Autism Spectrum Disorder

Case: Luca, 6. Five chapters follow: the case, epidemiology, presentation, functional neuroscience, and treatment.

The case: Luca, 6

Presenting concern

Luca is six and in kindergarten. His pediatrician's referral reads: "Teacher requests evaluation. Does not respond to his name, plays alone with toy buses, covers his ears, leaves circle time, 'tantrums' several times a week. Parents describe an affectionate, talkative boy at home. M-CHAT-R at 18 and 24 months low risk. Speech mildly late, now fluent. Hearing normal at three. Please evaluate for autism and ADHD."

Luca, asked about school: "Circle is too loud and too long. I want to play buses with the boys but they don't do the stops." Asked about the fire drill: "The bell goes in my ears and then my whole body is loud."

Developmental and family history

Born at term after an uncomplicated pregnancy; his mother took an antibiotic for a urinary infection in the second trimester and asks whether that "did something." Single words came at 15 months and phrases at two years and eight months. From about 20 months he pointed at trucks and looked back to check that his parents were looking. From age two he lined up shoes by the door and was upset if the order changed. His preschool teacher called him "bright but in his own world."

Tags and seams have bothered him since toddlerhood; dressing takes forty minutes on school days. He eats about ten foods and refuses mixed textures. He knows every bus route in the city by number and street, lines his toy buses in route order, and narrates the stops.

His father repairs city buses and brought home the route maps that started the interest. Spanish and English are spoken at home; Luca speaks mostly English. His father's older brother "was always in his own world"; he has never been assessed. His maternal grandmother has been treated for anxiety. The parents hesitated for a year. His mother: "Everything I see online is children who don't talk or hug. He does both."

Observations in session

Luca walked past the clinician without greeting, found two toy buses, and set them in a line. When she said, "Look, this one has a wheel missing," he glanced at her face, then at the wheel, and said, "Route 14 has the bendy ones." A moment later he held a bus up toward her with a brief glance. His name was called three times before he answered while arranging the buses; when the clinician sat at his level and waited for a pause, he answered on the first call.

His speech was fluent, with a sing-song cadence. He looked at the clinician more while telling than while listening. When she made a bus into a boat, he said, "No. It's a bus," and turned it back. Asked how a classmate feels when Luca tells him the routes, he said, "Happy," without looking up.

When his mother cried, he looked up, carried the tissue box to her, and returned to the buses; what he said and did afterward did not change. When the clinic phone rang he pressed his hands to his ears and rocked until it stopped. When the two adults talked over each other, he played on and responded to neither. He flapped his hands each time a bus door clicked open. At the door he asked, "Are you coming to my school?"

Collateral

The teacher's log records two "tantrums" in one week. In the first, a classmate knocked over his bus line; he screamed, hit the table, and could not resume for twenty minutes. In the second, during the fire drill, he dropped to the floor, hands over ears, and rocked until it stopped. He does not answer to his name in the cafeteria or on the carpet; in the library he answers. He runs in corridors, leaves circle time mid-story, and calls out answers. One to one with the reading specialist he stays twenty-five minutes. At recess he narrates routes to two boys, who drift away. One boy, Jaden, stays and follows Luca's directions. Luca talks about Jaden at home.

WISC-V testing gave an uneven profile: verbal comprehension borderline, visual-spatial and fluid reasoning average, working memory low average, processing speed low. Vineland-3 standard scores were 78 for communication, 74 for daily living, and 70 for socialization. An ADOS-2 (Module 3) gave a high comparison score in both domains. His parents asked "what this means for the parts of him that are fine."

Epidemiology

How common

A 2022 meta-analysis of 74 studies covering 30,212,757 participants, published between 2008 and July 2021, estimated the world prevalence of autism spectrum disorder at 0.6% (95% CI 0.4–1.0) [67]. Estimates differed by continent: 0.4% in Asia, 1.0% in the Americas, 0.5% in Europe, 1.0% in Africa and 1.7% in Australia, each with a wide confidence interval [67]. Two features of the abstract matter for reading these numbers. Its stated aim is the prevalence of ASD "in high-risk groups by continent," so the pooled figure may not describe unselected populations, and the age ranges and sex ratios of the included samples are not reported in the abstract [67]. The authors describe prevalence as increasing without giving a trend figure, and note that the diagnosis is usually made below the age of three [67].

