Interpersonal communication theories explain how people create, interpret, coordinate, and revise meaning while interacting. The central question is not simply how information travels from a sender to a receiver, but how two or more participants use words, timing, gesture, prior knowledge, social expectations, and responses to establish what an utterance means in a particular relationship and situation. Transmission models illuminate signals and noise; interactional and transactional models bring feedback and simultaneous interpretation into view; pragmatic and collaborative approaches explain how utterances perform social actions and how understanding becomes sufficient for the task at hand. Contemporary research treats face-to-face conversation as a jointly organized activity involving language, cognition, embodiment, and social context (Hadley, Naylor, and Hamilton, 2022; Holler and Kuhlen, 2026).
A useful theory identifies a process that can be observed or tested. For example, it should help explain why one person interprets “That was interesting” as enthusiasm while another hears polite disagreement, why a brief question can repair a misunderstanding, or why the same words may have different interpersonal consequences depending on authority, relationship history, and timing. A model can represent these processes schematically, while an empirical theory specifies mechanisms and predictions. Both have value when their scope and limitations are clear.
What Is an Interpersonal Communication Theory?
An interpersonal communication theory is an organized explanation of patterns through which people produce, perceive, negotiate, and respond to socially meaningful messages. Its units of analysis may be messages, turns, speech acts, perceptions, relationships, interaction sequences, or dyads. Its outcomes may include accurate reference, successful coordination, perceived understanding, solidarity, relationship change, or conflict. Because these outcomes differ, no single measure of “good communication” can validate every theory.
An ordinary exchange contains several analytically distinct events. A speaker forms an intention, produces an observable signal, and makes that signal available through a particular channel. The recipient notices selected elements, combines them with contextual expectations, and produces an interpretation. Their response then supplies further evidence to the first speaker. An agreement to act, a request for clarification, a delayed pause, and a visible withdrawal each changes the subsequent interaction. These elements can overlap in time and recur many times within a few seconds.
A communication model is a simplified representation of this system. A theory becomes more informative when it explains under what conditions specific forms of feedback, linguistic accommodation, prediction, or common-ground updating occur. The foundations and scope of the broader subject are treated in Interpersonal Communication in Psychology: Definition, Processes and Research; the present article concentrates on explanatory theories and how to evaluate them.
Message, meaning, and effect are different levels
The message is the observable communicative contribution: words, movement, writing, or another signal. Intended meaning is what the producer seeks to accomplish. Interpreted meaning is what a recipient infers. Interpersonal effect is what happens next, including changes in trust, behavior, emotion, or shared plans. A recipient may understand the intended literal content and still disagree with the underlying request. Conversely, two people may coordinate an action successfully even while holding somewhat different private interpretations of the event.
Consider “Can you close the window?” It can be analyzed as a grammatical question about ability, a conventional request, a demonstration of irritation, or an invitation to negotiate comfort. The phrase, the action performed, and the relationship implication depend on context. Pragmatic theories focus on what participants accomplish through such utterances; relational theories examine what those acts imply about closeness, entitlement, roles, and identities (Holtgraves, 2002; Clark, 1996).
A Short Intellectual History of Communication Models
Early schematic accounts of communication emphasized a source, an encoded signal, a medium, and a destination. Those distinctions remain useful when studying degraded sound, text delivery, competing signals, and perceptual access. Later interpersonal perspectives placed participants' responses, relationships, cultural conventions, and changing context closer to the center of explanation. The history is best understood as an expansion of the questions researchers ask: from how a signal reaches someone, to how a response influences the next signal, to how people achieve coordinated social action.
Mid-twentieth-century systems approaches also emphasized recurring interaction patterns and the relational implications of behavior. Watzlawick, Beavin, and Jackson's influential 1967 account situated communication within sequences of interpersonal behavior, although its broad axioms should be understood as theoretical propositions rather than universal diagnostic laws (Watzlawick, Beavin, and Jackson, 1967). Work on conversation and social interaction subsequently provided fine-grained observational evidence about how people organize turns and repair misunderstandings (Levinson, 2016; Dingemanse et al., 2015).
The next sections distinguish the central families of models without treating historical diagrams as competing claims about one identical outcome.
Linear and Transmission Models: Signals, Channels, and Noise
A linear model represents a source encoding a message, a signal traveling through a channel, and a receiver decoding it. Noise is whatever reduces fidelity in that process. A public announcement heard through a malfunctioning speaker, a sentence obscured by traffic, and a notification hidden by poor interface design illustrate the questions for which the framework remains useful. It directs attention to signal quality, channel constraints, coding conventions, and the recipient's opportunity to receive the message.
In interpersonal settings, noise includes background sound, divided attention, ambiguous terminology, competing demands, and mismatched linguistic knowledge. These factors differ causally: sensory loss of a word is not the same event as misunderstanding its social implication. A linear framework helps isolate the former; a theory of pragmatic inference is needed to investigate the latter. The companion scientific article Communication Barriers: Why People Misunderstand Each Other addresses barriers themselves in detail.
What the transmission diagram explains
If a colleague says “Thursday” during a noisy video call and the recipient hears “Tuesday,” improving acoustic fidelity or seeking explicit confirmation directly targets the failure. A repeated phrase, caption, or written summary can add redundancy. The transmission diagram maps this sequence clearly. Researchers can measure recognition accuracy, intelligibility, or response latency under controlled channel conditions.
The model becomes less complete when both people heard the same words correctly but understood different intentions. An employee may hear “We should talk later” as routine scheduling while interpreting it as a threat because of prior experiences with a supervisor. Signal fidelity is high; relational meaning remains uncertain. The relevant explanatory variables now include history, power, anticipated consequences, and social inference.
