The clouds are pillows. Time is a thief. Life is a journey. According to Groucho Marx a hospital bed is a parked taxi with the meter running. And I wish I could write a Psychonomics post as fast as a speeding bullet—but because I can’t, all other commitments today will be delayed by virtue of the domino effect.
We all use metaphors in our daily language. Some estimates run to as many as 50 per 1000 words of discourse. Just like other types of playful, non-literal and non-direct language—such as jokes and sarcasm—metaphors are readily understood, and are thought by some professional writers to add a layer of richness to story-telling.
Metaphors have also been the object of much psychological study: Cognitive scientists have examined how people can infer the non-literal meaning (hospital beds don’t look much like taxis) from the surface structure of the metaphor. Researchers have also examined why people speak in metaphors (time is no thief, but the metaphor is an efficient way to describe the inexorable loss of remaining life as time passes).
A recent article in Memory & Cognition, one of the journals of the Psychonomic Society, addressed a third psychological aspect of metaphors, namely their role in creating social bonds and understanding each other’s’ intentions. According to the philosopher Ted Cohen, “… there is a unique way in which the maker and the appreciator of a metaphor are drawn closer to one another.” The speaker is thought to issue “a kind of concealed invitation” to which the listener responds with “a special effort to accept the invitation.” This interactive process creates common ground and a shared understanding.
Researchers Andrea Bowes and Albert Katz set out to test this idea, that metaphors create intimacy, in a series of three experiments. The researchers were particularly interested in the question whether processing of a metaphor enhances a person’s ability to infer mental states in others. If metaphors create a shared understanding between speaker and listener, then this must be reflected in the listener’s enhanced knowledge of the speaker’s mental state.
To measure participant’s ability to infer mental states, Bowes and Katz used a series of photographs of the eye region of faces (i.e. from just above the eyebrow down to midway along the nose) that were accompanied by 4 mental-state descriptions. One of those descriptions was correct, the other 3 were foils. An example stimulus is shown below:
Is this person reflective, aghast, irritated, or impatient?
The ability to identify emotions from such partial information has been shown to be related to the more general ability to identify mental states in social situations. Performance in the task involves a network of brain regions often implicated with social cognition, and it also varies with levels of oxytocin—a neuropeptide associated with interpersonal closeness and attachment.
In one of the experiments, Bowes and Katz presented participants with brief passages involving an interaction between two people. The passage would contain either a literal statement (“Frank warned Kyle: ‘be careful what you say to him’ ”) or an equivalent metaphorical statement (Frank warned Kyle: ‘watch your back around him’ “).
After reading the passage, participants first answered some questions about its content. The item of greatest interest was the judged degree of intimacy or familiarity between the actors in the story. AS expected, intimacy was judged higher when the metaphor was used than when the story contained a literal sentence.
More intriguingly, when the researchers compared the ratings of intimacy with people’s ability to detect emotionality in partial faces—for example, the ability to identify the person in the picture above as being “reflective”—those two measures were found to be related.
This basic finding was shown in two further experiments. Of particular interest is the fact that even when metaphorical (“the price change was a major drop”) and literal sentences (“the bungee jump was a scary drop”) were read in isolation, people who read the metaphors were subsequently better at identifying emotions from partial faces than participants who read the literal statements.
Taken together, the three studies show that the mere act of reading metaphors orient people more towards interpersonal social information, and this orientation can be revealed on an unrelated task involving very different stimuli.
What processes might underlie this phenomenon? Bowes and Katz focus on the simulation-theory approach according to which people “…use their own, sometimes fragmentary, bodily reactions to make inferences about others.” That is, to understand what others are thinking or feeling we “simulate” their presumed responses using our own reference point, by imagining how we would react in that situation.
To enable comprehension of metaphors, people must rely on an extra-linguistic context consisting of past memories, thoughts, and emotions more than when they read literal text. Understanding how a hospital bed might resemble a taxi requires considerable processing and cultural context such as knowledge of the Marx brothers. In consequence, that additional processing primes people to “simulate” other people’s emotions later on when they inspect the test faces.
Bowes highlights the importance of those results by noting that “The research explains why we speak differently with friends and family than with strangers, and [it] shows how we make friends and meet partners simply with the style of language we use.”
“The results also stress the importance of literature in developing and understanding human empathy,” adds Katz. “Reading fiction indeed promotes people’s ability to identify the emotions or mental state of others.”