So perhaps that isn’t the best litmus test? I wonder if it’s written down that should work as part of the assignment.
Breaking default browser behavior because you find it unintuitive is generally a bad idea.
Now if you’re dealing with something that isn’t really a web form—as in you’re overloading input fields for some interactive non form like behavior—then I can see it.
In my case, if the person had some well thought out reason for doing it, I might let it slide. But the vast majority of times I’ve seen it, it’s because the person doesn’t even know how to use forms. Not understanding the underlying technology at at least a very basic level is a strike against you in my book.
Again, it’s not that I disagree with most of your premise, but in fact, as the comments (not just mine) show - people have issues with submitting too early, so it can be a good decision to break that depending on context. Fewer errors. Which is why I brought up how the problem is presented to the candidate. You might be filtering out people based on something less universally accepted and understood than you think.
If you accidentally break it because you don’t understand it, that’s a strike against you in my book. If you consciously break it for a good reason and can coherently defend that reason, that’s a different story.
At least you'd have to get a sample big enough to make a statistically significant conclusion.
And the burden of proof is on you, because you are changing the default, not the other way around.
Think of any other products, they all have established default behaviours, yet it's also very easy to find someone who's never used a product before and finds it "unintuitive".
And intuitiveness is only one part of usability anyway.