Why feedback changes the model
A one-way representation does not depict a recipient asking, “Did you mean Thursday?” or a speaker responding, “Yes, after lunch.” That acknowledgment supplies evidence about understanding and affects the next move. Feedback therefore extends the unit of analysis from an isolated message to a sequence. It may be verbal, such as confirmation or correction, or embodied, such as orienting toward a new object. A gesture or silence is only interpretable within the particular interactional setting; the same observable behavior has multiple possible explanations.
Interactional Models: Feedback and Alternating Roles
Interactional models depict participants as exchanging messages across successive turns. Each person alternates between producing a contribution and interpreting the other's response. These diagrams make room for feedback, context, and iterative correction. They are particularly useful for explaining questions and answers, clarification, turn-taking, and the accumulation of understanding.
For example, one person says, “Put it on the left.” The other asks, “Your left or mine?” The speaker answers, “To the left of the monitor.” A further response establishes whether this reference is adequate. The exchange involves a sequence of candidate meanings, checks, and repairs rather than a single transfer. Experimental work on reference has shown how interlocutors collaboratively refine noun phrases and reduce repeated descriptive effort as they build shared references (Clark and Wilkes-Gibbs, 1986).
An interactional account also clarifies why listening shapes subsequent speech. In experiments with personal storytelling, particular listener responses were associated with the narrator's ongoing performance; distracting listeners reduced context-specific responses and diminished how well stories were told. This is experimentally supported evidence for moment-to-moment listener participation in the tested storytelling setting, rather than proof that any particular nod or phrase universally improves relationships (Bavelas, Coates, and Johnson, 2000).
Feedback is information about reception, not proof of agreement
A listener saying “I understand” can indicate successful hearing, recognition of the other person's position, or agreement with a claim; these possibilities should be distinguished. Confirmation of understanding is different from endorsement of behavior. Likewise, asking for clarification can indicate engaged attention, limited information, a challenge to authority, or strategic resistance. Effective models keep the observable response distinct from the motive inferred from it.
When feedback is absent, uncertainty can rise, but the absence may result from delay, disability, silence norms, distraction, reduced access, or unwillingness to participate. The theory identifies the information gap without converting the gap into a personality judgment.
Transactional Models: Communication as a Simultaneous Process
Transactional models treat people as interdependent participants who continuously perceive, act, and interpret. Speaking and listening often overlap: a person adjusts a sentence while seeing a facial reaction, predicts a turn transition while the other is still talking, and evaluates a message against shared history. Context is not simply an external background; it enters the process through expectation, social roles, material environment, and the unfolding interaction.
This approach captures a familiar occurrence: while one person explains a plan, the other looks puzzled; before a question arrives, the first person supplies an example. The facial movement may have influenced the explanation, although its exact meaning cannot be inferred from the movement alone. Communication is transactional because each participant's behavior helps condition what the other does next, with effects accumulating over time.
Simultaneity does not imply perfect mutual understanding
People predict and respond to one another quickly, sometimes before a complete utterance has been heard. A synthesis of conversational turn-taking describes how the rapid pace of ordinary dialogue places concurrent demands on comprehension and planning (Levinson, 2016). A contemporary review likewise emphasizes the coordination of language, gesture, timing, and multimodal information in situated joint activity (Holler and Kuhlen, 2026).
But responsiveness is compatible with misreading. A participant can adapt to an apparent frown that actually reflects pain, effortful listening, or a distracting thought. Transactional diagrams depict reciprocal influence; evidence about specific inferences requires additional methods. Neither continuous feedback nor rapid synchrony guarantees correctness or healthy interaction.
Transactional models and relationships
Long-standing relationships contain expectations that alter the meaning of an utterance before the words are complete. Familiar jokes, routines, earlier conflict, and asymmetries in authority shape what counts as reassurance, criticism, or a request. Relationship history therefore changes both the interpretation of messages and the likelihood that participants will challenge one another's interpretations. To study this well, researchers must observe sequences and compare situations rather than infer an enduring relationship property from a single exchange.
Collaborative Theory: Grounding, Common Ground, and Joint Action
Herbert Clark's joint-action approach explains communication as an activity that participants accomplish together. People coordinate not only which words are said but also what those words count as for present purposes. Common ground is the information that interlocutors take to be mutually available and relevant; grounding is the process of establishing a contribution as understood well enough to proceed with the current activity (Clark, 1996).
Common ground need not mean identical private knowledge. Two friends planning a journey may both know that “the station” refers to the one near the river, while one also knows that it recently changed its entrance. The conversation works until that difference becomes relevant. A message can therefore be adequately grounded for an initial task but require renegotiation when the task changes. The standard is practical sufficiency within a particular joint activity.
Presentation, acceptance, and repair
In collaborative reference, a speaker proposes a description such as “the blue card,” and the recipient either proceeds or indicates trouble: “The dark blue one?” The speaker then adjusts the description. Repeated interaction can lead to shorter, locally conventional references because participants develop a history of successful joint use. Clark and Wilkes-Gibbs's laboratory study supplied influential evidence for collaborative refinement of reference, though the particular task was narrower than everyday interpersonal interaction (Clark and Wilkes-Gibbs, 1986).
A collaborative account predicts differences in communicative effort depending on available evidence. Participants in a face-to-face setting can use gaze, gesture, and immediate acknowledgment; asynchronous writers may need more explicit descriptions because delayed responses make repair costly. This is a contextual proposition, not a rule that more elaboration is always better. Excess detail can increase cognitive load and obscure the action the message is intended to accomplish.
