The Autistic Nervous System

The diagnostic manuals describe autism through what is missing: "deficits in social communication," "restricted, repetitive patterns of behaviour." Every criterion is written from the outside, measuring distance from a neurotypical baseline. This is the framework nearly everyone — parents, teachers, clinicians, autistic people themselves — first encounters autism through.

Autistic researchers and self-advocates have spent the last three decades building a different account, not to deny real difficulty, but to relocate where the difference actually sits. These accounts are theoretical frameworks with growing but still limited empirical testing. Two frameworks carry most of that weight: monotropism, which explains autism as a distinctive style of attention rather than a social defect, and the double empathy problem, which relocates "poor social communication" from a one-sided autistic failing to a two-way mismatch between different communication styles.

Monotropism was first articulated in the early 1990s by Dinah Murray, an autistic researcher, and developed with Wenn Lawson and Mike Lesser into a fuller theory in 2005 (Murray, Lesser and Lawson, 2005). Their central claim: attention is a finite resource, and people differ in how they distribute it. The neurotypical mind is generally polytropic — it spreads attention across many channels at once, tracking a conversation, the room, a half-formed plan for later, all simultaneously. The autistic mind is generally monotropic — it channels most of its available attention into very few interests at a time, but goes far deeper into each one than a polytropic system typically can.

Murray called the result an attention tunnel. Inside the tunnel, an autistic person can reach a state of total absorption — sometimes called autistic flow — in which time, hunger and surrounding noise all but disappear. Outside a tunnel, when nothing has claimed the available attention, small stimuli can pull focus unpredictably, since there is no dominant interest holding it steady. Both patterns follow from the same underlying architecture, not from two unrelated symptoms.

This single idea reframes several things usually listed as separate deficits. A "restricted range of interests" is what a deep attention tunnel looks like from outside. Difficulty with sudden change follows directly: an unanticipated interruption doesn't just redirect attention, it forcibly closes a tunnel that took real effort to enter, and re-entering — or opening a different one — takes time the neurotypical world rarely allows for.

Sensory processing differences are reported by the large majority of autistic people — around 90–95% in some surveys (Tomchek and Dunn, 2007; Crane et al., 2009) — and have been part of the formal diagnostic criteria (DSM-5) since 2013. The nervous system can register ordinary sensory input as hypersensitive (a fluorescent light or a scratchy seam registers as genuinely painful, not merely annoying) or hyposensitive (the nervous system under-registers input, producing a need to seek out intense sensation — spinning, deep pressure, loud sound — just to reach an ordinary baseline of feeling).

Research complicates the tidy hyper/hypo split: the same person is often hypersensitive in one sensory channel and hyposensitive in another, and can show both patterns in the same channel depending on context (Schauder and Bennetto, 2016). There is no single "autistic sensory profile"; there is a nervous system calibrated differently, channel by channel, from the neurotypical default.

Interoception — sometimes called the eighth sense — is the ability to perceive signals from inside the body: hunger, thirst, a racing heart, the need to use the bathroom, the first stirrings of an emotion before it has a name. Some autistic people report reduced awareness of internal body signals, and this can have real practical consequences: forgetting to eat or drink for hours, not registering pain that would stop a neurotypical person immediately, or reaching a full meltdown without any felt warning that stress was building.

The research picture is subtler than “autistic people can’t feel their bodies.” Findings are mixed, and several well-known studies found that interoceptive difficulty tracked alexithymia rather than autism itself (Shah et al., 2016). Alexithymia — difficulty identifying and describing emotions — is common but not universal among autistic people; a meta-analysis estimated that about half have it (Kinnaird et al., 2019). Some studies find autistic people without alexithymia show interoceptive accuracy similar to non-autistic controls. The honest picture: autism and alexithymia frequently travel together and are easy to conflate, but they are not the same thing, and treating every autistic person as automatically alexithymic overstates the evidence.

This distinction matters for the empathy question specifically. Alexithymia makes it harder to recognise and name an emotion in the moment — including one's own — which can look, from outside, like not having the emotion at all. It is a communication and recognition difference, not an absence of feeling.

Autistic academic Damian Milton proposed, in 2012, that the “social communication deficit” attributed to autistic people is only half of a two-sided story (Milton, 2012). Autistic and non-autistic people, Milton argued, are simply differently disposed social groups — each finds the other's communication style, body language and emotional expression somewhat opaque. The breakdown runs in both directions; it has simply always been described from the neurotypical side, because that side wrote the diagnostic criteria.

The theory does not claim every social difficulty is perfectly symmetrical, and it does not deny that autistic people can face genuine communication challenges. What it removes is the assumption that the autistic person is always the one who failed to understand. Experiments find that information passes as effectively between two autistic people as between two non-autistic people, and less effectively in mixed pairs (Crompton et al., 2020), and that non-autistic people are also worse at reading autistic people’s mental states than autistic people’s own (Sheppard et al., 2016; Edey et al., 2016). This supports the idea that the mismatch is relational rather than a one-sided deficiency, although the research is still modest in size.

Pathological Demand Avoidance — increasingly relabelled Pervasive Drive for Autonomy by those who find the older name stigmatising — describes a profile, most widely recognised in the UK, in which ordinary requests trigger intense anxiety and avoidance (Newson et al., 2003). PDA is not an official diagnosis in DSM-5 or ICD-11, and its status and boundaries are debated; some researchers view demand avoidance as a feature of anxiety and intolerance of uncertainty (O’Nions et al., 2014; Stuart et al., 2020). It is not stubbornness or a parenting failure: people with this profile and their families describe an everyday instruction like “please put your shoes on” provoking a fight, flight or freeze-like reaction, because loss of autonomy feels threatening.

The profile can obscure autism rather than announce it. Many people with a PDA profile appear socially fluent — good eye contact, easy conversational rapport, comfort with role-play and pretence — which leads to autism being missed entirely and the demand avoidance being mislabelled as oppositional defiant disorder, especially in children. PDA-informed guidance, based largely on clinical experience and family reports rather than controlled trials, suggests that collaboration and negotiated choice reduce perceived threat, while rigid structure — genuinely helpful for many autistic people — can feel like an imposed demand to some people with this profile.

Masking (also called camouflaging) is the conscious or unconscious suppression of autistic traits to pass as neurotypical — forcing eye contact, rehearsing scripted small talk, suppressing stims in public and releasing them only in private. It is thought to be one reason autism in women and AFAB individuals has been underdiagnosed: the true male-to-female ratio is probably closer to 3:1 than the 4:1 or higher once believed (Loomes et al., 2017), and women tend to be diagnosed later; camouflaging appears more common among autistic women (Hull et al., 2017; Lai et al., 2017).

Sustained masking is associated with poorer mental health, and autistic adults describe autistic burnout — a state of chronic exhaustion, reduced functioning and loss of skills (Raymaker et al., 2020). Some autistic writers describe a cycle of four overlapping states — burnout, inertia, meltdown and shutdown (BIMS); this is a community model rather than a research-validated one. Meltdown and shutdown in particular are frequently confused with each other and with simple bad behaviour. A meltdown is an involuntary overflow of overwhelm, not a tantrum aimed at getting something; a shutdown involves withdrawal, loss of speech or slowing. Polyvagal-informed writers liken a meltdown to a sympathetic fight-or-flight response and a shutdown to a dorsal-vagal freeze response; this is an analogy rather than something established by autonomic measurements. In either case, neither is a choice made in the moment.