There is certainly truth to the claim that the brain abstracts over the raw sensory input. Reading, for example, would be a much more intractable problem if you

There is certainly truth to the claim that the brain abstracts over the raw sensory input. Reading, for example, would be a much more intractable problem if you had to do it in terms of photons hitting your retina instead of first abstracting the patterns on your retina into shapes, letters and words. But the leap from there is some abstraction going on to reality is nothing like the abstraction seems not [well] justified to me. And we can check it, we can build all kind of sensors and machines that interact with the world out there, cut out consciousness in the process, and see if the sensors report a world very different from how we perceive it. And they don't. There is of course a complication to this, constructing those sensors and machines and looking at what they are doing always has to pass through our perception, so maybe we accidentally build sensors that yield the same distorted version of reality as our brain does or the sensors are actually reporting a very different world but when we look at the sensor readings they get distorted to match the perception through our senses. But given all the ways and angles from which we can measure things and then transform and represent the results before perceiving them, I have a hard time imagining how it could always be coherently distorted to match our normal perception without showing any inconsistencies. And our perception and intuition is of course not perfect. Time and space, for example, are different from what our perception and experience tells us, but we managed to measure it and uncover the differences. But even in this case our perception and intuition are not really wrong, they are a very good approximation for everyday situations and only shows noticeable discrepancies in extreme circumstances.