Ups and Downs: The Affective Consequences of Power in Different Contexts

Power dictates the allocation of, and access to, valued resources. It is perhaps not surprising then that people have a 'will' or 'lust' for power (Nietzsche, 1924; Russel, 1938), and believe that power is exciting and powerlessness depressing (Mondillon et al., 2005). The Approach-Inhibition Theory of Power echoes this belief, suggesting that high power brings positive mood and low power negative mood (Keltner, Gruenfeld, & Anderson, 2003). However, empirical data on power and mood is mixed. Inducing feelings of power sometimes elevates mood (e.g., Galinsky, Gruenfeld, & Magee, 2003), but other times does not (e.g., Berdahl & Martorana, 2006), suggesting that the relationship between power and mood may be more complex than initially thought.

The Situated Model of Power argues that the relationship between power and mood is dependent on the context (Guinote, 2007a). In this view, power attunes people to the current situation, elevating mood in positive contexts and depressing mood in negative contexts. This implies that power increases variability in mood between contexts of opposing valence (negative vs. positive; Guinote, 2007a), and is consistent with the fact that power increases variability in thought and behaviour (Guinote, Judd, & Brauer, 2002).

Five studies, informed by the Approach-Inhibition (Keltner et al., 2003) and Situated (Guinote, 2007a) Models of Power, looked at the impact of high and low power on self-reports, and physiological indices, of mood at baseline and in contexts of differing valence (negative vs. positive). A meta-analysis revealed that across studies (N = 1046) high power elevated, and low power depressed mood at baseline/in neutral and in positive contexts. However, neither high nor low power predicted mood in negative contexts. Furthermore, high power increased, and low power decreased variability in mood across contexts (although the former effect was marginally significant). Results reconcile disparate findings and are discussed in relation to theoretical models of power.