For a child like Luca, assessed at six in kindergarten (the case does not name a country), the continental figure that would apply if his setting is in the Americas is the 1.0% estimate, roughly one child in a hundred. The sources retrieved for this book do not supply a US surveillance figure, and the meta-analysis should not be read as one.

Who is affected, and how severity is rated

Neither the meta-analysis nor the other retrieved sources state a sex ratio, so this chapter does not give one. What the sources do describe is how the DSM-5 severity rating behaves in practice. Among 248 children evaluated at six Autism Treatment Network sites, higher severity ratings in both the social-communication and the restricted-repetitive-behavior domains were associated with younger age, lower IQ and greater domain-specific symptom severity on the ADOS-2, and were not associated with emotional or behavioral problems [74]. The authors suggest that clinicians may be folding cognitive functioning into the severity determination [74]. For epidemiology this means that a distribution of "Level 2" and "Level 3" ratings in a clinic partly reflects the IQ distribution of the children it sees.

Risk and heritability

Reviews of the genomics describe more than a thousand ASD-associated genes involved in brain development, identified through sequencing, epigenetic and model-system strategies [69]. That count describes the breadth of genetic contribution, not the share of children in whom a single variant explains the diagnosis, which the retrieved abstract does not give.

Prenatal exposures have been examined in a small case-control study of 215 mothers of children aged 0 to 10 (107 with ASD, 108 without), which found univariate associations with having a sibling or other family member with ASD, maternal medication use during pregnancy, smoking during pregnancy and the gestational age at which prenatal vitamins were started; in multivariable analysis, medication use and timing of prenatal vitamins remained associated, and the authors call the results preliminary and note the specific medications could not be identified with the sample size [68]. The family-history associations in that study are consistent with the genetic literature; the exposure associations are the kind of finding the ADHD chapter's review would expect to shrink in family-controlled designs [56].

Early identification and its trade-offs

A review of early diagnosis lists the screening instruments in common use (M-CHAT-R/F, SCQ, PEDS, CARS) and summarizes the case for early diagnosis as improved developmental and adaptive outcomes and reduced parental stress, set against the risks of labeling and stigma, over- or misdiagnosis, and a lengthy diagnostic process [70]. Luca's history illustrates the limit of screening: his M-CHAT-R was low-risk at 18 and 24 months, and the referral came from a teacher at six.

The public context

Public discourse shapes who is referred and how families receive a diagnosis. A content analysis of YouTube videos and comments found the dominant theme to be educational information about ASD characteristics, with mixed sentiment and two recurrent stigmatizing ideas: that people with ASD cannot understand emotion, and that ASD is a single, uniformly severe condition [71]. Among 44 adolescents with ASD compared with 56 clinical controls, social media use was associated with higher friendship quality, and the association was moderated by anxiety; no such association appeared among controls [72]. In self-disclosed adults with ASD, 84% used social media, and happiness rose with use up to a point and then fell [73]; that is adult data and is cited here only for the shape of the relationship.

What this case cannot establish

Luca's diagnosis and severity rating are one clinic's judgment about one child. The rating does not measure a fixed property; the studies above suggest it moves with age and IQ, and Luca's uneven WISC-V profile (verbal comprehension borderline, visual-spatial and fluid reasoning average) could pull it either way.

Source status

#67, #68, #69, #70, #71, #73, #74 from published abstracts; #72 from the publisher's abstract page; #56 from the publisher page. No source in the list gives a US surveillance prevalence or a sex ratio, and none was added.

Figures for this chapter

Presentation

Interest, inference, and coordination as three observations

"He plays alone" is the teacher's summary. It compresses three things that are separately visible in Luca. Social interest is present: he wants to play buses with the boys, talks about Jaden at home, carries the tissue box to his mother, and asks whether the clinician is coming to his school. Social inference is limited: he does not notice the two boys drifting away, and he answers "Happy" for Jaden without looking at him. Flexible coordination is limited: a bus cannot become a boat, a knocked-over line cannot be rebuilt for twenty minutes, and the shoes by the door have had one order since he was two. Keeping these apart, and asking of each behavior what Luca perceived, what it meant to him, and how he participated, is what allows later chapters to say what treatment acts on.

