Pragmatics in psycholinguistics is the scientific study of how people infer communicative meanings from words, sentences, speakers, and situations. A sentence supplies vocabulary and grammar, but the meaning a listener takes away also depends on what the participants know, what they are trying to achieve, and why the speaker chose one expression rather than another. “It is cold in here” can be a temperature report, a request to close a window, or an explanation for putting on a coat. Psycholinguistics asks how listeners select among such interpretations, which contextual cues they use, how fast the inferences occur, and why they sometimes fail. Goodman and Frank (2016); Papafragou and Grigoroglou (2025).
Pragmatic comprehension is an evidence-sensitive form of language processing. Linguistic representations, contextual expectations, alternative utterances, perspective information, and knowledge of communicative conventions contribute to interpretation. Different pragmatic tasks call on these resources to different degrees. Experiments with eye movements, reaction times, electrophysiology, and probabilistic models demonstrate rapid contextual influences, while leaving important questions about the precise computations unresolved.
What Is Pragmatics in Psycholinguistics?
Pragmatics concerns language use in context. Psycholinguistic pragmatics investigates the representations and processes that make such use possible. A semantic interpretation of “Some students passed” is ordinarily compatible with all students passing. A pragmatic interpretation may infer that not all passed, because a speaker who knew that everyone passed might have selected the stronger word “all.” Whether listeners derive that inference, and when, depends on the situation, available alternatives, and task. Breheny, Katsos, and Williams (2006); Degen and Tanenhaus (2015).
The field connects linguistics, experimental psychology, cognitive science, developmental research, social cognition, and computational modeling. Core phenomena include reference, implicature, presupposition, indirect speech acts, common ground, conversational repair, and the contextual interpretation of irony. These categories overlap in everyday conversation, but experiments can target distinct mechanisms.
Semantics, pragmatics, and communicative action
Semantics describes conventional lexical and grammatical contributions to meaning, including compositional meaning and truth conditions. Pragmatics explains how context and communicative goals shape the interpretation of utterances. The boundary can itself be theory-dependent: some approaches assign contextual enrichment to semantic representation, while others explain it through pragmatic inference. Researchers therefore evaluate observable predictions rather than assume a fixed boundary resolves every case.
A helpful framework distinguishes what an expression conventionally encodes, what proposition is communicated in a particular context, and what the speaker accomplishes by using it. “Can you reach the shelf?” may communicate a genuine question about ability or perform an indirect request. The same form supports both because the action intended by an utterance extends beyond its syntax.
Major kinds of pragmatic meaning
Reference identifies which person, object, or event a speaker means. Implicature concerns additional messages inferred from utterance choice. Presupposition concerns information treated as established background, as when “She stopped jogging” normally presents earlier jogging as given. Speech acts concern performing requests, promises, warnings, and other actions through language. Common ground concerns information participants treat as mutually accessible for coordinating their interpretations. Irony, politeness, humor, and figurative expression draw on several of these operations at once.
Why Literal Words Leave Meaning Open
Even grammatically complete utterances contain interpretation problems. “She is ready” invites the question ready for what. “The blue one” requires an identifiable comparison set. “We will meet there tomorrow” depends on speaker, time, and location. An apparently explicit sentence may also carry an unspoken consequence: “The elevator is broken” in response to “Shall we take the elevator?” communicates an answer without using yes or no.
Meaning is reconstructed under uncertainty. A listener combines the words with the ongoing question, background assumptions, social relationship, and estimates about what the speaker knows. The alternatives matter: saying “some” when “all” was an available, accurate, and relevant choice can convey more than saying “some” in a context where the stronger alternative would have been irrelevant. Degen and Tanenhaus (2015).
Context as evidence
Context includes preceding discourse, visual scene, speaker knowledge, perceptual access, shared history, conversational task, intonation, gesture, and the norms of the interaction. It changes during a conversation. A clarification can reveal that two participants have been using different referents; a response can indicate which assumption needs correction. This makes pragmatic meaning dynamic and interaction-sensitive. Accounts of conversational grounding explain communication as a collaborative coordination of evolving mutual knowledge. Clark and Brennan (1991).
