Learning from many: Nocebo effects following exposure to multiple social models

Observing others experience side effects can induce negative expectations and exacerbate symptoms via socially induced nocebo effects. While most experimental work has focused on dyadic model-observer interactions, real-world exposure to symptom information typically involves multiple models reporting symptoms of varying severity. The present study examined whether exposure to multiple social models reporting cybersickness induces nocebo effects, and whether the timing of an extreme symptom report within a sequence of models influences expectancy formation and symptom experience. Drawing on learning and memory research on order effects, we tested whether encountering a highly salient adverse exemplar early, middle, or late in a sequence moderates socially induced nocebo effects. One hundred and seventy participants observed five social models undergo a virtual reality (VR) rollercoaster in Zoom, with four reporting mild cybersickness and one reporting severe cybersickness. The position of the model reporting severe cybersickness was manipulated between groups as either first (primacy), middle, or last (recency). Observing symptom modeling resulted in greater cybersickness compared with control, and this effect was mediated by participants' expectations of cybersickness. However, the position of the severe model within the sequence did not influence symptom outcomes. These findings demonstrate that socially induced nocebo effects occur following exposure to multiple models reporting mixed symptom severity. While expectancy plays a central role in mediating these effects, the timing of exposure to an extreme symptom model does not appear to further amplify symptom experience. Interventions targeting expectancy are critical for reducing the burden of socially induced nocebo effects.