An interlocutor is simply a person who takes part in a conversation. The word comes from Latin and literally means “one who speaks between,” but in practice it refers to anyone you are actively talking with, whether that is a friend across the table, a colleague on a video call, or even a chatbot in a text window. Researchers across linguistics, psychology, and neuroscience use the term constantly because conversation is never a solo act: what you say, how you say it, and how well communication works all depend on who your interlocutor is and what you know about them.
How Interlocutors Build Shared Understanding
Conversation looks effortless from the outside, but underneath it runs a constant, mostly invisible process of establishing common ground. When two people start talking about something new, they tend to use long, tentative descriptions. As the exchange continues and both sides confirm they are on the same page, those descriptions shrink into quick, confident shorthand. Classic experiments on this showed that when a director described abstract figures to a partner, the descriptions went from lengthy and vague to short and definite as common ground accumulated between them.1Journal of Memory and Language. Coordinating beliefs in conversation That shift is not just about saving breath. It signals that both interlocutors have built a shared mental model of what they are discussing.
A related idea, called interactive alignment, proposes that interlocutors converge on a shared understanding by priming each other at multiple levels, from the sounds of words to sentence structure to the overall meaning of what is being discussed. Because the brain systems for producing speech and understanding speech overlap, hearing your interlocutor say something a certain way nudges you toward saying it the same way, and vice versa.2PubMed Central. Toward a neural basis of interactive alignment in conversation This mutual priming happens largely below conscious awareness. You do not decide to copy your interlocutor’s phrasing; it just starts happening as the conversation flows.
Tailoring What You Say to Who You Are Talking To
People do not speak the same way to every interlocutor. You automatically simplify your vocabulary for a child, add more context for a stranger, and skip the backstory with a close friend. Researchers call this audience design, and it turns out to be more sophisticated than it might seem. One framework distinguishes between adjustments you make before you even open your mouth, based on what you already know about the other person, and adjustments you make on the fly as you monitor how well your message is landing.3PubMed. A Mechanistic Framework for Explaining Audience Design in Language Production The first kind is like choosing a gift based on what you know someone likes; the second is like watching their face as they unwrap it and pivoting in real time.
Audience design gets more interesting when you are talking to a group. If you are addressing two listeners who know different things, a natural instinct is to “aim low” and design your speech for the person who knows the least. But research shows that when the group grows to three or four people, speakers shift strategy. Instead of pitching everything to the least informed listener, they balance the combined knowledge of the group, striking a compromise that works well enough for everyone.4PubMed. Audience Design in Multiparty Conversation That same study found that speakers can successfully track what up to four partners do and do not know, which is a nontrivial feat of mental bookkeeping.
There are limits to this tracking, though. Research on how instructors guide followers through a map task found that speakers often rely on quick verbal feedback (“uh huh,” “yeah”) rather than actually checking whether their interlocutor is following visually. Under time pressure, they become even less likely to verify the listener’s understanding before plowing ahead. The burden of keeping a conversation on track is genuinely shared between speakers and listeners, and when that sharing breaks down, miscommunication creeps in.
When Brains Start to Sync
One of the more striking discoveries in recent years is that conversation produces measurable synchronization between the brains of interlocutors. Using EEG to record the brain activity of speakers and listeners simultaneously, researchers found statistically significant increases in synchronization across multiple frequency bands during oral narrative. The effect appeared in over a hundred electrode pairings, suggesting that brain-to-brain entrainment during speech is widespread rather than confined to a single region.5Scientific Reports. Brain-to-brain entrainment: EEG interbrain synchronization while speaking and listening
A separate study using fMRI with pairs of close friends who took turns telling each other about personal life events confirmed a similar coupling: brain areas associated with speech production in the speaker tracked with auditory processing areas in the listener.6PubMed. Interindividual synchronization of brain activity during live verbal communication The synchronization is not just a curiosity. Research using near-infrared spectroscopy found that when pairs of people performed a coordinated task, brain-to-brain synchrony at the left middle frontal cortex was significantly higher than in pairs working independently. More strikingly, the degree of that synchrony correlated positively with how much the partners were inclined to help each other afterward.7PubMed Central. Brain-to-brain synchronization across two persons predicts mutual prosociality In other words, the more your brain syncs with your interlocutor’s, the more prosocial you feel toward them.
Bodies Follow the Same Pattern
The synchronization between interlocutors is not limited to the brain. People in social interactions tend to naturally synchronize their body movements, often without noticing.8PubMed. The interacting partner as the immediate environment: Personality, interpersonal dynamics, and bodily synchronization You might mirror your interlocutor’s posture, shift your weight at the same time they do, or match the tempo of their gestures. This is not imitation in any conscious sense. It appears to be a byproduct of mutual engagement.
Even in three-person conversations, bodily synchrony holds up. A study of triads engaged in casual interaction found that pairs within the group showed higher movement correlation than would be expected by chance. The effect was small in absolute terms but statistically robust, and it appeared regardless of where participants were seated.9PubMed Central. Body synchrony in triadic interaction The researchers also found periods of negative correlation, where one person’s movement went in the opposite direction of another’s. This makes sense if you think about how conversation works in a group: while one person leans in to make a point, another might lean back to listen. The dance is not always unison. Sometimes it is complementary.
