Bear tests were probably more an early proof of viability than anything else. The bears survived, proving that humans wouldn't be completely annihilated.
Animal testing cruelty wasn't really being thought about heavily in post WWII military aerospace development. It's unfortunate but a historical footnote reminding us of the importance of proper testing.
Fitting it in ejection seat might be another challenge but I'm sure they would have developed something before moving to live subjects.
Considering that the endpoint is machines killing actual people, it would be the height of hypocrisy in my books for them to care about "animal testing cruelty".
(Seriously, who said it has to be exactly "like" it? We use crash test dummies for testing car crash behavior, and they're nothing "like" humans either).
Maybe people were misconstruing "horrible science" as something other than low quality science? HN while being a seething pit of criticality, it has a low tolerance of anything that questions science or the march of technology.
It would only be "low quality" if it didn't fit their purpose. Which was some rough estimate, not to get the final word of whether humans survive an ejection.