Speaking a Different Language: Neurodivergent Communication and the Double-Empathy Problem

Why some conversations feel effortless while others feel like translating between two different languages.

I have pragmatic communication differences, and I have ADHD with many autistic-like traits. One of the experiences I have struggled to explain is that communicating with some neurotypical people can feel like speaking a different language. We may use the same words, understand the same definitions, and participate in the same conversation, yet somehow miss each other entirely.

Sometimes I find myself wondering why a conversation that seems straightforward to me becomes unexpectedly complicated. I may say exactly what I mean, only to discover that the other person has interpreted my words differently than I intended. I may approach a conversation looking for clarity, directness, or a practical exchange of information, while the other person seems to be attending to implications, social conventions, or meanings I did not realize were part of the conversation.

I have also noticed something different when interacting with other neurodivergent people. There can be an immediate sense of ease, a feeling that we are operating on a similar wavelength. I have heard other neurodivergent people describe similar experiences. Sometimes the conversation flows before either person knows much about the other's background or diagnostic history.

I have wondered whether we are recognizing something familiar in one another's communication styles. Is it possible that the interaction simply requires less translation?

As it turns out, there is research that helps us begin to explore this question. The evidence is more nuanced than the idea that neurodivergent people automatically communicate better with one another, but it challenges the assumption that communication difficulties belong exclusively to the neurodivergent person.

The double-empathy problem: What if communication is a two-way process?

In 2012, autistic researcher Damian Milton introduced the concept of the double-empathy problem. His work challenged the traditional assumption that autistic people struggle to understand non-autistic people while non-autistic people can readily understand autistic people.

Milton proposed that people with different experiences, perspectives, and ways of communicating may have difficulty understanding one another. The mismatch is reciprocal. An autistic person may misinterpret a non-autistic person's intentions, just as a non-autistic person may misinterpret an autistic person's intentions.

This is an important shift in perspective. Rather than viewing communication as a skill one person either possesses or lacks, the double-empathy problem asks us to consider what happens between people.

Imagine a person who values direct communication. They say what they mean, ask explicit questions, and expect others to communicate their needs clearly. Another person may rely more heavily on indirect language, subtle social cues, or shared assumptions about what should remain unsaid.

Neither person necessarily intends to confuse the other. They may simply be relying on different conversational expectations.

The first person may experience the second as vague, evasive, or unnecessarily complicated. The second may experience the first as abrupt, insensitive, or socially unaware. Each person interprets the other's behavior through their own expectations, and both may conclude that the other person is the source of the problem.

This is where the language metaphor becomes useful. We can share a spoken language while having different expectations about what words mean in practice, how much should be implied rather than stated, when someone should speak, and what constitutes an appropriate response.

The difficulty may not be that one person cannot communicate. It may be that the two people are communicating according to different rules.

What research tells us about autistic-to-autistic communication

One of the most relevant studies on this question was published by Catherine Crompton and colleagues in 2020. The researchers recruited 72 adults and organized them into groups consisting entirely of autistic participants, entirely of non-autistic participants, or a mixture of both.

Participants took part in a version of the familiar telephone game. One person received a story and passed it along to another person, who then passed it to someone else. Researchers examined how accurately information was transmitted through the groups and also measured participants' sense of rapport.

The results were striking. Autistic groups transmitted information as effectively as non-autistic groups. Mixed groups, however, showed greater information loss and lower rapport in this particular experimental task (Crompton et al., 2020).

These findings challenge the assumption that autistic people are inherently less capable of communicating effectively. The autistic participants demonstrated that they could share information successfully with one another. The difficulties emerged differently depending on the composition of the interacting group.

A related study by Crompton and colleagues, also published in 2020, examined interpersonal rapport. The researchers found that neurotype matching, rather than autism alone, influenced how participants and outside observers rated the quality of interactions. The findings suggested that shared neurotype could contribute to a greater sense of interpersonal connection (Crompton et al., 2020).

For someone who has experienced the ease of connecting with certain neurodivergent people, these findings are worth considering. They suggest that communication effectiveness and interpersonal comfort may depend partly on the compatibility of the people involved.

However, science requires us to remain open to findings that complicate an appealing explanation.

In 2025, Crompton and colleagues published a larger, preregistered study involving 311 adults across three research sites. This study did not replicate the earlier finding that mixed autistic and non-autistic groups transmitted information less accurately. The results also did not support a simple rule that matching neurotypes always produces better rapport across all measures (Crompton et al., 2025).

