We cannot be confident that "science usually" overcomes faulty models. The best we can say is that science has often been seen to succeed at this, in well publi

We cannot be confident that "science usually" overcomes faulty models. The best we can say is that science has often been seen to succeed at this, in well publicized cases. Many less visible fields might never overcome their biases. Usually a field cannot correct course until a whole generation trained on a false premise dies or retires. Economics is a field that has been particularly resistant to correction, but is far from alone. Geology and statistics are recovering from a similar handicap. As Max Planck is often quoted, "Science advances one funeral at a time." Often vindication is finally delivered only after all the opponents are dead, and the ultimate victor has retired from a career blighted by them. Probably much more often people are driven out of the field and never vindicated. Lynn Conway was driven out of computer architecture (where she invented out-of-order execution in the '60s, thus long delaying that advance) before finding success many years later in VLSI chip design methods.