Psychotic Disorders

Case: Ravi, 16. Five chapters follow: the case, epidemiology, presentation, functional neuroscience, and treatment.

The case: Ravi, 16

Presenting concern

Ravi is sixteen and in eleventh grade. His school counselor wrote to the clinic: "Ravi has become a different student this year. He stopped sitting with his friends in the fall, he has been found staring at the wall during class, and last week he told his chemistry teacher that the fire alarm was 'a signal.' Two teachers think he is hearing things. We would like him evaluated before we decide about placement." His pediatrician added a six-month decline in grades, an eight-pound weight loss, normal physical and neurological examinations and normal screening laboratory results, and asked for "psychiatric assessment of possible early psychosis."

Ravi's mother described the change at home: "He talks in his room at night. I stand outside the door and I can't tell if he's on the phone or not. He doesn't shower unless I tell him. Three weeks ago he asked me whether the neighbor's new doorbell camera was pointed at our house on purpose."

Ravi, seen alone, said: "Sometimes I hear my name when nobody said it. Mostly when it's quiet, like at night or in the library. It's outside my head, but I know no one's there. And there's this feeling in class that everyone knows something about me that I don't." Asked about the fire alarm: "It went off right when I was thinking about leaving. That's not nothing." He added, "I'm not hearing voices like in movies. It's more like thoughts that aren't in my voice." He described his mood as "flat, not sad exactly," with more bad days than good since the summer. He reported smoking cannabis "a few times a month" since tenth grade, most recently about five weeks before the visit, and no other substances. He denied wanting to die or to hurt anyone.

Developmental and family history

Ravi was born at term after an uncomplicated delivery. His mother recalled a second-trimester urinary tract infection treated with oral antibiotics, and no other illness during the pregnancy. He walked at fourteen months and said his first words at about twenty months; he used phrases by three and received speech therapy in preschool. Through elementary school he preferred one or two friends to groups, found the cafeteria overwhelming, and narrated aloud while building models. A third-grade teacher asked whether he might be autistic; no evaluation was pursued. His grades were average until this year. In ninth grade a group chat circulated photographs of him with mocking captions; he changed friend groups and, according to his mother, "went quiet for a month and then seemed fine." A maternal uncle was hospitalized in his twenties for what the family calls "a breakdown"; the details are unknown. Ravi's paternal grandmother was treated for depression. His father lives in another state and is in occasional contact.

Observations in session

Ravi sat with his hood up and answered after long pauses. His speech was reduced in amount but coherent and on topic. Twice he stopped mid-sentence and looked toward the corridor. The first time, a door had slammed; he said "sorry, what?" and resumed. The second time nothing audible had occurred; he stared for several seconds and then said he had "lost the thread." Asked about his 3D printer, he explained a recent print failure in organized, detailed sentences and smiled. Asked whether other people could hear his thoughts, he said, "I don't think so. But it feels like they might." His affect was flat for most of the interview; he laughed once, at a joke about the printer. Apart from the second pause, he did not appear to respond to anything the clinician could not see or hear.

Collateral

His mother reported that Ravi has been sleeping four to five hours a night since the summer and eating less. She also said that talking aloud while building "is something he has always done," and that she cannot say whether the nighttime speech is different. The school reported falling attendance and two abrupt exits from class; his chemistry teacher said he had asked whether alarm tests were "scheduled around him." His closest friend told Ravi's mother that Ravi stopped answering messages about two months ago.

Epidemiology

Symptoms are common; disorders are rare

The first epidemiological fact about psychosis in young people is that psychotic symptoms and psychotic disorders have prevalences an order of magnitude apart. A primer for clinicians summarizes the symptom literature: a meta-analytic prevalence of psychotic symptoms of 5% to 8% in the general population, 14% of 11-year-olds endorsing psychotic symptoms on interview, 9% to 14% of adolescents in interview-based studies and more than 25% in some self-report studies [87]. Those symptoms are not benign as a group: they were associated with a 5- to 16-fold increased rate of psychotic illness in early adulthood [87]. The disorder is another matter. Schizophrenia has a worldwide prevalence of 0.46%, and childhood-onset schizophrenia an incidence of less than 1 in 10,000 children [87]. A second review gives the prevalence of schizophrenia as less than 1 per 10,000 to 50,000 in children and early adolescents and 1 per 500 to 10,000 in mid-adolescents, a steep rise across a few years of age, although the two ranges meet at 1 per 10,000 [89].