Mapping the pattern to the criteria

Autism spectrum disorder is defined by persistent deficits in social communication and interaction (social-emotional reciprocity; nonverbal communication; developing, maintaining, and understanding relationships) together with at least two of four restricted, repetitive features (stereotyped movements or speech; insistence on sameness; highly restricted interests; sensory hyper- or hypo-reactivity). These must be present early, impairing, and not better explained by intellectual disability, although autism and intellectual disability can be diagnosed together when social communication is below that expected for general developmental level [111].

Reciprocity in Luca is present in some forms and absent in others. He responds to overt distress with a learned act of care. He initiates sharing: at 20 months he pointed and checked back, and in session he held a bus up with a glance. He does not sustain an exchange whose content is not his own, and he does not track whether his partner is still there. He looks at the clinician more while telling than while listening, and limited eye contact while listening does not tell us whether he is listening. Jaden is a friend. The friendship works because Jaden follows Luca's directions, which describes what it requires rather than judging its value.

All four restricted and repetitive features are visible, and each needs describing by function rather than form. Flapping at the click of a bus door and rocking during the fire drill are both repetitive movement; one accompanies pleasure, the other accompanies an aversive input and stops when it stops. Insistence on sameness appears in the shoes, the route order, the refused boat, and the knocked-over line. The bus interest dominates what Luca brings to any exchange; it is also a family interest, real knowledge, and the reason Jaden plays with him. Sensory features are prominent: the alarm and the phone, tags and seams, mixed textures.

The alarm and the two adults talking are the pair most easily merged. The alarm was perceived as painful and escaped. Competing speech interfered with selecting one speaker and changed nothing in his play; nothing suggests it hurt. Calling both "sensory processing" would lose the detail on which the school plan depends.

Five events behind two log entries

Observation	What Luca perceives	What it means to him	What he does	What it points to
Name called in the cafeteria	His name among overlapping voices	Nothing distinguishable arrives	No response	Competing input; attentional selection under load
Name called while arranging buses, quiet room	His name, audibly	An interruption of an absorbing plan	Answers on the third call	Absorbed goal outcompeting social input
Name called in the library, or after a pause	His name, nothing competing	An address he can answer	Answers first time	Capacity present; access depends on conditions
Bus line knocked over	An ordered arrangement destroyed	A plan he cannot restore	Screams, hits the table, twenty minutes to resume	Insistence on sameness; limited flexible coordination
Fire drill	Sound at painful intensity	Something to escape	Drops, covers ears, rocks until it stops	Sensory hyper-reactivity; rocking as regulation

The teacher's log records two kinds of event; the table records five. Later chapters use them as five.

Severity ratings

The diagnosis carries a severity level for each domain, from "requiring support" to "requiring very substantial support," and specifiers including accompanying intellectual impairment, language impairment, and a known medical or genetic condition [111]. The level describes support needed now, not prognosis.

In 248 children evaluated at six Autism Treatment Network sites, higher severity ratings in both domains were associated with younger age, lower IQ, and greater ADOS-2 domain-specific severity, and were not associated with emotional or behavioral problems; the strength of the IQ association led the authors to suggest that clinicians fold cognitive functioning into the rating [74]. For Luca this cuts two ways. Fluent speech and average nonverbal reasoning could pull a rating down; borderline verbal comprehension could pull it up. Neither is what the rating describes: what his social communication and his repetitive behavior require this year, rated separately and revisited. The two tantrums are behavioral problems; in that sample they did not track the rating, and a rating is not a measure of how hard the week was.

Co-occurring cognitive limitation

Luca's WISC-V [83] profile is uneven: verbal comprehension borderline, visual-spatial and fluid reasoning average, working memory low average, processing speed low. A single composite does not convey these differences. The verbal score raises a question the case cannot settle: a broad cognitive limitation, a specific language difficulty, an effect of a bilingual home tested in one language, or a combination. The specifier "with accompanying intellectual impairment" requires deficits in both intellectual and adaptive functioning [111]. At six, with this profile, the question is recorded as open and retested.

