With thanks to our member, Mark Redwood, for this blog.
Saying I am a neurodivergent therapist, is one of those statements which seems to say a lot, without actually revealing very much. One neurodivergent ‘feature’ I have is being blind to social hierarchies, having only a vague idea of how to behave in them. On one occasion I managed to upset a team, because I failed to ask one of the more senior staff about their holiday.
The ‘unwritten rules’
This is where everyone except you got the guide book on social rules. These rules are like trying to walk across constantly shifting sands, they change at a moment’s notice with little to no notice or reason, you were ‘just supposed to know’, and if you ask, the reasons you are given often make little sense. I heard the ‘unwritten rules’ described as ‘shibboleths’ once, a seemingly mundane word or phrase which when spoken reveals which social group people belong to.
A significant proportion of time in therapy can be spent decoding these ‘shibboleths’. I am fortunate, in that I have a number of neurodivergent friendly models which can explain much neurotypical social behaviour.
A lack of empathy?
Not understanding the ‘unwritten rules’ which guide social expectations is a common experience for autistic people, and is a factor behind the so-called ‘deficit in social reciprocation’, a key diagnostic criteria. On a related note, you may have read that autistic people can ‘lack empathy’.
This idea comes from research led by Simon Baron-Cohen (1985) which showed that autistic children lacked a ‘theory of mind’. If you aren’t familiar with this concept, it’s the ability to understand another’s mind from their perspective, so you can predict how they will behave. Simon Baron-Cohen hypothesised that it was this lack of a ‘theory of mind’ which was behind autistic social difficulties
When I first have a sense that someone might be neurodivergent, I will typically ask them if they have any neurodivergent friends, and the usual answer is yes. The commonest reason given is that they feel more “accepted and understood” by other neurodivergent people. Perhaps like me, you are having a ‘hang on a minute’ moment. Aren’t autistic people supposed to find understanding other people difficult?
There’s more. One of the common things autistic and neurodivergent clients talk about is being constantly “misread, and misunderstood” by…
...neurotypical people.
So wait, this neurotypical ‘theory of mind’ which on the one hand excels at decoding ‘shibboleths’ is somewhat rubbish at understanding autistic people. You would be right if you were wondering that there might be a big and important part of the story missing,
Damian Milton (2012), who is autistic himself, calls this ‘missing story’ the ‘double empathy problem’. He proposes that autistic people do not in fact lack empathy, rather autistic empathy works very differently to neurotypical empathy. When you understand it this way, it now makes sense why two people with very different empathy systems would have a hard time understanding each other. It’s a double problem because it swings both ways.
However, what Damian Milton says about our side of the problem, isn’t very comforting, “When [neurotypical] ‘empathy’ is applied toward an ‘autistic person’ however, it is often wildly inaccurate in its measure.” (2012, emphasis added by author). I personally found reading that in an academic paper shocking. Not just sometimes, but often, and not just a bit wrong, but wildly wrong. Does that mean, as a profession, we are ‘often wildly inaccurate’ about our autistic clients?
Double empathy in the therapy room
Damian Milton (2012) argues that there is a “growing complacency around […] what ‘good autism practice’ entails, for me this is an ongoing imperfect process of interaction and should never be seen as a given”. For my neurodivergent specialist colleagues and I, a typical client is someone who has tried several other therapists with limited success before coming to one of us.
This seems to be borne out by surveys, such as the one by the Autistic Mutual Aid Society Edinburgh (AMASE, 2018), which revealed 52% had an overall negative experience of mental health services, while 42% said they weren’t understood or listened to, which included one person whose practitioner missed they were suicidal. To me, this looks like a problem of inaccurate empathy leading to autistic people being ‘misread and misunderstood’.
An imbalance of power
Damian Milton adds a further nuance to the double empathy problem, when he notes it isn’t experienced equally by both sides.
An example is when someone is describing something they have experienced. A typical autistic empathic response is to demonstrate understanding by describing a similar situation that happened to them, which is often misinterpreted, usually as selfishness or wanting to turn the subject onto themselves.
What this creates is a break in the social connection, which Damian Milton describes as a ‘disjuncture’. On the neurotypical side it’s seen as a breach of a ‘social norm’, an implicitly agreed set of standards and expectations, while on the autistic side it is experienced as yet another potentially traumatising experience of ‘getting things wrong again’.
I personally like to show empathy by example sharing too, but I know that the ‘correct’ response is to give the person space to freely express their feelings. However, a common reason to talk about feelings for my neurodivergent clients is to seek validation that their feeling or behaviour was ‘normal’. Allowing space to express feelings has the potential to feel unanchored, and risks deepening shame, while a carefully matched piece of example sharing can be a better and more accurate empathic response.
Of course what determines which empathic response forms the social norm is the group holding the most social power, which generally is neurotypical people. My own professional journey was one of first encountering people who broke many of these ‘social norms’, and then finding I had to challenge many of my assumptions particularly around what I believed constituted ‘good therapy’. For instance I will typically put in more explanation, often using illustrative stories, with sessions sometimes being more like an ‘analysis of theory’ than an exploration of process.