Common ground can be mistaken
Assumptions about shared knowledge sometimes cause errors. In experimental referential tasks, listeners' eye movements suggested that objects outside the speaker's view could initially influence interpretation, even when common knowledge should have ruled those objects out. The result shows that perspective information is not invariably applied at the earliest stage of comprehension (Keysar et al., 2000). It qualifies idealized accounts in which partners automatically exclude everything the other cannot know.
Therefore, common ground is both a resource and a hypothesis held by participants. It is revised through evidence, questions, and action. A useful clinical or educational application is to ask what each person actually knows before assuming a failure reflects bad intentions or deficient social competence.
Pragmatic Theories: What People Do With Utterances
Pragmatic accounts investigate communication as action. Asking, promising, refusing, apologizing, hinting, warning, and inviting each accomplish something within a social arrangement. Literal sentence meaning supplies only part of the explanation; participants interpret what the act is doing in context. In everyday life, “It's getting late” can describe the time, suggest ending a meeting, or indirectly request someone to leave. A theory of pragmatic communication asks how listeners infer these purposes and how speakers select a form suited to the situation (Holtgraves, 2002).
Inference and implied meaning
Interpretation depends on conversational goals, shared conventions, relevance, and expectations about what a reasonable partner intends. If someone replies “I have an early flight” to an invitation, the recipient commonly infers a decline rather than treating the sentence as a mere travel report. The inference is an accomplishment of context, conversational sequence, and social knowledge. Alternative intentions are still possible, so a pragmatic analysis must examine the speaker's next actions and the response to the inference.
The practical importance is substantial: repeated literal clarification may be helpful when stakes are high, when culturally specific implications are unfamiliar, or when participants genuinely differ about implied intentions. It would be inappropriate to treat direct language as inherently less sophisticated or to assume that implicit language is inherently polite. Directness, indirectness, and the interpretation of hints depend on context, culture, accessibility, power, and individual preference.
Politeness, face, and interpersonal accountability
Goffman's account of face-work emphasized the interactional labor through which participants manage publicly presented identities and the consequences of embarrassment or challenge. Face concerns are relevant to why a request may be phrased indirectly, why a correction can feel humiliating, and why public and private disagreements unfold differently (Goffman, 1967). Such concepts describe social processes; they are not scales of mental health or grounds for diagnosing a person who communicates differently.
Pragmatics also clarifies accountability. A person can claim an utterance was “only a joke,” but the unfolding response shows how it functions in the relationship. A careful interpretation examines the content, prior understanding, audience, power, and opportunity to repair harm. The social meaning of an utterance can be contested by its participants, and those contests are themselves communicative actions.
Relational Theories: Communication Creates and Negotiates Relationships
Relational perspectives ask how exchanges define the relationship between participants, how roles and expectations develop, and why the same message carries different meaning across relationships. Statements simultaneously concern a topic and contribute to a pattern of treatment: asking for help may communicate trust, a superior's command may establish authority, and a public correction may affect status. These relational implications cannot be measured accurately by examining wording alone.
One major tradition, relational dialectics, examines how competing meanings and expectations are voiced and negotiated in relationships. Baxter and Montgomery treated autonomy, connection, openness, uncertainty, and their changing interpretations as elements of relational dialogue rather than fixed stages that every relationship must pass through (Baxter and Montgomery, 1996). The framework is valuable for interpreting how people negotiate relational meanings, but specific claims about outcomes require direct observational or experimental tests.
Relational meaning changes through repetition
Imagine a coworker who repeatedly says “I'll take care of it” after another colleague expresses concern. During the first conversation it may be reassurance. After several missed deadlines, the phrase may signal an attempt to end discussion without providing a verifiable plan. A relational interpretation requires sequence, history, evidence about consequences, and participants' accounts. The sentence does not carry a stable interpersonal significance independent of its use.
Communication can also change a relationship because acts establish expectations. A disclosed secret can alter the expectations surrounding trust; a boundary stated and respected changes what requests are permissible; a respectful disagreement can establish that dissent is allowed. These are interpretive examples, not claims that a particular behavior produces the same effect in every relationship. For operational evidence, researchers need temporal measurement of behavior and participants' subsequent experience.
Uncertainty reduction and first encounters
Berger and Calabrese proposed a developmental framework for initial interaction focused on uncertainty and the collection of information about others. Their early formulation specified axioms and researchable propositions about how communication and uncertainty may be related (Berger and Calabrese, 1975). This framework illuminates why early conversations contain questions, tentative impressions, observation, and adjustments of predictions.
It does not follow that people always seek maximum certainty, that learning more about someone always increases liking, or that a formal developmental sequence describes every encounter. Some uncertainty can preserve privacy and autonomy, while information seeking may be unwelcome in a hierarchical or threatening situation. Theory should specify when uncertainty is salient and when people have an opportunity to reduce it safely.
Communication Accommodation Theory: Adapting to the Other Person
Communication accommodation theory examines how people modify speech and other communication behaviors in relation to conversational partners and social identities. People may converge in terminology, pacing, tone, or forms of address, or emphasize distinctiveness through divergence. Such adjustment can convey affiliation or distance, but its social effect depends on motive attributed by the recipient, context, and group relations.
In a systematic review covering 149 articles and a meta-analysis of a subset of 76 studies totaling 18,382 participants, Soliz and Giles investigated patterns linking accommodation and relational outcomes in the studied settings. This is stronger evidence for the relevance of accommodation than a single classroom example, though the behaviors, populations, and outcomes varied substantially (Soliz and Giles, 2014). A later theoretical review traced accommodation research across a range of interactional contexts (Giles, Edwards, and Walther, 2023).
