This week, I read the Sumner paper, the Sumner and King paper, and the Sumner and Katoaka papers. The papers all addressed the topic of extracting semantics from the speech of a variety of speakers. The first Sumner paper discussed the topic of encoding and how voice cues and biases occur very early in speech processing and how accents themselves are not really the factor that aids understanding and retention of speech, but rather the key factor is social. People can easily understand an esteemed accented speaker when they don't understand a lower status accented speaker. I wonder whether there are any differences in comprehension between the speech of esteemed accented and unaccented speakers.
The Sumner and King paper discussed differences in word associations when listeners were asked for the first word to come to mind listening to a middle aged white woman or an elderly black man. They found that 22.5% of the words had different top associated words. It is intuitive that listeners would associate different words with the primer depending on the speaker, especially with the lack of context. For instance, men do not frequently talk about yeast infections, so it is unsurprising that "bread" rather than "infection" was the top associate for J (male) with the primer "yeast," whereas "infection" was the top associate for M (female).
The Sumner and Katoaka paper showed that recognition of variable pronounciation words did not vary with the frequency, and that priming in GA/BE improved recognition of related words, but priming in NYC did not. It was also found that NYC induced high false recall rates but no semantic priming. This suggests that NYC was not encoded as well as the other accents.
These findings have implications on society and might be able to explain why social mobility is difficult for people who do not speak in the preferred/typical way. If people who are of higher class fail to properly encode the words that other classes use, this is a large barrier.
I completely agree that social mobility can be difficult for people who do not speak in the preferred way. This reminds me of the moving Good Will Hunting, in which the character Will was a genius, but because of his Bostonian southside accent he was frequently overlooked. Without the accent that was so easily recognizable, perhaps long before he could have had more easily available opportunities by people who just recognized his intelligence, nothing more.
ReplyDeleteThe question about social mobility is interesting and related to a question I had about causality of this memory bias. That is, is it the memory bias from speaker accent that would hinder social mobility, or do our preconceived notions of a speaker's background influence the way we consolidate such memories (as well as other behavior, such as discrimination in hiring, etc.)?
ReplyDeleteGreat post!