A growing number of studies demonstrate high rates of subthreshold psychotic experiences, but there is considerable heterogeneity in rates due to study cohort and design factors, obscuring how prevalent psychotic experiences may or may not relate to rare psychotic disorders. In a representative general population sample (n = 4011) in Izmir, Turkey, the full spectrum of expression of psychosis was categorized across 5 groups representing (1) absence of psychosis, (2) subclinical psychotic experiences, (3) low-impact psychotic symptoms, (4) high-impact psychotic symptoms, and (5) full-blown clinical psychotic disorder and analyzed for continuity and discontinuity in relation to (1) other symptom dimensions associated with psychotic disorder and (2) proxies of genetic and nongenetic etiology. Results were tested for linear and extralinear contrasts between clinical and nonclinical and between disorder and nondisorder expression of psychosis. Demographic variables, indexing premorbid social adjustment and socioeconomic status, impacted mostly linearly; proxy variables of genetic loading (more or more severely affected relatives) impacted in a positive extralinear fashion; environmental risk factors sometimes impacted linearly (urbanicity and childhood adversity) and sometimes extralinearly (cannabis), occasioning a disproportional shift in risk at the clinical disorder end of the spectrum. Affective symptoms were associated with a disproportionally higher risk below the disorder threshold, whereas a disproportionally higher risk above the threshold was associated with psychotic symptom load, negative symptoms, disorganization, and visible signs of mental illness. Liability associated with respectively affective and nonaffective symptom domains, in interaction with environmental risks, may operate by impacting differentially over a quasi-continuous extended psychosis phenotype in the population.