Accommodation, convergence, and inappropriate imitation
Adapting one's terminology to a conversation partner may facilitate reference when both parties agree on the meanings of terms. Slowing down and using a different register may assist a person who requests it. The same adjustment can be experienced as condescending when it is based on an inaccurate stereotype or is performed without regard for the other person's preferences. Divergence may assert a valued identity or signal rejection; observation alone cannot establish which one.
Accommodation theory consequently invites multiple measurements: what changed in the signal, what the listener noticed, what motive the listener attributed, and what subsequent interactional outcome occurred. Matching pitch, accent, or gesture is not inherently a marker of healthy communication. Cultural, linguistic, disability-related, and power differences alter what counts as a considerate adaptation.
Models of Interactional Alignment and Coordination
Psycholinguistic alignment accounts propose that dialogue involves coordination in linguistic representations, including words, syntactic forms, and conceptual perspectives. Pickering and Garrod developed an interactive alignment account to explain how repeated patterns and priming-like mechanisms may support fluent dialogue. It offers a mechanistic theory of parts of language processing in conversation, not a guarantee that two participants endorse the same beliefs (Pickering and Garrod, 2004).
Two interlocutors may begin with different terms for an object, settle on one label, and subsequently reuse it. This can reduce ambiguity and effort. Alignment also occurs in multimodal behavior, where timing, gesture, and shared attention influence the production and reception of turns. Contemporary syntheses distinguish linguistic coordination, behavioral synchrony, and higher-level understanding, because these phenomena are measured differently and need not covary in all circumstances (Hadley, Naylor, and Hamilton, 2022).
Behavioral synchrony is not shared belief
People sometimes adjust their timing and movement in ways that become coordinated. It is tempting to equate synchrony with empathy, consent, trust, or accurate understanding. Such claims overreach the measure. Synchrony may be produced by a shared external rhythm, task constraints, familiarity, or mimicry; individuals can synchronize during an argument without agreeing on the topic. A rigorous study specifies the synchronized variable, its baseline and controls, and whether a distinct interpersonal outcome was measured.
Neuroscientific proposals about brain-to-brain coupling suggest that communication can create correlated activity across interacting people through shared sensory signals and coordination. Hasson and colleagues articulated this multibrain framework as a mechanism-oriented research program, whose interpretation depends on task design and the separation of genuine interaction effects from common stimulation (Hasson et al., 2012). Neural synchrony is therefore neither a direct reading of two minds nor a clinical measure of relationship quality.
Turn-Taking and Conversational Repair: Meaning in Real Time
Turn-taking is the organization through which participants coordinate opportunities to contribute. Conversation analysis investigates how a question projects an answer, how an interruption is managed, and how misunderstandings become visible in subsequent turns. These sequential details form an empirical bridge between abstract feedback models and naturally occurring behavior.
Cross-linguistic work by Stivers and colleagues documented both widespread organizational similarities and meaningful cultural variation in the timing of conversational responses (Stivers et al., 2009). The implication is more precise than a universal etiquette rule: certain coordination problems recur across languages, while the interpretation of response timing depends on language, activity, and local norms.
Repair as a cooperative operation
A repair sequence occurs when someone signals trouble in hearing, speaking, or understanding and the participants work to resolve it. For instance: “The blue one?” “No, the small green one.” “I see.” That brief cycle revises the reference, verifies sufficient understanding, and permits the activity to continue. The correct interpretation of the exchange is established by what participants do next, not merely by the analyst's judgment of their internal intentions.
Dingemanse and colleagues examined 48.5 hours of sampled conversation across 12 languages belonging to eight language families, identifying 2,053 instances of other-initiated repair and a mean frequency of approximately one repair initiation every 1.4 minutes. The study provides substantial comparative evidence for recurrent repair practices, while its sampling frame does not imply identical rates in every conversation or population (Dingemanse et al., 2015).
Kendrick's quantitative conversation-analytic study found that the timing of repair initiation can differ from ordinary responsive turns, suggesting that longer pauses may permit recognition, self-correction, and visible signaling of trouble (Kendrick, 2015). A delay can thus be part of competent interactional work rather than evidence of confusion or rejection.
Why repair is theoretically important
Repair integrates several frameworks. Transmission models identify the initial disturbance; interactional models represent the correction sequence; collaborative models explain the joint establishment of a workable referent; pragmatic models examine what the question and correction are doing socially. Relational models add whether it is safe for a participant to reveal uncertainty or challenge another person's wording.
Repair does not always end in full convergence. Participants may continue with provisional understanding, defer a disputed issue, or recognize that they interpret a value differently. Whether a conversation succeeds depends on its purpose. In a safety-critical instruction, a provisional guess is inadequate; in a casual exchange, perfect agreement about every nuance would be unnecessary.
Feedback Loops: Four Functions and Their Limits
Feedback is any subsequent behavior that provides information relevant to the preceding contribution. Analytically, it can serve at least four overlapping functions: evidence that a signal was received, evidence of a proposed interpretation, a response to the social action, and an indication of how the relationship or activity should continue. These are analytical categories rather than independently validated clinical scales.
Reception feedback includes repeating a word, nodding, or continuing the instructed activity. Interpretive feedback includes paraphrasing or checking a referent. Action feedback includes accepting, refusing, or negotiating a request. Relational feedback includes signals of affiliation, threat, respect, or withdrawal. A single response may serve several functions, and the meanings attributed to it must be checked against context.
