Can threat information bias fear learning?: Some tentative results and methodological considerations

Whereas it is widely recognized that both verbal threat information and stimulus pairings can install strong and persistent fear, few studies have addressed the interaction between these two pathways of fear. According to the expectancy bias of Davey (1992, 1997), verbal information can install expectancy biases for aversive events that can result in facilitated fear learning through stimulus pairings and can delay extinction of fear. However, these predictions of the expectancy bias account have not been explored fully. Following up on two earlier studies (Field & Storksen-Coulson, 2007; Ugland, Dyson, & Field, 2013), we investigated the impact of prior threat information on fear acquisition, extinction and reinstatement. To this aim, participants received instructions about four unfamiliar animals, two of which that were described as dangerous whereas the other two were described as harmless. One animal of each pair was subsequently paired with an electric stimulus. Our results indicated that threat information resulted in stronger fear responses prior to fear conditioning and in delayed extinction of fear. However, these effects of instructions were not very pronounced and not found on all measures of fear. We discuss several methodological and procedural considerations that may modulate the effects of (verbally installed) expectancy biases.