Education & family
School aides in Germany. What the planned 2028 reform means for autistic children.
In March 2026, Germany's Federal Ministry for Family Affairs published a draft of the Children and Youth Services Reform Act (KJHSRG). As of May 2026 it is not yet passed. Planned effective date: 1 January 2028. From that date, youth welfare offices would become responsible for all support services for children and adolescents with disabilities, regardless of whether the disability is classified as mental, physical, or intellectual. This affects roughly 80,000 school-aide cases in Germany today.
This article describes three things. First, what the draft contains. Second, what this shift means neurologically for autistic children. Third, how to recognise good school support mechanistically, regardless of who funds it.
What the planned reform sets out
Two German social codes currently share responsibility. SGB VIII § 35a covers integration support for children with mental disabilities through the youth welfare office. SGB IX § 112 covers participation in education for children with physical or intellectual disabilities through the social welfare office. Autistic children fall into either track depending on diagnostic classification. Parents experience this as a jurisdictional dispute.
The KJHSRG would dissolve this split from 2028. One agency, one application path, one approval logic. Existing approvals remain valid past the cutoff. The reform sits within the framework of the UN Convention on the Rights of Persons with Disabilities (Article 24, right to inclusive education) and Germany's Federal Participation Act.
Two points are contested. The first is the planned "infrastructural educational assistance". Instead of an individual legal entitlement, a pool of school aides would be maintained at each school and shared across multiple children. Disability associations such as Lebenshilfe and DBSV warn that this shifts the burden of proof: parents would have to explain why a pool solution does not work for their child. The second point is § 85 (5) SGB VIII-E, a regional opt-out clause allowing different standards across federal states. The federation of welfare associations (BAGFW) criticises that the draft prioritises administrative simplification over individual legal entitlement.
The classroom as a predictive-coding stress test
To understand the reform debate neurologically, it helps to look at what actually happens in the autistic nervous system in a classroom.
The autistic brain works through predictive coding. It continuously builds prediction models of what comes next: sounds, reactions, transitions, moods. When model and reality match, processing is efficient. When they diverge, processing cost rises. In autistic processing these models are more granular and more sensitive to deviation than in neurotypical processing. That is not a deficit, it is a different calibration.
Add monotropism. Attention bundles into a single channel rather than spreading broadly. This enables deep processing but makes interruptions costly. Every shift between task, person, and room incurs state-change costs that neurotypical processing does not pay at the same level.
School aides intervene in both mechanisms. A consistently present person becomes part of the prediction model. Reaction patterns, language, body proximity, transition cues. The brain no longer has to re-evaluate the aide constantly. A rotating person is structurally unpredictable. This is not "less comfort" but measurable additional processing load running in parallel to lessons.
The conversations below are taken 1:1 from the app, unedited and uncut. Translated from German.
User question
The federal government is planning a reform of school-aide funding. What does that mean for my autistic child in primary school?
Autistic Mirror (outside view, parent)
Current and specific information on a pending or planned reform of school-aide funding by the federal government is not available here. Legislative changes, reform proposals, and their concrete status are not part of the knowledge this app is based on.
What can be answered reliably here is the legal framework that applies today and how school-aide funding works for your child. The entitlement is anchored in SGB IX (integration support) and SGB VIII (children and youth services). It is not a discretionary offer but a legal entitlement when it is shown that your child cannot participate in lessons without support.
Whether and how a reform changes this framework would require a current source: the responsible federal ministry, VdK, Caritas, or Autismus Deutschland e.V. They are closer to ongoing legislative processes than this app.
The app deliberately keeps distance from current politics. The explanatory text in this article supplements the reform status. You can ask your own questions about your child in Autistic Mirror.
1:1 support or pool. What decides the question neurologically.
This is exactly where the dispute over "infrastructural educational assistance" begins. The question is mechanistically answerable, without a blanket verdict.
User question
Currently the welfare office funds school aides via integration support (SGB IX). Which neurological mechanisms decide whether my autistic primary-school child needs 1:1 support or whether a pool model with several children at once works?
Autistic Mirror (outside view, parent)
Predictive coding is the central mechanism. A school aide who is exclusively with your child becomes predictable: presence, reactions, communication style. The brain can build a stable model. A pool aide is structurally unpredictable. Who comes today? How long do they stay? Will they be needed elsewhere? Each open variable produces processing cost, even when "nothing is happening".
Autistic inertia amplifies this. Inertia means: an ongoing state is hard to interrupt or switch neurologically. When the aide suddenly moves to another child, this is not a neutral event but a forced state change.
Then there is sensory-filter exhaustion. The autistic nervous system filters sensory input less automatically than neurotypical peers. A classic classroom is an accumulating constant load: fluorescent light, chairs, voices, movement. The aide acts as a buffer in this context. In a pool model that buffer is only intermittently available.
