The lifelong, institutional arc for the neurodivergent — exclusion from school, to the street
Homelessness in neurodivergent people is not the start of a story, it’s the terminal outcome of a story that begins much earlier — at school, where the demand structure first encounters the substrate and starts to reject it. New research from the National Development Team for Inclusion, published in Neurodiversity on 4 June 2026, traces the arc explicitly: exclusion from young age, often beginning at school, compounding through unemployment, isolation, and absent social safety nets, until housing precarity becomes housing collapse. The end-point is not coincidental. It is the predictable result of an architecture that rejects the same substrate at every threshold of institutional life.
The headline statistic anchors the structural reality. Approximately 12% of people experiencing homelessness in the UK are autistic. The general-population figure is 1-2%. That is a six to twelve times overrepresentation. People with ADHD are also at significantly elevated risk. The 156% increase in households in temporary accommodation in England since 2010 — now at 134,760 households, with 172,420 children among them — is happening to a population in which neurodivergent people are vastly overrepresented relative to their share of the general population. The architecture that produces this is not hidden. It is the same architecture the Directory has been mapping in pieces for the last 18 months.
What this piece does is trace the arc itself, from the school exclusion that opens it to the street homelessness that closes it, while naming the pattern that makes the trajectory near-inevitable for neurodivergent people whose mechanism profiles cross diagnostic thresholds early. This is not about individual failure. It is about a system designed for a substrate that not everyone has.
What the research into neurodivergent homelessness found
The research was commissioned by a local authority and conducted by a neurodiverse team — Lauren Blood, Dr Gemma Williams (who is autistic), Paul Gutherson, and Dr Seb Shaw. The methodology was life-mapping with seven participants who had personal experience of neurodivergence, insecure housing, and unemployment. Life-mapping is a participatory qualitative method that allows participants to construct visual representations of their life trajectory in their own terms. The output is the lived shape of the arc rather than a set of discrete data points.
The arc that emerged across participants was consistent. Exclusion started young, often at school. Educational exclusion produced employment difficulty. Employment difficulty produced housing precarity. Housing precarity produced homelessness. At no point in the arc did institutional support catch the trajectory and redirect it. In fact, at several points it actively accelerated the decline.
The team identified specific barriers to accessing support during housing crisis: paperwork that participants couldn’t navigate, financial management that participants couldn’t sustain, interactions with services that participants couldn’t perform reliably, and sensory environments inside support services that participants couldn’t tolerate. The barriers are not exotic. They are direct restatements of autism-archetype mechanism mismatches — bottom-up perceptual construction colliding with sensory-overloaded service environments, social inference processing differences colliding with services that assume neurotypical social signalling, executive function load-sensitivity colliding with paperwork-heavy bureaucratic processes, working memory constraints colliding with finance management requirements. The architecture that catches people upstream — at school, at work, at GP surgeries — keeps catching them downstream at the very services nominally designed to help them out of crisis.
The participants also reported the absence of social safety nets. Neurodivergent people who have been socially excluded from school onwards rarely accumulate the dense social networks that act as informal safety nets when crisis hits. When the formal services fail, there is no informal layer to catch the fall. The arc to homelessness is therefore unbuffered. The downward trajectory accelerates rather than steadying.
The paper’s title — “Society Doesn’t Care About You” — is the participant-reported lived experience of being inside this arc. It is not a polemic the researchers added. It is what people described feeling, repeatedly, after years of institutional rejection at every threshold they encountered.
The 12% autistic homeless figure — and the architectural inevitability
A 12% autism overrepresentation in the homeless population, against a 1-2% general-population baseline, is not the kind of statistic that should be possible if homelessness were driven by individual circumstance distributed roughly evenly across the population. It is the kind of statistic that appears when an architecture is systematically rejecting one population at higher rates than another, at every threshold where housing security is at stake.
The architecture has been operating in plain sight. The pillar piece on ADHD and autism as archetypes identified “the round hole” — the demand structure built through post-feudal, mercantile, and post-industrial standardisation that requires interchangeable units of compliance behaviour the autism-archetype and ADHD-archetype substrates cannot reliably produce. The accommodation con piece showed how workplaces handle the threshold-crossing by constructing “adjustment” pathways that turn out to be abdication structures. The Right to Choose con piece showed how NHS care does the same with neurodivergent assessment and ongoing management. Homelessness is the same architecture at its most severe outcome.
