Some words affect us more than others. It boils down to how they sound

Effective communication lies at the heart of human connection. It helps us collaborate with each other, solve problems and build relationships. And communicating clearly is a major consideration for most of us in most aspects of life.

But what if the way some words sound makes them more impactful in communication?

New research I coauthored, published in Cognition, suggests this might be the case. And it could help us all understand how to communicate better with each other.

What makes some words special?

Psycholinguistics is the study of the use and acquisition of language. A key concept in the field is known as “surprisal”. Surprisal is a measure of how expected something – for example, a word in a sentence or a sound in a word – is relative to what typically happens in the language.

All English words are built from the same vowels and consonants and follow the same rules of combination. But despite this, not all words are created equal. Words made up of unusual sequences of vowels and consonants sound more surprising to listeners. That makes them stand out in speech and affects how we process them.

We use information theory to calculate how surprising a word is. Information theory works by calculating how much information each speech sound contributes to an individual word.

Speech sounds that occur in highly predictable environments contribute relatively little information. These include sequences like /st/ as in “stick” and “stone”, and /an/ as in “can” and “and”. This is because lots of other words share the same sequences of sounds.

Speech sounds that occur in unusual environments contribute much more information. These include /koi/ like in “coil” and “coin”, and /sv/ as in “svelte” and “svabite”. This is because more unusual sequences of sounds are shared with fewer words in the lexicon.

Highly vivid words are more surprising

We applied an information theoretical analysis to data from a very large corpus of 51 million words of spoken American English – many of which were repeated – taken from movie and television subtitles.

This allowed us to assign each word in English a score showing how surprising its sounds are in English.

We then took these scores and cross-referenced them with the results of a battery of word-processing experiments. These included an auditory lexical decision task which required participants to decide whether what they heard were real words or not; a reading task; and several memory recognition tasks.

This revealed that highly vivid words – those that are very specific or concrete – are more surprising. It also revealed that both vividness and surprisal improves memory recognition.

For example, words like “dog” and “flower” are more vivid than words like “stun” and “plot”, and they also sound more surprising.

Using highly surprising word forms ensures their meaning is processed deeply and remembered better.

Challenging modern linguistics

The fact that highly vivid words sound more surprising than other words challenges the assumption in modern linguistics that the relationship between a word and its meaning is arbitrary and conventionalised.

For example, while English speakers use the sound sequence /tri:/ to refer to the concept of a “tree”, French speakers use the sequence /aʁbʁə/ (arbre) just as successfully. This shows that language users agree on what names to use within each speech community. This is what it means for language to be conventionalised.

Similarly, small things can have long names (for example, “caterpillar”) and vice versa (for example, “bus”). Very different things can have very similar names – take “pig” and “pin”. This shows that the shape of a word is not normally linked to its meaning. In other words, it is arbitrary.

An exception to this is onomatopoeic words, which linguists refer to as being “iconic”. Well-known examples are animal noises such as “cockadoodledoo” and “miaow”, and words like “splash” and “boom”.

Iconic words are often highly surprising and include unusual sounds and sound combinations. They often sound similar across languages.

Read more: Kapow! Zap! Splat! How comics make sound on the page

Why are highly vivid words more surprising?

Previous research has shown that words with negative meanings tend to be surprising. This includes words such as “snake” and “tiger”, which some researchers argue is because they clearly communicate danger. Surprising sounds mean dangerous words are less likely to be confused with other words.

But this doesn’t explain why vivid words also show high levels of surprisal.

We propose that speakers “hack” into the linguistic structure to ensure that listeners’ attention and memory is drawn toward important concepts. They do so by unconsciously creating or maintaining surprising word forms, because these words grab our attention and stick in our memory more effectively.

In this way, the sounds of words are subtly guiding how we focus and learn from language.

How can we “hack” communication?

The fact that some words induce deeper processing and better recall potentially has significant impact on how we structure communication in many domains.

Using highly surprising words might enhance personal communication. But it could also be impactful for professional communication. And it is easy to imagine that any marginal improvement afforded by using highly surprising words might be valuable in advertising and public service messaging.

It might also be the case that knowledge of high-impact words could be leveraged in education.

And if we tested enough readers of this article, we would be able to demonstrate better recall for “flower” and “dog” than “stun” and “plot”, hinting that science communication can also benefit from using some words rather than others.

The author would like to acknowledge the crucial contribution of Dr Alexander Kilpatrick to this article.