Introduction

As described since ancient times, and by Hippocrates 2000 years ago, psychosis is a mental state characterized by impaired reality testing. It is manifest as a constellation of symptoms, including false perceptions and fixed false beliefs, called “hallucinations” and “delusions,” and through disturbed thinking evidenced by the use of language, including “loose associations,” and frequently through bizarre motor activity and behavior. Schizophrenia is the best-known psychotic disorder, and many of the classic symptoms of psychosis were originally observed in patients who would currently be diagnosed as having schizophrenia.

Eugen Bleuler (1857–1939), the Swiss psychiatrist who coined the term “schizophrenia” (from the Greek skhizein “to split” and phren “mind”), considered psychotic symptoms to arise from a “break” (disconnection) between the neural systems subserving thought and emotion. He proposed that this schism produced observable phenomena, later referred to as the four A’s, including autistic preoccupation, ambivalence, affect (flat), and associations (loose, which are characteristic of thought disorder). All of these, except ambivalence, continue to be key criteria for the diagnosis of schizophrenia in modern diagnostic procedures. Kurt Schneider (1887–1967), a German psychiatrist, later compiled a core group of “hallmark” psychotic symptoms termed “first rank symptoms” (FRSs), which he proposed could distinguish schizophrenia from other psychotic disorders. These were the experiences of audible thoughts, hallucinations of two or more individuals arguing or conversing, voices that commented on the patient’s actions, thought broadcasting, thought withdrawal or insertion, delusions of passivity, and delusional perception. But in more recent years, studies have shown that these are not specific to schizophrenia.

While psychosis is a defining feature of schizophrenia spectrum disorders, it is a variable feature in conditions such as substance use, developmental, endocrine, and brain disorders and diseases. Substance abuse including alcohol, amphetamines, hallucinogens, marijuana, and opioids can all cause isolated periods of psychosis. Substance-induced psychosis is diagnostically distinct from other psychotic disorders in that symptoms subside shortly after substance use, while psychotic symptoms in psychotic disorders continue and may be chronic. Psychotic symptoms are also observed in other disorders, including bipolar disorder, depression with psychotic features, post-traumatic stress disorder and obsessive-compulsive disorder (OCD), some personality disorders, and in subclinical quantities in some persons without mental illness. It has also been proposed that there could be psychoses associated with anxiety disorders such as social anxiety, panic anxiety, and OCD.

Psychosis has historically been split into two dichotomous categories: affective (including bipolar disorder and major depressive disorder with psychotic features) and non-affective psychosis (including schizophrenia and schizoaffective disorder). In recent years, the validity of the nosologic distinction between these two categories has been called into question and a transdiagnostic dimensional approach has been put forth as a more ecologically valid way of conceptualizing psychosis-related disorders. Research using factor analysis has shown that a general, transdiagnostic dimension that includes affective and non-affective symptoms in addition to five symptom dimensions (depression, mania, positive symptoms, negative symptoms, and disorganization) may better model the different forms of psychosis experienced by patients. This model is called the bifactor model.

The central thesis of this book is that psychosis proneness may interact with certain comorbidities (including OCD, panic disorder, social anxiety, major depressive disorder, mania, substance abuse, and medical illnesses) and other factors (including environmental and genetic) to produce different forms of psychosis. In this chapter, we will discuss the pathogenesis of psychosis and factors related to its development. The terms “psychosis” and “schizophrenia” will be used somewhat interchangeably due to the fact that the bulk of the extant research focuses on schizophrenia. Presently, we will review psychosis as it is featured in schizophrenia and other psychosis-related disorders.

Psychosis in Schizophrenia

Characteristic symptoms for a diagnosis of schizophrenia include profound deficits in emotion and behavior, interfering with the expression of emotion and with motivation, often referred to as flat affect and avolition. As these are the absence of normal functions, they are called negative symptoms, distinguishing these from the active psychotic symptoms, conversely named the positive symptoms. The schizophrenia diagnosis also includes deterioration in functioning, a phenomena that typically occurs before the emergence of psychotic symptoms. In the last century, it has become evident the positive and negative symptoms associated with schizophrenia are highly variable between individual patients.

