Summary: A large-scale study of 800 adults finds that pragmatic language skills — the abilities people use to read sarcasm, indirect requests, tone, and nonliteral meaning — organize into three distinct cognitive components. These components draw on social-rule knowledge, general world knowledge, and sensitivity to speech intonation, offering a new framework for studying how people interpret meaning beyond literal words.
Participants tended to perform consistently across tasks within each cluster, suggesting common underlying mechanisms for different kinds of nonliteral interpretation. The results provide a structured way to investigate communication differences across cultures and in neurodivergent populations.
Key Facts
- Three Core Systems: Pragmatic skills separate into social-context inference, physical-world causal reasoning, and tone-based interpretation.
- Context Shapes Meaning: Identical sentences can carry very different implications depending on timing, tone, and situational cues.
- Broad Applications: The framework can clarify communication challenges in autism and help explain cultural differences in indirect and direct speech.
Source: MIT
Understanding spoken language requires more than decoding words; it requires interpreting the context in which those words are spoken. If it’s pouring rain and someone calls it “lovely weather,” you recognize sarcasm because you account for context and intent.
Pragmatic language ability refers to the set of skills that help listeners infer a speaker’s intended meaning when it differs from literal wording. That includes detecting sarcasm, grasping metaphors, recognizing white lies, and interpreting indirect requests.
“Pragmatics is about reasoning why someone said something and what message they’re conveying by choosing that particular phrasing,” says Evelina Fedorenko, an MIT associate professor of brain and cognitive sciences and a member of the McGovern Institute for Brain Research.
In new research led by Fedorenko and colleagues, data from 800 participants revealed that pragmatic abilities cluster according to the kinds of inferences they require. The study found three reproducible groups: those relying on social conventions, those relying on world knowledge and causal reasoning, and those relying on intonation and prosody.
Fedorenko and Edward Gibson, MIT professor of brain and cognitive sciences, are senior authors of the study published in the Proceedings of the National Academy of Sciences. The lead authors are Sammy Floyd, now an assistant professor of psychology at Sarah Lawrence College, and Olessia Jouravlev, now an associate professor of cognitive science at Carleton University.
The importance of context
Much prior research has emphasized literal word meanings and sentence structure. But understanding what someone really intends requires integrating contextual cues — social, visual, and conversational — that guide interpretation.
As Gibson explains, a phrase like “people are leaving” can signal very different things depending on timing and situation. At the end of the night it may mean the party is ending; early on, it could imply the event is disappointing. The literal sentence stays the same, but the inferred meaning shifts with context.
About a decade ago, with support from MIT’s Simons Center for the Social Brain, Fedorenko and Gibson began developing methods to precisely distinguish the processes that support pragmatic language skills.
Rather than relying solely on brain imaging, they used an “individual differences” approach: testing large samples on a broad battery of tasks to determine whether performance patterns group together, which can indicate shared cognitive mechanisms.
If the same people perform well on social-convention tasks (like interpreting indirect requests or irony) but differently on tasks that require reasoning about physical causality or processing speech melody, that pattern suggests distinct underlying systems support each ability.
Jouravlev led the initial task development, assembling existing pragmatic tests and creating new ones to produce a 20-task battery. These tasks measured comprehension of humor, sarcasm, indirect and figurative language, and cases where intonation changes the sentence meaning — for example, emphasizing a particular word to imply omission: “I wanted blue and black socks” (stress on “black”).
“People find creative, indirect, and nonliteral ways to communicate, and this task battery captures that range,” says Floyd.
Components of pragmatic ability
Researchers recruited participants through an online platform. The initial group of 400 completed the full battery, which took about eight hours. Analysis revealed three distinct clusters of performance corresponding to social-context inferences, world-knowledge-based causal reasoning, and intonation-driven interpretation.
A second, preregistered sample of 400 participants replicated the three-cluster structure. The researchers also verified that general intelligence and basic auditory processing did not account for the clustering, supporting the view that these are separable pragmatic components.
Future work aims to link these behavioral components to neural systems using brain imaging and computational modeling. Prior research suggests that imaging findings often mirror distinctions revealed by individual-differences approaches and can map abilities to systems such as core language networks or theory-of-mind regions.
The task battery can be applied to study communication differences in autism and other brain disorders to pinpoint which pragmatic components are affected. It can also illuminate cross-cultural variation in communication styles — for example, how native speakers of more direct languages might process indirect requests differently than speakers of languages with different conversational norms.
Funding:
This research was supported by the Simons Center for the Social Brain at MIT, the National Institutes of Health, and the National Science Foundation.
Key Questions Answered:
A: Pragmatic skills include understanding sarcasm, indirect requests, metaphors, white lies, and any meaning that depends on context rather than literal wording.
A: Because each group relies on different inference systems: social norms and conventions, general world knowledge and causal reasoning, or interpretation of tone and emphasis.
A: It offers a structured approach to studying communication differences in neurotypical and neurodivergent individuals and across cultures with varied conversational norms.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- Journal paper reviewed in full.
- Additional context added by our staff.
About this cognition, language, and neuroscience research news
Author: Sarah McDonnell
Source: MIT
Contact: Sarah McDonnell – MIT
Image: The image is credited to Neuroscience News
Original Research: Closed access.
“Three distinct components of pragmatic language use: Social conventions, intonation, and world knowledge–based causal reasoning” by Evelina Fedorenko et al. PNAS
Abstract
Three distinct components of pragmatic language use: Social conventions, intonation, and world knowledge–based causal reasoning
Effective communication depends on making inferences that go beyond literal meanings — from understanding metaphors and jokes to detecting irony and interpreting intonation. This study asks whether such inferences draw on a single cognitive ability or fractionate into distinct components.
Using covariation analysis across diverse nonliteral comprehension tasks in two large samples, the authors identified three reliable components: (i) understanding social conventions, important for indirect requests and irony; (ii) interpreting contrastive and emotional intonation; and (iii) making causal inferences based on world knowledge.
In Experiment 1 (n = 376), participants completed an 8-hour battery of 20 tasks. Controlling for general cognitive ability, exploratory factor analysis revealed the three-cluster structure, which largely replicated in a preregistered Experiment 2 (n = 400) and remained robust to analytic choices.
This work reveals structure in the human communication toolkit and can guide further behavioral, neuroimaging, and computational studies to map the component processes that support linguistic and nonverbal communication.