You contend that the point of the difficulty in the levels is to "make the player think." Fair enough, but much of the "difficulty" in the game comes from questionable design principles.
The play control is not just an exercise in low-friction platforming: it is buggy(if I wanted to play a low friction challenge platformer, I would opt to play Pugsley's Scavenger Hunt...). This is actually noticeable from the first two stages on! If the player decides to bypass the first target platform in the stages, there is a good chance that the back of the character's sprite will "clip" against the side of the platform. because of this clipping, it will lose all of its momentum and careen downwards, missing the intended target. This has a good chance of unnerving the player during the third stage: if he clipped against the side of the platform in the first two stages, what's to prevent him from clipping against the platform as he slides off of it using his momentum ? Because the first two levels are the ones that will be most frequently visited, this glitch will inevitably serve to turn frustrated players into ones who are frankly apt to ragequit the game entirely.
Beyond this , though, the collision detection is awkward. Spikes have notably larger hitboxes than their actual sprites depict, but the player character can fall off of a block even when a good portion of its sprite is resting atop of it , something that makes level 10 absolutely dreadful. This isn't a case of the game being "merciless." I Wanna Be The Guy, long considered to be the pinnacle of rage-inducing platforming, featured spikes with hitboxes that absolutely matched their sprites, something a few of the segments in the game actually relied on.
Level 8 is particularly illustrative of the flaws in the game. The level features a sloped pattern, something that has not yet been encountered up to that point. Hanging overneath it is a blue wall that separates it from some spikes, which were intended to be a threat to the player earlier on in the stage. When the player encounters the slanted tile, he has no way to predict how it will operate. Will the player character shift to the left as it's resting on it? can it jump off of the platform? will it bounce off of it on contact? Will it stay in place on it? The player cannot possibly know this at that point. There goes "using your brain." Because your instructions guide informed us that the blue repel the block, one could assume that jumping upwards would be safe, and would not result in a collision with a spike. Instead, the box simply passes through the zone, defying past experience and striking against the spikes. This is fundamentally bad level design, and it punishes the player for actually remembering what blue zones are -supposed- to do.
You really should use level transitions, by the way . It's clear from the start of level 9 that you recognize the flaws in the game instantly shifting from level to level with no intervening period to prevent the player from careening off of the platform. You put a block on the starting platform to prevent the player from falling off. Nice as that is, you really need legitimate level transitions. The lack of transitions actually results in a nasty situation at the start of level 11: due to the disproportionately large hitboxes of spikes and doors, the player will actually spawn on top of a spike most of the time! This sort of situation should never occur, especially in a game that lacks any checkpoints whatsoever. Not only does it eject the player back to the start of the game...it also challenges the player's faith in the game. On top of that, the difficulty balance verges on the bizarre. Notably, level 6 is trivially easy, while level 4 is ,from experience, the most challenging of the first eight levels. Spikes are positioned over death pits in positions where they could not possibly do any further harm to the already doomed block without rhyme or reason. As a way of standing out on a limb and showing your ideas, this is fine. As a product, this is insufficient.