Better safe than sorry? Exploratory associations between fear-learning processes and later paranoia in threat-exposed adolescents in standardized Virtual Reality environments
Longitudinal study reveals altered fear learning linked to subsequent paranoia in threat-exposed adolescents, highlighting an adaptive but costly defense.
Key Points
To examine whether threat–safety discrimination and fear generalisation prospectively predict virtual-reality-assessed state paranoia 18 months later, and whether childhood threat exposure and gender moderate these pathways.
120 adolescents aged 12–16 completed a classical conditioning and fear generalisation task alongside childhood threat exposure assessments at Wave 1.
At Wave 2 (18-month follow-up), 97 participants completed a standardized virtual reality social environment measuring state paranoia.
Linear regression models evaluated main effects of threat–safety discrimination and fear generalisation, moderation by childhood threat exposure, and three-way interactions with gender.
Neither threat–safety discrimination (β = 0.14, p = .198) nor fear generalisation (β = −0.07, p = .530) showed a significant main effect predicting paranoia at Wave 2.
Childhood threat exposure significantly moderated the link between threat–safety discrimination and paranoia (β = 0.21, p = .035), with higher threat exposure and better discrimination predicting elevated paranoia.
A significant interaction between fear generalisation, threat exposure, and gender emerged (β = 0.53, p = .044), showing higher fear generalisation predicted lower paranoia in boys (F(1, 84) = 7.34, p = .008) but not girls.