The authors describe development of a survey measure that they used to examine characteristics of human cognitive processing (such as cognitive biases) and assess the results for individual resistance to some aspects of Truth Decay, including blurring of the line between opinion and fact and diminished trust in formerly respected institutions as sources of factual information.
Individual Differences in Resistance to Truth Decay
Exploring the Role of Reasoning and Cognitive Biases
- What cognitive biases and reasoning processes are associated with greater resistance to Truth Decay?
In this report, the authors address one of Truth Decay's proposed drivers: characteristics of human cognitive processing, such as cognitive biases. The authors describe development of a survey measure that they used to examine characteristics of human cognitive processing (such as cognitive biases) and assess the results for individuals' resistance or susceptibility to Truth Decay. The authors focused on six Truth Decay measures: endorsement of scientific consensus, endorsement of verifiable facts, rejection of false conspiracy theories, distinguishing fact from opinion, willingness to accept expert recommendations, and philosophical positivism versus skepticism. The survey used six measures of cognitive biases and reasoning: numeracy, scientific reasoning, magical reasoning, availability bias, unjustified confidence, and ingroup bias.
Generally speaking, greater resistance to Truth Decay on each of the six scales was predicted by greater numeracy, greater scientific reasoning, and less magical reasoning. Among the cognitive biases, greater availability bias was associated with greater susceptibility to false conspiracy theories but also greater trust in experts. Greater unjustified confidence by individuals in their own knowledge was associated with greater trust in experts. Ingroup bias was at times associated with greater susceptibility to Truth Decay (lower endorsement of scientific consensus and verifiable fact, lower philosophical positivism) and at other times associated with greater resistance to Truth Decay (rejecting false conspiracy theories, distinguishing fact from opinion). In terms of demographics, resistance to Truth Decay was most consistently associated with those who had a higher income, those who were White, and those who voted for Hillary Clinton in 2016.
- The most consistent finding across models was that greater numerical and scientific reasoning and lower magical reasoning were associated with greater resistance to Truth Decay.
- No strong or consistent associations were found between resistance/susceptibility to Truth Decay and well-known cognitive biases (for example, availability bias, unjustified confidence). Rather, the greatest predictors for resistance/susceptibility to Truth Decay were reasoning processes that are developed over an individual's lifetime and are all at least somewhat adaptive within their proper context.
- A concerning finding across models was that self-reported non-White respondents were consistently more susceptible to Truth Decay. Although race was not always significant, it often was significant even after controlling for other variables, such as education, political party, and biases or reasoning processes. This finding potentially reflects distrust of traditional sources of factual information among groups that have, at times, been systematically persecuted by societal institutions for government, medicine, and science.
- The models also highlight how perceptions of key issues, worldviews, and even ways of processing information in the United States are now split according to partisanship and religiosity. Controlling for these variables in a multiple regression often eliminated the significance of many of the reasoning processes and biases seen in bivariate correlations.
Table of Contents
Identifying Indicators of Individual Resistance/Susceptibility to Truth Decay
Identifying Reasoning Processes and Cognitive Biases Relevant to Resistance/Susceptibility to Truth Decay
Survey Administration and Analytic Approach