Closed-loop communication in high-stakes settings
When the consequence of error is substantial, a reasonable practice is to make instructions specific, invite a read-back, and verify observable execution. This follows from principles of grounding and correction, but the success of a particular protocol requires evidence in the relevant occupation or setting. The conversational rule “always repeat everything back” would create unnecessary burdens in everyday relationships and can become controlling when imposed without consent.
Feedback can be intentionally misleading. A subordinate may acknowledge an order without feeling permitted to question its safety. A person may say “okay” to escape an intimidating encounter. Interpersonal theories therefore need power and safety variables; an apparent completed feedback cycle can coexist with coercion or unresolved disagreement.
Negative feedback and interpersonal regulation
In systems language, negative feedback can refer to a response that counteracts a deviation and stabilizes a process, while positive feedback may amplify it. This technical usage differs from everyday “negative feedback” meaning criticism. A tense exchange can contain reinforcing cycles: accusations increase defensiveness, which is read as proof of bad faith, leading to further accusations. Such narratives are plausible process hypotheses; identifying a causal cycle requires longitudinal or experimental evidence rather than one participant's retrospective account.
In constructive disagreements, clarification and explicit recognition of different perspectives may interrupt one particular cycle. Their effectiveness depends on mutual willingness, behavioral options, and the surrounding conditions. Where there is intimidation, threatened retaliation, or coercive control, asking the vulnerable party to provide more accommodating feedback may increase danger. Communication analysis must preserve the distinction between a reversible misunderstanding and an unsafe exercise of power.
Shared Meaning: Levels of Agreement and Coordination
Shared meaning is a negotiated, context-sensitive achievement in which participants establish sufficient overlap in reference or interpretation to support a particular joint action. It need not entail identical internal experiences, identical emotional evaluations, or agreement on every proposition. This definition draws on collaborative and pragmatic traditions, especially the conception of common ground and grounding as purpose-relative (Clark, 1996).
A group planning a meeting may share the meaning of “next Friday” after identifying the date and time zone. They can then coordinate attendance despite holding different expectations about how enjoyable the meeting will be. A friend who says “I understand why you feel betrayed” may demonstrate recognition of the other's perspective without confirming the accusation as fact. A contract requires a more formal and precise level of mutual understanding than casual storytelling. In every case, the criterion for sufficiency is tied to the activity and the consequences of error.
Five questions for locating a meaning problem
When participants report that they “are not communicating,” an analyst can examine which layer is failing. Did they hear or see the contribution? Did they identify the same referent? Did they infer the same speech act? Did they understand the other's reasons or emotional significance? Did they agree on a next action? Each question defines a distinct hypothesis with different observable evidence. Jumping immediately to a presumed personality defect obscures these distinctions.
Such an analysis can also uncover genuine disagreement. If two people accurately understand different goals, more repeated explanation may not resolve the issue; negotiation, boundary setting, changed incentives, or a decision to remain apart may be more appropriate. A successful communication process does not require that one party persuade the other.
The Role of Culture, Language, and Context
Theories grounded primarily in English-speaking university samples must be evaluated against other cultural and linguistic settings. Turn timing, address forms, politeness conventions, permissible questioning of authority, and the meaning of silence can vary. At the same time, research on repair and turn-taking has identified organizational similarities that recur across diverse languages (Dingemanse et al., 2015; Stivers et al., 2009). Neither absolute cultural uniformity nor unlimited interpretive variation describes the evidence.
A gesture may be helpful to one recipient, inaccessible to another, and socially ambiguous to a third. Eye contact can signal attention in some situations while being uncomfortable, inappropriate, or difficult in others. An effective theory specifies which cues are available and which interpretation is supported, instead of turning dominant local norms into universal criteria of competence.
Pragmatic differences and neurodevelopmental diversity
People differ in how they use implicit requests, repair misunderstandings, manage turn transitions, or prefer verbal and nonverbal information. Neurodevelopmental differences, sensory conditions, language experience, trauma histories, and situational anxiety can alter interactional opportunities. Ordinary variation on these dimensions does not amount to a diagnosis. Descriptions of communication difficulty should indicate the actual behavior, situation, and impact rather than assigning an unsupported clinical label.
Models built from interaction between two people are particularly useful because interaction outcomes depend on both participants and the environment. A conversation can become easier when someone changes the pace, provides written clarification, or makes expectations explicit. Such adaptations should be negotiated with the person concerned, rather than prescribed as compulsory eye contact, extroversion, or conformity to one conversational style.
Power, status, and consent
Participants may have unequal opportunities to interrupt, correct, disagree, or withdraw. A person with authority may interpret silence as agreement even when subordinates anticipate consequences for speaking. A partner may demand repeated explanations while refusing to accept any answer. Systems and transactional models gain explanatory accuracy when combined with power relations and realistic options for refusal. Shared meaning is meaningful only where the observed response can be interpreted in light of these constraints.
The scientific distinction between negotiated understanding and coerced compliance matters especially when communication is discussed in clinical, workplace, or intimate contexts. Techniques intended to increase conversational harmony cannot substitute for safety, legal protection, informed consent, or organizational accountability when these are required.
Technology and Mediated Interpersonal Communication
Interpersonal theory also applies to messages exchanged through phones, video calls, email, and platforms, but each medium redistributes communicative resources. A video conversation allows rapid audible feedback and some visual orientation while presenting latency, framing, and acoustic limitations. Email supports revision and a durable record while delaying responses and removing real-time repair. Text messaging has linguistic conventions for tone and timing, yet those conventions are partly platform-specific and relationship-specific.
Walther's relational perspective on computer-mediated communication challenged the assumption that fewer immediate nonverbal cues necessarily prevent the development of interpersonal relationships. Accumulated textual exchanges can support individuating impressions and relational messages under some conditions (Walther, 1992). The framework highlights adaptation to the medium rather than equating physical presence with good communication.