Ravi is 16. At his age the disorder has moved from vanishingly rare to merely uncommon, and the primer's figures for the adult illness apply to his future: up to 20% of adults with schizophrenia became ill before 18, onset is generally between 14 and 35 with half diagnosed before 25, and up to 30% make a lifetime suicide attempt [87].

What a psychotic presentation turns out to be

Among 215 children with psychosis referred to the NIMH childhood-onset schizophrenia study, 30% received a diagnosis of schizophrenia, 4.6% schizoaffective disorder, 12% major depression with psychotic features and 15.2% bipolar disorder [89]. Psychotic features occur in 16% to 87.5% of youth with bipolar disorder depending on the sample, more often in adolescent-onset than childhood-onset cases, and in about 4% of youth with major depressive disorder [89]. Among young people identified as ultra-high-risk, on average 76% do not transition to psychosis over 6 to 40 months of follow-up [89]. For the differential in Ravi's case these numbers carry a single message: in a referred adolescent with psychotic symptoms, a primary psychotic disorder is one of several outcomes, and even schizophrenia, the most common named diagnosis in the NIMH series, accounted for fewer than a third of referrals.

Substances

Substance abuse in first-episode psychosis ranges from about 25% to more than 60% across studies, and psychotic symptoms are twice as common in young cannabis users [89]. Ravi's occasional cannabis use is therefore neither an explanation nor a coincidence to be dismissed; it is a common companion whose contribution the presentation chapter leaves open.

Prenatal exposure

In a prospective US cohort of 15,421 pregnancies from 1959 to 1966, followed for about 40 years and yielding 116 offspring with confirmed psychotic disorders, maternal bacterial infection during pregnancy was associated with offspring psychotic disorder with an adjusted odds ratio of 1.8 (95% CI 1.2–2.7); multisystemic infection carried an odds ratio of 2.9 (1.3–5.9) versus 1.6 (1.1–2.3) for localized infection, and male offspring were significantly more vulnerable than female [90]. The absolute risk implied by these odds ratios is small because the base rate is small; the finding is a clue to developmental mechanism, which the neuroscience chapter uses, not a screening tool.

Antecedents

In the Dunedin cohort, adult schizophreniform disorder was preceded by a broad array of juvenile disorders rather than by a single childhood precursor [59]. Ravi's late language and long-standing preference for small groups are the kind of nonspecific antecedents that population studies record and that no individual history can convert into a diagnosis.

Cognition

The European Psychiatric Association guidance on cognitive impairment in schizophrenia reports that impairment across a wide range of domains is consistently found and underlies functional disability, but it is adult-focused and gives no child or adolescent figures [88]. It is cited in the treatment chapter for what to assess, not here for how often.

What this case cannot establish

The base rates say that a reported perceptual experience in a 16-year-old is common, that the disorder is rare, that most high-risk adolescents do not convert, and that affective diagnoses (27.2% combined) are nearly as common as schizophrenia-spectrum ones (34.6% combined) in a referred sample [89]. None of that settles Ravi's diagnosis, which the case chapters leave, correctly, to time and observation.

Source status

#87 and #89 from open-access full text (their figures are those reviews' summaries of cited sources); #90 and #88 from published abstracts; #59 from the abstract.

Figures for this chapter

Presentation

Three levels in one sentence

The counselor's letter says, "Two teachers think he is hearing things." That sentence holds three kinds of claim. The teachers observed Ravi staring at the wall. They inferred that he was perceiving something. They assumed that the perception, if there was one, meant psychosis. Assessment begins by separating them.

What others observed is limited: staring, abrupt exits, the alarm statements, reduced speech and affect, poor sleep, weight loss, unanswered messages, and speech behind a door. What Ravi alone reported is different in kind: his name heard in quiet, thoughts "not in my voice," a feeling that others know something about him, and a coincidence that felt meaningful. What is inferred is that Ravi has anomalous perceptual and cognitive experiences. What is assumed, if anyone says "psychosis," is a pattern and a mechanism. Ravi's account is the primary evidence for the perceptual experiences, which is usual in this domain and should be said plainly rather than treated as a weakness.

