We are interested in the subjective nature of social perception and how it can be harnessed to achieve desired outcomes. We consider a variety of dimensions on which subjective perspectives can vary, and we explore implications for the self, social judgment, decision-making, emotional well-being, and interpersonal and intergroup relations. For details on some of these projects, read on...

The role of imagery perspective in representing events and the self

When people think about events in their lives—recalling the past or imagining the future—they often "see" those events in their mind's eye. An intriguing fact about this imagery is that people don't always see events from their own first-person perspective; sometimes they use an observer's third-person perspective so that they see themselves in the image. Our research attests to the power that this subtle feature of subjective representation has in determining responses to recalled and imagined events.

For example, we have shown that manipulating imagery perspective can influence psychotherapy clients’ perceptions of progress in treatment (Libby, Eibach, & Gilovich, 2005), voters’ likelihood of turning out at the polls (Libby, Shaeffer, Eibach, & Slemmer, 2007), students’ feelings about life regrets (Valenti, Libby, & Eibach, 2011), intimates’ resilience to past relationship transgressions (Marigold, Eibach, Libby, Ross & Holmes, 2011), individuals’ proneness to shame (Libby, Valenti, Pfent, & Eibach, 2011), forecasters’ reliance on implicit versus explicit attitudes in predicting future behavior and feelings (Libby, Valenti, Hines, & Eibach, 2011), and decision-makers’ sensitivity to comparison standards in forecasting their satisfaction with choice alternatives (MacGregor, Valenti, Libby, & Eibach, 2011). These findings clearly demonstrate that imagery perspective matters. But, why does it matter? And, how?

According to our model (Libby & Eibach, 2011), imagery perspective functions to shape the style of processing individuals use to understand and subjectively represent events. From the first-person perspective people understand events bottom-up, in terms of the phenomenology evoked by concrete features of the pictured situation. From the third-person perspective people understand events top-down, in terms of abstractions that integrate the pictured event with its broader context. Thus, an action (e.g., “wiping up a spill”) is understood in terms of constituent aspects (e.g., “using a sponge”, “moving my hand”) from the first-person perspective, but in terms of connections to causes, consequences, traits, goals, and identities from the third-person perspective (e.g., “cleaning up after the kids”, “being a responsible parent”).

Evidence supporting this model reveals that these links are bidirectional in nature: perspective influences the level of meaning people see in actions, and the level of meaning they seek influences the perspective they use (Libby, Shaeffer, & Eibach, 2009; Libby & Eibach, 2011). Thus, imagery perspective is functionally involved in event representation, not epiphenomenonal. Further, the bidirectional relationship holds regardless of whether the image is internally or externally generated, whether the self or other is the actor, and when the objects in the image and distance to the action are controlled across perspectives (e.g., using photographs like those shown above; Libby et al., 2009). Thus, the relationship between perspective and level of meaning depends on visual perspective per se, and not on other dimensions that may sometimes covary with it. Finally, the effect of perspective on construal of pictured actions carries over to the construal of subsequently encountered actions not depicted in the images, consistent with the idea that the effect reflects a shift of processing style according to perspective (Shaeffer, Libby, & Eibach, 2011).

We apply a dual-faceted model of the self to interpret the implications of these findings for the role of imagery perspective in determining the impact of recalled and imagined life events on present self-judgment, emotion and behavior. This approach distinguishes our model from alternative accounts of imagery perspective, which treat the self as a unitary construct. These accounts predict that picturing a life event from the third-person perspective depersonalizes it, detaching it from the self and reducing its emotional power (e.g., McIsaac & Eich, 2004; Sanitoso, 2008; Williams & Moulds, 2007). This reasoning is based on evidence that third-person imagery reduces simulation of the internal components of event experience (e.g., sensory-motor activation, sense of “reliving”; Jackson, Meltzoff, & Decety, 2005; Nigro & Neisser, 1983). However, such logic overlooks the fact that experiential awareness defines only one facet of the self.

According to our account, there is not a one-to-one relation between perspective and perceived self-ownership. We propose that first-person imagery represents an event in terms of the experiential I, and third-person imagery represents the event in terms of the conceptual me. If so, when an event is pictured from the third-person perspective reactions should reflect the event’s meaning in the context of one’s life more broadly, as defined by the general knowledge structures that comprise the self-concept (e.g., self-schemas, self-theories, life narratives). When pictured from the first-person perspective reactions should reflect the phenomenology evoked by concrete features of the situation, apart from these general beliefs about the self. This line of reasoning allows us to explain and predict effects that alternative accounts do not.

