ResearchPod Summary
As the classic Cognitive Reflection Test (CRT) suffers from widespread item familiarity, its ability to measure reflective reasoning—the tendency to override intuitive, System 1 responses—has been compromised. This study introduces the Behavioural Reflection Test (BRT) and a new, low-exposure bespoke CRT (bCRT) to determine if these tools can better capture how reflective individuals navigate complex, real-world decisions involving moral and epistemic stakes.
Researchers recruited 473 online participants to complete the new four-item bCRT alongside the traditional CRT2. The BRT required participants to provide open-ended justifications for decisions in two scenarios: a social conflict involving a restaurant booking and a health-related decision regarding medication. The study analyzed whether bCRT scores predicted evidence-based choices, high-quality information search strategies, and specific linguistic markers (using LIWC) associated with reflective thinking, such as risk-attentiveness and emotional engagement.
The bCRT demonstrated superior predictive power compared to the legacy CRT2. Higher bCRT scores were significantly associated with evidence-weighted decision-making, a greater willingness to enforce social norms, and more sophisticated information-seeking plans. Linguistically, higher bCRT scorers used more risk-related and emotionally engaged language. Crucially, these associations remained robust after statistical correction, whereas the familiarity-adjusted CRT2 failed to recover these behavioral signatures, suggesting that the bCRT is a more effective tool for assessing reflective tendencies in ecologically valid contexts.
This research provides a practical, time-efficient alternative to legacy cognitive measures. By moving beyond decontextualized brainteasers, the BRT offers a way to observe how reflective reasoning manifests in everyday life, such as how people evaluate health information or manage social obligations. It demonstrates that reflective thinking is not just a trait measured by puzzles, but a style of engagement that shapes how individuals interact with the modern, information-dense environment.
AI-generated third-party summary by ResearchPod. Not official content or an endorsement by the paper authors or affiliated organizations.