Adaptive functioning is a different measurement. The Vineland-3 [85] places communication, daily living, and socialization below average, socialization lowest. Daily living at 74 includes the forty-minute dressing routine, and dressing depends on tags and seams, a learned sequence, the clothing available, and what help is present, as well as on cognitive ability. The scores set a baseline; they do not say which component is limiting.

Co-occurring ADHD

Both diagnoses may be given [111]. The question is whether Luca's inattention, restlessness, and impulsivity exceed what autism and the setting explain. In favor: he runs in corridors, calls out, and leaves circle time. Unresolved: circle time carries competing voices, sustained comprehension in his weakest domain, and a duration he calls "too long." One to one with the reading specialist he stays twenty-five minutes; with buses, longer. Interest improves access to sustained activity without removing every difficulty regulating it, and ADHD does not require that attention fail uniformly. The answer is deferred until supports are in place and teacher and parent ratings are collected under supported conditions. The severity rating is not a proxy; it was unrelated to behavioral problems in the sample above [74].

Why diagnosis came at six

Early screening instruments include the M-CHAT-R/F, the SCQ, the PEDS, and the CARS [70]. Luca screened low risk at 18 and 24 months, consistent with what he did at 20 months: he pointed and checked back. A screen samples a few behaviors at one age, and a child with early sharing, fluent speech, and affection for his family is the child a screen most easily passes. "Bright but in his own world" at four was not followed up.

Early diagnosis and intervention are associated with better developmental outcomes and adaptive skills and with reduced parental stress; the risks named are labeling and stigma, overdiagnosis or misdiagnosis, and a lengthy diagnostic process [70]. Luca's family met the stigma before the diagnosis. A content analysis of YouTube discourse on autism found the themes his mother describes: autistic people as unable to understand emotion, and autism as a monolithic, only-severe condition, with most viewer comments anecdote [71]. The tissue box answers the first. The uneven profile answers the second.

Differentials and what else to ask

A language disorder would not produce the alarm response or the collapse at the line, though comprehension needs its own assessment. Social (pragmatic) communication disorder is not diagnosed when restricted and repetitive features are present, currently or by history [111]. ADHD alone does not explain the line, the boat, or the sensory pattern. Anxiety is a live question: his grandmother has been treated for anxiety, distress at change can be anxious, and the two diagnoses are not exclusive; for now his distress follows specific violations rather than anticipated ones. Sleep, bowel habit on a ten-food diet, and episodes suggesting seizures are asked about.

What this case cannot establish

One assessment cannot establish whether Luca's borderline verbal comprehension is a cognitive limitation, a language disorder, or an artifact of testing a bilingual child in one language; whether his inattention at circle time would persist in a quieter room with a shorter story; whether the anxiety in his family history already contributes to his distress at change; or whether Jaden's friendship will hold when Jaden wants a turn. Every label in the table was inferred from what he did under particular conditions, and a behavior that fits a label does not demonstrate the mechanism the label names.

Figures for this chapter

Functional neuroscience

Weighting the world, modeling other people, changing the plan

Three things vary across Luca's case: how strongly incoming sensory input is weighted, how the unobserved states of other people are inferred and used, and how an active plan changes when the world changes. Autism is a label for a pattern that spans all three. Each construct below is anchored to something Luca did, not to the diagnosis, and the account is what treatment acts on.

Functional layer

Social orienting and joint attention

Social orienting is a Process: the allocation of processing to social input (a face, a voice, one's own name) among competing inputs. Joint attention is a Process that runs on top of it: coordinating one's own attention with a partner's on a third thing. Responding to joint attention means following a partner's gaze or point; initiating it means showing or pointing to share.

Luca initiates. At 20 months he pointed and looked back to check that his parents were looking; in session he held a bus up with a glance. He also responds: "Look, this one has a wheel missing" produced a glance at the clinician's face, then at the wheel, then a return to his own topic. The response was present and brief, and the shared object did not become a shared exchange.

