Understanding others is central to the development of positive and satisfying relationships. However, despite likely association between reflecting on one's own thoughts, feelings, and experiences and imagining those of others, the relationship between self-reflection and perspective taking has received surprisingly limited research attention. It has, for example, been argued that self-reflection and perspective taking ‘be seen as part of the same general domain — that of constructs having to do with awareness of and sensitivity to psychological states, motives, and behavioral tendencies: in one case . . . the attention is self-directed, and in the other . . . it is other-directed’ (Davis & Franzoi, Reference Davis and Franzoi1991, p. 72). Indeed, self-awareness has been shown to lead to a number of beneficial outcomes (e.g., understanding of emotions, cognitions, and behaviours) which, in turn, are inherent in the process of perspective taking. The aim of this review is to develop thinking in this area further by reviewing current evidence that self-reflection plays a key role in the perspective-taking process. In particular, the focus will be on understanding the importance of reflecting upon and developing insight into one's own past experiences in order to take the perspectives of others in similar situations.
Dymond (Reference Dymond1950), in defining empathy, effectively captured the meaning of perspective taking as ‘the imaginative transposing of oneself into the thinking, feeling, and acting of another’ (p. 343). Later conceptualisations, which place less emphasis on the process-nature of this core cognitive component of empathy, define perspective taking as ‘the tendency to spontaneously adopt the psychological point of view of others’ (Davis, Reference Davis1983, pp. 113–114), or more specifically as ‘cognitively trying to understand the other's internal mental states’ (Eisenberg & Okun, Reference Eisenberg and Okun1996, p. 158). While the ability to take another's perspective clearly requires an appreciation or understanding of the psychological vantage point of the other person, strategies used to take another person's perspective often involve reflecting upon and utilising one's own perspective and experiences. These include: mentally switching places with the other person (e.g., Batson, Early, & Salvarani, Reference Batson, Early and Salvarani1997; Stotland, Reference Stotland and Berkowitz1969); using one's own perspective as a starting point or ‘anchor’ to understand the other person (e.g., Epley, Keysar, Van Boven, & Gilovich, Reference Epley, Keysar, Van Boven and Gilovich2004); projection or stereotype use to infer the mental states of others (e.g., Ames, Reference Ames2004); relating what the person is going through to a similar experience in one's own past (e.g., Batson et al., Reference Batson, Sympson, Hindman, Decruz, Todd, Weeks and Burris1996); and using heuristics for understanding another's experiences (e.g., Karniol & Shomroni Reference Karniol and Shomroni1999). More general explanations of how we accomplish an understanding of others, including simulation theory (e.g., A.I. Goldman, 1989/Reference Goldman, Davies and Stone1995; R.M. Gordon, 1986/Reference Gordon, Davies and Stone1995) and theory of mind accounts (e.g., Gopnik & Meltzoff, Reference Gopnik and Meltzoff1997; Gopnik & Wellman, Reference Gopnik and Wellman1992, Reference Gopnik, Wellman, Hirschfeld and Gelman1994) have also highlighted the importance of the self to the process.
Recent social psychological work (e.g., Gerace, Day, Casey, & Mohr, Reference Gerace, Day, Casey and Mohr2013) has reconceptualised perspective taking as a process whereby self-information (e.g., imagining what it would be like to be in another's situation) and other-information (e.g., considering what led up to another's situation) are used either simultaneously or alternately by the empathiser to understand the other's perspective. This is supported by data from studies in which participants in either an imagine-self or imagine-other condition report imagining their own feelings as well as the feelings of the target person (e.g., Batson et al., Reference Batson, Early and Salvarani1997; Davis et al., Reference Davis, Soderlund, Cole, Gadol, Kute, Myers and Weihing2004). Indeed, both shared and unique neural networks are activated when a person focuses on their own versus another's perspective (Ruby & Decety, Reference Ruby and Decety2004).
