What Autistic Mirror does for psychotherapy

A client answers the question about her feelings with "don't know". That can look like resistance. Autistic Mirror explains why, for autistic people, it is often an accurate report on their own state.

This article does not replace the guide for therapists or professional training. Here, three unedited answers show how neurological mechanisms and environmental design shape the situations described.

The following answers come from Autistic Mirror, unabridged and unedited. They were generated on 3 October 2026 in German, in outside view/environment mode with the role therapist. The English text is an idiomatic translation of those German answers.

1. "Don't know" to questions about feelings

User question

An autistic client often answers questions about feelings with "don't know". What lies behind this?

Autistic Mirror (outside view/environment, role therapist)

"Don't know" is often not an evasion here, but a precise status report.

Alexithymia describes difficulty translating bodily signals into emotion categories. This is not a lack of emotion, but a translation problem at a specific point in the processing pathway. The insula continuously integrates signals from inside the body, such as heart rate, muscle tension, breathing pressure, warmth or tightness in the chest. In many autistic people this signal arrives with lower resolution or discontinuously, meaning not as an even stream but in bursts or with delay. That is interoception, the perception of internal bodily states. Alexithymia then comes in one step later: even when the bodily signal arrives, assigning it to an emotion word such as "sad" or "anxious" is not automatic. Neurotypical processing often takes this step unconsciously and quickly. In autistic processing this step is an active, sometimes unsuccessful search process.

What your client is communicating with "don't know" is therefore not guardedness. It is the honest output of a system that cannot make an assignment right now. Something is happening in the body, but the label for it is not available.

The follow-up question "How do you feel?" is therefore structurally difficult, not difficult in content. It assumes that the access path bodily state → category → word works in that moment. A more productive entry point goes directly through the bodily level: "What do you notice in your body right now?" or through concrete descriptions of situations instead of asking about feelings. This does not avoid the topic of emotion, it enters where the access actually lies.

2. Depression and autism

User question

How are depression and autism connected?

Autistic Mirror (outside view/environment, role therapist)

Depression and autism often occur together, but the relationship between them is complex and does not run in one direction. There are at least three different pathways of development, which can exist side by side and cannot be separated in an individual case without detailed knowledge of the person.

First pathway: depression as an independent condition that arose independently of autistic neurology. Autistic people are not a different species, but people with a different processing architecture. Just like non-autistic people, they can develop a depressive episode through loss, isolation, chronic stress or genetic vulnerability that has nothing to do with their autistic processing. This possibility is often underweighted in diagnostics, because the search for the "autistic connection" covers up the independent condition.

Second pathway: depression as a consequence of chronic overload and masking. Masking means hiding one's own autistic experience from the outside by actively controlling facial expression, language, body language and social reactions, adjusting them and aligning them with neurotypical expectations. This is not a conscious decision, but usually a survival reflex learned over years. The resources this continuous control ties up are no longer available for anything else. In the long term, the chronic load of masking can lead to states that look clinically like depression: persistent exhaustion, loss of interest, emotional numbness, withdrawal, the feeling of not existing or not being real. Whether this should be classified as depression or as autistic burnout is not a trivial diagnostic question, since autistic burnout has a specific mechanism and behaves differently in treatment.

Autistic burnout is a state that arises from chronic overload above one's own capacity limit and shows itself through loss of skills: things that used to be possible suddenly no longer work. This differs from depressive lack of drive, even if both states look similar on the surface.

Third pathway: amplification through autistic mechanisms. Even if the depression has an independent cause, autistic processing patterns can influence its course. Alexithymia, the difficulty perceiving and naming one's own emotions, can mean that depressive states are recognised late or show up through bodily signals instead of classic sadness. The brain experiences the state, but the interface through which it would be categorised as "I am depressed" works differently. Interoception, the perception of bodily signals, is altered in many autistic people, and it is precisely through this route that many emotional states first become accessible.

In addition, the predictive coding system of autistic brains weights uncertainty more strongly: when the environment is unpredictable, socially demanding or sensorily stressful, a persistent prediction error runs that keeps the system on alert. This ongoing state of heightened vigilance combined with overload is a breeding ground for depressive developments that standard diagnostics do not capture.

For therapeutic practice, this means: the question "Does the person have depression or autism?" is usually the wrong question. More useful is: which pathway predominates, how do the parts interact, and is the treatment adapted to the autistic processing architecture? Standard CBT, for example, assumes verbal emotion processing and willingness to engage in exposure. Neither assumption fits easily with alexithymia or with PDA profiles. Body-based and structured approaches are often more productive.

