The impact of negative emotions on self-concept abstraction depends on accessible information processing styles.

Abstract

Research suggests that anger promotes global, abstract processing whereas sadness and fear promote local, concrete processing (see Schwarz & Clore, 2007 for a review). Contrary to a large and influential body of work suggesting that specific affective experiences are tethered to specific cognitive outcomes, the affect-as-cognitive-feedback account maintains that affective experiences confer positive or negative value on currently dominant processing styles, and thus can lead to either global or local processing (Huntsinger, Isbell, & Clore, 2014). The current work extends this theoretical perspective by investigating the impact of discrete negative emotions on the self-concept. By experimentally manipulating information processing styles and discrete negative emotions that vary in appraisals of certainty, we demonstrate that the impact of discrete negative emotions on the spontaneous self-concept depends on accessible processing styles. When global processing was accessible, individuals in angry (negative, high certainty) states generated more abstract statements about themselves than individuals in either sad (Experiment 1) or fearful (Experiment 2; negative, low certainty) states. When local processing was made accessible, however, the opposite pattern emerged, whereby individuals in angry states generated fewer abstract statements than individuals in sad or fearful states. Together these studies provide new insights into the mechanisms through which discrete emotions influence cognition. In contrast to theories assuming a dedicated link between emotions and processing styles, these results suggest that discrete emotions provide feedback about accessible ways of thinking, and are consistent with recent evidence suggesting that the impact of affect on cognition is highly context-dependent.

Affect and cognition: three principles

“…This positivity should therefore empower accessible thoughts and inclinations, just as positive affect does. A recent study examined the broad-abstract vs. narrow-concrete nature of people’s self-descriptions when angry, sad, or fearful [41]. Anger led to self-descriptions that were more abstract or less concrete than sadness and fear.…”

Feeling and thinking: An affect‐as‐cognitive‐feedback account

“…Relevant evidence for this idea comes from research examining the influence of anger and fear on the degree of abstraction with which people describe themselves (Isbell, Rovenpor, & Lair, ). In this research, when a global focus was accessible, anger led individuals to describe themselves using more abstract terms, and fear led individuals to describe themselves in more concrete terms.…”

Differentiating the effects of anger and sadness: A perspective of spontaneous trait inference

“…In contrast, moods with low certainty, such as sadness, will make individuals feel uncertain about a subsequent context, thus activating a systematic, bottom‐up information processing style. In other words, although anger and sadness have the same valence, they actually induce different information processing styles (Isbell et al, ; Keltner et al, ). Therefore, when the trait‐implying behaviours were presented to the participants, the angry participants tended to represent the actors of the behaviours in terms of abstract traits, while the sad participants seemed to represent those actors via specific behaviours.…”

“…It should be noted, in a previous study examining the effect of mood on STIs, the researchers contrasted the effects of a positive versus negative mood rather than distinguishing the influences of specific mood states on STIs. However, past research has indicated that different emotions of the same valence actually can produce different effects on judgements and choices (e.g., Isbell, Rovenpor, & Lair, ; Keltner, Ellsworth, & Edwards, ; Tiedens & Linton, ). For example, Keltner et al () found that, although sadness and anger are both negative emotional states, they had different influences on causal judgements.…”