Tell the Riddle RightPrint
Subtle linguistic manipulations can change how we interpret questions
By Jessica Love
June 6, 2013
When the indefatigable canvasser stops me on the street and asks, “Do you care about animals?” she is really saying, “Will you commit to making a sizeable donation—right here, right now—to an organization you’ve never heard of, specifically the one that is paying me to stand here on the corner?” The canvasser knows that wording matters. And science backs her up. In the 1970s Elizabeth Loftus and John Palmer famously asked participants to provide testimony after watching videotaped car accidents. Asking, “About how fast were the cars going when they bumped each other?” prompted very different responses than asking, “How fast were the cars going when they smashed each other?”
Nor are such results the province of contrived experiments. Last year, for instance, Gallup reported that 44 percent of Americans approved of “the financial bailout for U.S. automakers that were in danger of failing,” while Pew reported that 56 percent of Americans deemed the government’s “loans to General Motors and Chrysler” to be good for the economy. As The Hill columnist Mark Mellman put it, “It should surprise no one that Americans prefer ‘loans’ to ‘bailouts.’”
Now a new study suggests that much subtler tweaks—boring tweaks, geeky English-major type tweaks we’d never suspect can make a difference—actually can make a difference.
We all are comfortable with verb tense, but verbs are also marked by aspect, which specifies whether an event is ongoing (“imperfect”) or completed (“perfect”). Tense and aspect are related but distinct. Present or future events can be represented as perfect: when William Carlos Williams confesses, I have eaten the plums that were in the icebox, he is doing so in the present tense, but those plums are long gone. Similarly, past events can be represented as imperfect (e.g., I was eating the plums that were in the icebox, but you told me to save the rest for breakfast).
Researchers Meghan Salomon, Joe Magliano, and Gabriel Radvansky wondered: might a verb’s aspect influence—for better or for worse—what people pay attention to when trying to solve a riddle? According to earlier research, the imperfect aspect encourages people to focus on the nature of the actions themselves; they are, after all, still happening! The perfect aspect, on the other hand, focuses our attention elsewhere—on, for instance, who is doing the acting, and why.
The researchers presented hundreds of participants with a riddle like the following: A woman was traveling for the weekend. She was checking her ticket, boarding, and placing her luggage above her seat. The pilot did not show up, yet she and the other passengers were arriving at their destination without a delay. How?
Some people read the riddle with imperfect verbs (as in the example above), while others read it with perfect verbs (A woman traveled for the weekend. She checked her ticket, boarded, and placed her luggage above her seat…)
Amazingly, about 60 percent of participants who received the imperfect version identified the correct solution—that the woman was traveling by train or bus—while only about 40 percent who received the perfect version succeeded. In other words, the riddle was easier when the verb’s aspect encouraged you to focus on a part of the problem (the actions undertaken by a traveler) that was relevant to its solution (the mode of travel).
It works the other way too. In a second experiment, participants read a riddle like, The jockey was riding in a race. Their horse was taking the lead, and it was clear that the jockey would win first place. Even though the horse was crossing the finish line without a man sitting on its back, they still won. How? As before, participants read either an imperfect version or a perfect one. But this time, attending to actions like “crossing the finish line” did not lead to the correct solution—that the jockey was a woman. And indeed this time, participants who’d read the imperfect version were less likely to solve the puzzle than participants who’d read the perfect version.
Magliano was quick to caution that, due to the small number of items, and the small amount of data collected from each person (each participant saw just one of three possible “action” or “gender” riddles, and in just one of the two aspect conditions), the effect is really much smaller than raw percentages would suggest. And the effect will need to be replicated on a larger, more diverse set of riddles before we can take it at face value—something lead researcher Salomon tells me is in the works.
But still, the data give reason to pause. Our education system relies on the assumption that responses to specifically worded questions are something other than responses to specifically worded questions—namely, that they constitute learning or achievement. To some extent, this is true, but as Magliano points out, we don’t yet know whether verb aspect affects some of us more than others. Might it allow advanced readers to home in on the most important elements of a passage? Alternatively, might a quirky prompt send advanced readers—or more confusingly still, a subset of advanced readers—down the wrong path, while everyone else plows forth obliviously?
Jessica Love recently received a Ph.D. in cognitive psychology and is currently a science writer and editor at Northwestern University.