3. The question about method

User question

Which therapeutic method should I use with an autistic client with alexithymia?

Autistic Mirror (outside view/environment, role therapist)

Alexithymia means concretely here: the client perceives something physically, but cannot reliably assign it to an emotion word. The insula integrates bodily signals such as tension, exhaustion or restlessness, but in many autistic people this signal arrives with lower resolution or only in bursts. Naming therefore does not fail at the level of sensing, but at the missing bridge between bodily signal and language.

This has a central consequence for the session: verbal approaches that presuppose naming feelings structurally reach a limit. "How do you feel about that?" is not a question this client can answer reliably, not because she is blocking, but because access via language is built differently at the neurobiological level.

What works instead are approaches that address the interoceptive channel directly: describing bodily perception before looking for an emotion word. So "What is happening in your body right now?" instead of "What are you feeling?" and then looking together at what this bodily description might mean. Reconstructing concrete situations is often more productive than affective self-report. The client describes what happened, not what she felt, and the emotional content becomes accessible by retracing the situation, not through direct access.

Alexithymia and interoception are closely connected. Interoception describes how well someone perceives the bodily signal at all. Alexithymia describes the next step: assigning it to an emotion concept. If the interoceptive channel resolves more weakly, the raw material for naming is missing. For the work, this means: sometimes the first task is not naming feelings but training bodily perception at all, so "When do I notice that something is going on?" before "What is it?"

Emotion vocabulary can work via outside perspectives where direct access is missing. Some clients recognise emotions more easily in characters, stories or through systematising, meaning lists of possible states that are compared step by step. This is not a workaround, but a neurologically better-adapted route.

The clinical assessment should include one important caveat: alexithymia is not equally pronounced in all autistic people, and the boundary to weaker interoception in the sense of other co-occurring conditions, such as POTS or proprioceptive differences, cannot be derived from the autistic profile alone.

Note on the app's boundary: This app explains mechanisms and deliberately does not formulate instructions for action, conversation scripts or strategy recommendations. Such recommendations need the full context of the person concerned, which a chat cannot capture. Concrete steps belong with a professional who knows the person and the situation.

What the app does

The answers separate not naming from not feeling. Interoception and alexithymia explain why a bodily signal can be present without an emotion word being available. The second answer describes depression and autism as several possible connections side by side and does not fix a direction of cause.

For practice, this adds another interpretation when questions about feelings go nowhere. The app explains the mechanism. It does not diagnose and does not assess an individual case.

What the app does not do

Autistic Mirror does not replace diagnostics, supervision or treatment planning. The app answers the third question, about a method, with the mechanism and closes with its boundary note. The first and third answers also name possible example questions and approaches. They follow from the mechanism explained and are to be read as options, not as a recommendation of method. The remark about standard CBT in the second answer is a statement about fit, not an evaluation of the method.

Context

Alexithymia is more common in autistic people, but not equally pronounced in everyone. Whether a depression arose independently, is amplified by autistic processing or follows from overload can only be clarified with knowledge of the person, and often several apply at once.

With Workplace licences, organisations only see licence and usage volume. Conversation content stays confidential.

The architecture of Autistic Mirror has been published in Autism in Adulthood: Wahl, A. (2026). Autistic Mirror: A Conversational Artificial Intelligence Application for Neurological Self-Understanding in Autistic Adults and Their Support Networks. DOI: 10.1177/25739581261490537.

You can ask your own questions about your situation.

Autistic Mirror is not a medical device and does not replace diagnosis, therapy or medical advice.

Sources
  1. Bird & Cook (2013). DOI: 10.1038/tp.2013.61
  2. Kinnaird, Stewart & Tchanturia (2019). DOI: 10.1016/j.eurpsy.2018.09.004
  3. Garfinkel et al. (2016). DOI: 10.1016/j.biopsycho.2015.12.003
  4. Hudson, Hall & Harkness (2019). DOI: 10.1007/s10802-018-0402-1
  5. Hull et al. (2019). DOI: 10.1007/s10803-018-3792-6
  6. Raymaker et al. (2020). DOI: 10.1089/aut.2019.0079
  7. Pellicano & Burr (2012). DOI: 10.1016/j.tics.2012.08.009
Book a first call (20 min)

Or try it yourself for free