The pool model can work when the child has a stable internal regulation mechanism, processes transitions without significant dysregulation, and the sensory load in school is moderate. It fails neurologically when the aide functions primarily as a regulation anchor and safety buffer, not just as practical help with concrete tasks.
Recognising good school support
Regardless of the funding model, the person and their stance decide. The question is not "kind or strict", but: does the aide adapt to the child's nervous system, or does she expect the child to adapt to her.
User question
From a neurological perspective, how does good school support differ from bad, and what should I, as a mother, look for mechanistically when meeting an aide for the first time?
Autistic Mirror (outside view, parent)
Predictive coding explains why reliability is central. A good aide becomes predictable over time: reaction patterns, language, body proximity, how she announces transitions. A bad aide stays unpredictable, switches reactions by mood, announces nothing. That produces persistent background stress, regardless of whether anything "bad" is currently happening.
The second mechanism is masking. A good aide reduces the pressure to mask. She accepts stimming, does not require eye contact, does not evaluate. A bad aide raises the masking pressure, often without noticing: through corrections ("Look at me when I'm speaking to you"), through normalisation impulses ("Can't you stop that?"). The result is a child who functions in school and collapses at home.
The third mechanism is interoception. A good aide learns to read external signs of overload before the child notices being at the limit. A bad aide waits for a meltdown and then reacts.
One question you can ask directly: "What do you do when my child seems overloaded but has not yet reached a meltdown?" The answer shows whether the person thinks preventively or reactively.
Risks of a centralised reform
The strength of the planned unification lies in removing the jurisdictional dispute. That is real and relieves families. The weakness lies where standardisation overlooks the heterogeneity of autistic profiles. Sensory profile, executive functions, masking budget, co-occurring conditions such as ADHD or ARFID, language processing. These variables differ substantially from child to child. A pool model that replaces this variability with infrastructural blanket solutions risks exactly what SGB IX § 112 and Article 24 of the UN CRPD are meant to secure: an entitlement that fits the concrete person, not the average classroom.
The regional opt-out clause sharpens this. What counts as "appropriate" in Berlin can look different in a rural state. The right to participation is federal, the implementation becomes regional. In practice, parents become translators between federal law, state implementation, and municipal approval practice.
Thinking school support neuroaffirmatively
The final perspective is the aide's own. From the outside-view mode, profile school aide.
User question
I work as a school aide for an autistic child in primary school. Which neurological mechanisms explain why "neuroaffirmative support" is more than just reducing demands, and how does support that preserves autonomy differ neurologically from support that undermines it?
Autistic Mirror (outside view, school aide)
The autistic brain processes the environment from the bottom up. Bottom-up processing means: stimuli, details, and signals are first processed in full and unfiltered before context is added. This costs more compute than in a neurotypical brain that relies heavily on prediction and filtering. When an external person takes over decisions, this breaks the predictability of one's own action. The brain loses control over its own models. That is not abstract discomfort but concrete neurological destabilisation.
Here PDA enters. PDA stands for "Pathological Demand Avoidance" or, more precisely, "Pervasive Drive for Autonomy". The nervous system registers demands and external control as a threat signal, not as an invitation to cooperate. That is not defiance. It is an automatic, anxiety-driven protection mechanism.
An autonomy-undermining aide structurally produces what masking produces. The child adapts behaviour to an external person's expectations rather than to its own neurological needs. When a school aide is present six hours a day as an implicit source of expectations, she can unintentionally become the main masking instance. That explains why some children with intensive support collapse completely after school.
Neuroaffirmative support is therefore structural: it creates conditions under which the autistic nervous system can use its own capacities. Not less school. A different relationship between the child's brain and its environment.
Sources
- German Federal Ministry for Education, Family, Senior Citizens, Women and Youth. Draft act to strengthen children, adolescents and their families (KJHSRG). 23 March 2026.
- Lebenshilfe Federal Association. Statement on the KJHSRG draft: risks of infrastructural educational assistance. 2026.
- BAGFW (Federal Association of Free Welfare Care). Statement on KJHSRG: focus on administrative simplification rather than individual legal entitlement. 2026.
- German Social Code Book IX (SGB IX), in particular § 112 (participation in education).
- German Social Code Book VIII (SGB VIII), in particular § 35a (integration support for children with mental disabilities).
- UN Convention on the Rights of Persons with Disabilities, Article 24 (right to inclusive education).
- gegen-hartz.de: background reporting on the planned 2026 school-aide reform.
One bright spot
The reform debate runs largely at the administrative level. What runs in the background is a different question: whether an education system actually accounts for the autistic mechanisms described here. A law alone does not answer that question. It is answered when school support is understood as a regulation anchor rather than a control instance, and when pool models are used where they fit, not everywhere they are cheaper.
Parents who know the mechanisms can substantiate applications more precisely. Aides who understand predictive coding and monotropism make different daily decisions. Both are effective regardless of the funding model.
Autistic Mirror explains autistic neurology individually, in relation to your situation. Whether for yourself, as a parent or as a school aide.