In several sets of studies we have found that when people picture life events, biases previously documented to reflect the influence of general self-beliefs emerge only when individuals use the third-person perspective and not the first-person (Libby et al., 2005; Libby, Valenti, Pfent, & Eibach, 2011; Marigold et al., 2011). For example, previous work demonstrates that low self-esteem tends to predict more extreme negative reactions to failure: low self-esteem individuals (LSEs) overgeneralize more and experience greater shame than high self-esteem individuals (HSEs) do. This difference has been attributed to self-esteem differences in global self-views, which either exacerbate (LSE) or minimize (HSE) the negative impact of failure (e.g., Kernis, Brockner, & Frankel, 1989; Brown & Marshall, 2001). In a series of studies, we (Libby, Valenti, Pfent, & Eibach, 2011) found that when individuals recalled or imagined failure LSE predicted greater overgeneralization and shame only when individuals pictured the incident from the third-person perspective. From the first-person the reactions of LSEs and HSEs could not be distinguished. Further, as a result of the greater influence of general self-views from the third-person, using the third-person perspective was beneficial for HSEs, significantly reducing shame relative to the first-person, but detrimental for LSEs, significantly increasing shame.

Such findings demonstrate that third-person imagery does not uniformly reduce the personal and emotional connection to a pictured event, as detachment accounts would predict. Instead, the impact of a pictured event is reduced by third-person imagery only when framing that event in relation to the self-concept, rather than in terms of concrete details, has such an effect. In cases where framing an event in relation to the self-concept suggests a more powerful impact, picturing the event form the third-person rather than first-person perspective increases its impact. Further, this process can produce important effects on present behavior. For example, voting seems more personally significant when framed in terms of its broader meaning (e.g., being a good citizen, participating in the democratic process) than in terms of its constituent actions (waiting in line, pulling a lever). In support of our model, a study conducted on the eve of the 2008 US Presidential Election found that picturing voting from the third-person as opposed to first-person perspective strengthened registered voters’ personal connection to the action, increasing their commitment, making them feel more excited, and actually causing them to be more likely to turn out to the polls on Election Day (Libby & Eibach, 2011; Libby et al., 2007).

The explanatory power of our model demonstrates the value of conceptualizing the self as a dual-faceted structure. Through this approach we have revealed new insights into the representational function of imagery, and the role of memory and imagination in defining the self and shaping present self-judgment, emotion, and behavior.

Changing beliefs and behavior through experiential self-other merging

Other lines of work in the lab apply the distinction between experiential and conceptual facets of the self to elucidate the mechanisms by which individuals understand and are influenced by other people. Narrative fiction provides an interesting context in which to explore these questions. Novels allow readers to expand the scope of their experience to include contact with individuals they might never have the opportunity to encounter in everyday life. Thus, novels hold the potential to create a unique pathway of social influence. When do readers take on the attitudes and behaviors of protagonists they encounter in narratives? Our research demonstrates that readers are more likely to do so when narratives are both told in the first-person voice and provide access to the protagonist’s thoughts and feelings. These features of narrative have this effect because they cause readers spontaneously to simulate story events as if they were happening to the self, and this accounts for readers’ greater tendency to take on the character’s attitudes and behaviors (Kaufman & Libby, 2011). 

This mechanism distinguishes the process responsible for our findings from the mechanism responsible for effects of explicit instructions to adopt a target’s perspective. Evidence suggests that explicit instructions to perspective-take operate by activating the self-concept as a vehicle for conceptually merging the self and other (e.g., Galinsky, Ku, & Wang, 2005). In contrast, activating the self-concept interferes with the spontaneous experiential merging that accounts for the effects in our experiments (Kaufman & Libby, 2009). These effects appear to result from merging with the protagonist on an experiential level. And, our research shows that this experiential merging can be a powerful tool for changing attitudes and behavior—e.g., reducing prejudice and increasing civic engagement with voting and volunteerism (Kaufman & Libby, 2011).

A new line of work extends these findings from the context of fictional narratives to explore implications for perceptions of real events—in particular, instances of discrimination. Members of historically victimized groups see more present-day discrimination against their group than others do. In previous work, my colleagues and I have found that one contributing factor is different impressions of related historical injustices. These feel closer in time to members of the victimized group, and greater subjective recency of historical injustices causes increased perceptions of present-day discrimination, as well as increased support for related public policies (e.g., affirmative action) (Libby, Eibach, & Ross, 2011). Our new work is applying the techniques shown to promote experiential merging with characters in fiction in an effort to promote experiential merging with targets of historical discrimination. Initial evidence suggests this may be an effective tool for reducing group differences in perceptions of present day discrimination and improving intergroup understanding (Libby, Rha, & Kaufman, 2011).