Delay changes the evidence available to participants
A delayed response may have several plausible causes: competing commitments, a technical problem, deliberate reflection, indifference, or an intention to end communication. The recipient rarely observes the cause directly. Transactional models encourage checking the sequence and context before treating timing as a definitive personal evaluation. Collaborative models predict greater need for explicit disambiguation when the cost of clarification is higher.
Digital and face-to-face interactions also differ in privacy, searchability, platform affordances, and the permanence of messages. These features matter for social interpretation but belong to separate empirical questions. Broad claims that a particular medium improves or harms every relationship would require studies with appropriate comparison groups and long-term outcomes.
How Scientists Test Interpersonal Communication Theories
Communication research draws on experiments, naturalistic recordings, conversation analysis, questionnaires, experience sampling, and longitudinal observation. Each method makes different variables visible. A study of listener memory after a scripted message measures a different process from a study of repair in naturally occurring conversations. A questionnaire about perceived understanding measures experience, not necessarily the accuracy of an interpretation. A theory earns support when its predictions survive tests matched to its actual claims.
Controlled experimental tasks
Researchers can manipulate shared visual access, referential ambiguity, response delays, partner feedback, or availability of a communication channel. Outcomes may include selection errors, processing time, gaze patterns, or whether a referent is correctly identified. Experimental control can distinguish alternative mechanisms, while artificial tasks and short interactions limit generalization. Clark and Wilkes-Gibbs's collaborative reference tasks and Keysar and colleagues' perspective-taking experiments illustrate different aspects of meaning coordination (Clark and Wilkes-Gibbs, 1986; Keysar et al., 2000).
Naturalistic interaction and sequential coding
Recorded conversations preserve how particular contributions create opportunities for subsequent responses. Conversation analysis describes participants' displayed orientations within the sequence; quantitative coding can measure turn intervals, repair frequency, and the conditional occurrence of behaviors. This combination is especially informative because analysts need not infer a partner's private motive from an isolated facial expression. Broad samples across languages can test which practices generalize and which vary by activity and culture (Dingemanse et al., 2015).
Surveys, diaries, and longitudinal studies
Self-reports illuminate perceived validation, satisfaction, misunderstanding, relationship norms, and the meanings participants retrospectively assign to an event. Repeated measures can track whether these perceptions covary with actual interactional events. However, two people may evaluate the same exchange differently, and recollections can be influenced by later conflict. Longitudinal designs improve temporal ordering but still require careful treatment of unmeasured influences and selective participation.
Multimodal and neural measurements
Eye tracking, motion capture, acoustics, physiological signals, and neural recordings permit fine-grained investigation of attention, coordination, and prediction. Their validity depends on whether a physiological pattern corresponds to the social process a theory proposes. Correlated signals may reflect common stimuli rather than a distinct interpersonal mechanism. The 2022 review by Hadley and colleagues and the 2026 review by Holler and Kuhlen emphasize the need to choose methods capable of separating theories rather than collecting rich data without discriminating predictions (Hadley, Naylor, and Hamilton, 2022; Holler and Kuhlen, 2026).
Measuring change and generalization
An intervention that improves performance during a role-play may have limited effects in difficult real-life conversations. Changes in a communication questionnaire may reflect increased confidence rather than more accurate perspective taking. Meaningful evaluation can combine behavioral tasks, reports from different people, everyday sampling, and longer follow-up. For interventions, control groups and active comparison conditions help separate the contribution of specific techniques from attention, practice, and expectancy effects.
What the Evidence Establishes and What Remains Debated
Established: communication routinely involves feedback, sequential organization, and opportunities for repair; reference can be collaboratively adjusted; interpersonal interpretation is sensitive to access, task, and context. These conclusions have support from experimental and observational evidence, including cross-linguistic conversation studies and reviews of face-to-face interaction (Clark and Wilkes-Gibbs, 1986; Dingemanse et al., 2015; Hadley, Naylor, and Hamilton, 2022).
Established within studied settings: listeners influence certain storytelling outcomes, perspective-taking processes can show early egocentric interpretations, and accommodation is associated with relational judgments under multiple circumstances. Those findings are supported by experiments or quantitative reviews, but their exact magnitudes, mechanisms, and scope should not be exported mechanically to every culture or relationship (Bavelas, Coates, and Johnson, 2000; Keysar et al., 2000; Soliz and Giles, 2014).
Preliminary or mechanism-specific: neural coupling and multimodal synchrony research identifies promising tools for examining interdependence, but interpretation remains sensitive to shared stimulation, analysis pipelines, and external validity. Their observation alone does not establish empathy, affection, or understanding (Hasson et al., 2012).
Limited or contested: claims that one transactional diagram exhaustively explains communication, that all misunderstanding can be solved through clarification, or that any single style of listening is universally effective are too broad for the evidence. Relational dialectics and uncertainty reduction offer important theoretical perspectives, while their component propositions need interpretation and validation in specified settings rather than acceptance as universal psychological laws (Baxter and Montgomery, 1996; Berger and Calabrese, 1975).
These levels of confidence concern particular propositions. No model receives a single scientific grade that applies equally to every application, construct, and population.
Putting the Models Together: A Detailed Conversation Example
Imagine that two colleagues are preparing a presentation. One says, “The client needs a simpler version.” The other replies, “Do you mean fewer slides, less technical language, or a different conclusion?” The first answers, “The conclusion is fine. I mean we should explain the evidence without assuming the audience knows the technical terms.” They agree to rewrite two slides and keep the supporting data in an appendix. The interaction appears ordinary, yet each theory identifies a different explanatory layer.