Perceptual reports

Hallucinations at the edge of sleep, as people fall asleep or wake, are common in the general population [222]. Ravi's description has features worth recording: it happens mostly in quiet, it is "outside my head," and he knows no one is there. Location inside or outside the head does not by itself sort clinical from nonclinical experiences [223]. His recognition that no one is present is often called insight; more precisely, he compares the experience with the evidence and reaches the ordinary conclusion.

"Thoughts that aren't in my voice" is a different report. It describes a change in the felt ownership of some thoughts, not in where they come from. The same words could describe inner speech that has lost its sense of being self-generated, an intrusive thought, or an auditory experience he is reluctant to call a voice. The questions that narrow this are concrete: how often, at what times, whether the thoughts comment or command, whether he can stop or steer them, and whether they continue in the weeks after cannabis stops. Ravi's report establishes an experience he finds anomalous and can describe precisely. It does not establish that it meets a diagnostic definition, or how often it occurs.

Beliefs and their evidence

The alarm sounding "right when I was thinking about leaving" is an idea of reference: a coincidence experienced as directed at him. His conviction was hedged. "That's not nothing" is an inference offered for discussion, not a belief held against evidence. The camera was raised as a question, which means he was still weighing it. The feeling that "everyone knows something about me" has a history: in ninth grade classmates did circulate photographs of him with captions, and others did know things about him that he did not. For a period, a representation of others as knowing and sharing information about him was accurate.

A delusion is conventionally a fixed belief, not shared by the person's culture, maintained despite contrary evidence. Ravi's ideas currently lack fixity. Conviction, preoccupation, and distress are rated separately [224], and action is recorded beside them as a fourth clinical dimension. Action is the observable part: leaving class, avoiding friends, and asking about the camera are the consequences others can see.

Reduced speech, flat affect, and withdrawal

Several explanations produce the same appearance. Reduced speech and flattened expression can be negative symptoms of a psychotic disorder, features of a depressive episode, guardedness in a young person who expects to be judged, or the effect of sleep loss and intermittent cannabis. On a topic he knew and valued, Ravi produced organized, detailed sentences and smiled, which shows that fluent speech and expressive affect are available under a particular condition. It does not show that the flatness elsewhere was chosen, any more than the enjoyment of a duet in the Depressive Disorders module (Jonah) showed that he could have started practicing alone.

Withdrawal needs the same care. Ravi has preferred one or two friends to groups since elementary school; that is a trait. His closest friend dates a change to two months ago, when Ravi stopped answering messages; that is a change from baseline. The two look alike from outside and are functionally different: one describes how Ravi has always participated, the other a loss of participation he previously maintained.

The two pauses

In session, Ravi stopped mid-sentence twice and looked toward the corridor. The first pause followed a slammed door; he reoriented and resumed. The second followed nothing the clinician could hear; he stared and then said he had "lost the thread." The first is attentional capture by a real stimulus. The second could be a lapse in holding the sentence's content, a response to an internal experience, distraction, or a brief interruption of thought. It could also be an absence seizure, which is why the medical evaluation does not end with screening results. From outside the pauses are identical. The only way closer to the second is to ask immediately what happened, and the answer will still be a report.

The same discipline applies to speech behind a closed door: lifelong narration while building, and a nighttime speech his mother cannot characterize, are two observations, not one. She can note whether the speech narrates, is addressed to someone, pauses as if for replies, and whether Ravi says he is talking with anyone.

Observation	What it can establish	What it cannot establish
Name heard in quiet (reported)	A recurrent anomalous experience he can describe and evaluate	A hallucination in the diagnostic sense; frequency
"The alarm was a signal"	A coincidence acquired significance; conviction hedged	A fixed belief; a mechanism
Second mid-sentence pause	Something unseen interrupted his speech	Memory, attention, internal experience, or seizure
Speech in his room at night	Ravi talks aloud when alone	Whether this is new or conversational
Organized speech about the printer	Fluent speech available under that condition	That reduced speech elsewhere is voluntary

Change over time and the developmental background

Ravi's early history includes late first words, preschool speech therapy, a preference for small groups, cafeteria sensory discomfort, and a teacher's question about autism. In a prospective birth cohort followed back from adult diagnosis, adults with schizophreniform disorder had received a broad array of juvenile diagnoses rather than one specific antecedent [59]. The antecedents raise the question of continuity between Ravi's early profile and his current change without answering it; they are common in young people who never develop a psychotic disorder. Whether Ravi is autistic is a separate question, evaluated once the current change has been characterized.

