Context is key in how people understand language
[post_content]
Disclaimer: This article has been automatically aggregated from
When describing someone running, most people would say exactly that: “The person is running.” They probably wouldn’t say the person is moving, even though both descriptions are technically correct. But what if that person were the only one moving in a room full of people standing still?
According to research by Anna Papafragou, a professor in the Department of Linguistics in the School of Arts & Sciences, and her group, the context in which words are used can significantly influence how people interpret language and determine whether they deem a description appropriate.
Their findings suggest that people evaluate verbs not simply according to their literal meanings but also according to how effectively they communicate information in a particular situation.
“The kind of description you use is going to vary depending on context and how informative and relevant that description is for the conversation,” Papafragou says.
This study builds on decades of psychological and linguistic studies examining how people categorize objects and choose words to describe them. The findings could inform clinical approaches to language difficulties and the development of artificial intelligence systems designed to interpret human communication.
People generally prefer what linguists call “basic-level” nouns, such as dog or chair, over broader—”superordinate”—terms such as “animal” or “furniture,” explains Papafragou. For example, when shown a photograph of a dog, most people will identify it as a dog rather than an animal or a Dalmatian, even though all three descriptions might be accurate.
To investigate whether similar preferences apply to verbs, linguistics Ph.D. student Ruidi Huang and undergraduate student Izzy Chapman worked with Papafragou to conduct two experiments involving 80 adults.
Half of the participants viewed photographs of people performing different actions. A cartoon character named Lion described each action using either a basic-level verb, such as “A woman is running,” or a broader verb, such as “A woman is moving.”
They evaluated the descriptions by rewarding Lion with a small, medium or large cupcake, depending on how appropriate they considered his statement. The researchers found that participants consistently preferred basic-level verbs, giving significantly lower rewards to broader descriptions. A second experiment examined whether introducing additional visual context changed those preferences.
The other half viewed groups of four photographs depicting different actions. In some cases, all four actions could be described using the same broad verb. In others, only one action fit that description.
For example, describing someone as moving provided little useful information when the other photographs also showed people running, jumping or dancing. However, the description became more informative when the other photographs showed people sitting, standing or lying down.
Although participants continued to favor more specific verbs overall, the preference was significantly weaker when the broader description helped distinguish the target action from the alternatives.
The findings could also have implications for how researchers study children’s language development, Papafragou says. Previously, a study led by her former graduate student June Choe involving children ages 4 and 5 found that the children demonstrated an ability to recognize when broader descriptions were appropriate. This challenged the belief that broader categories require more abstract thinking, potentially making them harder for young children to understand.
For Papafragou, the broader significance lies in understanding how people recognize not only what words mean but what speakers are trying to communicate.
Provided by
University of Pennsylvania
Citation:
Context is key in how people understand language (2026, October 19)
retrieved 19 October 2026
from https://phys.org/news/2026-10-context-key-people-language.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
part may be reproduced without the written permission. The content is provided for information purposes only.
for informational purposes only. We do not claim ownership, accuracy, or liability for the content provided. All rights belong to the original publisher.