flaccon:
Pay close attention to the bolded part of Pixel42's message.
When skeptics tell you you are experiencing paredolia, they are not insulting you and telling you you are stupid or crazy. They are merely improving your vocabulary. It would be really irritating to have to write "the well-known tendency of the human brain to find patterns even where there are none" over and over again when talking about human perception. So a word was invented to cover the concept. That word is paredolia.
It's extremely common. It allows us to 'see' faces and bunny rabbits or turtles in clouds. And whether we 'see' it or not is influenced by our expectations. You might not see a particular cloud as being a turtle until someone points out its turtle-ness to you, and then, ping, it jumps right out at you.
Think of the smiley emoticons people use in text. A colon : followed by a close parenthesis ) is
not a smiling face, and a semicolon followed by a close parenthesis is not a winking smiling face. But paredolia allows you to
perceive it as such.
Even the graphic emoticon I'll get here on the forum by typing

isn't really a face, is it? A yellow circle, two black dots, and a curved line. But paredolia lets/makes us see it as a face.
In my job, I have to pick zeros and ones sent with a very small amplitude out of a large signal a very noisy environment. To do that, we use filters that amplify just the part of the signal that the zeros and ones are sent in, and reduces everything else. Because of this,
I receive zeros and ones even when I know there's nothing being transmitted. That isn't because the filters are bad. It's because the filters do their job very well. The only waveform that comes out of the filter is something that could be data.
So I have to use additional tests to see if it actually is transmitted data.
Paredolia is like that. We live in a world that presents us with a lot of different stimuli. We've evolved to filter out all the stuff that it isn't important, and to amplify the bejeezus out of anything that
might be. The selection process is very good, but as with the signals in my job, that also means we'll get false positives. Data where there isn't any. Faces where there's just punctuation.
In my job, I use things like start patterns that can only happen once in a transmission, and additional 'stuff' bits that occur periodically to break up any accidental occurrences of that start pattern, and additional bits at the end that are mathematically generated from all the data bits in the transmission. If all those things match, there's a pretty good chance the data I receive is correct. It's not certain, but it's very likely. Without the checks, it
could be right, but it's more than likely wrong, because if it was likely to be right, I wouldn't need all that filtering in the first place.
Skepticism is kind of like that additional checking beyond just filtering.