The three name events describe social orienting under three conditions. In the cafeteria his name arrived among overlapping voices; attentional selection, selecting information from competing inputs, and distractor suppression, reducing the processing of irrelevant input, were strained by the room before any social operation was engaged. In the quiet room his name arrived clearly while he was arranging buses; a strongly maintained goal outcompeted a social input, and the third call won. In the library, and after the clinician waited for a pause, he answered at once. The capacity is present; what varies is access. None of the three events shows that social input carries no weight for him; together they show the conditions under which it wins.

Mental-state inference and the representation it runs on

Mental-state inference is a Process: inferring a specified unobserved state of another person from behavior, context, and represented information. It operates on a Representation: a model of the partner, holding what the partner wants, knows, and feels. The process and the representation can be limited separately.

When his mother cried, Luca detected a displayed state and produced a learned response. What he said and did afterward did not change; the detected state did not update his plan. Asked how Jaden feels, he said "Happy" without looking: supplied from a default, not inferred from Jaden. When the two boys drifted away, nothing in his narration changed, because their leaving was not represented as information about what they wanted. Detecting an overt state, inferring an undisplayed one, holding one's own plan, and changing the shared activity impose different demands. Luca's affection and his difficulty coordinating play coexist without contradiction, as they do in the Intellectual Disability module (Maya).

The partner contributes to the task. Jaden follows Luca's directions, so coordination succeeds without Luca inferring anything. The other boys expected turns, made no explicit request, and left. Milton's double empathy problem frames misunderstanding between autistic and non-autistic people as mutual [321]. Crompton and colleagues found greater loss of story detail in mixed autistic and non-autistic communication chains than in chains of either group alone, in 72 adults with IQs in the typical range [322]; applying an adult experiment about partner composition to a six-year-old requires further evidence. For Luca the usable question is how each partner recognizes and responds to the information the other offers.

Sensory precision as a State-like parameter

Sensory precision, or gain, is how strongly an input in a given channel is weighted in perception and in driving a response. It behaves like a State: it varies with arousal, fatigue, context, and expectation, and it changes access to capacities rather than being one. The phone ring and the fire alarm were weighted so heavily that they were aversive in their own right. Tags and seams are the same phenomenon in a tactile channel; mixed food textures may be a third.

The two adults talking over each other are not this phenomenon. Luca responded to neither, and nothing suggests it hurt; that was competing input interfering with selection. Reducing an aversive sound improves comfort even when performance was already accurate; removing competing speech improves access to an instruction. Sensory-integration capacity, combining cues about a common feature, is a further construct, and Luca's aversions do not demonstrate a deficit in it. High gain in one channel does not establish high gain in all: the click of a bus door is a sound he seeks out.

What predictive-processing accounts explain, and what they do not

Predictive-processing accounts describe perception as the comparison of incoming input with predictions from an internal model, with prediction errors weighted by their expected precision. Proposals in autism hold that sensory prediction errors are weighted too highly relative to prior expectations, or that priors are too weak or too inflexible, so that input is experienced with less contextual smoothing, unexpected change is costly, and predictable environments are preferred [112, 113].

Applied to Luca, this gives one vocabulary for several observations. The alarm is input with nothing to dampen it. The knocked-over line is a large error against a precise expectation. Bus routes are a maximally predictable domain. "No. It's a bus" is a refusal to let a pretend frame override sensory evidence. That is the account's strength: it connects sensory reactivity, insistence on sameness, and restricted interest under one parameter.

It does not explain why social input should be affected in particular; the suggestion that social signals are inherently noisy is a proposal, not a finding. It does not specify how or where precision is set in a six-year-old's brain. The parameter cannot be measured in Luca, and more than one setting could produce the same behavior. It is a hypothesis about a variable, not a finding about him.

Restricted and repetitive behavior: regulation, preference, habit

Insistence on sameness can serve different functions across autistic children. Repetition can be enjoyable, regulating, or an expression of preference, and it can also be habit: a learned sequence that runs with little cost because it has run many times. Luca shows each. Flapping at the click of a bus door accompanies visible pleasure; it is expression, and needs nothing done about it. Rocking during the fire drill occurred under aversive input and stopped when the input stopped; it is regulation, an action that changes state. The shoe order is closest to habit: a sequence that needs no reconstruction, enjoyed, and costly only when broken. The route order is preference as well as habit, since he chooses it and narrates it. The bus line is a held Representation of an ordered arrangement; restoring it required task switching, from "maintain the line" to "rebuild it," which he could not do for twenty minutes. The functions overlap, and which is dominant is found by changing conditions: remove the input and see whether the rocking stops; change the order and see what the distress is about. Which function a behavior serves decides what an adult does: leave it, reduce the input, or teach a repair.