Major Theories of Pragmatic Interpretation
Grice's account of cooperation and implicature
H. P. Grice proposed that conversational participants normally assume recognizable cooperative purposes. His familiar maxims concern appropriate quantity of information, truthfulness, relevance, and clarity. These provide an explanation for why the choice of an utterance can suggest more than its asserted content. They describe inferential expectations, not literal behavioral rules that every speaker obeys. Grice (1975).
Suppose one colleague asks, “Was the proposal approved?” and the other replies, “The slides were beautifully designed.” The answer could imply a disappointing decision because the listener seeks a relevant explanation for the absence of a direct reply. The implication is cancelable; the speaker might later add that the proposal was also approved. That defeasibility helps separate implicature from an entailment.
Relevance theory
Relevance theory describes communication as an inferential process guided by expected cognitive effects and processing effort. Under this framework, listeners seek an interpretation worth the cognitive effort required to construct it, given the utterance and accessible contextual information. A warning that “the roads are icy” may indirectly advise delaying a trip when a departure is being planned. This interpretation depends on the practical goal, rather than an invariant extra meaning of the words. The theory provides an influential explanation of contextual guidance, while particular processing claims require empirical tests. Sperber and Wilson (1995).
Rational speech act models
Rational speech act models, often abbreviated RSA, treat communication as reciprocal reasoning between speakers and listeners. A simplified literal listener interprets which states or referents fit an expression. A pragmatic speaker selects utterances that help a listener identify the intended referent, taking communicative usefulness and cost into account. A pragmatic listener then updates the probability assigned to different speaker intentions after hearing the chosen expression. This makes predictions explicit and measurable rather than purely verbal.
Referential language games have shown that probabilistic pragmatic models can fit systematic patterns in human choices. Such a fit supports the model as an explanatory tool; it does not establish that every participant consciously executes Bayesian calculations or that identical computations generate all pragmatic phenomena. Frank and Goodman (2012); Goodman and Frank (2016).
Interactive and constraint-based processing
Interactive theories describe interpretation as the integration of graded constraints, including lexical meaning, linguistic alternatives, information about speakers, question relevance, and prior discourse. A listener might rapidly enrich “some” under one display and hesitate under another. Constraint-based accounts attempt to explain this variability through the salience and reliability of the available evidence. Degen and Tanenhaus (2015).
These approaches answer different explanatory questions: Grice provides a logic of communicative inference; relevance theory emphasizes context selection; probabilistic models quantify inference; processing theories test the time course and resource constraints. A successful scientific account must specify where these ideas make distinct predictions.
Conversational Implicature and the Unsaid
Conversational implicature is an inferred message conveyed through the communicative use of an expression beyond what the sentence conventionally asserts. “Some reviewers liked the book” may communicate that not all did. “I have a deadline” in response to a dinner invitation may convey a refusal. Both interpretations are sensitive to the question, alternatives, and presumed speaker goals.
Scalar implicature: the some–all problem
Scalar implicature involves expressions that differ in informational strength. Some versus all and possible versus necessary are familiar examples. A speaker's weaker choice can suggest that the stronger statement is unsupported. Yet that inference can disappear when the speaker lacks the relevant knowledge or when the stronger alternative would have been pragmatically inappropriate.
Experiments show a complex time course. Breheny, Katsos, and Williams found that scalar inferences depend on context, challenging a fully autonomous default mechanism. Grodner and colleagues observed rapid enrichment in visual-world studies under suitable conditions. Degen and Tanenhaus demonstrated that interpretations vary with graded features of the alternatives and situation. The established conclusion is that pragmatic enrichment can be rapid and contextual; the general processing architecture remains contested. Breheny, Katsos, and Williams (2006); Grodner et al. (2010); Degen and Tanenhaus (2015).
Particularized implicatures
Particularized implicatures rely on local circumstances. Asked whether a manuscript is ready, someone answers, “The figures are finally finished.” That may suggest other sections remain unfinished, but a different question could make it simply a progress report. Experimenters can manipulate the preceding question while keeping the answer constant to determine whether the inference responds to context. Establishing contextual sensitivity is easier than uniquely identifying whether intention reasoning, discourse coherence, or lexical prediction generated it.
