Perceived Variance: A Unified Model of Hope, Entrapment, and Manufactured Certainty
Year: 2026
Type: preprint
Abstract: In this paper I propose perceived variance (PV), a person’s subjective estimate of the probability that their individual situation will change, and actual variance (AV), the objective probability of that same change, as a unified account of three literatures that have developed independently: clinical forms of entrapment and suicidal hopelessness, relational false hope in abusive partnerships, and structural entrapment produced by low autonomy environments. Each literature has treated its central phenomenon as a distinct and independent mechanism. I argue instead that all three reduce to the sign and magnitude of a single calibration error, Δ = PV − AV. An accurate low PV under low AV is the signature of clinical entrapment and is dangerous because it removes any perceived route to relief. An inflated PV under low AV sustains engagement with a static, harmful environment and characterizes relational entrapment with high fidelity. An accurately low PV under a deliberately engineered low AV characterizes structural entrapment, where individual level intervention fails because the calibration error being treated does not truly exist. A low PV under a high AV, which I call manufactured entrapment, is the case in which hopelessness is a cognitive distortion and is in principle correctable. This framework generates a falsifiable and clinically counterintuitive prediction: the same intervention, increasing a person’s expectation that change is possible, should reduce suicidal ideation in the clinical cell while increasing harmful attachment in the relational cell. I formalize PV and AV, propose a measurement approach, extend the model to a time varying process capable of producing hysteresis between environmental change and belief update, and connect the framework to reward prediction error accounts from computational psychiatry. Finally, I close with the clinical, relational, and organizational interventions the model implies, and the boundary and ethical conditions under which each should and should not be applied.
Source: OSF Preprints (OSF Preprints)
Author: Thompson, Aiden
Language: English
Subfield: Applied Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7224092222