Biased Assimilation

Attention, Perception, and Assimilation Biases

Following Bruner (1957), legions of researchers have demonstrated that, in disambiguating stimuli, people “go beyond the information given.” They perceive things as they have been led by experience or suggestion to expect them to be, and their perceptions are further biased by their hopes, fears, needs, and immediate emotional state. Investigators concerned with attitude change similarly challenged simple learning theory formulations by arguing that the recipients of persuasive arguments often prove to be rationalizing rather than rational animals, and as such are influenced less by logical rigor or objective evidence than by the interests and preconceptions that they bring to their task (see Katz, 1960).

Disagreements and misunderstandings about capital punishment, abortion policy, how to achieve peace in the Middle East, or the wisdom of trying a deposed political leader for war crimes, thus become inevitable. Even when (indeed, especially when) people earnestly attend to the facts and arguments offered by those on the “other side,” their opinions become even more polarized. This “polarization” reflects the tendency for partisans to accept at face value arguments and evidence congruent with their interests and beliefs, while critically scrutinizing arguments and evidence that threaten those interests and beliefs (see Edwards & Smith, 1996; Ross & Lepper, 1980). Moreover, we suggest, such biased processing of information fosters harsh evaluations of individuals on the other side whose perceptions and arguments, in the eyes of the opposing partisan, appear biased and self-serving.

Intergroup enmity can also arise from simple availability and representativeness biases (Kahneman & Tversky, 1973; Tversky & Kahneman, 1974) – biases with which readers of this book are likely to be quite familiar. Once again, illustrative examples are commonplace. Estimates of the severity of social ills like poverty, unemployment, or discrimination are bound to be influenced by availability biases reflecting where one lives, whom one talks to, and what one reads. Our media-based notions of what a “representative” INS agent, follower of Islam, Texan, lesbian, or homophobe looks and acts like influence our expectations, and thus help determine which claims about particular group members we find credible or noncredible. They also determine whether accounts about individual deeds by group members are treated as stereotypical or dismissed as unrevealing aberrations. The present thesis is that blindness about the role that such biases play in shaping our own political views, and a penchant for seeing self-serving or ideologically determined biases in other’s views, exacerbates group conflict.