The differential, held open

The primers on youth with psychotic symptoms treat such symptoms as common across many conditions and insufficient on their own to establish a psychotic disorder [87, 89]. Medical evaluation addresses delirium, seizures, endocrine and metabolic causes, medication and substance effects, and autoimmune and neurological conditions, guided by history and examination [87]. Five differentials remain live for Ravi, each predicting something different.

An affective psychosis would place the perceptual experiences inside a mood episode; their timing relative to the low mood since summer, and any interval of reduced need for sleep or activation, decide its weight. A substance-induced presentation would fade with abstinence; the experiences have persisted five weeks after his last reported use, which weighs against it without excluding it, so urine testing and a defined period of confirmed abstinence are needed. Autism with atypical perceptual reports would predict literal, precise descriptions of inner experience, aversion to the alarm on sensory grounds regardless of meaning, and lifelong narration; Ravi's account has some of these qualities. A trauma-related pattern following the group chat would predict hypervigilance about being watched and discussed, intrusive memories, and avoidance of those involved; the alarm-as-signal and the name in quiet fit it less well. Reported-but-not-observed psychosis names the possibility that the school's "hearing things" is an inference from staring, that the clinician saw one unexplained pause, and that the perceptual account rests on Ravi's testimony. His testimony is careful and consistent, a reason to take it seriously; corroborating observations are still needed.

Ravi's retained ability to evaluate his experiences and his hedged conviction place him closer to an attenuated than a frank psychotic pattern. In general-population samples, most people with subthreshold psychotic experiences do not develop a psychotic disorder, although their relative risk is raised [225]. Those samples were not help-seeking; the finding does not transfer directly to clinical-high-risk cohorts, which a referred adolescent like Ravi more closely resembles. Safety is assessed regardless of staging: sleep loss, substance use, and any sense that he must escape or defend himself against what he believes are asked about directly.

What this case cannot establish

The presentation cannot establish whether Ravi has a psychotic disorder, whether the name he hears is a hallucination in the diagnostic sense, whether cannabis initiated or maintains any experience, whether he is autistic, whether sleep loss is cause or consequence, or whether the speech in his room is new. It establishes that he has described anomalous experiences precisely, that his participation in school and friendship has fallen from his own baseline over roughly six months, and that several look-alike observations remain to be separated.

Figures for this chapter

Functional neuroscience

Functional layer

Perception is inference, and inference has settings

Hearing is not the passive receipt of sound. The auditory system combines what arrives at the ear with what it expects to hear, weighting each by its current reliability, which predictive-processing accounts call precision. Perceptual inference is the Process that produces a percept from expectation and input together. Prior precision and sensory precision are its settings: how much confidence is placed in expectation and how much in input. Perceptual expectation is the Representation on which the process runs, here a familiar and self-relevant word. When arousal, sleep loss, or a drug shifts the settings, that is a State: the capacities are unchanged, the balance has moved. In this computational account, hallucinations arise when prior precision dominates. Experiments support that route without establishing it as the cause of every hallucination [226].

Ravi's name in quiet fits this vocabulary without anything unusual. In a silent library, sensory precision is low because little input constrains the percept; a person's own name is the expectation with the highest prior precision in the lexicon. A percept can then be assembled from expectation with almost no input. Percepts without a corresponding input are common in the general population as people fall asleep or wake [222]. Paradigms that condition a tone to a visual cue and then present the cue alone find that people who hallucinate in daily life more often report the absent tone, consistent with heavier prior weighting [227].

For Ravi the account yields questions. Does he hear his name when expecting to be called, or also when nothing predicts it? Only in quiet? After sleep? A percept that tracks expectation and silence describes ordinary inference with little input. One that arrives regardless of expectation, in noise, and increasingly often describes a change in the settings. Neither identifies the mechanism that changed them.