The arc plays out like this. School: the neurodivergent substrate cannot reliably comply with the demand structure’s expectations of attention, behavioural regulation, social-norm performance, and sensory tolerance. The school, then, excludes — formally through suspensions and removals, informally through social ostracism, structurally through outcomes that close down further educational opportunity. Employment: the substrate cannot reliably comply with the demand structure’s expectations of consistent output, social performance, and executive function load tolerance. The workplace, then, excludes — through redundancy, through bullying that prompts resignation, through the inability of “adjustment” pathways to actually redesign demand. Housing: the substrate cannot reliably sustain the financial-administrative-social work required to keep housing in place. The system, then, excludes — through eviction, through inability to navigate housing applications, through gaps in informal support that for neurotypical adults would prevent fall. At each stage the same substrate to externally-imposed-demand-structure mismatch produces the same exclusion. The arc, then, compounds.
What the data captures, when read structurally, is the cumulative consequence of an architecture that excludes the same population at every transition. By the time the architecture has finished, the population catchment is so concentrated that 12% of those in the most severe outcome — homelessness — are autistic, with ADHD likely similarly overrepresented. This is not noise. It is signal.
The 134,760 households currently in temporary accommodation in England are not all neurodivergent. But neurodivergent people are vastly overrepresented inside that population. And neurodivergent people in the broader UK population are at substantially elevated risk of joining it. The arc this piece traces is the architectural inevitability that produces the overrepresentation, not the random misfortune that produces individual cases.
The bureaucratic-neurodivergent substrate mismatch
When neurodivergent people approach homelessness services for help, they encounter the same architecture again. The services are designed for substrates that can navigate bureaucratic processes — fill out forms accurately and on time, manage scheduling, perform social interaction with caseworkers, tolerate the sensory environment of council offices and emergency accommodation, sustain follow-through across multi-stage application processes that depend on attention, memory, executive function, and emotional regulation.
The bureaucratic demand structure is not a substrate-neutral mechanism. It assumes the same baseline compliance capacity the workplace and the school assumed, deployed in the most pressured circumstances imaginable — a person in housing crisis, often with co-occurring mental-health distress, asked to perform organised paperwork and consistent appointment attendance to qualify for the support they need to stop being homeless.
The cruelty here is not malicious. It is architectural. The same mechanism inventory that made it hard to comply with school, to sustain employment, to navigate NHS shared care, makes it hard to navigate homelessness services. The substrate is consistent across contexts; the demand structure is consistent across contexts; the mismatch is consistent across contexts. The services that nominally exist to redirect the arc back upward are configured in exactly the same way as the systems that produced the downward trajectory in the first place. Asking a neurodivergent person in crisis to perform bureaucratically reliable behaviour to qualify for housing support is asking them to do precisely what their substrate has been unable to sustain at every other threshold in their life.
This is why the participants in the Blood et al. research described services as confusing, inaccessible, and exhausting. The services were not designed for them. They were designed for the substrate the architecture assumes.
Terminal abdication — and what the State funds downstream of crisis
Local authorities in England spent £2.8 billion on temporary accommodation in 2024-25, a 25% rise on the previous year and more than double the spend of a few years earlier. The State pays for this. The State has the money. What it does not do is fund prevention upstream of the crisis — the educational adjustments at school that might keep the substrate in education, the workplace redesign that might keep it employed, the primary-care capability that might keep it healthy, the social housing capacity that might keep it housed. The downstream spend is enormous. The upstream investment is minimal. The State is paying for the consequences of an architecture it is actively choosing not to redesign.
This is the same pattern across the structural critique the Directory has been analysing. The workplace pays for tribunal awards and absence and lost productivity downstream of the friction that proper redesign would have prevented upstream. The NHS pays Right to Choose providers downstream of GP capability investments that have not been made. Local authorities now pay £2.8 billion in temporary accommodation downstream of the educational, employment, and social-care investments that would have kept neurodivergent people housed in the first place. The State has the resources. It has chosen to deploy them downstream of crisis rather than upstream of prevention. Crisis is more expensive than prevention; and the spend reflects the choice, not the absence of money nor he incapacity for management.
The terminal abdication is housing — the social safety net abdicating where the workplace and the NHS have already abdicated. The substrate that gets rejected by the round hole at school is then rejected by the workplace, then rejected by NHS care continuity, then rejected by the social safety net itself. “Society doesn’t care about you” is not a metaphor; it is the architectural truth, played back by the people inside it. The architecture is what it does. It rejects at every threshold, and the State funds the consequences of that rejection at every stage while refusing to redesign the architecture that produces them. If you are inside this trajectory and want to share your experience for ongoing work in this area, please reach out via the Directory. Stories meeting stories make architectures visible. And visible architectures can be challenged.
Citations
Blood, L., Williams, G., Gutherson, P. & Shaw, S. (2026) — “Society Doesn’t Care About You”: A Qualitative Life-Mapping Study of Neurodivergent People’s Experiences of Homelessness
National Development Team for Inclusion — Blog: Research on homelessness and neurodiversity
Ministry of Housing, Communities and Local Government — Statutory homelessness in England, October to December 2025
