The short answer is, you can't. At least not about anything that you can't test for yourself by personal experience.
The longer answer is to ask why you care about whatever particular aspect of "reality" you are thinking about testing. What difference would it make to what you choose to do, if reality were one way rather than another? If the answer is that it would make no difference, then the correct thing to do is to not care how reality is in that respect. Have no opinion at all. (There's an old engineer's joke about running tests. Before you run a test, ask yourself two questions: What will I do if the test passes? What will I do if the test fails? If both answers are the same, don't bother doing the test.)
The problem is that our minds did not evolve to be comfortable with having no opinion about something. Our minds evolved to seek answers, not to leave questions hanging. That probably made sense in the hunter-gatherer environment in which we originally evolved, but it doesn't make sense in our world now. There are simply too many questions, too much information, and too little time for anyone to check it all, and there are no "trusted" sources of information we can turn to to just tell us the answers. But it's very difficult to accept having unanswered questions; our minds keep sending us alarm signals even though we might have convinced ourselves intellectually that we should leave those questions unanswered. So many people end up accepting some answer even if it's wrong, and even if the answer makes no difference to anything they actually do.
The short answer is, you can't. At least not about anything that you can't test for yourself by personal experience.
The longer answer is to ask why you care about whatever particular aspect of "reality" you are thinking about testing. What difference would it make to what you choose to do, if reality were one way rather than another? If the answer is that it would make no difference, then the correct thing to do is to not care how reality is in that respect. Have no opinion at all. (There's an old engineer's joke about running tests. Before you run a test, ask yourself two questions: What will I do if the test passes? What will I do if the test fails? If both answers are the same, don't bother doing the test.)
The problem is that our minds did not evolve to be comfortable with having no opinion about something. Our minds evolved to seek answers, not to leave questions hanging. That probably made sense in the hunter-gatherer environment in which we originally evolved, but it doesn't make sense in our world now. There are simply too many questions, too much information, and too little time for anyone to check it all, and there are no "trusted" sources of information we can turn to to just tell us the answers. But it's very difficult to accept having unanswered questions; our minds keep sending us alarm signals even though we might have convinced ourselves intellectually that we should leave those questions unanswered. So many people end up accepting some answer even if it's wrong, and even if the answer makes no difference to anything they actually do.