What Can a Voice Convey?

In Thursday’s class, we read an experimental setup from Lev-Ari and Keysar (2010). They conclude that comprehension difficulty is inversely proportional to trustfulness for speakers rated by listeners. There is no connection between the voice characteristics of a speaker and comprehension difficulty. Listeners tend to make their judgement after the process of understanding is finished.

However, this week’s readings reveal to us that the voice characteristics of a speaker, cueing the speaker’s social attributes (such as age, gender, race, etc.), have impact on listeners’ understanding the speech. Sumner (2015) points out that listeners start to process the information of the speaker’s social attributes even before they process the words spoken by the speaker. Although it might help listeners better understand the speech, this “early access to social representations” easily leads to listeners’ biases, thus creates difficulty in listeners’ attention to the speech.

Sumner & Kataoka (2013) study that how phonetic variation influence semantic encoding in word recognition and recalling. They suggest that frequency may not be the only factor that lead to strong encoding; the social weighting, contained in phonetic variation, also plays a significant role in the word recognition process.

King & Sumner (2015) investigate the effect of speaker-specific phonetic cues on how listeners interpret the words spoken by the speaker. They conclude that listeners tend to have different responses to the same word spoken by different speakers with different voice characteristics, cueing their age, race, gender, etc.  

We can find many examples in our lives to prove the conclusion stated in this week’s readings. Take my experience of learning English as an example. Listening comprehension is essential for foreign students in their English learning. We need to listen to a variety of recordings from time to time. I like listening to TED talks every day. As we know, not every presenter is English native speaker in TED talks. Every time I hear a voice that has a strong accent, I will label it as “non-English native speaker” and tend to skip the talk. A strong feeling usually comes to my mind at the beginning of the speech that this speech is very difficult to understand. This example proves that the process of social information is even before the process of the word recognition. Also, consider the following sentence, “I have not done nothing wrong in my life.” If this sentence is spoken in a very young voice (it gives a cue that the speaker is a child), we tend to feel uncomfortable. It also takes more time for us to process the meaning, compared to the situation that it is spoken by an old voice. Why? Because the very young voice gives us an expectation that the speaker doesn’t have much experience. The conflict between the sentence’s meaning and the expectation from the cues given by the very young voice delay the processing of the meaning.

In the future, I would like to know more about how word processing is related to the voice characteristics, such as age, race and gender. In addition, a voice contains a variety of characteristics. Which characteristic plays a more significant role in the word processing? Is it influenced by the context?