Coherence plays an important role in psychology. In this article, I suggest that coherence takes two main forms in humans’ cognitive system. The first belong to ‘system 1’. It relies on the degree of coherence between different representations to regulate them, without coherence being represented. By contrast other mechanisms, belonging to system 2, allow humans to represent the degree of coherence between different representations and to draw inferences from it. It is suggested that the mechanisms of explicit coherence evaluation have (...) social functions. They are used as means of epistemic vigilance—to evaluate what other people tell us. They can also be turned inwards to examine the coherence of our own beliefs. Their function is then to minimize the chances that we are perceived as being incoherent. Evidence from different domains of psychology is briefly reviewed in support of these hypotheses. (shrink)

Taber and Lodge's 2006 paper provides powerful evidence that one's prior beliefs shape one's reception of new evidence in a manner that can best be described as “inadvertently dogmatic.” This is especially true for people who are well informed, which dovetails with findings going back to Converse (1964) showing political beliefs to be ideologically constrained (rigid) among the relatively well informed. What may explain the coincidence of dogmatism and knowledgeability is the very process of learning about politics, which must use (...) theories, schemas, ideologies, or Lippmannesque “stereotypes” to target certain political information as germane by putting it into an interpretive framework. This interpretive process is likely to create for each of us a growing database of information that is congruent with our extant convictions but that excludes incongruent information: in light of the data we have already processed, incongruent information seems increasingly implausible (if not incomprehensible), and is therefore rationally ignored or dismissed. But this does not necessarily mean, as Taber and Lodge follow Robert Abelson in suggesting, that people are “motivated” to be dogmatic rather than being unintentionally closed minded as a result of the plausibility they involuntarily accord to their priors. Recognizing the inadvertent (unmotivated) nature of dogmatism is essential if political science is to take seriously political actors' beliefs—and to assess the gravity of the problem posed by dogmatism. (shrink)

A recent critique of hierarchical Bayesian models of delusion argues that, contrary to a key assumption of these models, belief formation in the healthy (i.e., neurotypical) mind is manifestly non-Bayesian. Here we provide a deeper examination of the empirical evidence underlying this critique. We argue that this evidence does not convincingly refute the assumption that belief formation in the neurotypical mind approximates Bayesian inference. Our argument rests on two key points. First, evidence that purports to reveal the most damning violation (...) of Bayesian updating in human belief formation is counterweighted by substantial evidence that indicates such violations are the rare exception—not a common occurrence. Second, the remaining evidence does not demonstrate convincing violations of Bayesian inference in human belief updating; primarily because this evidence derives from study designs that produce results that are not obviously inconsistent with Bayesian principles. (shrink)

ABSTRACTIn everyday situations, people regularly receive information from large groups of people and from single experts. Although lay opinions and expert opinions have been studied extensively in isolation, the present study examined the relationship between the two by asking how many laypeople are needed to counter an expert opinion. A Bayesian formalisation allowed the prescription of this quantity. Participants were subsequently asked to assess how many laypeople are needed in different situations. The results demonstrate that people are sensitive to the (...) relevant factors identified for determining how many lay opinions are required to counteract a single expert opinion. People's assessments were fairly good in line with Bayesian predictions. (shrink)