Skip to main content

Research Repository

Advanced Search

Linguistic indicators of the effects of social disclosure on the fading affect bias

Muir, Kate; Brown, Charity; Madill, Anna

Authors

Kate Muir

Charity Brown

Anna Madill



Abstract

Introduction
The intensity of negative emotions associated with everyday unpleasant events often fade in memory over time, whilst positive emotions associated with everyday pleasant events remain relatively constant. This phenomenon is known as the fading affect bias or FAB (Walker, Skowronski & Thompson, 2003). One promising mechanism involved in the development of the FAB concerns the social disclosure of event memories (i.e., talking to other people about experienced events). Previous research has found that with increasing social disclosure frequency, negative emotions associated with the disclosed event fade more and positive emotions fade less (Ritchie et al., 2006; Skowronski et al., 2004). However, it is unclear whether the beneficial effects of social disclosure are due to mere verbalisation of the event memory in itself, or if the presence and behaviour of the listener during social disclosure is an important factor. This research therefore utilised a novel experimental paradigm in which listener presence and behaviour during social disclosure are manipulated, to examine the resultant effects upon the self-rated emotional intensity of the disclosed events.
Studies into the fading affect bias often obtain written event descriptions from participants along with ratings of positive and negative emotional intensity (i.e., Gibbons, Lee & Walker, 2011; Skowronski et al., 2004). Research within the FAB literature has yet to examine the linguistic characteristics of such event descriptions. However, research using computerised linguistic analysis has illustrated that where the emotions associated with events change over time or after an experimental manipulation, such changes in event emotionality can be correlated with changes in the linguistic characteristics of event narratives (Pennebaker & Francis, 1996; Schwartz & Drotar, 2004). Further, there are differences in the linguistic characteristics of the narratives of socially disclosed versus not disclosed events (Pasupathi, 2007). Thus, this research also involved linguistic analysis of written event descriptions provided by participants, to examine if the effects of social disclosure and listener behaviour are evident in the linguistic characteristics of participants’ event descriptions.
Method
One hundred and forty participants recalled three pleasant and three unpleasant events and wrote a description of each event. Each event was also rated for its emotional intensity upon event occurrence and recall, to provide a pre-existing, baseline level of the fading affect bias in the sample. Participants then came into the laboratory for the experimental manipulations. The events recalled at baseline were allocated to one of three experimental conditions: no disclosure (control), private verbal disclosure without a listener, or social disclosure to another participant whose behaviour was experimentally manipulated. During social disclosure, participants in the feedback group received verbal feedback from their conversational partner, whereas participants in the no feedback group were instructed not to provide verbal feedback in response to the social disclosure. Participants again wrote an event description and rated the emotional intensity of the events immediately after these manipulations and after a one week delay. The main dependent measures utilised were the emotional intensity ratings provided by participants at baseline, immediately after and one week after the experimental manipulations. Additional dependent measures were a set of linguistic indicators, compiled from the Linguistic Inquiry and Word Count Program (LIWC; Pennebaker, Booth & Francis, 2007) for each event description provided by participants at baseline, immediately after and one week after the experimental manipulations.

Results
Verbal disclosure alone was not sufficient to enhance the FAB. However, the effects of social disclosure and listener behaviour upon emotional intensity ratings were evident both immediately after social disclosure and after a one week delay. Compared to non-disclosed events, socially disclosing pleasant event memories made the events feel more intensely positive, regardless of the behaviour of the listener. Such increases in positive emotional intensity were maintained one week later, but only if the listener had provided verbal feedback during social disclosure. In comparison, social disclosure with feedback led negative emotions to fade in intensity, but social disclosure to a non-responsive listener actually increased the negative emotional intensity of unpleasant events. Further, the effects of social disclosure and listener behaviour were evident in changes in linguistic characteristics of written event descriptions. Social disclosure with feedback was associated with increased emotional expression in written event descriptions whereas social disclosure without feedback was associated with decreased emotional expression in written event descriptions. Thus, social disclosure and listener behaviour influenced the emotionality of event memories in ways that private verbal disclosure did not.

Conclusions
This research found that verbalisation of an event memory alone is not a convincing explanation for the relationship between social disclosure and the fading affect bias. Rather, there is novel evidence that interactions between speakers and listeners during social disclosure are important for enhancing positive and minimising negative emotions. Talking to an interactive listener (as opposed to a non-responsive listener) during social disclosure could be conceived as encouraging or facilitating the speaker’s use of emotional regulation processes in order to deal with emotional responses to events (Taylor, 1991). Further, the results of the linguistic analysis can be interpreted in several ways, providing insight into possible mechanisms underlying the effects of social disclosure on the FAB. In particular, the increased emotional expression in event descriptions after social disclosure with feedback could be interpreted as evidence of heightened emotional processing, or changes to event memory encouraged by the listener’s verbal responses. This study demonstrates new and important findings for research into the FAB phenomenon, and illustrates that utilising computational linguistic analysis has the potential to provide a useful additional dimension for research into the effects of social disclosure on autobiographical memory.

Citation

Muir, K., Brown, C., & Madill, A. (2013, November). Linguistic indicators of the effects of social disclosure on the fading affect bias. Paper presented at iCog Inaugural Conference: Interdisciplinarity in Cognitive Science, Sheffield, UK

Presentation Conference Type Conference Paper (unpublished)
Conference Name iCog Inaugural Conference: Interdisciplinarity in Cognitive Science
Conference Location Sheffield, UK
Start Date Nov 29, 2013
End Date Dec 1, 2013
Publication Date Nov 29, 2013
Peer Reviewed Peer Reviewed
Keywords linguistic, indicators, effects, social disclosure, fading affect bias
Public URL https://uwe-repository.worktribe.com/output/925840
Publisher URL http://www.mindcogsci.net/?p=2139
Additional Information Title of Conference or Conference Proceedings : iCog Inaugural Conference: Interdisciplinarity in Cognitive Science


Downloadable Citations