Two different kinds of processing
One key difference between autistic and neurotypical people is the kind of processing that dominates, which I think helps explain this ‘double empathy problem’. Autistic, and many other neurodivergent people, primarily process from the bottom-up, where as neurotypical people primarily process from the top-down. This idea is not a singular theory, developed by one person, rather it has emerged as a framework derived from multiple sources to explain autistic experience.
Starting with top-down processing. It’s called that because neurotypical brains are capable of processing lots of complex information to create rapid top-down global understandings. It’s like they are able to go straight to the conclusion without knowing exactly how they got there. Things like reading a room, knowing what to say or do in social situations, reading people’s emotions and motives, participating in back and forth group conversations, is relatively effortless. Neurotypical brains are like sponges for nuanced social rules, they seem to inhale them without ever being told what they are.
If this all sounds a bit mysterious, it’s only that way because to me, it seems like some magical process, which I have to work very hard at to even get close to. I happen to know that those social rules are learned intuitively over time by trial and error, which also explains why explanations of how a neurotypical person knows something, often doesn’t make much sense to me, because understanding unconscious processes is usually a matter of guesswork. Neurotypical brains are generally tuned to understand other neurotypical minds, hence why they can be ‘wildly inaccurate’ when applied to autistic minds.
Another factor which can contribute to this double empathy problem is that top-down big picture processes come with a feeling of ‘rightness’, which anchors any responses which follow. It thus limits the information which can be accommodated, and anything which is beyond this limit will typically get discarded. Being misread and then being disbelieved when you explain your side, is a common experience for autistic people. For example, try explaining that you weren’t just talking about yourself, you were in fact trying to empathise. This is not a conversational line which usually goes very well.
Neurodivergent, and autistic people in particular, have fewer of these intuitive systems. Instead we are relying much more on conscious cognitive systems, and so are largely processing from the bottom-up. Now, I am not thinking through each thing I say in a live conversation, because my thinking is a bit too slow and clumsy for that. What I am actually doing is picking from pre-prepared ‘scripts’. So for example, I have dog-walking scripts, how-are-you scripts, and listening scripts. In my ‘downtime’ I then develop, amend and rehearse these scripts, and I am almost always processing something. If I have something emotionally very significant, then this process can become extremely intense; delivering just one 4 hour workshop can consume upwards of 50 hours developing, amending, and rehearsing the ‘scripts’ I will need. With these kinds of processing demands it’s no surprise many neurodivergent people struggle with burnout.
A few weeks ago I went to a yoga laughter workshop. After the introductory outline, our trainer then began setting us group exercises. I was initially very uncomfortable, and if you had asked me, I would have said, “I don’t know what I am supposed to be doing”. When I used this phrase I didn’t mean I didn’t know how to physically do the exercises, what I meant was I didn’t know how to connect to the laughter. After our trainer had set the second exercise I realised they were both childhood games with some extra absurdity thrown in. So now I get it, I am laughing at how absurd this is, which is something I can do.
Although it’s quite subtle, this is also an example of top-down verses bottom-up processing. The reason my colleagues seem not to be struggling with the exercises is because their top-down system is able to rapidly and unconsciously process the ‘unwritten rules’ of this game, whereas my bottom-up system has to consciously discover those ‘unwritten rules’ to arrive in a similar place.
It’s important to say that my experience of bottom-up processing is just one example, and there is a lot of variation between different neurodivergent people, however there are some recognisable features which bottom-up processing will typically have. One is a strong focus on details, conclusions come later. As a consequence, not knowing and confusion can be common states. It’s generally a pattern matching type process, which is strongly cognitive and based on rules. It’s also not unusual for the person to be able to describe their process in detail, often producing long and complex reasoning chains.
What I like about this way of thinking, is that it explains why communication difficulties can happen between neurodivergent and neurotypical people. It makes sense when you realise that if one group is coming at a problem from opposite directions, it’s not a given they are going to arrive at the same conclusion.
Bridging the gap
Like many people who realise they are neurodivergent later in life, I came with many inbuilt neurotypical assumptions, which I have had to unlearn. And even though I am a long way from that place, being neurodivergent doesn’t mean I automatically understand another’s internal world. Understanding ideas such as double empathy and processing differences, underpins my work, helping me bridge the gap between what can be very different worlds.
References
Autistic Mutual Aid Society Edinburgh. (2018) Too complicated to treat? Autistic people seeking mental health support in Scotland. Available at: https://amase.org.uk/wp-content/uploads/2018/11/AMASE-autism-and-mh-report-full-ver.pdf (Accessed: 28th August 2025)
Baron-Cohen, S., Leslie, A.M., Frith, U. (1985) ‘Does the autistic child have a “theory of mind”?*’, Cognition, 21, pp 37-46.
Milton, D. E. M. (2012) ‘On the ontological status of autism: The 'double empathy problem' ’, Disability & Society, 27(6), pp 1-5.
Updated September 2025