This newer research matters. It does not erase the earlier findings or invalidate individual experiences of communication compatibility. Instead, it tells us that the relationship between neurotype, communication, and rapport is complex. The task, the individuals involved, and the context may all influence what happens.

It is also important to distinguish between autistic and non-autistic participants and neurodivergent and neurotypical people more broadly. Non-autistic people may have ADHD or other neurodevelopmental differences. Research comparing autistic and non-autistic groups cannot automatically tell us how every combination of neurotypes communicates.

Where ADHD enters the conversation

Autism research provides some direct evidence for the role of neurotype matching in communication. ADHD research offers another important piece of the puzzle, although the evidence answers a somewhat different question.

Pragmatic language refers to how we use and interpret language in social contexts. It includes conversational turn-taking, staying on topic, organizing information for a listener, recognizing what another person already knows, and interpreting indirect or nonliteral language.

These abilities are not simply about knowing words or understanding grammar. They involve coordinating attention, timing, context, and expectations during an interaction.

ADHD can affect some of these processes. For example, a person may interrupt because an idea comes to mind before the other person has finished speaking. They may talk at length about a topic they find interesting, struggle to organize their thoughts quickly enough for conversation, or lose track of what someone is saying while trying to formulate a response.

A 2025 study by Köder, Rummelhoff, and Garraffa examined self-reported pragmatic communication abilities among 179 multilingual adults, including 91 adults with ADHD and 88 without an ADHD diagnosis. Adults with ADHD reported more communication difficulties, particularly involving excessive talking, interrupting, and speaking before fully considering what they wanted to say (Köder et al., 2025).

The study offers useful adult-specific evidence that ADHD can influence the way people manage spoken interactions. However, it did not directly compare conversations between people with matching and differing neurotypes. It also did not collect information about co-occurring autism, making it impossible to isolate the effects of ADHD from potentially overlapping autistic characteristics.

This is important because ADHD is not a single communication style. Some people with ADHD are highly talkative, while others are quieter or need additional time to formulate a response. Some communicate very directly, while others rely heavily on context and implication. Individual differences, co-occurring conditions, lived experiences, and learned strategies all influence communication.

The available research supports the idea that ADHD can affect pragmatic communication. It does not yet establish that adults with ADHD consistently experience greater conversational ease with other people with ADHD or with autistic people.

That is an important question for future research.

Why might some neurodivergent conversations feel easier?

Research has not yet established one explanation for the immediate familiarity I and others have described. Still, several possibilities are worth exploring.

Shared expectations may reduce the need for translation. If two people have similar preferences for directness, conversational pacing, or explicit explanations, they may spend less time interpreting what the other person really means. They may be more comfortable asking questions, clarifying misunderstandings, or moving directly to the substance of a conversation.

Differences may be interpreted more generously. An interruption might be understood as enthusiasm rather than disrespect. A detailed explanation might be welcomed as useful information rather than interpreted as overexplaining. A pause might be accepted without assuming the person is disengaged. These are possibilities, not universal neurodivergent traits, but they illustrate how the same behavior can acquire different meanings depending on the interaction.

There may be less pressure to perform conventional social expectations. Some neurodivergent people spend considerable effort monitoring eye contact, facial expressions, tone, conversational timing, and other social expectations. When interacting with someone who shares their communication preferences, they may feel less pressure to consciously manage every aspect of the exchange.

Recognition may emerge from familiarity rather than diagnosis. I sometimes wonder whether what feels like an almost instinctive recognition of another neurodivergent person is actually the recognition of a familiar communication rhythm. The conversation may feel comfortable before either person knows whether the other has ADHD, is autistic, or identifies as neurodivergent at all.

These possibilities are consistent with the broader question raised by the double-empathy framework. They should not be mistaken for proven explanations of why neurodivergent people recognize or connect with one another. Research specifically examining this experience, particularly among adults with ADHD and people with overlapping ADHD and autistic traits, remains limited.

The difference between communicating effectively and feeling understood

One of the most important lessons from this research is that communication has multiple dimensions.

A conversation can transmit information accurately without creating a sense of connection. Two people may understand every word the other says but still feel misunderstood, judged, or uncomfortable. Conversely, two people may feel an immediate connection even when they occasionally misunderstand one another.

