PHIL38 reading quiz biased assimilation

Biased assimilation

Tendency to accept evidence that confirms one's prior beliefs at face value while critically scrutinizing evidence that disconfirms them

Attitude polarization

The gap between opposing groups' views widens (rather than narrows) after both are exposed to the same mixed evidence

Polarization hypothesis

Belief polarization will increase, not decrease, when proponents and opponents of opposite viewpoints assimilate mixed or inconclusive findings

Why might mixed evidence be expected to moderate views, according to a logical view?

Mixed evidence should logically produce some moderation (or at worst be ignored), but the authors predicted the opposite

What social issue did the study use, and why?

Capital punishment and its deterrent effect on murder; chosen because people hold strong views, it is heavily publicized, and both sides cite the same inconclusive correlational research

How were proponents and opponents defined in the study?

Proponents favored capital punishment, believed it deters, and thought research supported them; opponents opposed it, doubted deterrence, and thought research supported them

Why couldn't the researchers measure polarization with simple before-after difference scores?

Participants were already at or near the ends of the selection scales (ceiling effect), so they used reported change, cumulative change, and running-record measures instead

Key finding on study evaluations (Table 1)

Proponents rated the prodeterrence study as better conducted and more convincing; opponents rated the antideterrence study as better and more convincing, even though the procedures were identical

Key finding on overall attitude change

After both studies, proponents reported being more in favor of capital punishment and opponents less in favor, with the same pattern for beliefs about deterrence (net polarization)

Rebound effect

After briefly shifting toward disconfirming results, participants reverted to (and in 23% of cases exceeded) their original views once they saw methodological details and critiques; no comparable rebound followed confirming studies

Why did the authors say biased evaluation is not entirely irrational (normative issue)?

Prior knowledge legitimately shapes skepticism (e.g., doubting miracle cures or a speed-of-light method that gives the wrong answer); the real error is using biased-processed evidence to bolster the very theory that justified the bias

What was the subjects' main 'sin' according to the authors?

Not biased processing itself, but using already-biased evidence to strengthen their initial belief, making their hypotheses effectively unfalsifiable

Main limitation: 'Polarization: real or merely reported?'

Participants may have reported what they thought a rational response would be rather than real private change; future work could use less extreme samples or measure willingness to act

Belief perseverance

Beliefs survive both the complete discrediting of their original evidence (earlier research) and the addition of nonsupportive evidence (this study)

Top-down vs. bottom-up processing (implication for attribution models)

People are not impartial, data-driven processors; they interpret data through the hypotheses and implicit theories they bring, so models need a top-down component

Implication for social science and policy

Researchers cannot expect objective data on burning social issues to produce rationality or consensus; inconclusive research will often fuel rather than calm debate