Source monitoring: whose thought is this

Source monitoring is the Process that attributes an event to an internal or external origin. Ordinary inner speech carries a sense of having produced it. One proposed basis is a predictive signal sent with each self-generated action that dampens its sensory consequences, so that one's own voice registers as self-caused. That dampening has been measured when people speak aloud; its extension to inner speech and to hallucinations is a mechanistic hypothesis [228]. If that dampening fails, an internal event can be experienced as arriving from elsewhere. Electrophysiological studies in schizophrenia report reduced suppression of the auditory response to self-generated speech, consistent with this account; they are not a measurement of Ravi [236].

"Thoughts that aren't in my voice" describes a loss of ownership rather than a change of location, and describes it well; that is description, not explanation. Ravi's report establishes that the felt origin of some thoughts has changed, not that a predictive signal is impaired: a report cannot distinguish a failure of tagging from an unusual thought correctly tagged as his own but foreign for other reasons. The report is the observation, the experience is the inference, the failed signal is the assumption.

Salience: when a coincidence demands explanation

Salience attribution is the Process that marks an event as significant and in need of explanation. Every alarm is loud; not every alarm feels aimed at the person hearing it. The aberrant salience account proposes that in emerging psychosis ordinary events acquire significance they would not normally carry, and beliefs form to explain it [229]. The alarm sounding as Ravi thought of leaving is a coincidence that acquired significance, and "that's not nothing" is the beginning of an explanation.

Two cautions follow. Ravi found the cafeteria overwhelming as a child; an alarm can be aversive on sensory grounds before it means anything. And the observable fact is a statement about the alarm, not a measurement of salience. Increased significance is an inference; a dopaminergic basis is a further assumption.

Belief updating and the model of other people

Belief updating is the Process of revising a representation in light of new evidence. People with delusions have, on average, reached decisions on less evidence in probabilistic reasoning tasks, a pattern called jumping to conclusions [230]. Ravi has so far done the opposite: he asked about the camera rather than asserting it, and offered the alarm inference for discussion. Updating remains open, which is itself an observation to revisit.

The belief that "everyone knows something about me" runs on a Representation, a model of others' knowledge and intent. Ravi did not invent it: in ninth grade other people did know things about him and did share them. What he holds now is a prior learned from real evidence and generalized to new classmates and rooms, associative generalization from a trained context to similar ones. A prior accurate for one setting and applied to many is not a prior that was never grounded. Mental-state inference, introduced in the Intellectual Disability module (Maya), reads what a classmate knows or intends from behavior and context; Ravi's current model biases it toward hostile knowledge. The content of the belief and the process of holding it are separate treatment targets, which matters when the belief has a true history.

Cognitive capacities and access to them

Processing speed, working-memory maintenance, and verbal learning and memory are Capacities: what Ravi can do under good conditions. Cognitive impairment is treated as a core feature of schizophrenia, present at first episode, and European Psychiatric Association guidance recommends systematic assessment across defined domains including processing speed, attention, working memory, verbal and visual learning, reasoning, and social cognition [88]. That guidance was written for schizophrenia in general, not for adolescents, and it concerns assessment: a measured deficit does not establish a psychotic disorder.

Ravi's "lost the thread" pause, falling grades, and reduced speech could each reflect reduced capacity, or reduced access to it. Four to five hours of sleep, intermittent cannabis, preoccupation with what classmates know, and the effort of monitoring a room for signals all change the conditions under which capacity is used. The printer conversation showed organized, detailed speech under conditions of interest and familiarity: capacity is present, and access varies with State. An assessment after sleep improves and abstinence is confirmed gives a truer estimate; an early baseline still has value for comparison.

Withdrawal and the motivational account

Not answering messages could be avoidance driven by the belief that friends know something, reduced anticipation of benefit as in a depressive state, reduced drive as a negative symptom, or the fatigue of five hours' sleep. The distinctions the Depressive Disorders module (Jonah) draws among anticipatory reward, effort cost, and action initiation apply without importing his diagnosis, and each function predicts something different: avoidance lifts when the belief is tested, anhedonia with the mood state, a negative symptom persists across settings, fatigue follows sleep.