Entailment, presupposition, and implicature
Entailment is a logical consequence of asserted content. Presupposition is information presented as already accepted or accommodated, often surviving questions and negation. Implicature is additional meaning inferred from speaker choice in context. “Did Maya stop smoking?” ordinarily presupposes earlier smoking, while “Some guests arrived” can pragmatically suggest that some did not. These categories can interact; researchers separate them using targeted judgment, memory, and time-sensitive comprehension tasks.
Referential Communication: How We Identify Intended Objects
Reference resolution is particularly amenable to experiment because researchers can control the candidate objects. Listeners may hear “Pick up the tall glass” while looking at several glasses. Eye movements during the phrase reveal how alternatives compete as information arrives. Sedivy and colleagues found that contrasting objects in a display can influence the interpretation of an adjective while it is still unfolding, supporting rapid contextual integration. Sedivy et al. (1999).
Common ground versus privileged knowledge
Common ground includes information participants treat as mutually available. Privileged knowledge is available to only one participant. In a classic arrangement, a listener sees an object behind a screen that the speaker cannot see. A strictly common-ground interpretation should exclude that object from the speaker's possible referents. Keysar and colleagues documented temporary consideration of privileged objects. Hanna, Tanenhaus, and Trueswell found early effects favoring common-ground candidates even when privileged alternatives also competed. Together these results support graded perspective constraints rather than either perfect early exclusion or universal early egocentrism. Keysar et al. (2000); Hanna, Tanenhaus, and Trueswell (2003).
Audience design
Speakers adapt descriptions to their interlocutors. An unfamiliar object may receive an extended description, while a previously established referent may be identified by a short expression. Such audience design depends on memory for prior interactions, partner-specific knowledge, and the communicative task. It is graded and susceptible to interference. Research must distinguish a speaker's intention in selecting a phrase from a listener's process of resolving it. Papafragou and Grigoroglou (2025).
Speaker Intentions and Indirect Speech Acts
An utterance can do something besides describing a state of affairs. “I promise to return it” creates a social commitment, and “Can you pass the salt?” ordinarily functions as a request. Successful pragmatic comprehension identifies the intended conversational action in addition to the wording. Familiar requests may be recognized through established conventions; novel hints may require a more elaborate inference about what the speaker wants.
An indirect request can soften a demand, acknowledge the listener's choices, or convey politeness. “This room could use some fresh air” might be a request to open a window among people who share a routine, yet a new visitor might hear only a statement about comfort. Work on linguistic politeness examines how forms express social relationships and obligations. Its claims about recurring patterns are theoretical contributions to be assessed across languages and cultural settings, not evidence for a universal single mental algorithm. Brown and Levinson (1987).
Recognizing the intended action
The listener integrates sentence type, conventional usage, ongoing activities, prosody, practical possibilities, and the social relationship. No one cue is infallible. A literal ability question may be genuine in a setting that tests physical reach; the same form may function as a request during dinner. Experiments must compare contexts that preserve words while changing the plausible communicative purpose, or use novel forms to distinguish learned conventions from flexible intention recognition.
Common Ground and Grounding Across Conversation
Dialogue requires mutual coordination across turns. Participants introduce objects, make claims, acknowledge information, check whether a message was understood, and update assumptions. Clark and Brennan's treatment of grounding emphasizes that shared understanding is continually established and revised through joint activity. Common ground is therefore a working coordination resource rather than a complete inventory of two identical minds. Clark and Brennan (1991).
A request for clarification such as “Which report?” reveals that an expression did not identify its intended referent. “Do you mean the version from Tuesday?” proposes a candidate interpretation. These repair moves can reveal the context in which earlier understanding broke down. A laboratory study can measure precisely which alternative was chosen, while a naturalistic dialogue study can show how real participants repair emerging misunderstanding. Both forms of evidence are valuable, but observational correlations do not uniquely identify the mental causes of an error.
