Even on mute, TV can perpetuate racial bias

ByEd Yong
December 17, 2009
8 min read

Those of us who have been on the receiving end of racial abuse know all too well that words can hurt. But they’re also the tip of the iceberg. According to a study of popular US television, we’re exposed to the spectre of racial bias on a regular basis, all without a single word being uttered.

When scenes are muted, body language and facial expressions are enough to convey more negative attitudes towards black characters compared to white ones. This bias is so subtle that we’re largely unable to consciously identify it, yet so powerful that it can sway our own predispositions. In some ways, racial bias acts as a contagion and television as one of its vectors.

These nonverbal cues could have many origins. Actors could act slightly more negatively towards black colleagues, even if they have no explicit racial biases themselves. Their actions could be written into scripts or they may be directed to behave in a certain way, again without any conscious effort on the part of the writers or directors. Whatever the cause, it’s clear that audiences of millions are regularly exposed to very subtle forms of racial bias that can affect their own behaviour. 

Max Weisbuch from Tufts University showed volunteers a series of clips taken from episodes of 11 popular shows, including CSI, Grey’s Anatomy, Heroes and House, with an average weekly audience of 9 million Americans. Each show has a racially diverse cast, and Weisbuch focused on an important black character from each show. He chose his clips systematically, taking the first scene where the chosen black character interacted with a white one of roughly equal status within the first, middle and last 5 minute bursts of each episode. Matching for status is important – it obviates the fact that black characters are sometimes less prominent or important to a show. 

Weisbuch cut the audio and the featured character from each clip, leaving behind just the reactions of their conversational partner. Each altered clip was shown to 23 white students who had never seen any of the 11 shows. Without any clues from tone of voice or choice of words, the students judged that responses to unseen white characters were significantly more positive than those to unseen black characters.

Another 17 students who did watch most of the shows felt that the black characters that Weisbuch picked were weren’t negatively portrayed – they were seen to be just as attractive, intelligent, sociable and kind as their white counterparts. Another group of 13 students saw transcripts of the muted audio from each clip and, again, saw no difference in the words spoken to characters from each racial group. Nonetheless, stripping out these other elements reveals that biases can still be hidden in non-verbal behaviour.

LIMITED TIME OFFER

Get a FREE tote featuring 1 of 7 ICONIC PLACES OF THE WORLD

And that, in turn, affects people’s own biases. Weisbuch recruited 53 fresh students and asked them which of the 11 shows they watched. Each student completed an implicit association test (IAT) designed to reveal hidden racial biases. The test asks people to use two keys to categorise words from different groups (e.g. black and white, or good and bad). It works on the basis that category combinations that contradict our own biases should subtly slow our reaction times. Higher scores mean stronger pro-white biases.

Students who watched shows with a stronger non-verbal racial bias to them had higher IAT scores, meaning that their implicit biases were also greater. And once again, this trend has nothing to do with what the characters on the shows actually said, or how attractive they were. Of course, this is just a correlation. It could be that people with stronger personal biases gravitate towards shows that pander to their prejudices, even at an unconscious level. Weisbuch addressed that in his next experiment.

He showed 97 students silent clips with both characters still in. In the “pro-white” clips, white characters elicited more positive responses than black characters, while the opposite was true for the “pro-black” clips. Some of the clips featured the same character in both sets, while others used different characters who were matched in terms of attractiveness, sociability, kindness and intelligence. After watching the clips, the students completed an IAT. As expected, those who saw the pro-white clips had significantly higher IAT scores than those who saw the pro-black clips.

The fact that our own racial bias can be swayed by the physical actions of people we see on TV is even more alarming because we’re largely unaware of it. We may know a thing or two about expressions and body language but we still have trouble in identifying patterns of racial bias, encoded in looks or postures. Weisbuch dramatically demonstrated this by asking 22 volunteers to find a “hidden pattern” in the silent clips from the previous experiment. Afterwards, he explicitly told them to say if black characters had been treated better or worse than white ones.  They failed – 55% said that the clips were pro-white, no different from what you’d see if they were just guessing.

In a final experiment, Weisbuch repeated the previous study with pro-white and pro-black clips, but added a third group of control clips where expressions and body language were equally positive to all characters, regardless of race. And this time, instead of an IAT, his recruits did an “affective priming test” where they had to group image as either positive or negative after seeing subliminal images of black, white or Asian faces. Here, the idea is that people who are biased towards a specific race would respond more quickly to the positive images.

Even with these tweaks, the same trends emerged. People who saw the pro-white clips showed stronger white biases in the affective priming test, while those who saw the pro-black clips had stronger black biases. Biases towards Asians, who didn’t feature in the clips, went unchanged.

These subtle shifts in bias even translated to more outspoken attitudes. After seeing the silent clips, the recruits said that they liked white characters more if they’d seen the pro-white material, and black characters more of they’d seen the pro-black clips. And if you’re thinking at this point that the two effects would balance out, remember from the first study that the majority of the 11 TV shows are, on the whole, slightly pro-white.

All in all, these four studies suggest that racial biases are hidden in the unspoken actions of characters in popular US TV programmes. These biases can affect those of viewing audiences, and they’re very hard to consciously pin down. Turned around, they could even help to reduce racial prejudice. In Weisbuch’s final study, when white recruits filled in a survey that measured their attitudes towards black people, they showed significantly fewer signs of racial prejudice if they had seen the pro-black clips. 

Bringing this phenomenon to light should hopefully galvanise people in the entertainment industry to give it due consideration. This is especially important because American TV is hardly confined to America. It travels all over the world, picking up audiences far greater than the 9 million quoted in Weisbuch’s paper. It’s a potent international cultural force and it needs to be careful if it isn’t to promote the globalisation of racial bias.

Go Further