• Citer

False Words Seem True: The Power of Truth Bias in shaping Memory and Judgment 

par Pantazi, Myrto
Promoteur Klein, Olivier
Co-Promoteur Kissine, Mikhail
Publication Non publié, 2017-02-01

Thèse de doctorat

• Accès en ligne
• Détails
• Contenu
• Statistiques

Résumé :

Language is one of the main means of acquiring information about the world. An important debate in social psychology, linguistics and philosophy is how we come to believe information contained in statements we hear and read. Are we capable of assessing it and filtering it out, in case it is erroneous? Or do we rather tend to believe it? The experimental studies described in this thesis suggest that we strongly tend to believe statements we hear and read, even if we are aware of their falsity. Truth bias, as this tendency has been called, was detected both at the level of people’s memory and at the level of their judgments. Specifically, in a fake judicial context participants who read or listened to statements explicitly designated as true or false about a perpetrator tended to misremember false statements as true. Additionally, they were influenced by the false statements’ content in judgments they made about the perpetrators. Chapter 1 encompasses 5 Studies, all pointing to a strong truth bias, that, contrary to what is often assumed, may operate even in the absence of increased cognitive load (Studies 1–2). Studies 3–5 in Chapter 1 were methodologically-oriented, primarily aimed at testing the validity of the generalized truth bias established in Studies 1–2. Chapter 2 examines potential vigilance triggers that may reduce the truth bias. Manipulating the source of the false information (by informing participants that the speaker is either a defense attorney or prosecutor; Study 1), rendering participants accountable for their judgments (Study 2) or asking real judges to accomplish our experimental task (Study 3) did not reduce the truth bias. Nevertheless, offering participants financial incentives for providing an accurate judgment, eliminated both the memory-based and the judgment-based truth bias (Study 4). Based on the present experimental findings, I develop a model predicting that the truth bias is an intrinsic element of linguistic communication and hard to override.