This observation led to the concept of “the group of schizophrenias” by Bleuler, but the differential determinants of etiology and optimum precision treatments for schizophrenia remain unclear. Most patients are initially affected in young adulthood; 50% go on to experience some disability throughout their lives, and an additional 25% never recover and require lifelong care. The lifetime risk for schizophrenia is commonly reported to be approximately 1% and to be similar for men and women, with an earlier onset in males. However, recent epidemiologic reviews demonstrate a prominent geographic variation in incidence, with a fivefold variation around a median incidence of 15.2/100,000 persons (7.7–43.0/100,000), and more affected males than females (1.4:1).

Distinguishing Psychotic Depression and Bipolar Disorder

Among persons with psychosis and deterioration, schizophrenia spectrum conditions are traditionally distinguished from mood-related psychosis, including bipolar and other affective disorders. However, the diagnostic boundary between the affective and non-affective (schizophrenia) psychoses is an active topic of debate. Emil Kraepelin (1856–1926) was the first to dichotomize affective and non-affective psychotic conditions. He provided the earliest rigorous descriptions of dementia praecox (now denoted as schizophrenia) and observed differences between these cases and those with manic-depressive insanity (now denoted as bipolar disorder) in the episodicity of their symptoms and long-term outcome. His idea was that these were dichotomous conditions with different underpinnings. This was one of the earliest attempts to categorize mental illnesses based on observable behavioral phenomena. The separateness of these conditions, while still retained in separate chapters of the DSM-5, is challenged by diagnostic instability and overlap in their epidemiology, genetics, neuroanatomy, and neuropsychiatric finding.

Schizoaffective Disorder

Psychosis can occur during an episode of depression or mania, or other psychiatric disease in which it is diagnosed as a component of the primary psychiatric condition. However, persons with schizophrenia can also experience mood episodes. Of course, as above, if the full mood syndrome is always present or treated during the psychotic episodes, the diagnosis is not schizophrenia. It has been more problematic to name a condition that includes periods of psychosis without any mood disturbance, of at least 2 weeks, with other periods of psychosis including full manic, mixed, or depressive episodes. When such mood syndromes occur, less than half of the total duration of all periods during which there are psychotic or residual symptoms, the DSM-5 diagnosis is schizophrenia and a separate mood diagnosis. If full mood syndromes have been present for more than half of the total duration of all psychotic plus residual periods (> 50%), which the clinician estimates, then the diagnosis is schizoaffective disorder, either depressed or bipolar.

Psychosis in Community Samples

Notably, some members of the general population have odd experiences that verge on hallucinations or delusions, others enter the psychosis prodrome, and still others develop full-fledged psychosis. Studies done in community samples indicate that psychotic-like experiences are not uncommon in the general population. Indeed, the incidence of psychotic symptoms occurring in the general population is approximately between 4.8% and 8.3% depending on the particular symptom being examined. These psychotic-like symptoms may represent variations on normal personality traits or they may be forme frustes indicative of an underlying vulnerability to a psychotic disorder.

One theory posits that positive psychotic symptoms exist on a continuum, with full-fledged psychosis on one end and psychotic-like symptoms on the other end. Two models of continuity have been put forth: the quasi-dimensional model and the fully dimensional model. The quasi-dimensional model theorizes that psychotic-like symptoms represent partial penetrance of psychosis, and those exhibiting such symptoms would be at an increased vulnerability for the development of a psychotic disorder. This model implies a discontinuity from the normal population. Paul Meehl supported this theory, coining the term “schizotaxia” to describe the symptomatic, cognitive, and behavioral vulnerabilities that are likely to eventuate in a diagnosis of schizophrenia. By contrast, the fully dimensional model of psychosis posits no discontinuity from a healthy population suggesting that psychotic-like symptoms are a feature of personality.

