A Study on Communicating Biases

Meryl Streep made headlines last month with her Golden Globes acceptance speech that included a response to “one performance this year that stunned me,” i.e.,  Donald Trump’s imitation of a disabled reporter during the presidential campaign. Her words garnered attention, just as Trump’s nonverbal behaviors had. This calls to mind the warning of a character Meryl Streep once played, the witch in Into the Woods. “Careful the things you say. Children will listen. Careful the things you do. Children will see and learn.” When considering the lessons we are teaching children, it should be noted that these lessons are being conveyed not just verbally, but nonverbally as well. 

A recent study built upon past research suggesting that:

a) implicit biases can be communicated through nonverbal signals

b) infants avoid or respond negatively toward objects after observing negative responses to them; and

c) evidence of social bias and prejudice has been observed in children as young as in preschool

Researchers at the University of Washington showed four- and five-year-olds a video of a person in a black shirt and a person in a red shirt, each target person being treated differently by an actor. The actor displayed either positive nonverbal signals (e.g., smiling, leaning in, using a warm tone of voice, eagerly giving a toy) or negative nonverbal signals (e.g., scowling, leaning away, using a cold tone of voice, giving the toy reluctantly) toward the targets. The actor then introduced a novel object. Each target labeled it with a made-up word and used it in a particular way. The video also explained that one target was part of “the dark-red group” and the other was part of “the black group,” and a friend from each group was introduced.

The children were asked which target person they preferred, what the novel object should be called, and how the object should be used. The latter questions were asked because imitation, or failure to imitate, has been shown in the past to reflect children’s feelings about a person. The children were also asked which friend they preferred, and an imitation measure was taken.

Results showed that children explicitly preferred the person who was treated with positive nonverbal signals, and that they were more likely to use that person’s label for an object, even though they were not significantly more likely to imitate the displayed use of that object. Further, the children showed more positive attitudes toward the friend identified as being in their preferred person’s group, and they were more likely to imitate this friend.

While the implicit biases actors displayed in this experiment may be less subtle than many displayed toward individuals and groups in natural settings, the implications are certainly interesting and worthy of further exploration. Adults have great potential to teach children many things, including — it seems — biases. It is likely wise to heed the words of the witch.