26 May, 3.30pm CEST
Room 117, Institute of Philosophy, Heidelberg University
Abstract. Modern societies rely fundamentally on the production, circulation, and recognition of reliable knowledge. Yet despite the normative and institutional prominence of knowledge, we know surprisingly little about what citizens themselves count as knowledge, to whom they attribute it, and on what grounds. A dominant philosophical account defines knowledge as Justified True Belief, requiring that a proposition be true, believed, and adequately justified. In this talk, I present a large-scale empirical test whether ordinary knowledge ascriptions adhere to this normative standard. In a preregistered conjoint experiment with a nationally quota-matched U.S. sample (N = 1,295), participants judged whether an agent “knows” propositions across a politically contested domain (climate change) and an uncontested domain (astrophysics). We fully crossed Justification (six levels varying strength and source), Truth (true vs. false), and Belief (strong vs. weak). Knowledge ascriptions systematically diverged from Justified True Belief across both domains. Belief exerted the strongest causal influence (Average causal effects: AMCE ≈ −0.42 for weak vs. strong belief), Truth was helpful but not necessary (AMCE ≈ 0.18 for true vs. false), and Justification contributed little or not at all (AMCE range across levels ≈ 0.00–0.05). This asymmetry had striking consequences: more than half of participants attributed knowledge even to false propositions when belief was strong, whereas only about one quarter attributed knowledge to true, strongly justified propositions when belief was weak. Across both domains, participants thus heavily prioritized conviction over truth and justification when judging whether others “know”. By showing that ordinary knowledge ascriptions more closely follow a model of “Strong Belief with optional Truth” than the normative account of Justified True Belief, these results help explain why low-justification and even false propositions can be treated as knowledge in public discourse.
Helen Fischer is an incoming Heisenberg professor at Heidelberg University. A trained cognitive psychologist, she has been a researcher at the Leibniz Institute for Knowledge Media in Tübingen since 2022. She also held the visiting professorship “Science and Society” at the Karlsruher Institut für Technologie as well as research positions at the Stockholm Resilience Centre and Max-Planck-Institut für Bildungsforschun, Berlin. She works primarily on metacognition, our insight into the reliability and limits of our own knowledge for beliefs, e.g. about politicized science of climate change or COVID 19. Her research sheds light on the role of metacognition in recognizing one’s own thinking errors such as motivated information processing and information sharing in social networks.
