Autism: when the frame of interpretation gets in the way of the person

We often talk about autism in terms of degrees, severity and functioning. We describe what can be seen: speech, eye contact, adaptation, interaction, sometimes the ability to work or to live independently.

These observations are necessary. They make it possible to describe certain difficulties and to identify support needs.

But describing what we see does not necessarily tell us what is happening.

Between the two lies something we easily forget: the gaze of the person observing.

And that gaze never arrives alone.

It carries with it a way of understanding what a successful interaction is, what normal communication is, what it means to be autonomous, present, engaged or available.

In other words, it carries a norm.

Describing is not understanding

Clinical classifications of autism necessarily rely on observable manifestations and on their impact in a person's life. They make it possible to build a shared language, to establish a diagnosis and, in some cases, to open access to support.

So the problem does not come from observation.

It appears when we forget the distance between what we observe and what we infer from it.

A person looks away.

We observe a gaze turned away.

When we speak of avoidance, lack of interest, anxiety, sensory overload or a strategy for processing information better, we have already changed level. We are no longer only describing a behaviour: we are proposing an explanation.

Written down, this distinction seems almost trivial.

In everyday life, however, it disappears very quickly.

A silence becomes withdrawal. A request for clarification becomes rigidity. Low emotional expressiveness becomes an absence of emotion. Difficulty answering immediately becomes a lack of understanding.

The observable effect ends up taking the place of the mechanism that might produce it.

And from there, something can easily escape us.

An observation is never entirely neutral

We always observe from within a frame.

The sociologist Erving Goffman devoted a significant part of his work to these frames that silently organise social interaction. We know, often without being able to explain how we know it, when to speak, how long to look at someone, how to show that we are listening, how to soften a criticism or how to signal that we have understood an expectation that was never explicitly stated.

As long as these rules are widely shared, they become almost invisible.

They simply look like reality.

It is when someone does not draw on them in the expected way that we begin to see them.

The silence suddenly seems too long. The answer too direct. The eye contact insufficient. The question needlessly precise. The emotion not expressed at the right moment or in the right way.

The other person's behaviour becomes visible precisely because it departs from the frame.

The frame itself often stays out of shot.

That is where our gaze can start to shift without our noticing: we gradually measure the distance between a person and a norm, then risk attributing that distance entirely to the person.

The subject and the social self

George Herbert Mead proposed distinguishing, within the self, between the “I” and the “Me”.

The “I” refers to the spontaneous dimension of the subject, while the “Me” is built in relation to others, through the gradual internalisation of social expectations.

We learn very early that we are not only what we feel or think. We are also someone for others.

Part of our behaviour then becomes oriented towards that legibility.

We soften a sentence because we know how it might be received. We smile to signal that everything is fine. We add a few words to an answer that might sound too blunt. We show that we are listening. We produce, often without thinking about it, the signs that allow others to understand our presence.

This process is so integrated that it can become difficult to consider it work.

And yet it requires constant coordination between what we are experiencing and the way we make it socially legible.

When presence becomes a performance

Goffman described social interaction using, among other things, the metaphor of the stage.

We present ourselves to others. We adjust what we show depending on the situation and the people involved. This does not mean that we are lying or that every relationship is artificial. It is simply an ordinary property of social life: we learn to make our presence understandable within a shared frame.

I use the term social ego to describe this dimension of the self turned towards its own legibility: the part that anticipates how it will be seen, adjusts the form, understands that the content of a message is not the only information transmitted, and gradually learns the codes that make it possible to exist within an interaction.

I am not proposing the social ego here as a clinical category, nor as a feature that would simply set autistic and non-autistic people apart. It is a frame of interpretation I use to question something my experience of autism has made particularly visible: the potential cost of this shaping when social rules are not intuitively shared.

What is almost automatic for one person can become a conscious operation for another.

You then have to observe, understand, memorise, reproduce.

And sometimes start again, because the context has changed.

Speech does not serve the same function for everyone

Language makes this difference particularly visible.

We obviously speak to convey information, but not only. Speech also maintains the relationship. It reassures, positions, softens, suggests, and shows that we belong to the frame of the exchange.

A sentence can therefore be perfectly accurate in informational terms and still be considered socially inadequate.

Conversely, a sentence can convey very little new information while accomplishing something essential in the relationship.

For some autistic people, the balance between these functions can be different. Precision of content may take up more space; relational markers may be less spontaneous or require conscious attention.

A silence is therefore not necessarily an absence of relationship.

It can be processing time, a way of reducing the amount of information to handle, or simply a moment when nothing seems to need adding.

In the same way, a very direct question is not necessarily an attack. It can be a sincere attempt to understand exactly what the other person means.

The problem appears when we use a single function of language to interpret all the others.

We can then confuse the absence of an expected relational performance with the absence of a relationship.

When two frames look at the same interaction

This difficulty connects with what Damian Milton proposed through the double empathy problem.