The Time Course of Pragmatic Comprehension
Pragmatic information is not invariably postponed until a whole sentence has been decoded. Visual contrast can affect the interpretation of an adjective, common-ground assumptions can bias reference selection, and discourse context can alter the processing of unusual word combinations. The timing of a response depends on what kind of evidence becomes available, the cognitive operation required, and the measurement used. A fast effect on eye movements is evidence of graded influence, not automatically proof that the listener has fully determined the speaker's intended meaning.
Visual-world eye tracking
The visual-world method records where participants look as they hear language and interact with visible referents. Fixation patterns can reveal competing objects before the participant responds. Researchers estimate whether speaker-access information, contrast sets, or scalar alternatives affect attention early in the unfolding phrase. Interpretation requires care: eye movements have perceptual and motor delays, visually salient objects can attract looks for independent reasons, and looking at a candidate does not establish final endorsement. Sedivy et al. (1999).
Reading times and pragmatic judgments
Reading-time measures identify locations where expectations increase processing difficulty. Truth-value and picture-selection tasks indicate which interpretations people accept. But different judgments can reflect different standards. Accepting “some” when all objects satisfy the property could mean the person recognizes the literal truth, tolerates underinformativity, or follows a task instruction differently. Careful research manipulates acceptance criteria and compares measures rather than treating a single response as a transparent window into pragmatic competence. Katsos and Bishop (2011).
Event-related potentials and discourse
Electroencephalography provides time-sensitive neural responses during language processing. The N400 is sensitive to several aspects of semantic and contextual integration. Nieuwland and Van Berkum showed that discourse support can substantially influence the processing of otherwise improbable expressions in fictional contexts. This challenges models in which context is always used only after a context-independent semantic computation. However, the N400 is not a unique biomarker of a particular pragmatic inference or of a speaker's intention. Nieuwland and Van Berkum (2006).
Intention Recognition and Theory of Mind
Theory of mind concerns the representation of others' beliefs, knowledge, desires, and intentions. Many pragmatic tasks invite this sort of reasoning: the listener must determine what a speaker could see, which alternative message was available, or why a certain remark was indirect. Yet a common pragmatic response may be produced through a well-learned convention without a detailed explicit calculation of another person's mental state.
A 2025 synthesis by Papafragou and Grigoroglou reviews evidence about the relation between pragmatic communication and theory of mind, especially for reference and implicature. Their analysis emphasizes the need to distinguish rich mental-state reasoning from simpler cues and learned communicative expectations. Some tasks require more explicit belief reasoning than others. Behavioral success on an indirect request therefore does not automatically reveal the specific representation used to generate the response. Papafragou and Grigoroglou (2025).
Designing a test of speaker beliefs
A useful experiment can vary what a speaker knows while preserving the words and perceptual scene available to the listener. If interpretations change reliably with the speaker's informational access, the result supports perspective-sensitive processing. Further controls are required to decide whether the effect arose from belief representation, learned object-selection strategies, or other contextual cues. The distinction between behavioral prediction and explanation of mechanism is fundamental.
Figurative, Ironic, and Evaluative Meanings
Metaphor, hyperbole, understatement, idiom, and irony can depend on pragmatic inference, although their detailed lexical and conceptual processing constitutes a neighboring scientific intent. “Wonderful timing” uttered after an hour-long delay can convey criticism when situation and tone make literal praise implausible. A listener must integrate the utterance with shared expectations and the speaker's stance.
Irony is especially revealing because people can identify an implied evaluation while disagreeing about the speaker's emotional attitude. A review of decades of children's irony research reports that irony comprehension often emerges later than other nonliteral competencies, while the estimated age depends on the form of irony and the structure of the test. No universal developmental age or single processing theory can be inferred from one study design. Fuchs (2023).
Prosody, Gesture, and Multimodal Pragmatics
Speech supplies rhythm, emphasis, timing, and intonation in addition to lexical content. Contrastive stress can identify which alternative matters; a pause can imply hesitation or invite another participant's turn; a rising contour can express uncertainty or inquiry. Gesture and gaze can help establish intended referents. Their meanings depend on language and context: eye direction may reflect attention, hesitation, or social convention rather than a stable universal message.
