Social Categories and Perceived Emotion

This week, I read the papers The social weight of spoken words, Effects of phonetically-cued talker variation on semantic encoding, and Voice-specific effects in semantic association.  I was surprised to learn that the social associations of speech attributes are processed simultaneously to the content of the speech, and even influence the way we listen and how readily we recognize words.  This phonetic processing gives our speech an additional, often subconscious, layer of content beyond word meaning. 

The paper Voice-specific effects in semantic association gives the following example of how linguistic cues subconsciously alter semantic content: when we hear a woman say the word ‘clothes’ we are more likely to think of female-specific clothes than when a man says the word.  It seems like a similar process might subconsciously influence how we interpret the emotional content of speech. This idea reminded me of an interview with Hillary Clinton in which she discussed how people react differently to her speeches versus a male politician saying the same thing in the same way.  She said,

“You have to communicate in a way that people say: ‘OK, I get her.’ And that can be more difficult for a woman. Because who are your models? If you want to run for the Senate, or run for the Presidency, most of your role models are going to be men. And what works for them won’t work for you. Women are seen through a different lens. It’s not bad. It’s just a fact. It’s really quite funny. I’ll go to these events and there will be men speaking before me, and they’ll be pounding the message, and screaming about how we need to win the election. And people will love it. And I want to do the same thing. Because I care about this stuff. But I’ve learned that I can’t be quite so passionate in my presentation. I love to wave my arms, but apparently that’s a little bit scary to people. And I can’t yell too much. It comes across as ‘too loud’ or ‘too shrill’ or ‘too this’ or ‘too that.’

Maybe the concept of exemplar-category resonance and the way that social categories affect word association can help explain why people react so differently to male versus female public speakers.  We are used to hearing male politicians speak, and this association between political rhetoric and male speakers can throw people off when a woman utters almost the same words.  Based on the results reported in Voice-specific effects in semantic association, listeners may even be interpreting the contents of Hillary’s speech differently because of her gender. 

Similarly, Obama is known for being calm and collected at public speaking engagements, even in situations that might inspire other politicians to yell.  His poise under pressure even inspired a series of Key & Peele skits featuring Obama and his ‘anger translator.’ One potential reason for Obama’s reserve is that his identity as a black man might cause people to process the emotional content of his speeches differently – in particular, adding in anger that isn’t there. Obama and Clinton are just two examples, but it would be interesting to study whether social categories affect our perception of emotion in the same way they affect word association and interpretation.