Misunderstandings between autistic and non-autistic people cannot always be understood as a difficulty located solely in the autistic person. When experiences of the world and communicational habits differ, both parties can struggle to interpret the other correctly.

More recent experimental work, notably by Catherine Crompton and colleagues, has brought interesting elements in this direction by showing that the quality of certain interactions also depends on neurotype matching.

This does not mean that all relational difficulties are perfectly symmetrical.

It simply requires looking at the relationship itself.

I develop this question more precisely in Autism and double empathy: when two people don't experience the same interaction.

The cost of becoming legible

When a person gradually understands that their spontaneous way of communicating, moving, looking or expressing emotion regularly produces negative reactions, they can learn to modify what they show.

This is where work on camouflaging, or masking, becomes particularly important.

Research by Laura Hull and colleagues has described various strategies used by autistic adults to mask certain characteristics, compensate for perceived difficulties or produce socially expected behaviours.

From the outside, the result can look like successful adaptation.

The person takes part. They answer. They look in the right place. They have learned what to say. Sometimes they even give the impression of being particularly comfortable with social rules.

But this observable success does not necessarily tell us anything about its cost.

Arlie Hochschild's work on emotional labour, developed in another context, offers an interesting parallel here. Hochschild shows how certain situations require individuals to produce, modify or contain emotional expressions in order to match the expectations of the role they occupy.

This is not about conflating emotional labour and autistic masking. The two concepts do not describe exactly the same phenomenon.

But they make a common question visible: what does producing the expected expression cost?

An organisation can measure the final behaviour.

It measures far less easily the energy required to produce it.

Observable effect ≠ underlying mechanism

This is probably where our way of looking at autism deserves the most caution.

Two identical behaviours can have very different origins.

Two different behaviours can also answer the same need.

A person may stay silent because they do not understand the question, because they are looking for a sufficiently precise answer, because they are overloaded, because they do not wish to answer, or simply because they have nothing to add.

Observation alone cannot settle the matter.

Yet the further a behaviour moves from what we expect, the greater the temptation to look for its explanation immediately within the person.

This is particularly visible in the world of work.

Someone who performs well alone but struggles in meetings may be considered to lack interpersonal skills. An employee who asks for very precise criteria may be seen as lacking autonomy. Someone who withdraws after several hours of interaction may seem poorly integrated into the team.

These interpretations are sometimes right.

Sometimes not.

What is often missing in between is one further question:

what does the context require of this person in order to produce the behaviour we consider normal?

This question takes nothing away from individual responsibility. It simply brings the environment back into the analysis.

What if our gaze were simply misplaced?

For a long time, a large part of interventions around autism logically sought to help the person function better in their environment.

Learning the codes. Developing strategies. Anticipating situations. Improving communication. Compensating for certain difficulties.

All of this can be useful.

But if we never look at the frame in which this adaptation becomes necessary, we risk seeing only half the problem.

The person's behaviour is visible.

The norm they are responding to is far less so.

This may be one of the things autism reveals most forcefully: a large part of what we consider natural in human interaction is in fact learned, shared and collectively maintained.

That does not make these rules bad.

It simply makes them open to discussion.

And a rule that can be discussed can also be changed when it produces more cost than value.

This question takes on a very concrete dimension in companies. Meetings, feedback, job interviews, expectations of social availability or the way autonomy is defined all rest, in part, on this kind of implicit norm.

That is also why talking about neurodiversity at work cannot consist solely of teaching managers a few characteristics of autism.

Sometimes you have to look at management itself.

Turning the gaze around

I obviously did not arrive at these questions only by reading Mead, Goffman, Milton or Hull.

I also speak from lived situations.

Misread silences. Words judged too direct. Moments when a question meant to understand was received as a challenge. Situations where I thought the problem necessarily lay in my way of being, because that was always what I was asked to change.

For a long time, the conclusion was fairly simple: do better.

Speak better. Decode better. Anticipate better. Explain better what I meant.

Over time, something shifted.

Not towards the opposite idea that the problem always comes from others. That would simply move the error.

But towards a question I had almost never allowed myself to ask:

from which frame are we deciding that there is a problem?

That question changes a great deal.

It removes neither the difficulties linked to autism, nor disability, nor the need for support. Nor does it turn every social norm into an arbitrary construction to be discarded.

It simply puts the frame back into the picture.

And perhaps that is where a fairer inclusion begins.

Not when difference is asked to become invisible.

But when we finally agree to look at the norm that makes it visible.

You cannot understand difference without understanding the norm from which you look at it.

Going further

On double empathy

To explore how two different frames can produce different interpretations of the same interaction, read Autism and double empathy: when two people don't experience the same interaction.

On management

To see how these differences of interpretation play out concretely inside a company, read How to manage an autistic person at work.

To place these questions in an organisational perspective

The Guide — Neurodiversity at work: understand, manage, transform

It covers more broadly the implicit rules of work, masking, managerial practices, recruitment and the transformations possible at the scale of an organisation.