A transmission account asks whether the words were audible, legible, and correctly received. An interactional model tracks the question as feedback and the subsequent revision as the speaker's response. A transactional model considers how both participants are interpreting and adjusting during the unfolding exchange, including the possible influence of hesitation, facial expression, and previous joint work. A grounding account identifies the progression from an underspecified description to a mutually accepted working reference. A pragmatic explanation shows that “simpler” is a request to modify an artifact and that the follow-up question is a socially appropriate means of specifying the requested action. A relational perspective asks whether the exchange reinforces a pattern of collaborative authority, whether either person feels undermined, and what their previous exchanges have taught them to expect. An accommodation approach considers whether the first colleague adapts technical vocabulary to the other's knowledge.
The case also illustrates how partial common ground is often enough for action. Both may agree about which slides to rewrite without holding identical private views about the presentation's ultimate purpose. The immediate task provides a criterion for coordination. If they later discover conflicting understandings of “simpler,” additional repair becomes possible. A theory that predicts how this discrepancy will be detected is more useful for research than a vague injunction that both people should communicate better.
Practical Applications Derived From the Research
Clarify the decision or action that depends on the message
The most productive question is often specific to the purpose of the exchange: what does the other person need to know, decide, recognize, or do? In a scheduling conversation, repeat the actual date and time. In a discussion of a shared document, point to the exact passage. In an emotionally significant conversation, check whether the speaker is asking for advice, recognition, or a concrete change. These different kinds of coordination require different forms of evidence that a message has been understood.
Treat feedback as data and use targeted repair
A puzzled expression, a delayed answer, or “I thought you meant Friday” can signal the need for clarification. Feedback alone does not identify the precise cause of a misunderstanding. A targeted question—“Which part sounds different from what I intended?”—can reveal whether the difficulty concerns terminology, factual reference, implied request, or relationship meaning. Repair is collaborative when both participants have opportunities to amend their contributions. In a coercive setting, however, requests for clarification may be used to prolong surveillance or assert control; interpersonal safety and freedom to decline participation take priority over optimizing the conversation.
Make relevant assumptions visible
Shared background knowledge should be treated as a working hypothesis that can be updated. A speaker might say, “I'm assuming you've seen the revised deadline; is that right?” Such a question identifies the premise on which the rest of the conversation depends. It can be particularly useful across institutional roles, languages, generations, or specialized fields, where apparently ordinary words may carry different practical meanings. The objective is sufficient shared context for the current purpose, rather than complete agreement about every interpretation.
Attend to relational consequences without guessing hidden motives
Descriptions of another person's silence or speaking style easily become claims about their character. More accurate accounts begin with observable sequences: “After I described the problem, there was a long pause, and I interpreted it as disagreement.” That formulation leaves room for the other person's account of the pause. Communication theory supports careful inquiry into how meanings arise, while the scientific evidence does not authorize inferring deception, affection, dominance, or pathology from a single vocal or bodily cue.
Choose a channel that fits the interpretive task
A written checklist can stabilize exact requirements and dates; a synchronous discussion may offer faster repair when ambiguity is high. Asynchronous communication allows considered responses but can extend the time before misunderstanding is detected. Accessible channels, written alternatives, and reasonable processing time matter for participants whose sensory, cognitive, linguistic, or communication preferences differ. Selecting a channel is therefore part of communication design, not a mere afterthought.
Frequently Asked Questions
What is the difference between an interactional and a transactional model?
An interactional model highlights alternating messages and feedback: one contribution influences the next. A transactional model treats communicative behavior, interpretation, and the evolving interpersonal situation as mutually interdependent and frequently overlapping. Both can describe the same exchange at different levels of abstraction. Neither guarantees that a speaker's intention becomes the listener's interpretation.
Which theory explains shared meaning most directly?
Collaborative grounding and common-ground approaches specifically analyze how people present, acknowledge, question, revise, and provisionally accept contributions. Pragmatics adds an account of socially situated intentions and actions. Relational theories examine the significance these contributions acquire within an ongoing connection. A study should select the framework that makes clear predictions about its particular meaning outcome (Clark and Wilkes-Gibbs, 1986; Clark, 1996).
Does shared meaning require agreement?
Shared meaning may concern the referent of a word, the practical force of a request, or the next agreed action. People can understand one another accurately while continuing to disagree about values, interests, or conclusions. Conversely, verbal agreement may disguise different assumptions. Researchers therefore specify whether the outcome is factual comprehension, coordination, interpersonal recognition, or agreement about a proposition.
Can communication theory tell whether someone has good social skills?
Communication theories explain processes; evaluations of social skills require separate evidence about context, goals, opportunities, behavior, and consequences. Effective participation varies with cultural norms, disability, power, language, and the participants' own aims. Difficulty in one situation establishes neither an enduring personal deficit nor a psychiatric diagnosis. The assessment of social competence is covered in Social Skills in Psychology: Definition, Components and Assessment.
Do communication models apply to text messages and AI-mediated exchanges?
Their core questions about signals, inference, feedback, and shared context remain applicable, but delayed replies, platform affordances, and automated contributions change what participants can observe and what they may reasonably infer. Evidence from face-to-face dyads must be tested rather than automatically transferred to mediated, multiparty, or human–AI interaction. The psychological consequences of interacting with an AI system also depend on that system's actual functions and on the user's expectations; perceived reciprocity alone does not establish an AI system's subjective experience.
