Building Cognitive Resilience to Vaccine Misinformation Using Conversational AI: Evidence From a Multi-Country European Randomised Trial
Building Cognitive Resilience to Vaccine Misinformation Using Conversational AI: Evidence From a Multi-Country European Randomised Trial
This multi-country trial evaluates an AI chatbot designed to counter cognitive biases and inoculate parents against vaccine misinformation, embedded within a large survey across five European countries (n=5,000). Scaled from a 2026 single-country study, outcomes compare behavioural and attitudinal responses against standard-of-care materials (e.g., UNICEF infographics) to identify effective, scalable misinformation interventions.
The aim of this multi-country trial is to identify and quantify the effectiveness of inoculation strategies against vaccine-related misinformation. Building on a single-country study implemented in early 2026 [NCT07488962], the study evaluates an AI chatbot designed to counter cognitive biases, embedded within a larger survey to enable integrated assessment of engagement, behavioural responses, and system performance at scale.
The study will be conducted across five European countries-the United Kingdom, the Netherlands, Croatia, Latvia, and Slovakia-with a total sample of 5,000 parents or caregivers (n=1,000 per country). Participants will complete a structured survey assessing childhood vaccination behaviours, vaccine confidence, trust in information sources, exposure to misinformation, and relevant psychological and sociodemographic factors.
Participants will be randomly assigned to either the intervention or control arm. The intervention arm will receive the AI-based chatbot, while the control arm will receive standard-of-care materials-such as national or international health institute communications (e.g., UNICEF infographics)-that address the same misinformation messages, enabling a direct and rigorous comparison.
The primary outcome is parents' ability to correctly distinguish true from false childhood vaccine statements. Quantitative analyses will include pooled and country-specific models to identify determinants of vaccine confidence and estimate intervention effects. Findings will clarify drivers of declining vaccine confidence and assess whether a scalable AI-based chatbot can reduce vulnerability to misinformation and support informed vaccination decisions.
Inclusion Criteria:
Exclusion Criteria:
Leesa.Lin@lshtm.ac.uk+44 (0)20 7636 8636
Kana.Kobayashi@lshtm.ac.uk