You may wish to read Schrödinger's original article. It is about whether quantum mechanics could possibly be a complete description of reality. He considers two options: either that the wavefunction is a statistical description of an ensemble of possible states, or that the properties of microphysical systems are objectively undefined prior to observation. The cat shows up (in part 5) as a demonstration that the second option is absurd. This is the part where it may sound (to a modern reader, versed in MWI-think) that he is talking about a physical process of wavefunction collapse, but he's not. He's discussing a situation where the cat goes from "neither alive nor dead" to one or the other, at the moment of observation. At this stage, the wavefunction is not being proposed as a description of the cat's physical state.
Parts 7 through 9 are where he addresses the Copenhagen interpretation, and contrasts it with wavefunction realism: "psi-function as expectation catalog", versus "psi-function as description of state".
If the probability distribution collapses, why is that not a physical process? Is it an immaterial, spiritual probability distribution?
Flip a coin but don't look at it. Then look at it. The probability distribution of the coin just "collapsed". The collapse of the wavefunction in the Copenhagen interpretation is the same thing. A wavefunction is like a prior; you adjust it on the basis of information acquired.
Suppose you believe in the Copenhagen interpretation of quantum mechanics. Schroedinger puts his cat in a box, with a device that has a 50% chance of releasing a deathly poisonous gas. He will then open the box, and observe a live or dead cat, collapsing that waveform.
But Schroedinger's cat is lazy, and spends most of its time sleeping. Schroedinger is a pessimist, or else an optimist who hates cats; and so he mistakes a sleeping cat for a dead cat with probability P(M) > 0, but never mistakes a dead cat for a living cat.
So if the cat is dead with probability P(D) >= .5, Schroedinger observes a dead cat with probability P(D) + P(M)(1-P(D)).
If observing a dead cat causes the waveform to collapse such that the cat is dead, then P(D) = P(D) + P(M)(1-P(D)). This is possible only if P(D) = 1.
If you don't say that only conscious agents can collapse waveforms, then you have to agree that something in the box collapses the waveform as seen from inside the box, while it's still uncollapsed to Schroedinger. And Schroedinger's opening the box collapses that waveform for him; but it is still uncollapsed for someone outside the room. This seems like it might be equivalent to many worlds - all possibilities already exist; you just haven't chosen which one you're going to access until you open the box.
But if you do say that only conscious agents can collapse waveforms, then it's something about their mental processes that does the collapsing. This could mean their beliefs matter. And then, the cat is always dead.
ADDED: People. Read the entire post before responding. I am not claiming that the cat is always dead. I am not claiming that consciousness collapses waveforms. I am claiming that there are only 2 known alternatives:
If you can't produce another alternative, and you don't believe in many-worlds, you owe me an upvote.
Finally, this post is supposed to be fun! You are crushing all whimsy and playfulness on LessWrong when you pile downvotes like bricks on anything playful because it does not provide a complete and satisfactory resolution.