Attention, working memory, and the ADHD question

Circle time asks Luca to keep the goal "listen for the story" active against voices, movement, and his own interests, with comprehension in his weakest domain. Working-memory maintenance preserves the goal; attentional selection acts on it. In a small experiment, Kofler and colleagues found that observed inattention in children with ADHD varied with working-memory demands [320], a reason to specify what Luca must hold at circle time before calling what he does there inattention. One to one he stays twenty-five minutes; with buses, longer. Interest improves access to sustained activity without removing every difficulty regulating it. ADHD describes a developmental pattern of impairment; it does not require that attention fail uniformly across activities or that every failure originate in one operation.

Construct	Kind	What it names in Luca
Social orienting	Process	Which call to his name wins, and under what conditions
Joint attention	Process	The bus held up with a glance; the glance at the wheel
Mental-state inference	Process	"Happy" without looking; not noticing the boys leave
Model of the partner	Representation	What he holds about Jaden: someone who follows
Sensory precision	State-like parameter	The alarm, the phone, tags and seams
Attentional selection	Process	The cafeteria; two adults talking
Task switching	Process	From "maintain the line" to "rebuild it"

Circuits, development, and genes

Systems for social attention and sensory weighting

These operations are implemented by interacting systems. In a prospective eye-tracking study, eye fixation in infants later diagnosed with autism was initially typical and declined between 2 and 6 months of age [114]. Face and biological-motion processing recruit fusiform, superior temporal, and amygdala regions, with group differences reported in autism that vary with task, sample, and developmental stage [115]. Two separate accounts address sensory reactivity: one proposes an increased ratio of excitation to inhibition in key neural systems, and the predictive-processing account proposes altered precision, or gain, in weighting prediction errors [116, 113]. Neither establishes the mechanism of an individual child's sensory reactivity. Each is a group finding or a proposal; none was measured in Luca.

Development

On average, behavioral signs in prospective infant-sibling cohorts emerge during the second year, after a period in which social attention looks typical, though timing varies from child to child [117]. Luca's pointing at 20 months and his low-risk screens are consistent with gradual divergence, but a retrospective history cannot date it. A subset of children show early brain overgrowth on imaging [118]; nothing in the case says whether Luca is among them.

Genes and exposures

More than a thousand genes associated with autism have been identified, many involved in brain development [69]. Satterstrom and colleagues' exome study linked rare variation to genes involved in developmental regulation and neuronal function [308]; Fu and colleagues then showed overlapping and differing associations between autism and developmental-delay cohorts [309]. Variant effects cross clinical boundaries, and similar functional difficulties can arise through different developmental routes. The uncle "in his own world" is consistent with familial liability and establishes nothing about a shared variant.

Luca's mother asks about the antibiotic. A case-control telephone survey of 107 mothers of children with autism and 108 without found univariate associations with a family member with autism, maternal medication use in pregnancy, smoking in pregnancy, and gestational age at the start of prenatal vitamins; in multivariate analysis, medication use (not delineable by drug in that sample) and vitamin timing remained; the authors call the results preliminary [68]. The family-history associations are what genetic liability would produce. A preliminary association between "any medication" and diagnosis cannot be applied to one antibiotic in one pregnancy, and it is not a basis for the answer his mother fears.

Moving between levels of evidence

A task experiment tests specified operations under controlled conditions. A developmental group comparison describes a distribution across children. A treatment trial estimates benefits and harms on its measured outcomes. Connecting these findings to one child's functioning requires an additional inference. Neither an eye-tracking trajectory nor a risk association by itself identifies Luca's functional difficulty or the support that will help him.