Psychosis Prodrome

The criteria for the prodromal phase of psychosis is a period of functional deterioration before the onset of the full-blown active symptoms of psychosis. Individuals within this phase are considered to be at clinical high risk for a transition to psychosis. The diagnosis of clinical high risk is warranted when there is a presence of at least one of the following: (1) functional impairment (social and occupational) in the context of familial risk; (2) brief limited intermittent psychotic symptoms (BLIPS—an episode of severe psychotic symptoms, which resolves spontaneously within 1 week and occurred during the past 12 months); or (3) attenuated psychotic symptoms (APS). According to the DSM-5, APS include perceptual disturbances (subthreshold hallucination), overvalued ideas (subthreshold delusions), and disorganized speech, all with relatively unimpaired reality testing. Approximately 36% of clinical high-risk individuals will transition to first episode frank psychosis after 3 years, follow-up, with half of the transitions (18%) occurring after the first 6 months.

Schizotypy

Schizotypy represents a latent and persisting set of social oddities, cognitive problems, and personality disturbances that can lead to an array of schizophrenia-related phenotypes. Unlike the prodromal state of psychosis, schizotypy is more enduring. These phenotypic outcomes can manifest as overt personality disorders (e.g., schizotypal personality disorder, schizoid personality disorder, paranoid personality disorder, and avoidant personality disorder) or as a more tenuous subclinical presentation (e.g., interpersonal aversiveness, perceptual aberrations, referential thinking, magical ideation). Schizotypy can also manifest as non-psychotic endophenotypes invisible to the naked eye but detectable with applicable technologies. These endophenotypes include sustained attention deficits, eye-tracking dysfunction, working memory impairments, motor dysfunction, thought disorder (secondary cognitive slippage), and psychometric deviance (PAS).

Schizoid and schizotypal personality disorders are both cluster A personality disorders that are characterized by odd or eccentric thinking and behavior. Although these disorders resemble schizophrenia, they are diagnostically distinct categories. Schizotypal personality disorder encompasses a pervasive pattern of social deficits that may include cognitive or perceptual distortions, eccentric behavior, and unusual thinking. These symptoms may resemble positive schizophrenia symptoms. Individuals with schizoid personality disorder tend to prefer being alone, have difficulty understanding social cues, and may lack motivation in their professional lives. These symptoms may resemble negative schizophrenia symptoms. While many of the cluster A personality disorder symptoms overlap with schizophrenia symptoms, they never progress into full-fledged hallucinations or delusions. For example, while a person with schizotypal personality disorder may believe that the news anchor that night was broadcasting a story about their own life, when they are confronted with evidence contradicting their belief, they are able to modify their beliefs, whereas an individual with schizophrenia will not be dissuaded that the news story was not about them.

While genetic claims regarding schizoid personality disorder’s link to schizophrenia remain inconclusive, there is compelling evidence demonstrating the genetic overlap between schizotypal personality disorder and schizophrenia. Indeed, 40% of those who receive a diagnosis of schizotypal personality disorder progress to some psychotic disorder within the span of 2 years. The last decade has brought a burgeoning body of genetic research, including twin studies and linkage studies, that highlight the overlap between schizotypal personality traits and schizophrenia.

The outcome and clinical course of schizotypy are still not fully understood. Many individuals with the phenomenon will never develop full-blown, active symptoms of psychosis (this is also true of individuals with prodromal psychotic symptoms). As conceived by Lenzenweger, individuals at the prodromal phase of psychosis who do not transition to frank psychosis are harboring schizotypy and will subsequently demonstrate a level of impairment in personality throughout a life span. A recent study, however, does suggest premorbid schizotypy as an independent risk factor for transition to psychosis in the clinical high risk. In their study, Kotlicka-Antczak et al. demonstrated that premorbid schizotypy, as elicited through mothers of individuals at clinical high risk, might be time-independently associated with the transition to psychosis. This finding was much more notable in clinical high-risk individuals with high levels of premorbid negative schizotypy (social isolation, restricted affect, and social anxiety/suspiciousness). It is also of importance to note that this positive correlation to psychosis transition was not only evident in clinical high risk with premorbid core schizotypy but was significantly present with antisocial aspects of schizotypal expression as well.

