Notes
Analysing the role of processing fluency in the relationship between attributes of online media and political opinion
142Attributes of online media (AOM) like personalized environments and repetition of content are assumed to be related to opinion formation and potentially extremization and polarization. In this context, processing fluency and its effects on judgments, like the repetition-induced truth effect, have often been discussed as mechanisms behind these relations. However, to our knowledge, previous research has not specifically examined the mediating role of processing fluency and related judgments in the association between AOM and general attitudes toward a politically relevant topic. The present work aims to close this gap.
143In the first experiment, the effect of personalization will be explored as a between-subjects factor (3 levels). To this end, participants will be presented with a majority of statements that are either congruent or incongruent with their previously assessed attitudes towards the use of autonomous weapons, or neutral regarding the topic. The time spent processing each statement will be recorded as a measure of processing fluency, and participants will be asked to judge how true, familiar, likeable, and trustworthy they perceive each statement to be. Thus, processing fluency and judgment type (judgment of truth, trust, likeability, and familiarity) are assessed as within mediator variables. Participants’ opinions on the topic will be assessed after the experimental manipulation, again. In a second experiment, the effect of repetition will be examined using the same experimental design. Here, participants will be presented with statements they will be previously exposed to and “new” statements. For this set of statements, the majority of “old” statements will be either in favour of or against the use of autonomous weapons. In the control condition, for pro, contra, and neutral statements, the same amount of “old” and “new” statements will be presented.
144To analyse the serial mediation in each experiment, we plan to use multilevel structural equation modelling.
145Fluency / political opinion / repetition-induced truth effect / personalization / online environments