What this case cannot establish

The case cannot establish whether Luca's social orienting is reduced or simply outcompeted; whether his mental-state inference is limited by the process, by the representation of the partner, or by attention to the partner's face; whether sensory precision is high in several channels or only the ones he has complained about; whether a predictive-processing parameter, if it can be measured, differs in him; or whether his inattention at circle time belongs to a second disorder. Each construct was inferred from what he did under particular conditions. None was measured, and a behavior that fits a construct does not demonstrate the mechanism it is named after.

Figures for this chapter

Treatment

What Luca and his family want

Luca wants to play buses with the boys and wants circle time shorter and quieter. His parents want mornings that do not take forty minutes, a school that understands him, and an answer to "what this means for the parts of him that are fine." His teacher wants fewer collapses and a child who stays on the carpet. These are different goals from "reduce autism symptoms," and each can be counted: minutes of joint play at recess, minutes from waking to dressed, transitions completed, alarms survived. Participation is what matters to him, and it can improve before any score changes.

Neurodiversity, in the sense the Intellectual Disability module (Maya) uses it, directs attention to the value of Luca's interests, his ways of communicating, and the relationships he wants. His disability, need for teaching, and need for support remain part of the same account.

Does the support make knowledge available or supply the decision?

The central distinction from Maya's care applies directly. A picture of the bus stops and a card that says whose turn it is make coordination available: Luca uses what he knows about routes while the card carries the information he does not infer. An aide who tells him what to say next supplies the decision. Both can be useful; they lead to different next steps. The question is asked of every support in the plan: which operation does it assist, and does Luca's contribution to the success grow?

What the diagnosis makes available

Diagnosis is a label for the pattern; the functional account is what the plan acts on. Early diagnosis and intervention are associated with better developmental outcomes and adaptive skills and with reduced parental stress; the risks named are labeling and stigma, over- or misdiagnosis, and a lengthy process [70]. At six the same trade applies: school eligibility, a shared explanation, and services, against a label his parents have seen used for a condition described online as monolithic and only severe [71]. The plan makes the label specific.

Severity is rated separately for each domain, by the support needed this year, and revisited. In a large clinical sample, ratings tracked age, IQ, and observed symptom severity and not emotional or behavioral problems [74]; Luca's rating is not raised by his tantrums or lowered by his fluent speech, and it is not a prognosis.

Social communication: the dyad and the playground

In the PACT trial, a preschool-onset intervention worked through parent-child communication; long-term follow-up found benefits on the trajectory of measured autism symptoms and on interaction, while the language composite showed no clear group difference [303]. Luca is older than the PACT children, and applying its findings to him is an inference; the specificity of what changed is the lesson. A randomized adaptive trial extended the same principle to minimally verbal children aged 5 to 8 [119]. It delivered a joint-attention and play intervention (JASP) with enhanced milieu teaching, with or without a speech-generating device.

For the playground, coordination is made concrete: a shared picture of the route, a visible division of the buses, a short turn-taking sequence, and a reliable way to ask for a break. The other boys are taught the game with stops, because coordination is mutual and the double empathy account puts their expectations in the explanation. Whether a sequence learned with Jaden transfers to the others is observed, not assumed. A structured social-skills group is considered later. A meta-analysis of group programs for youth found benefits that varied substantially by reporter and outcome, and how far they apply at six is uncertain [120].

Adaptive skills

The Vineland-3 [85] is the baseline and the outcome measure. Applied behavior analysis has review-level evidence for adaptive living skills in people with autism [86]; that review does not settle what intensity Luca needs or whether the same methods suit his social communication.

Dressing is separated into its components. Tags and seams are sensory: seamless socks and tagless shirts remove the input. The sequence is learning: a visible order of steps moves part of the burden from internal maintenance to perception, as Maya's schedule does, with its own reading cost, so it starts short. The negotiation is routine: the same order every day.

Feeding is reviewed with growth: ten foods and refused textures warrant a dietary history and weight monitoring [121].

Sensory features: two problems, two responses

The alarm and the cafeteria get different plans because they are different events. For the alarm and the phone, the input is reduced: advance warning of drills, ear defenders, a known exit, and an adult who does not require him to stop rocking. That improves comfort. For circle time and the cafeteria, access is improved: a seat near the teacher, a shorter circle with a visible agenda, and his name followed by a pause before the instruction. Evidence for Ayres sensory integration therapy as a standalone treatment is mixed: a small manualized trial was positive, while the larger SenITA trial did not improve its primary outcome over usual care [122, 123, 124]. The environmental changes here are sensory-based accommodations, not that therapy.

