The Point of Pitch

Bite-Sized Research Report

BLIP Lab research published last week shows that the language a person grows up hearing changes their sensory experience of the world. And it’s not just the sense of hearing – connections between the senses are also affected.

We have known for a long time that people who don’t speak Chinese have difficulty learning how pitch is involved in the meanings of Chinese words. In Mandarin Chinese for example, “mā” (said with a high, steady pitch) means ‘Mum’. But don’t confuse it with “mǎ” (said with a low dipping pitch) or you might accidentally say “Horse”. And that’s not even to mention “mà” (high, falling) or “má” (mid, rising), which mean ‘to bother’ and ‘to scold’*.

When the pitch of the voice is used as part of the meaning of words like this, linguists call it ‘lexical tone’. Lots of languages in the world have tone-systems like this, with some estimates suggesting that more than half of the population of the world speak a Tone language. These kinds of languages are especially common in Asia and Africa. You can see the four tones of Mandarin Chinese in the diagram below.

People who don’t speak a tone language have difficulty tracking these properties of words because their brains are not used to paying attention to pitch in this way. This difference in hearing emerges during childhood, when we are learning our languages for the first time. During infancy, the developing brain is bathed in acoustic information, and this triggers structural changes in the way our brains represent different kinds of sounds – we become highly skilled at hearing the sounds we hear most often, and we become less sensitive to sounds we don’t hear very much. This means that the developing brain grows and adapts in ways that help us hear the patterns in our own language. We begin this process as babies without any explicit ‘teaching’, simply by hearing some kinds of sounds more often than others.

Without this ‘tuning’ in childhood, processing lexical tones is quite difficult: It may be relatively simple to decide if one tone is higher or lower than another, but much harder to hold that representation in mind over time, and even more difficult to hold a representation for tones that change their pitch during a single word. For this reason, English speakers find it easier to tell the difference between pairs of tones that are high-versus-low (especially if the difference appears at the start). For speakers of Mandarin Chinese, on the other hand, the way the pitch changes is critical for word meaning, so their brains have ‘tuned in’ to this feature. Chinese speakers find it easier to tell the difference between pairs of tones that differ in how much they change. You can see the patterns in the diagram below.

We wanted to know if the brain’s linguistic ‘tuning’ is leaking out beyond the sense of hearing, and creating differences in the way our senses are connected.

In this study, we asked people who spoke Mandarin Chinese and English to match up sounds with shapes. The sounds we used were “ee” and “oo ” pronounced using the four tones of Mandarin Chinese.  (The audio files, and other materials used in the study are collected in the Open Science Framework repository for this project: https://osf.io/364fm/)

We asked people to tell us which of these shapes (below) ‘goes best’ with each of the different sounds.

In two experiments, we compared how many people chose shapes for different sounds, and found an interesting pattern: Chinese-dominant bilinguals who grew up and went to school in mainland China made different choices when compared to English speakers who don’t speak any tone languages: English speakers matched tones with a high pitch to pointy shapes, and tones with a low pitch to curvy shapes. Chinese speakers matched tones that were steady to smooth shapes, and tones that changed dramatically to pointy shapes.

So what happens when a tone is both high and steady – like Tone 1, the tone in “mā” (‘Mum’)? The two groups gave exactly opposite choices! So the Chinese-speakers’ match for Tone 1 was curvy, and the English-speakers’ match for the same tone was pointy!

We also tested Chinese-English bilinguals who grew up and went to school in bilingual Singapore – these participants were much more ‘balanced’ in their two languages, meaning they were highly fluent in both from a young age. These Singaporeans swapped between the English pattern and the Chinese pattern, depending on whether we presented the Tones one by one (Chinese pattern), or in pairs (Chinese pattern for Chinese (steady/dynamic) pair; English pattern for English high/low pair).

This pattern swapping might seem peculiar to people who only speak one language. However, it’s a kind of flexibility that appears to be quite natural for bilinguals – swapping between languages depending on who you are talking to,what you are trying to say, or even which language has a better word to match your mood. Although this new type of language swapping is conceptually similar to other bilingual types of switching, we have never seen sensory mapping switch like this before. It shows that the language-specific sensation can have more than one form within a single person.

This is the first time in more than 80 years that groups of people have shown exactly opposite choices in this kind of task, and that a group of people have shown different preferences under different conditions. Up until recently, previous research into this question has led researchers to believe that people all over the world have the same linkages between the senses, and that an individual’s matching pattern is more-or-less stable over time.

Earlier this year, we showed that when you do this kind of test with people whose language doesn’t have the sound combinations you are trying to test, then they don’t show the expected effect. This time, we have shown that if your language causes you to hear different properties of the same sounds (pitch height/pitch change) then the sensory connections between hearing and vision also follow this pattern. This means that whatever kind of sensory connections we are born with, they are shaped by the sensations we experience through our lives – including the sensory structure of language. It also means that people who speak different languages inhabit different “sensory worlds” that subtly shape our experience of reality.

It’s comforting to think that we share our sensory world with people who speak the same language – but we can also step, however briefly, into another, when we learn to listen to other tongues. Some of us may stand staring at the shores of an unfamiliar sensory world, barely able to dip our toes into the watery depths that divide us. Others will paddle, ankle deep in the curious foam, or dive head-first into the waves, and begin the slow process of learning to swim. I’ll be right here on the shorelines with my trousers rolled up, exploring the inter-tidal zone. I’ll tell you all about it as I go!

Suzy J Styles

 

More BLIP Bites?

SOURCES. The article described in this BLIP Bite is:

Shang, N. and S. J. Styles (2017). “Is a high tone pointy? Degree of pitch-change in lexical tone predicts of sound-to-shape correspondences in Chinese bilinguals.” Frontiers 8(2139): 1. Open Access Link: https://doi.org/10.3389/fpsyg.2017.02139

Footnotes:

* People familiar with  Chinese will no doubt be aware that if a Chinese learner says a word with an incorrect tone, there is usually enough information in the context to figure out what the person was trying to say. This is similar to when people mix their words up in other languages or makes a slip of the tongue (for example, saying “The river biter” instead of “The River Tiber”, or something similar) – context can usually help us to solve the puzzle of what the person meant to say, but everyone would agree that there was an error.