The Personalized Bedtime Story Nobody Has Tested: Why generic sleep stories may have failed their first active-controlled test — and how a theory-designed, personalized, screen-free reading intervention could be tested next

Narrated "sleep stories" are a flagship feature of consumer wellness apps, yet the first active-controlled randomized trial of app-delivered bedtime stories (USleep, 2026; N = 495) found no significant benefit over a digital control for sleep disturbance (Hedges' g = 0.12). This paper argues that the null result marks the limit of the generic format rather than of bedtime reading itself. Drawing on evidence on pre-sleep cognitive arousal, imagery-based distraction, worry concreteness, screen-light effects, and the psycholinguistics of arousal, it specifies an intervention that remains untested: a theory-designed, personalized, screen-free bedtime reading text, evaluated with multi-night sleep-onset latency as the prespecified primary endpoint. The paper operationalizes a low-arousal text specification, presents a sample stimulus with norm-based scoring, proposes a brief preference-based personalization intake derived from validated constructs, and outlines a preregistered three-arm randomized trial (personalized designed text vs. non-personalized designed text vs. expectancy-matched neutral text) with falsifiable predictions, prespecified safety outcomes, and open materials. The proposal is framed as an adjunct and access bridge to first-line cognitive behavioural therapy for insomnia, not a treatment substitute. This is a theoretical/methodological contribution (a research proposal and intervention specification). It does not report new empirical data; it synthesizes existing evidence and specifies a falsifiable trial design with open materials, intended for preregistration and testing by any research group.