Confirmation Bias in UFO Investigations

This simulator models a specific, well-documented failure mode in pseudoscience investigation: confirmation bias in evidence weighting. A stream of ambiguous, mostly-explainable field reports arrives over time, each carrying a small true signal toward or away from the "genuine anomaly" hypothesis. A perfectly rational investigator would update belief in direct proportion to that signal. Here, instead, each report's influence is distorted by how well it agrees with the investigator's current belief — confirming reports are amplified, disconfirming reports are discounted — so belief drifts toward an extreme even when the underlying evidence, averaged honestly, is genuinely weak or neutral. Adjust confirmation-bias strength, evidence ambiguity and arrival rate to see how quickly and how far the feedback loop runs away.