In face-to-face communication, listeners integrate auditory and visual evidence while interpreting an utterance. In signed languages, reference, information structure, and conversational coordination similarly operate in a visual-manual modality. This makes pragmatics a property of language-in-use across modalities, rather than a property tied exclusively to spoken sound.
How Pragmatic Abilities Develop
Pragmatic development involves linguistic learning, communicative experience, cognitive resources, perspective sensitivity, and familiarity with social conventions. Young children already participate in meaningful interaction, yet different inference tasks develop at different rates. A child's performance on an artificial laboratory judgment should not be treated as a complete measure of their everyday ability to communicate.
Nadig and Sedivy reported that five- to six-year-old children used common-ground information during online reference resolution, challenging descriptions of uniformly egocentric early comprehension. Katsos and Bishop found evidence that children can recognize informativity while tolerating pragmatic violations more readily than adults. Hence accepting an underinformative statement does not necessarily show inability to derive its implication. Nadig and Sedivy (2002); Katsos and Bishop (2011).
Why different pragmatic skills have different trajectories
A 2023 research review by Pouscoulous describes apparently different developmental routes for implicit, figurative, and ironic meanings and highlights interactions with language knowledge and executive functions. Differences across experiments may reflect the complexity of stories, vocabulary demands, response formats, and the exact mental states participants must evaluate. Researchers should control these factors before attributing age variation to a single underlying pragmatic capacity. Pouscoulous (2023).
Individual Differences and Cultural Context
Pragmatic interpretation depends on personal experience, attention, memory, language proficiency, familiarity with the communicative partner, and the conventions of the interaction. Directness, silence, politeness, and conversational timing can convey different implications across settings. A response that is efficiently polite in one community may sound uncertain or evasive in another. Cross-cultural experiments need equivalent tasks and locally valid interpretations rather than assuming that a single English-language judgment captures universal practice.
Multilingual communication introduces additional sources of variation. Direct translation can preserve literal sentence meaning while shifting the practical force of an indirect request. A multilingual participant may draw an inference differently because of context familiarity, linguistic proficiency, or social conventions. Those findings do not establish a universal bilingual advantage or deficit in pragmatic cognition. Research should specify the particular inference and the language of testing.
Ordinary Pragmatics and Clinical Communication
Communication styles and occasional misunderstandings are part of normal variation. Persistent difficulties with language use may also arise in clinical conditions, but a single unexpected conversational response is not diagnostic evidence. The American Speech-Language-Hearing Association distinguishes social communication, social interaction, and social cognition in its clinical framework for social communication disorder. It recommends culturally and linguistically appropriate assessment and differential consideration of other conditions. ASHA Social Communication Disorder Practice Portal.
Autism, developmental language disorder, acquired language impairments, and other clinical conditions can affect pragmatic behavior through different pathways. Diagnosis depends on professional evaluation, developmental and functional context, validated instruments, and differential assessment. Laboratory measures of implicature or perspective sensitivity contribute to scientific understanding but cannot independently determine a clinical disorder. This article examines general cognitive mechanisms; clinical diagnostic and treatment questions require dedicated clinical evidence.
Research Methods and Causal Inference
Scientific pragmatics compares interpretations under systematically varied conditions. A well-controlled experiment changes the speaker's perspective, contextual question, set of alternatives, or discourse information while preserving as much linguistic material as possible. The dependent measures may be choice probabilities, eye movements, reading times, or neural responses. Each measure speaks to a particular aspect of processing and has its own limitations.
Controlled referential games
Language games specify a finite set of referents and potential utterances. They permit precise manipulations of what a speaker can know and how informative different descriptions are. They can also be fitted with rational speech act models to generate quantitative predictions. The same feature that produces experimental control—a small, explicitly defined world—limits generalization to richer natural conversations. Frank and Goodman (2012).
Naturalistic dialogue and corpora
Recorded interaction reveals how referents are introduced, misunderstandings repaired, requests made, and shared context updated across turns. Conversation corpora offer ecological richness and show phenomena that researchers might overlook when constructing stimuli. Because interlocutors self-select expressions and situations, however, a correlation between one construction and one interpretation cannot by itself prove the underlying mechanism. Experiments and observational methods answer complementary questions.