Conclusion
Interpersonal communication theories are most powerful when used as complementary, testable explanations for different parts of social interaction. Transmission models trace signals and noise; interactional and transactional frameworks organize feedback and interdependence; collaborative and pragmatic theories explain how participants make references and perform actions; relational and accommodation frameworks investigate social identity, relationship patterns, and adaptation. The observable process of shared meaning is iterative: people present contributions, infer intentions, respond, detect discrepancies, and revise what can be treated as understood. The scientific task is to identify which mechanisms operate, under which conditions, with which measurable outcomes, and with what limits on generalization.
Related Articles
Interpersonal Communication in Psychology: Definition, Processes and Research — the broader psychological foundations, processes, and research designs.
Verbal and Nonverbal Communication: How Meaning Is Conveyed — speech, gesture, voice, facial expression, and multimodal inference.
Communication Barriers: Why People Misunderstand Each Other — obstacles to interpretation and opportunities for clarification.
Active Listening: Skills, Examples and Psychological Evidence — listening responses and the behavioral evidence for specific techniques.
Social Skills in Psychology: Definition, Components and Assessment — distinguishing explanatory communication models from context-specific assessments of social behavior.
Social Perception and Misinterpretation in Everyday Conversation — Review attribution, expectancy, bias and interpersonal interpretation.
References
Bavelas, J. B., Coates, L., and Johnson, T. (2000). Listeners as co-narrators. Journal of Personality and Social Psychology, 79(6), 941–952. https://doi.org/10.1037/0022-3514.79.6.941
Baxter, L. A., and Montgomery, B. M. (1996). Relating: Dialogues and dialectics. Guilford Press. Publisher
Berger, C. R., and Calabrese, R. J. (1975). Some explorations in initial interaction and beyond: Toward a developmental theory of interpersonal communication. Human Communication Research, 1(2), 99–112. https://doi.org/10.1111/j.1468-2958.1975.tb00258.x
Clark, H. H. (1996). Using language. Cambridge University Press. https://doi.org/10.1017/CBO9780511620539
Clark, H. H., and Wilkes-Gibbs, D. (1986). Referring as a collaborative process. Cognition, 22(1), 1–39. https://doi.org/10.1016/0010-0277(86)90010-7
Dingemanse, M., Roberts, S. G., Baranova, J., Blythe, J., Drew, P., Floyd, S., Gisladottir, R. S., Kendrick, K. H., Levinson, S. C., Manrique, E., Rossi, G., and Enfield, N. J. (2015). Universal principles in the repair of communication problems. PLOS ONE, 10(9), e0136100. https://doi.org/10.1371/journal.pone.0136100
Giles, H., Edwards, A. L., and Walther, J. B. (2023). Communication accommodation theory: Past accomplishments, current trends, and future prospects. Language Sciences, 99, 101571. https://doi.org/10.1016/j.langsci.2023.101571
Goffman, E. (1967). Interaction ritual: Essays on face-to-face behavior. Anchor Books. WorldCat
Hadley, L. V., Naylor, G., and Hamilton, A. F. de C. (2022). A review of theories and methods in the science of face-to-face social interaction. Nature Reviews Psychology, 1, 42–54. https://doi.org/10.1038/s44159-021-00008-w
Hasson, U., Ghazanfar, A. A., Galantucci, B., Garrod, S., and Keysers, C. (2012). Brain-to-brain coupling: A mechanism for creating and sharing a social world. Trends in Cognitive Sciences, 16(2), 114–121. https://doi.org/10.1016/j.tics.2011.12.007
Holler, J., and Kuhlen, A. K. (2026). Psycholinguistic perspectives on face-to-face conversation. Nature Reviews Psychology, 5, 240–255. https://doi.org/10.1038/s44159-026-00538-1
Holtgraves, T. (2002). Language as social action: Social psychology and language use. Lawrence Erlbaum Associates. Publisher
Kendrick, K. H. (2015). The intersection of turn-taking and repair: The timing of other-initiations of repair in conversation. Frontiers in Psychology, 6, 250. https://doi.org/10.3389/fpsyg.2015.00250
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. https://doi.org/10.1111/1467-9280.00211
Koudenburg, N., Postmes, T., and Gordijn, E. H. (2013). Conversational flow promotes solidarity. PLOS ONE, 8(11), e78363. https://doi.org/10.1371/journal.pone.0078363
Levinson, S. C. (2016). Turn-taking in human communication—Origins and implications for language processing. Trends in Cognitive Sciences, 20(1), 6–14. https://doi.org/10.1016/j.tics.2015.10.010
Pickering, M. J., and Garrod, S. (2004). Toward a mechanistic psychology of dialogue. Behavioral and Brain Sciences, 27(2), 169–190. https://doi.org/10.1017/S0140525X04000056
Soliz, J., and Giles, H. (2014). Relational and identity processes in communication: A contextual and meta-analytical review of communication accommodation theory. Communication Yearbook, 38(1), 107–144. https://doi.org/10.1080/23808985.2014.11679160
Stivers, T., Enfield, N. J., Brown, P., Englert, C., Hayashi, M., Heinemann, T., Hoymann, G., Rossano, F., de Ruiter, J. P., Yoon, K. E., and Levinson, S. C. (2009). Universals and cultural variation in turn-taking in conversation. Proceedings of the National Academy of Sciences, 106(26), 10587–10592. https://doi.org/10.1073/pnas.0903616106
Walther, J. B. (1992). Interpersonal effects in computer-mediated interaction: A relational perspective. Communication Research, 19(1), 52–90. https://doi.org/10.1177/009365092019001003
Watzlawick, P., Beavin, J. H., and Jackson, D. D. (1967). Pragmatics of human communication: A study of interactional patterns, pathologies, and paradoxes. W. W. Norton. WorldCat