Restricted and repetitive behavior: function first

Flapping at the click of a bus door needs nothing done. Rocking at the alarm is regulation; the plan reduces the alarm, not the rocking. The knocked-over line is rehearsed: a photograph of the finished line, a place to store it, and a short repair script ("line broken, rebuild from Route 1") that turns a catastrophe into a practiced event. The routines the family already has are kept.

ADHD: supports first, then decide

Medication for ADHD symptoms is considered when inattention and impulsivity continue to interfere across appropriately supported activities. For Luca the supports are not yet in place, so the question is deferred and rating scales are collected once they are. If impairment persists, the evidence the Intellectual Disability module (Maya) reviews applies: methylphenidate in the RUPP crossover trial [314]; extended-release guanfacine, which improved hyperactivity ratings without a treatment-group difference on a brief cognitive battery [315]; and atomoxetine, which with parent training improved ADHD symptoms relative to placebo [316]. Their response percentages describe different populations and definitions and cannot rank expected benefit for one child. The order is fixed: change the room, then measure, then decide.

His tantrums have identified triggers and are not "irritability associated with autism" in the sense that warrants an antipsychotic. Two atypical antipsychotics, risperidone (ages 5 to 16) and aripiprazole (ages 6 to 17), carry US labeling for that indication rather than for core social-communication features [125, 126], and neither is indicated here.

Medical and genetic evaluation

Hearing is rechecked if concern recurs. Sleep, bowel habit on a narrow diet, and any episodes suggesting seizures are asked about, and new deterioration gets its own medical assessment rather than attribution to the diagnosis. Genetic evaluation is offered. The American Academy of Pediatrics' 2020 clinical report recommends chromosomal microarray and fragile X testing in autism [121]. The separate 2021 ACMG guideline supporting exome or genome sequencing applies to defined presentations of developmental delay, intellectual disability, or congenital anomaly [312]. More than a thousand genes have been associated with autism [69]. A molecular result can inform surveillance and counseling; it answers a different question from which support helps him at recess. His mother's antibiotic is addressed directly: the study she may have read is preliminary by its authors' account [68].

Looking ahead: friendship, media, anxiety

His parents ask about the future. In 44 adolescents with autism, social media use was associated with higher friendship quality, moderated by anxiety, with no such association in clinical comparison adolescents [72]. In adults with autism, users of one platform were happier than non-users [73]; that is adult data and says nothing about a six-year-old. Anxiety disorders are the most common class of disorder in adolescents [53], Luca's grandmother has been treated for one, and his distress at change could become anticipatory; it is asked about at every review. For now the friendship is Jaden and the bus game, and it is protected.

School

The plan is written into an individualized education program. The teacher's log is changed to record what the Presentation chapter separated: an exit during an alarm from a collapse over a broken arrangement, and a name unanswered in noise from one unanswered in absorption. Fire drills are announced to him. The reading specialist's one-to-one time is kept, because it is where he shows what he can sustain.

Measuring what changes

Joint play minutes, morning duration, transitions completed without collapse, alarms managed, and the separated log are tracked separately. The Vineland-3 is repeated at a year and the severity ratings revisited. Luca's own criteria, whether the boys "do the stops" and whether circle is "too loud," are asked each visit. A support that improves access can remain if Luca finds it useful; independence is one legitimate outcome and a continuing aid is another.

What this case cannot establish

The case cannot establish which component of the plan will help Luca, whether he will need medication for ADHD symptoms, whether a turn-taking sequence learned with Jaden will transfer to anyone else, or whether the borderline verbal comprehension will resolve into a language disorder, a broader limitation, or neither. If mornings shorten, the case cannot show whether seamless socks, the picture sequence, or practice did it. A response to a support does not demonstrate the operation it was designed around, and an absence of response does not show the account was wrong. The account offers a way to arrange support so that Luca's contribution to each success can be seen.

---