Why adolescence, and what came before

A neurodevelopmental framework places the usual onset of psychotic disorders in the years when prefrontal circuits are remodeled, synaptic density in association cortex falls, dopamine signaling matures, and social demands rise sharply [231]. Each is a plausible reason for timing; none has been shown to be the reason for any individual. Cohort studies find later psychotic disorders more common among children with early language delay, social difficulty, and lower childhood cognitive scores [232]. A prospective birth cohort followed back from adult diagnosis found that adults with schizophreniform disorder had received a broad array of juvenile diagnoses, not one characteristic antecedent [59]. Ravi's late first words, speech therapy, and preference for small groups fall within a common and nonspecific range. The history raises the question of neurodevelopmental continuity; it cannot settle it, and should not become a prior that outweighs current observations.

Circuits, development, and genes

Dopamine, the striatum, and the meaning of a signal

Positron emission tomography studies report elevated striatal presynaptic dopamine synthesis capacity in schizophrenia, and higher values in a small group of high-risk individuals who later developed a psychotic disorder than in those who did not [233, 234]. The finding concerns dopamine synthesis rather than receptors. It anchors the aberrant salience account: distinct populations of dopamine neurons signal motivational value, motivational salience, and alerting events [235], and excess or mistimed release is proposed to lend significance to events that have none. Antipsychotics act at dopamine D2 receptors, and their efficacy for positive symptoms is consistent with, without proving, this account. None of this measures Ravi. A group difference in a tracer study describes a distribution; it does not identify one adolescent's striatum or explain why an alarm, rather than a bell or a bus, became the signal.

Glutamate, the NMDA receptor, and a medical differential

In a controlled challenge in healthy adults, ketamine, an N-methyl-D-aspartate receptor antagonist, produced transient states resembling positive, negative, and cognitive symptoms [237]. The result is consistent with a hypothesis of NMDA receptor hypofunction in schizophrenia but does not by itself establish that hypofunction causes the disorder. The Depressive Disorders module (Jonah) describes this receptor's voltage dependence: its response depends on the receiving cell's electrical state as well as the transmitter, so the same change in signaling has different consequences in different circuits. For Ravi the relevance is a differential rather than a treatment. Antibodies against the NMDA receptor cause an encephalitis that can present with psychotic symptoms in adolescents, often with seizures, movement abnormalities, autonomic instability, or reduced consciousness; a screening panel does not exclude it [238]. The evaluation primer places neurological and autoimmune causes within the initial assessment of new psychotic symptoms in youth [87].

Development and prenatal exposure

Longitudinal imaging in childhood-onset schizophrenia has reported exaggerated loss of cortical gray matter through adolescence relative to typically developing peers [239]; gray matter loss on MRI is not a direct measure of synaptic pruning. Variation in a complement gene has been linked to schizophrenia risk in humans, and complement-mediated synaptic pruning has been examined experimentally in mice [240]. These describe groups and models at the level of tissue and molecule.

At a further level, in a birth cohort maternal bacterial infection during pregnancy was associated with increased risk of psychotic disorders in offspring, varying by severity of infection and offspring sex [90]. Ravi's mother recalls a treated second-trimester urinary tract infection. A cohort association cannot be applied as a cause in an individual, and cannot be excluded either; it belongs in the history as a recorded exposure, not an explanation.

Genetic liability

Liability to schizophrenia is distributed across many common variants of small effect, and a small number of rare copy-number variants, including the 22q11.2 deletion, carry substantially higher risk [241, 242]. A maternal uncle hospitalized for "a breakdown" raises the prior probability of familial liability without specifying it; the phrase can cover a mood episode, a psychotic episode, or an intoxication. Asking the family what they know, and considering genetic evaluation if dysmorphic features, cardiac history, or developmental findings suggest a syndrome, are separate steps.

Moving between levels of evidence

A conditioning experiment tests a specified process under controlled conditions. A tracer study describes a distribution across groups. A cohort study estimates a risk association in a population. A trial estimates benefit and harm on its measured outcomes. Ravi's report of his name in quiet sits at none of these levels; connecting them to him requires an additional inference at every step, and the connection weakens with each.

