To discern whether a specific opinion is true, is it important to engage in discussions with those who hold that opinion?
In what cases it is, and in what cases it is not? What is the demarcation line?
(Specific examples in a comment to the OP).
Philosophy Stack Exchange is a question and answer site for those interested in the study of the fundamental nature of knowledge, reality, and existence. It only takes a minute to sign up.Sign up to join this community
It's too vague. Why would a clear demarcation be expected? How can you reasonably generalise, or list criteria all fields & cases?
Discussion with others, as opposed to looking up references or using reasoning, has a couple of specific strengths: non-explicit queues like strength of conviction; being challenged on your own views/disagreements with; and perhaps most importantly glimpse other mindsets/worldviews. So where they are important, discussion will be important. In topics of historical record or scientific consensus, not so much, and unless you are a specialist in the field it won't be you that should be in the discussion anyway.
Discussions can vary in quality and usefulness. A Socratic dialogue, with mutual respectful inquiry into truth without preconceptions, is likely always useful. A partisan debating match, is unlikely to change anyone's mind about the truth, but it may help distill pursuasive arguments, and share them. An absolute minimum requirement, should be to respect society norms & standards, eg no resorting to violence. In politics the norms may be explicit in law, and tolerating a lack of commitment to them in a candidate, risks eg insurrection by a candidate unable to pursuade a majority of 'their truth'.
That just leaves what 'true' is, which is discussed here: Why is a measured true value “TRUE”? You say specifically "To discern whether a specific opinion is true", implying it's something not easily verified, which may be unfalsifiable, and could be about something controversial in a partisan way. You can still apply reasoning. And look up verifiable subclaims. Cognitive biases & motivated/post-hoc reasoning may show up too. But overall, it's about pursuasiveness, which can be quite subjective.
We don't all live in the same 'world'. Jonathan Haidt found through research, different groups have different moral foundations. He linked the origins of that to different agricultural practices, herding cultures like 'cowboys' or Afghan hill tribes, vs those in river valleys who sow & harvest together. The correct moral foundations for an opinion, may depend on what future we expect - eg Mad Max vs Star Trek futures.