There's an old joke about various animals — sometimes about cows, or foxes, or whatever — that says, basically, they're simple. They see something and put it into one of four categories: 'things that can eat me', 'things I can eat', 'things I can fuck', and 'things I can ignore (i.e., rocks)'.
For cows in particular, this is untrue. They primarily divide the world into two foundational categories: 'things I can push over' and 'things I can't push over'. This may seem like a strange and not-very-useful categorisation system to you, but your opinion isn't going to carry much weight for most cows. After all, they could push you over quite easily, so what could you possibly have to contribute to the discussion?
Much like the human mind is a highly evolved system for throwing objects (with some unplanned and largely useless side functions, such as 'written language' and 'economics'), the cow mind is a precision-engineered machine that effortlessly calculates shoveability, pushability, tippability, and (most importantly) heft.
Humans who have worked with cows for a short while will likely say something like "That's nonsense. These cows are terrible at working out what they could push over: they go exactly where this tiny ten-pound yappy sausage in a fur coat wants them to go, despite the fact that they could easily push it over."
Those humans are in no place to talk about nonsense, however, because they are just throwing machines with delusions of economy that do things like breed a bunch of ten-pound yappy sausages to have 80% less leg and 180% more neuroses than their ancestors, and then give them day jobs moving thousands of kilograms of cow from place to place. The fact that this arrangement actually works quite well (except for the cows, in the long term) makes it no less of a nonsense idea.
People who have worked with cows for a longer time will likely say something like "Well, I suppose sometimes a cow does push over one of the dogs instead of letting itself be herded, but it's not a behaviour that spreads, on account of how we promptly send those particular cows to be slaughtered. Now stop bloody talking about it; you'll give the young 'uns ideas."
This selective pruning of entirely reasonable (to the cow) pushing behaviours has, over generations, produced a very keen sense in most farmed cow breeds for the aforementioned idea of 'heft', which is not a measure of any physical characteristic but a purely instinctual estimation of how likely pushing something over is to have consequences. Cows do not like consequences and are therefore, by now, very good at calculating heft.
Prey animals intuitively understand the phrase "just because you can, doesn't mean you should" and in this sense (among many others) are vastly more sensible than humans. It is one of the great injustices of the universe that this sensible nature has not paid out for cows at the species scale (given the industrial meat-slaughter arrangement), but they also do not have to do economies basically ever, so one might call that a draw in the grand scheme of things.
In terms of the grand scheme of things, actually, it may surprise you to learn that most cows are, in a technical sense, religious. They are fully aware that at least one god exists. Through a complex series of pushing-related calculations (kind of like algebra, but by and for cows and therefore involving more rumination), they have largely come to the conclusion that this god isn't very important and have firmly categorised it as 'push-overable'.
It may surprise you yet further to know that they are entirely correct. The last time this god tried to materialise in the mortal world, it was comprehensively pushed over and trampled by a herd of Devon cattle and has spent the last few hundred years having a bit of a sulk about it. As far as the cows are concerned, this serves the god right for not taking a more robust and less push-overable form (such as one with four legs and at least four stomachs) and generally being a brief-chewing horizon scanner.
It should be noted that this phrase (brief-chewing horizon scanner) is quite descriptive if you have a cow-adjacent perspective. One who insufficiently chews their food is overly short-sighted, causing future problems while rushing for no real gain. One who spends too much time looking at far-away things is overly long-sighted, allowing worries to distract from the important things in front of them (usually, grass). For cows, this term, with its combination of both short- and long-term foolishness, carries intrinsic implications of being easily push-overable and functions as their collective name for humans, gods, and large flightless birds.
Some more edge-case interpretations of cows suggest they may also conceptualise a third pushing-related category (tentatively labelled by researchers as 'things that can push me over'), but studies have been few and inconclusive as this category would comprise remarkably few things for most adult cows (i.e., construction equipment, military hardware, and some (but not all) tornadoes), making most potential tests unethical.
For those wondering why all those rural stories about drunken cow-tipping don't count for more here: they're generally not true. If you are in any doubt, you are invited to try and upend a sleeping cow at night while you're sober.
After you've come to a more comprehensive understanding of your categorisation regarding push-overableness (and laundered all the hoof marks out of your clothes), you might then be in a better position to understand how "tipping over one thousand kilograms of inertia and stubbornness on four legs while in the dark", which is already a fool's errand, would be even harder while drunk and uncoordinated.
In any case, cows have generally refused to cooperate with tip tests, scientific or otherwise, and show no inclination to fill out surveys, questionnaires, or rating scales. This should be taken as another indication of their sensibleness.