Computational model comparison
Models can make assumptions explicit about speaker knowledge, referential priors, communication costs, and listener inference. Predictions should be tested on new conditions and compared with simpler alternatives, not evaluated only by fit to the data on which parameters were tuned. A model that predicts referential choices need not also explain irony, conversational repair, and discourse comprehension. The degree of generalization remains an empirical question. Goodman and Frank (2016).
Separating competence from task strategies
A participant may understand an implicated meaning but select a literal response because instructions emphasize logical truth. Conversely, guessing an expected response does not establish a rich mental-state representation. Converging methods, graded judgments, comprehension checks, and careful manipulation of task demands support stronger conclusions. Experimental outcomes should be interpreted at the level actually measured.
Pragmatics and Predictive Language Processing
Language comprehension exploits expectations shaped by linguistic and discourse evidence. Speaker knowledge can change which referent is likely, while a conversational goal can alter which reply is expected. Yet prediction is an umbrella concept covering probabilistic expectations, anticipatory activation, prediction-error responses, and more elaborate actively generative accounts. A contextual benefit in comprehension does not discriminate among these mechanisms by itself. Kuperberg and Jaeger (2016).
The processing of extended discourse further involves tracking characters, propositions, questions, and communicative commitments. Context can make an otherwise unusual utterance meaningful. ERP evidence for discourse sensitivity shows that larger narratives can influence relatively early processing responses, although those responses do not uniquely identify a particular model of pragmatic intention. Nieuwland and Van Berkum (2006).
Pragmatics in the Artificial Era and Large Language Models
Large language models can generate context-sensitive answers and perform selected tasks involving implication, reference, and indirect meaning. These are measurable linguistic behaviors. The mechanisms that produce them may differ from human communication, where joint attention, shared perception, lasting social commitments, and assumptions about participants' knowledge shape interaction. Pragmatic task performance should be evaluated independently of claims about human-like cognition, intentions, or subjective experience.
A 2025 survey of datasets for evaluating language-model pragmatics identified methodological challenges in assessing implicatures, references, and realistic interaction. A 2025 experimental comparison reported high scores for a particular model on Gricean dialogue tasks. Such findings support conclusions about performance under specified evaluation conditions, including the prompts, answer formats, population, and model version; they do not establish equivalence to the entire human pragmatic system. Ma et al. (2025); Bojić, Kovačević, and Čabarkapa (2025).
Grounding versus conversational imitation
Human interlocutors often construct shared understanding through action, joint perception, correction, and continuing social relationships. A language model may instead condition its output on text, images, learned statistical structure, and its system design. Similar surface responses do not establish identical internal computations. Questions of artificial understanding require careful theoretical definitions and independent evidence. Questions of artificial subjective experience belong to a distinct philosophical and scientific inquiry.
Evidence Strength and Scientific Disagreement
Established empirical findings include rapid effects of visual contrast on reference, systematic influence of shared perspectives, contextual modulation of scalar inference, and differences between pragmatic sensitivity and the willingness to reject an underinformative statement. These findings have been investigated through multiple paradigms. Sedivy et al. (1999); Hanna, Tanenhaus, and Trueswell (2003); Katsos and Bishop (2011).
Mechanism-contested issues include whether a general-purpose recursive inference procedure explains all implicatures, the degree to which speaker beliefs are explicitly represented, and whether particular types of contextual meaning are computed by one shared architecture. Evidence is limited when generalizations are drawn from narrow language samples, artificial displays, or single response types. Findings about large language models are often preliminary relative to broad claims about interaction across contexts and model versions. Theoretical and philosophical interpretations should be identified at their appropriate evidential level.
Practical Implications
Understanding pragmatics helps explain why concise utterances can require clarification and why the same sentence changes function across settings. Educators can separate vocabulary comprehension from interpretation of a speaker's intentions. Multilingual teams can check whether apparently shared expressions carry the same practical force. Designers of dialogue systems can make assumptions visible and provide opportunities for repair. Each application still requires direct evaluation before strong claims about effectiveness are justified.
