Feeling good and feeling truth: The interactive effects of mood and processing fluency on truth judgmentsIntroductionProcessing fluency and truth judgmentsProcessing consequences of affectMethodOverview, participants, and designProcedure and materialsTruth judgmentsFluency manipulationDebriefing and manipulation checksResults and discussionPreliminary analysesMood validationFluency validationTruth judgmentsEvidence for processing differencesJudgmental confidenceMood-congruent effectsGeneral discussionTheoretical implicationsPractical implicationsLimitations and future prospectsConclusionReferencesReportsFeeling good and feeling truth: The interactive effects of mood and processing fluencyon truth judgmentsAlex S. Koch, Joseph P. Forgas⁎University of New South Wales, Sydney, Australiaabstractarticle infoArticle history:Received 4 April 2011Revised 5 October 2011Available online 19 October 2011Keywords:MoodProcessing fluencyTruth judgmentTruth effectCan mood states influence the perceived truth of ambiguous or novel information? This study predicted andfound that mood can significantly influence people’s reliance on processing fluency when making truth judg-ments. Fluent information was more likely to be judged as true (the truth effect), and consistent with Blessand Fiedler's (2006) assimilative vs. accommodative processing model, negative mood eliminated, and pos-itive mood maintained people's reliance on processing fluency as an indication of truth. Post hoc analysesconfirmed the predicted mood-induced differences in processing style, as judges in a negative mood adoptedmore accommodative processing and paid greater attention to external stimulus information. The relevanceof these results to contemporary affect-cognition theories is discussed, and the real-life implications of moodeffects on truth judgments in applied areas are considered.© 2011 Elsevier Inc. All rights reserved.IntroductionMuch of the information we come across in everyday life is am-biguous, confusing, and potentially unreliable. How do we decidewhether a particular claim or statement is true or false? We needto steer a careful course between excessive gullibility – acceptingfalse claims – and excessive skepticism, rejecting true claims. As athorough investigation of every claim is inherently impossible (Fiedler& Wänke, 2009), we often rely on simple heuristics to determinewhether to believe or disbelieve new information (Fiedler, 1996).This experiment investigated the interactive effects of two variableson truth judgments: ease of processing (fluency), and the affectivestate of the judge. Based on prior affect-cognition theories, we pre-dicted that negative affect should reduce, and positive affect shouldpromote reliance on processing fluency as a relevant cue in truthjudgments.Processing fluency and truth judgmentsSubjective ease of processing, or fluency, is one of the most influ-ential cues in truth judgments (Unkelbach, 2006). Easy to process orfluent information is more likely to be accepted as true, and disfluentclaims are more often judged as false (Begg, Anas, & Farinacci, 1992;Reber & Schwarz, 1999). This so-called truth effect (see Dechêne,Stahl, Hansen, & Wänke, 2009) occurs regardless of a statement'scontent (Schwarz et al., 1991). The experience of fluency itself isinfluenced by a variety of factors, such as the frequency of priorstimulus exposure, previous primes, and the linguistic complexityas well as the visual clarity of the target information (see Alter &Oppenheimer, 2009; Unkelbach, Bayer, Alves, Koch, & Stahl, 2011).It is this last fluency manipulation that will be used here.Despite growing evidence for the truth effect, its boundary condi-tions remain poorly understood. It seems that people discount fluencyas a diagnostic truth cue “once they explicitly or implicitly recognizethat it stems from an irrelevant source” (Alter & Oppenheimer, 2009,p. 231). Interestingly, more elaborate processing can also eliminatethe truth effect (Hawkins, Hoch, & Meyers-Levy, 2001). As negativemoods typically recruit a more vigilant, externally focused cognitivestyle (Bless & Fiedler, 2006; Forgas, 1998; 2010; 2011), mood mayalso be a significant moderator of the truth effect, as discussed below.Processing consequences of affectAffect can have a significant impact on both the content and pro-cess of cognition (Forgas & Eich, in press). In addition to contenteffects such as affect congruence in memory, judgments and socialbehaviors (Forgas, 2002), positive and negative moods also influencehow information is processed (Bless, 2000; Bless & Fiedler, 2006;Fiedler, 2001). In particular, Bless and Fiedler's (2006) assimilative/accommodative processing model argues that moods perform asubconscious regulatory function. Positive mood signals a benignenvironment, promoting top-down, assimilative processing wherepeople “impose internal structures on the external world” (p. 66). Incontrast, negative mood signals a problematic situation, recruitingmore bottom-up, accommodative thinking where people focus onJournal of Experimental Social Psychology 48 (2012) 481–485⁎ Corresponding author.E-mail address: [email protected] (J.P. Forgas).0022-1031/$ – see front matter © 2011 Elsevier Inc. All rights reserved.doi:10.1016/j.jesp.2011.10.006Contents lists available at SciVerse ScienceDirectJournal of Experimental Social Psychologyjournal homepage: www.elsevier.com/locate/jespnew information and “modify internal structures in accordance withexternal constraints” (p.66).Numerous experiments support this mood-induced processingdichotomy, showing that positive mood increases, and negativemood decreases the tendency to rely on internal information in avariety of cognitive tasks (Bodenhausen, Sheppard, & Kramer, 1994;Forgas, 1998; 2011; Park & Banaji, 2000; Unkelbach, Forgas, &Denson, 2008). For example, happy persons were more influencedby easily retrieved (fluent) arguments, but negative mood reducedreliance on the ease-of-retrieval heuristic (Ruder & Bless, 2003;Tversky & Kahneman, 1973). Further, Wyland and Forgas (2010)demonstrated that when judging others, happy people paid moreattention to heuristic nonverbal cues (e.g., direct vs. avertedeye gaze), whereas sad people tended to ignore such peripheralinformation.Extrapolating from this research, this experiment sought todemonstrate for the first time that moods can also modera te people'sreliance on processing