Accuracy, rapport, emotional safety, and a sense of being understood are related, but they are not interchangeable.

This distinction helps explain why the experience of communication can be difficult to capture through conventional measures alone. If we evaluate communication exclusively by whether information was conveyed correctly, we may miss the experience of effort, exhaustion, or connection that accompanies the interaction.

For many people, the question is not simply whether they can communicate with someone. It is how much effort communication requires, whether they feel accepted during the exchange, and whether they can express themselves without constantly monitoring how they are being perceived.

This is particularly relevant when considering the experiences of neurodivergent adults who have spent years adapting to environments organized around other people's expectations. Someone may become highly skilled at navigating conventional social interactions while still experiencing those interactions as effortful.

Successful adaptation does not necessarily mean that the interaction feels natural or reciprocal.

Moving beyond the idea of a communication deficit

I believe this research invites us to ask a different question when communication breaks down.

Instead of immediately asking, "What is wrong with this person's communication?" we might ask, "What is happening between these two people, and what assumptions are each of them bringing to the interaction?"

This does not mean that communication difficulties are imaginary or that no one ever needs support with conversational skills. ADHD can affect turn-taking, attention, and impulse regulation. Autistic people may experience genuine difficulties interpreting certain social conventions or communicating in environments that do not accommodate their needs. Individuals can benefit from learning strategies that make communication more accessible and effective.

The point is that support should not automatically assume that one communication style is the correct standard and every other style is a deviation from it.

A more reciprocal approach considers the needs of both people. Can expectations be made explicit? Can direct questions be welcomed rather than interpreted as confrontational? Can someone ask for clarification without being judged? Can conversational differences be discussed without assigning blame?

In therapy, relationships, workplaces, and everyday interactions, these questions can open up possibilities that a deficit-only explanation might overlook.

They also invite neurotypical people to participate in improving communication rather than placing the entire responsibility for adaptation on their neurodivergent counterparts.

What I take away from this research

My experience of communicating across neurotypes is not proof of a universal pattern, and research has not yet fully explained why some neurodivergent interactions feel immediately familiar to me. The evidence is strongest for particular aspects of autistic-to-autistic communication, while adult ADHD research has begun to clarify how ADHD can influence pragmatic language and spoken interaction.

Nevertheless, the research offers an important alternative to the assumption that communication difficulties necessarily originate within one person.

Perhaps what feels like speaking a different language is sometimes a mismatch in expectations, processing, conversational rhythm, or the meaning assigned to behavior. Perhaps the ease I experience with certain people reflects a greater compatibility between our ways of communicating. And perhaps the most useful question is not which person communicates correctly, but what allows two people to understand one another.

I am interested in what happens when we stop treating conventional communication as the universal standard and begin to understand communication as a shared process. That shift does not require us to deny differences or overlook the challenges they can create. It simply asks us to consider the possibility that understanding is a responsibility we share.

Communication is not only about how well one person expresses themselves. It is also about what happens when another person receives, interprets, and responds to that expression.

Sometimes, the most meaningful change begins when we stop asking one person to learn a different language and start learning how to meet each other halfway.

References

Crompton, C. J., Ropar, D., Evans-Williams, C. V. M., Flynn, E. G., & Fletcher-Watson, S. (2020). Autistic peer-to-peer information transfer is highly effective. Autism, 24(7), 1704–1712.

Crompton, C. J., Sharp, M., Axbey, H., Fletcher-Watson, S., Flynn, E. G., & Ropar, D. (2020). Neurotype-matching, but not being autistic, influences self and observer ratings of interpersonal rapport. Frontiers in Psychology, 11, Article 586171.

Crompton, C. J., Foster, S. J., Wilks, C. E. H., Dodd, M., Efthimiou, T. N., Ropar, D., Sasson, N. J., Lages, M., & Fletcher-Watson, S. (2025). Information transfer within and between autistic and non-autistic people. Nature Human Behaviour, 9, 1488–1500.

Köder, F., Rummelhoff, C., & Garraffa, M. (2025). Comparing pragmatic abilities across multiple languages in adults with ADHD: Insights from a self-report questionnaire. Clinical Linguistics & Phonetics, 39(9), 827–842.

Milton, D. E. M. (2012). On the ontological status of autism: The ‘double empathy problem.’ Disability & Society, 27(6), 883–887.