Related scientific territory is treated in the published Communication, Conflict and Social Skills research chapter. The distinct effects of wording on judgment and choice belong to Framing Effect: How the Wording of Choices Changes Decisions. This article's scientific focus remains the comprehension of intended communicative meaning.
Open Research Problems
Important unresolved questions include how listeners represent uncertainty about what speakers know, which pragmatic processes are shared across modalities and languages, how conventions develop through repeated interaction, and how laboratory effects scale to genuine multi-turn communication. Replication across populations and tasks is important because pragmatic inference is sensitive to wording, display design, and testing assumptions. Researchers should distinguish the existence of a contextual effect from the stronger claim that one mental mechanism has uniquely caused it.
Frequently Asked Questions
What is an everyday example of pragmatics?
“Could you close the window?” is grammatically a question about ability but commonly functions as a request. The intended meaning follows from the social action and the context in which it is spoken.
How is pragmatics different from semantics?
Semantics studies conventional contributions of words and grammar to meaning. Pragmatics investigates how the utterance is interpreted and used in context. The interaction between them is an active theoretical research topic.
What is conversational implicature?
It is additional communicative meaning inferred from an utterance choice and its context. “Some people agreed” can suggest “not all” when a stronger alternative was salient, but the inference depends on speaker knowledge and relevance.
Is pragmatic inference always slow?
No. Context and common ground can influence comprehension incrementally, and some scalar inferences arise quickly under supportive conditions. The time course varies with the inference, experiment, and accessible evidence.
Does pragmatics require theory of mind?
Many tasks involve assumptions about speaker intentions or knowledge. Some familiar conventions may be interpreted without explicit, detailed belief reasoning. Research distinguishes degrees and types of mental-state involvement. Papafragou and Grigoroglou (2025).
Do children understand implied meaning?
Children demonstrate pragmatic sensitivity in appropriately designed studies, though performance depends on vocabulary, task instructions, context, and the specific inference. Accepting an underinformative statement does not prove that its pragmatic weakness went unnoticed. Katsos and Bishop (2011).
Are unusual communication styles disorders?
Communication varies across people and communities. Clinical conclusions require persistent functional evidence and culturally appropriate professional evaluation; a style difference or isolated misunderstanding cannot establish a diagnosis. ASHA.
Can an AI system understand implicature?
Models can be tested on pragmatic tasks, and some perform well under particular conditions. Whether the processing resembles human intention recognition or supports broader claims about understanding remains a separate question. Ma et al. (2025).
Conclusion
Pragmatics in psycholinguistics explains how linguistic forms become contextual communicative meanings. Experiments establish that listeners use alternative expressions, shared knowledge, and discourse information during comprehension. Models of cooperation, relevance, probabilistic reasoning, and interactive processing provide distinct but complementary perspectives. The most productive scientific approach identifies precisely which inference is under investigation, measures its time course and conditions, and separates robust behavior from theoretical interpretation.
Related Articles
Figurative Language Comprehension: Metaphors, Idioms, and Meaning — the published companion on processing metaphors, idioms, and other nonliteral expressions.
Discourse Comprehension: How We Understand Connected Language — scientific mechanisms of referent tracking, connected meaning, discourse inference, and coherence.
Prosody and Intonation: How Speech Conveys Meaning — a published account of how spoken pitch, stress, and phrasing contribute to communicative interpretation.
Sentence Processing: How We Understand Grammar in Real Time — the distinct grammatical-processing foundation for pragmatic interpretation.
Cluster Language, Thought and Psycholinguistics — the verified parent scientific chapter on language processing, meaning, comprehension, and related research.
Cluster Communication, Conflict and Social Skills — the adjacent published chapter on interpersonal communication and social interaction.
Framing Effect: How the Wording of Choices Changes Decisions — the independently owned topic of how expression framing influences judgment and choice.
References
American Speech-Language-Hearing Association. (n.d.). Social communication disorder. ASHA Practice Portal. Official clinical source.