Risk Factors for Psychosis

Only a small proportion of the population-attributable risk for schizophrenia can be explained by our current knowledge. Interactions between genes and exposures, particularly in certain developmental windows, will almost certainly be important. Increasingly, it is clear that gene-environment interactions must be considered for illness risk, with changes in gene expression as the likely mechanism. Data that support models of interaction between genetic vulnerability and the environment result from adoption studies. For adopted-away children of mothers with schizophrenia, adversity and poor family functioning in the adoptive home increase the risk of schizophrenia; however, no such effect is seen for adopted-away children of healthy mothers. In an Israeli study, children of mothers with schizophrenia had a higher risk of developing the illness themselves if raised in a kibbutz instead of a family home; again, this effect was not seen for children without genetic risk. Among the interesting findings of gene-environment interaction for schizophrenia was the observed interaction of a specific allele of a susceptibility gene, val-COMT, with an exposure (specifically cannabis abuse), in increasing schizophrenia risk. While val-COMT’s primary role is the enzymatic degradation of adrenaline, noradrenaline, and dopamine, it is also thought to play a secondary role in the expression of both anxiety and depression. Whether or not the observation of the gene-environment interaction with val-COMT stands up, it is a useful approach and may help inform future studies examining gene-environment interaction.

Global Genetic Risk Factors

Evidence from family, twin, and adoption studies suggests that the cause of schizophrenia is, at least in part, inherited. First-degree relatives of schizophrenic probands have an increased risk of schizophrenia, with estimated risk rates of 6% for parents, 10% for siblings, 13% for children with one affected parent, and 45% for children with two affected parents. Identical twins have a concordance rate of 53%, whereas fraternal twins have a concordance rate of 15%. The disparity in these rates suggest that 60% to 90% of schizophrenia liability can be attributed to genes. Of interest, offspring of affected and unaffected (discordant) monozygotic twins have equal risk for schizophrenia (about 16%–18%), consistent with either incomplete penetrance and/or gene-environment interaction. Adoption studies also support inheritance of schizophrenia liability, though cannot discriminate if such effects are genetic or environmental (in utero) if mothers are the parent of comparison. When fathers are the parent in common (paternal half-siblings), however, the same increased risk in biological (vs. adopted) relatives holds. Adoption studies also provide evidence for gene-environment interaction, as adopted-away children of biologic mothers with schizophrenia are even more likely to develop schizophrenia if they are raised in an adverse environment.

The consistent evidence of inherited vulnerability revealed by a century of family twin and adoption studies of schizophrenia risk encouraged genetic research in schizophrenia. Dozens of candidate studies, genetic linkage, and association research studies were published, but not replicated. It is only with the advent of the hypothesis-independent genome-wide association studies (GWAS) approach that the Psychiatric Genetics Consortium uncovered scores of common loci that are significantly associated with schizophrenia, with increasing numbers of loci identified as sample sizes increased. These studies highlight the importance of immune factors, calcium signaling, N -methyl- d -aspartic acid (NMDA) glutamatergic function, and synaptic elements for schizophrenia risk, but the majority of common gene findings are not specific for schizophrenia. Notably the genetic risk architecture for schizophrenia also includes copy number variants (CNV), estimated to account for 5% of cases. These intermediate-sized gene variants can influence the expression and function of many genes, with certain very rare CNVs associated with a manifold increase in the risk for schizophrenia.

