The only (limited) success I've had moving emotionally-held models of belief is specifically exploring what signals would be present if the competing models wer

The only (limited) success I've had moving emotionally-held models of belief is specifically exploring what signals would be present if the competing models were incorrect. By focusing on falsifiability and reconciling it with reality, I've _sometimes_ managed to have my counter-party downgrade their argument from scientific surety to human behavioural expections in some key ambiguous situation.