Bojić, L., Kovačević, P., and Čabarkapa, M. (2025). Does GPT-4 surpass human performance in linguistic pragmatics? Humanities and Social Sciences Communications, 12, 794. Publisher.
Breheny, R., Katsos, N., and Williams, J. (2006). Are generalised scalar implicatures generated by default? An on-line investigation into the role of context in generating pragmatic inferences. Cognition, 100(3), 434–463. DOI.
Brown, P., and Levinson, S. C. (1987). Politeness: Some Universals in Language Usage. Cambridge University Press. Publisher.
Clark, H. H., and Brennan, S. E. (1991). Grounding in communication. In L. B. Resnick, J. M. Levine, and S. D. Teasley (Eds.), Perspectives on Socially Shared Cognition (pp. 127–149). American Psychological Association. University-hosted text.
Degen, J., and Tanenhaus, M. K. (2015). Processing scalar implicature: A constraint-based approach. Cognitive Science, 39(4), 667–710. DOI.
Frank, M. C., and Goodman, N. D. (2012). Predicting pragmatic reasoning in language games. Science, 336(6084), 998. DOI.
Fuchs, J. (2023). 40 years of research into children's irony comprehension: A review. Pragmatics and Cognition, 30(1), 1–30. DOI.
Goodman, N. D., and Frank, M. C. (2016). Pragmatic language interpretation as probabilistic inference. Trends in Cognitive Sciences, 20(11), 818–829. DOI.
Grice, H. P. (1975). Logic and conversation. In P. Cole and J. L. Morgan (Eds.), Syntax and Semantics, Vol. 3: Speech Acts (pp. 41–58). Academic Press. Bibliographic source.
Grodner, D. J., Klein, N. M., Carbary, K. M., and Tanenhaus, M. K. (2010). “Some,” and possibly all, scalar inferences are not delayed: Evidence for immediate pragmatic enrichment. Cognition, 116(1), 42–55. DOI.
Hanna, J. E., Tanenhaus, M. K., and Trueswell, J. C. (2003). The effects of common ground and perspective on domains of referential interpretation. Journal of Memory and Language, 49(1), 43–61. DOI.
Katsos, N., and Bishop, D. V. M. (2011). Pragmatic tolerance: Implications for the acquisition of informativeness and implicature. Cognition, 120(1), 67–81. DOI.
Keysar, B., Barr, D. J., Balin, J. A., and Brauner, J. S. (2000). Taking perspective in conversation: The role of mutual knowledge in comprehension. Psychological Science, 11(1), 32–38. DOI.
Kuperberg, G. R., and Jaeger, T. F. (2016). What do we mean by prediction in language comprehension? Language, Cognition and Neuroscience, 31(1), 32–59. DOI.
Ma, B., Li, Y., Zhou, W., Gong, Z., Liu, Y. J., Jasinskaja, K., Friedrich, A., Hirschberg, J., Kreuter, F., and Plank, B. (2025). Pragmatics in the era of large language models: A survey on datasets, evaluation, opportunities and challenges. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (pp. 8679–8696). DOI.
Nadig, A. S., and Sedivy, J. C. (2002). Evidence of perspective-taking constraints in children's on-line reference resolution. Psychological Science, 13(4), 329–336. DOI.
Nieuwland, M. S., and Van Berkum, J. J. A. (2006). When peanuts fall in love: N400 evidence for the power of discourse. Journal of Cognitive Neuroscience, 18(7), 1098–1111. DOI.
Papafragou, A., and Grigoroglou, M. (2025). Pragmatic communication and theory of mind. Philosophical Transactions of the Royal Society B, 380(1932), 20230502. DOI.
Pouscoulous, N. (2023). More than one path to pragmatics? Insights from children's grasp of implicit, figurative and ironical meaning. Cognition, 240, 105531. DOI.
Sedivy, J. C., Tanenhaus, M. K., Chambers, C. G., and Carlson, G. N. (1999). Achieving incremental semantic interpretation through contextual representation. Cognition, 71(2), 109–147. DOI.
Sperber, D., and Wilson, D. (1995). Relevance: Communication and Cognition (2nd ed.). Blackwell. Publisher.