Other Family Risk Factors

Environmental Exposures

Presently, we will review the putative environmental exposures associated with schizophrenia risk. These risk factors include prenatal adversity, childhood trauma, childhood infections, traumatic brain injury (TBI), discrimination, immigration status, urban birth, and cannabis use. It is important to note that most exposures have small effects, with typical odds ratios or risk ratios of approximately 2. Questions of causation remain about some premorbid or prodromal syndromes that could increase risk of exposure (i.e., to head injury or drug use).

Pathophysiology of Psychosis

Predictors of Psychosis

Premorbid Abnormalities Support Developmental Origins

Many types of premorbid deviations can be found at each age in some portion of subjects, including minor physical abnormalities, soft neurologic signs, delayed and suboptimal motor and language, social interest and capacity, and neurocognition. It was, indeed, nearly a half century ago that Fish and colleagues (1977) coined the phrase “pan-developmental retardation,” to refer to the widespread subtle abnormalities in multiple domains, observed in children who later developed schizophrenia, including motor, sensory, cognitive, and cerebellar function.

In line with models of psychotogenesis, many studies confirm the presence of early and subtle abnormalities in at-risk children, such as compromised psychomotor performance, and cognitive and motor dysfunction by age 10. An innovative blinded review of home videos showed that raters could accurately identify which children would later develop schizophrenia, based on motor skills. Other precursors of schizophrenia (albeit nonspecific) include delays in developmental milestones (i.e., walking and talking), more isolated play at ages 4 and 6, speech problems and clumsiness at ages 7 and 11, and poor school performance and social anxiety during the teen years. Global attentional dysfunction may be a biobehavioral marker for genetic liability to schizophrenia, as attentional deficits exist in nearly half of all schizophrenia patients, and in the offspring of parents with schizophrenia. Abnormal social behavior also may be another nonspecific indicator of risk for schizophrenia, i.e., trouble making friends, disciplinary problems, and unusual behavior in childhood and adolescence. Early symptoms of premorbid anxiety disorders could also help explain abnormal social behavior seen in schizophrenia.

Later Insults Associated with Schizophrenia

Early trauma has consistently been linked to the development of psychosis-related disorders. One study found that people who are victims of child abuse are three times more likely to develop psychosis than their counterparts who had never been abused. Early trauma has also been associated with an earlier onset of schizophrenia, greater symptom severity, and the need for more aggressive treatment. There is also evidence, however, that patients are not “doomed from the womb” and that a number of postnatal exposures may be associated with the risk for schizophrenia, including TBI and the use of drugs, such as cannabis. For some or many cases, there is evidence that progressive deterioration occurs in adulthood, so that any developmental lesion affects neural function across adulthood as well as in early stages. Evidence that neural damage may be ongoing is clearly demonstrated by changes beyond the onset of illness : for example, the superior temporal gyrus decreases in volume after onset of first psychosis.

While earlier family, twin, and adoption studies pointed to a likely pattern of heritability in schizophrenia, the last decade has been marked by significant advances in our understanding of the genetic underpinnings of schizophrenia, including the identification of specific genes and chromosomal regions likely implicated in the pathogenesis of schizophrenia. While such research has pointed to a likely genetic predisposition to schizophrenia, converging evidence from the last decade has also highlighted the role environmental factors that influence psychotogenesis, including early childhood experiences, drug use, identity, and physical environment. In addition to environmental factors, accompanying comorbidities may also help determine endophenotype. Comorbidity subtypes of particular interest—panic disorder, social anxiety disorder, melancholic depression, atypical depression, and OCD—are all conditions that frequently occur alongside psychosis. Although not formally demonstrated, these comorbidity subtypes may correspond with the five specific factors seen in the bifactor model of psychosis (see introduction): positive symptoms, negative symptoms, depression, mania, and disorganization. These comorbidity subtypes and the bifactor model will be explored further in subsequent chapters.