What this case cannot establish

Nothing in Ravi's presentation measures prior precision, source monitoring, salience, or striatal dopamine. Those are functional descriptions and group-level findings that give his reports a vocabulary and the clinician questions. His history cannot establish that his early language delay, his mother's infection, or his uncle's hospitalization contributed to his current experiences, only that each is a recorded antecedent. His pause in session cannot establish a neural event. What the functional account establishes is a set of distinctions his treatment can act on: expectation versus input, ownership versus location, a grounded prior versus an ungrounded one, and capacity versus access to it.

Figures for this chapter

Treatment

What Ravi and his mother are asking for

Ravi wants to sleep, to stop the feeling in class, and not to be "crazy." His mother wants to know whether he is safe and what the speech behind his door is. The school wants a decision about placement. These are different goals with different timescales, and the plan has to serve all of them while the diagnosis remains open. For an emerging presentation, the evaluation is the first treatment, because most of what will change the plan over the next three months is information that does not yet exist.

Evaluation as the first intervention

The primers on evaluating and managing youth with psychotic symptoms set out the components: a full psychiatric and developmental history with collateral, a medical evaluation guided by history and examination, assessment of substance use, and a safety assessment [87, 89]. For Ravi, the screening panel and examination were normal. What remains includes urine toxicology, a defined period of confirmed cannabis abstinence, and attention to features that would prompt neurological or autoimmune investigation: fever, seizures, abnormal movements, autonomic instability, fluctuating consciousness, or rapid deterioration [87].

Cognitive assessment belongs early. The European Psychiatric Association's guidance recommends systematic measurement of cognition in people with schizophrenia across domains including processing speed, working memory, verbal learning, and social cognition, using standardized instruments [88]. Ravi does not have that diagnosis, and the guidance was not written for adolescents. Its value here is a baseline: a profile obtained now and repeated when he is sleeping and abstinent shows whether his falling grades reflect capacity or access to it, and gives the school something concrete to accommodate.

Three further pieces of information are cheap and decisive. His mother can keep a nightly note on the speech in his room: does it narrate, address, or reply. Ravi can log the name experience: time, setting, whether he was expecting to be called, hours slept. And the ninth-grade group chat can be asked about directly, including whether any of it continues, because a belief with a true history is treated differently from one without.

Staging before prescribing

Ravi's hedged conviction, retained ability to evaluate his experiences, and absence of observed disorganization place him closer to an attenuated than a frank psychotic presentation. The evaluation framework distinguishes these because they carry different treatment recommendations [89]. For an attenuated presentation, the plan is monitoring at defined intervals, treatment of current impairment, and family psychoeducation. UK national guidance for children and young people advises against antipsychotic medication for subthreshold psychotic symptoms or solely to prevent psychosis [243]. What is impairing Ravi now is sleep, school attendance, isolation from his friend, and distress in class, and each of these has a treatment that does not depend on the diagnosis.

If the picture changes, the staging changes with it. Perceptual experiences that persist after two to three months of confirmed abstinence and improved sleep, beliefs that become fixed and acted on, disorganized speech, or further decline in self-care and function would each move the formulation toward a psychotic disorder and open the question of antipsychotic medication. A sustained interval of reduced need for sleep, pressured speech, and expansive plans would move it instead toward an affective psychosis, for which the medication evidence is different. Neither shift is predicted by current observations; both are watched for.

Antipsychotic medication: what the trials show and what mechanism adds

If Ravi meets criteria for a schizophrenia-spectrum disorder, antipsychotic medication is the treatment with the strongest evidence for positive symptoms. Comparative trial evidence in early-onset schizophrenia spectrum disorders, summarized in the primers, found no clear efficacy advantage among the first- and second-generation agents compared, but substantial differences in adverse effects, particularly weight gain and metabolic change with some second-generation agents and extrapyramidal effects with others [87, 89]. Adolescents are more sensitive than adults to weight gain and metabolic effects, and monitoring of weight, body mass index, fasting glucose and lipids, prolactin, sedation, and movement is part of prescribing rather than an addition to it [87, 89]. Clozapine has evidence in early-onset schizophrenia that has not responded to other agents, and its use requires hematologic monitoring and attention to seizures, myocarditis, and metabolic effects [89].

