[spookylukey]

Working out whether someone is smart or not on the basis of how an entirely informal conversation goes makes you extremely vulnerable to biases.

Read "Thinking, Fast and Slow" by Daniel Kahneman and you will become much more suspicious of sentiments like "I think it’s pretty easy to tell whether someone’s smart in casual conversation". We are actually extremely bad at guessing people's competences on first encounter, and very prone to cognitive biases such as the halo effect.

With no objective measures, it seems this method is likely to be subject to massive bias towards people that you just get on with or like for some reason.

[arsenide]

Do you think, perhaps, after being aware of these biases it makes it easier to determine someone's "smartness" based on a sequence of interactions as described in the article?

[learnstats2]

Kahneman explicitly studies this and concludes that it makes no difference. The biases apply just as much to people who are aware of them.

[jwdunne]

Side anecdote: I get the feeling I'm judged as less capable mentally in person. I wonder whether it's because of my appearance, my voice or perhaps because I am less capable mentally. Confusing world to live in when I get such conflicting information.

I don't care either way. It doesn't get in the way of my curiosity, which takes me on plenty of rides regardless of my ability.