Download pdf
Andrews-Hanna et al. (2022).Journal of Experimental Psychology: General, 151(3), 628–642.Bellana et al (2022).Nature Communications, 13(1), Article 1.De Deyne, S. et al. (2018).Manuscript submitted for publication.Gable et al. (2019).Psychological Science, 30(3), 396–404.Gray et al. (2019).American Psychologist, 74(5), 539–554.Honey etl al. (2023).Directions in Psychological Science, 09637214221143053.Klinger (1978).The Stream of Consciousness: Scientific Investigations into the Flow of Human Experience(pp. 225–258).Paszke (2019).Advances in Neural Information Processing Systems, 32.Pennington et al. (2014).EMNLP 2014, pages 1532–1543, Doha, Qatar. Association for Computational Linguistics.Pennebaker et al. (2015).Austin, TX: University of Texas at Austin.Spinhoven et al. (2018).Journal of Affective Disorders, 241, 216–225.Wolf et al. (2020). EMNLP 2020, pages 38–45, Online. Association for Computational Linguistics.REFERENCESMeasure Personal Concerns by directly eliciting them.Why does mental content persist? Identify rational/normativecosts and benefits.What neural mechanisms underlie persistence of content?NEXT STEPSFree association word chains reliably predict participants' creativity (Gray et al., 2019)Post-story word chains provide an objectivecorrelate of the subjective experience of lingeringthoughts.Narrative content persisted in spontaneous thoughtfor 5 minutes or more.We did not find a link between pre-story freeassociations and lingering thoughts orpersistentnarrative content.CONCLUSIONS* mean_wr_pre -> linger_rating* theme influence* pre (lstm) -> linger_rating prediction* liwc -> linger_rating predictionCurrent conclusion: not so easy!PersonalConcernsNarrativeContentHighLingeringHigh overlapPersonalConcernsNarrativeContentLowLingeringLow OverlapHYPOTHESISTheme influence: SWOWenPennebaker et al. (2015)Wolf et al. (2020), Paszke et al. (2019)LIWCDe Deyne et al. (2018)Transformer / LSTMw1w2w3w4e1e2e3e4e1.2e2.2e3.2e4.2e1.ne2.ne3.ne4.nStory-RelatednessNone of our methods using pre-story associates predicted the subjectiveexperience of lingering thought or persisting narrative content.estimate overlapto narrative contentEstimate Personal-Concerns with pre-story free associates.(Andrews-Hanna et al. 2022)ANALYSISEXPLAINING INDIVIDUALVARIABILITY1234567−2024−2024−2024−2024−2024−2024first 10last 100.00.50.00.50.00.50.00.50.00.50.00.5Word Position(5 min)(0 min)Change in Story-Relatedness (Post - Pre Cohen's D)Story-Relatedness of Word Chains (Z-scored)Subjective LingeringWORD-SCRAMBLEDORIGINALSENTENCE-SCRAMBLEDREPLICATIONSENTENCE-SCRAMBLEDORIGINALNEUTRAL CUEORIGINALINTACTREPLICATIONINTACTORIGINALORIGINALREPLICATIONORIGINALREPLICATIONNEUTRAL CUEREPLICATIONORIGINALINTACTSENTENCE-SCRAMBLEDWORD-SCRAMBLEDR2= .00R2= .18R2= .18R2= .20R2= .26R2= .18R2= .01R2= .14R2= .17R2= .13R2= .14R2= .07HUMAN RATINGSGLOVE RATINGSTo what extent did thetext linger in your mindafter reading it?1 (Not At All) to 7 (Very Much)73Subjective experienceof lingering thoughtObjective measureof mental contentSUBJECTIVE AND OBJECTIVEMEASURES OF MENTAL PERSISTENCEfloatboatswimdrownwaterdeaddrowningmurderwifehusbandkillercrimerobberyrapeassualthomicidedestroyevidencegroupsecretfoundriverbankmoneystealrobberprisonjusticecerealmilkmanannouncegravitasencourageadministrationallowdisappearpillowpillowpearls5101520510152012345671234567Word PositionStory−RelatednessMEASURING MENTAL CONTENTPennington et al. (2014)Reading7654321How related is the wordto aspecific momentwithin the story?RatingQuestionsTimeHUMAN RATINGSGLOVE RATINGSPARADIGMFreeAssociation5 MinutesSelf-pacedReadingText ManipulationFreeAssociation5 MinutesPost-taskQuestionsI. Theme word generationII. Self-reported lingeringTimeFree AssociationBODYstory-relevant cuemindparticipant-generated associateMINDassociate becomes next cueTime (5 Minutes)INTACTOriginal story orderWORD-SCRAMBLEDWords in scrambled order,working memory taskSentences in scrambled orderSENTENCE-SCRAMBLEDREADING CONDITIONSWe hypothesize that content is more likely to linger if it resonates with participants'personal concerns (Andrews-Hanna et al., 2021; Klinger, 1978).Why do individuals differ in the mental persistenceof narrative content?In Bellana et al. (2022) lingering biases were only detectable for about 2minutesfollowing the story’s end. Subjective and anectdotal reportsindicatenarrative content persists longer.For how long is narrative content detectable infree-association word chains?Is the subjective experience of lingering thoughtsreflected in free-association word chains?Past experiences shape our current thoughts, actions, and decisions. Butwhich experiences resonate in mind and for how long? Persisting thoughtsenable us to find creative solutions to problems (Gable et al., 2019) and areassociated withdepression and anxiety (Spinhoven et al., 2018). Werecently introduced aparadigm characterizing what contentlingersinspontaneousthought after participants read a story (Bellana etal. 2022).INTRODUCTIONHOW ARE LINGERING THOUGHTS MODULATED BY TIMEAND PERSONAL CONCERNS?Gabriel Kressin Palacios1, Buddhika Bellana2& Christopher J. Honey11Department of Psychological & Brain Sciences, Johns Hopkins University;2Department of Psychology, Glendon Campus, York University5We thank Dr. Jeff Bowen for statistical consultation.