Theory of Mind and Reading Your Interlocutor
To be a good conversationalist, you need to model what your interlocutor is thinking, a capacity psychologists call theory of mind. Everyone does this to some degree when they anticipate what the other person knows, infer what they meant by an ambiguous remark, or sense when they are confused. But the link between this capacity and actual conversational success has been hard to pin down empirically.
A study of people recovering from right-hemisphere stroke offered unusually direct evidence. Researchers measured patients’ ability to explicitly figure out another person’s perspective, especially when that perspective conflicted with their own, and then looked at how well those patients performed in real conversations. The ability to manage that conflict turned out to be the strongest predictor of conversational success, independent of demographics and baseline cognitive abilities.10bioRxiv. Theory of mind predicts conversational success in early right hemisphere stroke recovery The finding makes intuitive sense: the hardest moments in conversation are not the ones where you and your interlocutor agree, but the ones where you have to hold two conflicting pictures in mind at once and still respond sensibly.
How Your Interlocutor Shapes Language Learning
If you have ever tried to learn a new language, you know that who you practice with matters enormously. Research on second-language learners confirms this. A study comparing interactions between pairs of learners and interactions between a learner and a native speaker found that when a native speaker was the interlocutor, language-related problem-solving moments came up more often and were more likely to be resolved successfully.11Applied Linguistics. Collaborative Dialogue in Learner–Learner and Learner–Native Speaker Interaction The native speaker, simply by being a richer source of the target language, created more opportunities for the learner to notice gaps in their knowledge and fill them in real time.
That does not mean learner-to-learner conversation is useless. Those same pairings still generated collaborative problem-solving, just less of it. And there are social benefits to practicing with someone at a similar level: less anxiety, more willingness to experiment, more balanced participation. The key insight is that the identity and skill of your interlocutor shape not just what you say, but what you learn from saying it.
Autism and the Challenge of Dynamic Adaptation
Conversation requires constant adjustment. You monitor your interlocutor’s reactions, revise your assumptions about what they know, and shift your approach accordingly. Research on autism spectrum conditions has revealed a specific pattern in how this adjustment process can differ. One study found that autistic participants were just as motivated and capable as non-autistic participants in adjusting their speech based on broad expectations about a partner’s abilities, for instance using greater emphasis when speaking to someone they assumed was less capable. However, they were less likely to update those initial adjustments when interaction-based evidence suggested the partner was actually more competent than expected.12PubMed Central. Partner-dependent communication without dynamic adaptation in autism Non-autistic participants, by contrast, dynamically adapted their communication to treat both partners equivalently once the evidence warranted it.
This distinction between initial calibration and ongoing recalibration matters for understanding what makes reciprocal conversation difficult. Pragmatic challenges in conversation are widely documented in autism research, with differences in how individuals manage the back-and-forth of reciprocal exchange.13PubMed. Partner Perceptions of Conversations with Individuals with Autism Spectrum Disorder The picture that emerges is not one of inability to consider an interlocutor’s perspective, but rather a difference in how flexibly that perspective is revised mid-conversation. Initial calibration can work fine; the friction shows up when the situation demands a mid-course correction based on subtle cues.
Chatbots as Interlocutors
With AI chatbots becoming a common feature of daily life, researchers have started asking how people behave when their interlocutor is a machine. A study comparing human-to-human online conversations with human-to-chatbot conversations found some telling differences. People chatted with the bot for longer stretches but sent shorter individual messages. The vocabulary they used was less rich, and they swore more.14Computers in Human Behavior. Real conversations with artificial intelligence: A comparison between human–human online conversations and human–chatbot conversations That last detail is revealing: when you know your interlocutor cannot be offended, some social filters switch off.
The broader takeaway is that your conversational skills transfer to human-chatbot interaction, but the social contract changes. You still take turns, you still form sentences, you still try to be understood. But the audience-design machinery described earlier, all that careful modeling of what the other person knows and feels, scales back. People seem to intuit that the chatbot does not need to be managed the way a human interlocutor does, and their speech becomes blunter and more transactional as a result. As chatbots become more human-like in their responses, it will be interesting to see whether people start treating them more like real interlocutors or continue to exploit the absence of social consequences.
Evolutionary Roots of Turn-Taking
The machinery of interlocutor interaction did not spring up from nowhere. Turn-taking, the basic structure of alternating contributions in a conversation, appears to have deep evolutionary roots. A study of bonobos and chimpanzees found that mother-infant pairs frequently engage in cooperative turn-taking sequences during gestural communication, with many structural similarities to human conversational turn-taking.15Scientific Reports. Unpeeling the layers of language: Bonobos and chimpanzees engage in cooperative turn-taking sequences The researchers argued that cooperative communication has a more ancient evolutionary origin than previously assumed, and that human conversation represents an ensemble of abilities layered over time, with turn-taking among the oldest.
The evidence extends beyond great apes. A study of sooty mangabeys, an Old World monkey species, found that their grooming interactions featured key elements of conversational turn-taking: rapid responses, paired sequences resembling adjacency pairs in human talk, and participation frameworks. Gestural turn-taking, a form previously documented only in great apes, turned out to be the most common type of signal exchange among the mangabeys as well.16PubMed Central. Turn-taking in grooming interactions of sooty mangabeys (Cercocebus atys) in the wild The implication is that the structural scaffolding of conversation, the back-and-forth rhythm that makes someone an interlocutor rather than merely a bystander, likely predates language itself. Before our ancestors had words, they had turns.