The functional account gives this evidence a vocabulary without adding to it. Antipsychotics act at dopamine D2 receptors, and the salience account proposes that reducing striatal dopamine signaling reduces the significance coincidences acquire, so that an alarm is again only an alarm [229]. This is a plausible route from a receptor to Ravi's experience of the chemistry room. The trials did not measure salience; they measured symptom ratings, and they do not show that reduced salience mediated the ratings, or which adolescents will respond. The guidance on cognition treats cognitive impairment as a target requiring its own assessment and interventions rather than something expected to resolve with antipsychotic treatment [88]. A receptor action can change the effectiveness of signaling within a circuit; it does not explain the individual outcome, and it does not restore a friendship or teach Ravi what to make of his name in a quiet room.

If medication is started, the dose begins low, the target is the positive symptoms and their behavioral consequences, and the success criterion is Ravi's, not the scale's: fewer exits from class, a room that feels unwatched, sleep. A sedated adolescent who no longer leaves class has not necessarily improved.

Psychosocial treatment acts on the processes named in the formulation

Family psychoeducation lowers relapse rates in schizophrenia in adult trials [244], and for Ravi's mother it also answers her question about what she is hearing at night. It teaches her what to observe and what to do, and reduces the chance that ordinary narration while building becomes evidence of deterioration.

In adult trials, cognitive behavioral therapy for psychosis has small effects on positive symptoms, and these shrink in trials with blinded assessment [245]. Its methods map onto the functional layer directly. The camera belief and the alarm belief are predictions that can be tested, and Ravi has already shown that his updating is open. The belief that everyone knows something about him has a true origin, which means the work is not to dismiss it but to help him distinguish the setting in which it was accurate from the settings to which it has generalized. The name in quiet can be normalized with the information that hallucination-like experiences at the edge of sleep are common in the general population [222], that his own experience tracks silence and expectation, and that his own log becomes the evidence.

Coordinated specialty care for first-episode psychosis combines medication management, psychotherapy, family involvement, and supported education or employment, and showed better functional outcomes than usual care in a large US trial [246]. Whether Ravi would be eligible depends on staging; the components apply regardless. Cognitive remediation has evidence for improving measured cognition in schizophrenia [247], though not equivalent evidence for every program or for adolescents, and would follow from his cognitive profile rather than his diagnosis. School accommodations, informed by that profile, keep him in class while the evaluation proceeds.

Cannabis use is associated with earlier onset of psychotic disorders, and continued use with higher relapse rates, though both associations are open to confounding [248, 249]. Its cessation is both a treatment and a diagnostic test: if the experiences fade with abstinence, the formulation changes. Sleep is treated directly, because a night of total sleep deprivation increased anomalous perceptual experiences in healthy adults [250] and because five hours a night changes the State under which every capacity is used. If the group chat proves to be a live source of hypervigilance, trauma-focused treatment is considered on its own merits.

Returning to the chemistry room

The plan that follows is concrete. Ravi sleeps more, with a schedule and, if needed, short-term help. He stops cannabis and is tested. His mother keeps her nightly note; he keeps his log. His cognitive profile is measured and shared with the school, which allows a quiet place to work and a way to leave class without it being an event. He and a therapist test the camera and the alarm and trace the ninth-grade history of the belief about what others know. His closest friend is told, with Ravi's permission, that a short message is welcome. He is reviewed at fixed intervals, and the questions at each review are the ones the formulation generated: does the name track silence and expectation, has updating stayed open, has the pause become a pattern, has anything new appeared. If a psychotic disorder declares itself, medication is added to this plan, not substituted for it.

What this case cannot establish

The treatment plan cannot establish that Ravi will not develop a psychotic disorder, that abstinence and sleep will resolve his experiences, or that any psychosocial method acts through the functional process it is described as targeting. It cannot establish, if medication is eventually needed, which agent will suit him, or that a receptor mechanism explains his response. What the plan establishes is a set of observations that will answer those questions over time, a set of treatments for his current impairment that are justified regardless of diagnosis, and defined signals that would change the plan.

Figures for this chapter

---