The problem with trusting your perceptions without testing them is outlined nicely by Michael Shermer (yet again) here.
At least we’re able to figure out that our perceptive powers are suspect! Of course, you could always argue that if we can’t trust our percpetive powers, then how do we know we’re right to not trust our perceptive powers.
…riiiight. And luckily you came along with your “other ways of knowing” to save the day!!
Or, we could just acknowledge that tested results are the only results worth paying any attention to, and that we gravitate to bad explanations in the face of none (or partial) ones.
Put on a brain condom and get on with it already! Safety first people! 
