the school’s graduate program. Cohn’s ideas were just too far off the map.
After four years of such replies, Cohn was close to giving up. But then he heard that Jackendoff had just moved from Brandeis to Tufts to direct its Center for Cognitive Studies. Cohn worked up the courage to send a second letter. To his surprise, Jackendoff invited him to earn a Ph.D. with him.
In 2006 Cohn came to Tufts and began his experiments. To test his hypothesis that the brain uses a visual grammar to make sense of comics, he designed a series of experiments that updated some classic language studies. In one set of experiments, subjects read sentences in different forms while wearing an EEG cap. Some of the sentences were grammatically correct, while others were scrambled.
When subjects encountered a sentence with an unexpected grammatical error, EEG readings picked up a distinctive signal: The voltage in an area of the left brain briefly dropped. Scientists call this drop, which takes place directly over the left temple,
the “left anterior negativity” and see it as a sign the brain is struggling to make sense of a sentence with a scrambled structure.
The Peanuts strips that Cohn showed me also came in different forms. Some combined Charles Schulz’s original panels to make strips that made sense. (Cohn selected comics with little language in them and stripped out any speech bubbles, so that his subjects wouldn’t be processing language as well as the images.) In other cases, Cohn jumbled panels from different strips so that they had none of the ordinary structure found in comics--the visual equivalent of stringing together a randomly chosen set of words. Cohn found that subjects experienced the left anterior negativity effect when they were shown the random panels, just as if they were reading sentences with grammatical errors.
The most tantalizing result of Cohn’s study (recently published in Cognitive Psychology) emerged when he showed subjects panels arranged so that they had a narrative arc but didn’t add up to a meaningful story. They were the equivalent of grammatically correct but meaningless sentences. The linguist Noam Chomsky, who published pioneering ideas about language in the 1950s, offered a famous example of such a sentence: “Colorless green ideas sleep furiously.”
Cohn found that when he showed subjects such “colorless green” panels, they experienced a weaker left anterior negativity response than when they read garbled panels. The result suggests that although participants struggled to understand the panels, they still recognized an underlying logic to them, supporting the idea that we depend on a visual grammar in comics in order to make sense of them.
Cohn has gotten similar results from other experiments in which people had to press a button every time they saw one particular panel in a strip. It took them longer to hit the button if the strip was a random jumble than if it was in its original order. If the strip was “colorless green,” on the other hand, people could press the button at an intermediate speed.
“People are able to predict what’s coming next,” Cohn says, “even if there’s no meaning to it.”
Cohn hopes his research will bring a deeper appreciation of comics as more than entertainment. He sees them as windows to the evolution of fundamental structures in the human mind, which is wired to use strings of signals to communicate. Sometime more than 50,000 years ago, our ancestors evolved the ability to string spoken words together. Much later, they used that ability to create written language. But Cohn has shown that the same cognitive processes also gave rise to a visual language--one that can be found on the comics pages in American newspapers, in Japanese manga, and in many other forms.
English and Japanese are not identical, of course; despite being based on the same underlying rules of grammar, they have developed into different languages. Cohn sees the same cultural changes driving different styles of graphic storytelling. When he is not recording brains, Cohn is comparing comics from different cultures to learn
how this diversity springs from a universal biology.
“My goal is to create a field of study for this stuff,” he says.
Copyright 2012 Carl Zimmer