Understanding and reflecting on past experiences is also likely to be important in many of the strategies used to apprehend another's perspective. Reflection on one's own past experiences (e.g., relationship problems, bereavement) makes it easier to take the viewpoint of another person in a similar situation, but this requires both increased perceptions of similarity and cognitive processing of the thoughts, feelings, and behaviours that accompanied one's own experience (see Gerace, Day, Casey, & Mohr, Reference Gerace, Day, Casey and Mohr2015). However, while this process is subjectively easier, evidence is mixed as to whether having experienced a similar situation to another person leads the empathiser to form a more accurate understanding of that person's thoughts and feelings (e.g., Hodges, Kiel, Kramer, Veach, & Villaneuva, Reference Hodges, Kiel, Kramer, Veach and Villanueva2010). In this article, it will be argued that to the extent that reflection is motivated by self-concerns or a ruminative focus on past events, perspective taking is inhibited and the positive aspects of using self-knowledge and past experiences to understand another are diminished. By contrast, when self-reflection is driven by a need to develop insight and understanding into one's own experiences, the person is likely to be more successful in understanding the perspective of another person who finds himself or herself in a similar situation.
We begin by examining the concept of self-reflection, applying two dominant theoretical models. We then examine the relationship between self-reflection and perspective taking in theoretical and empirical work. In this work, self-reflection has been found to be accompanied by increased self-knowledge in understanding one's own thoughts, feelings, and behaviours, as well as a more accurate understanding of the other person's unique and separate perspective. From this examination, it is apparent that considering the type of self-reflection in which an individual engages is important for understanding its role in perspective taking. To this end, we present factor analytic work on measures of self-reflection and investigations of the relationship of perspective taking to different types of reflection. In these studies, perspective taking was negatively associated with rumination and positively associated with more positively framed abstract self-reflection. In order to understand the ways in which people may move beyond ruminating on problematic experiences, we outline clinical research and psychological models of change that suggest that focusing on past problematic experiences involves a process of reflection, assimilation, and understanding of these experiences. It is at this insightful and distanced thinking about past experiences that we believe the person is able to more successfully take the perspective of another person. Finally, we discuss potential directions for future research, as well as general approaches to therapy and intervention, to provide ways that self-reflection and perspective taking can be increased in close relationships. While perspective taking, and we would argue self-reflection, play a central role in all interpersonal interactions (Vorauer & Cameron, Reference Vorauer and Cameron2002), unique aspects of closer relationships are likely to influence how self-reflection and perspective taking may be related.
What Is Self-Reflection?
For the past 40 years, two conceptualisations have guided investigations into self-reflection: self-awareness theory (Duval & Wicklund, Reference Duval and Wicklund1972) and self-consciousness theory (Fenigstein, Scheier, & Buss, Reference Fenigstein, Scheier and Buss1975). According to the theory of self-awareness, often known as objective self-awareness theory, attention at any given time is either self-directed or externally oriented. Self-directed or objective self-awareness is a state where ‘consciousness is focused exclusively upon the self and consequently the individual attends to his conscious state, his personal history, his body, or any other personal aspects of himself’ (Duval & Wicklund, Reference Duval and Wicklund1972, p. 2). In contrast, externally oriented or subjective self-awareness, suggested to be the standard attentional state of individuals, involves attention being focused on the external environment. Unlike the state of subjective self-awareness, which is characterised by feelings of direction and control (since the individual is the ‘subject’ and his or her environment is the ‘object’), objective self-awareness as originally proposed often has somewhat negative undertones (for more recent conceptions, see Silvia & Duval, Reference Silvia and Duval2001). This is said to occur because self-focused attention has an evaluative component: the self is compared with standards regarding various parts of the self (e.g., behaviour), and an evaluation is made regarding the degree of discrepancy between the self and those standards. When discrepancy and negative affect are apparent, attempts are made to resolve the discrepancy (e.g., through a change in behaviour) or avoid objective self-awareness and, consequently, the evaluation and resulting affect (Duval & Wicklund, Reference Duval and Wicklund1972; for another conception, see Carver & Scheier, Reference Carver and Scheier1981, Reference Carver and Scheier1998). Self-focus, whereby the person sees themselves as an ‘object’ in the environment, can be facilitated with the use of mirrors, photographs, tape recorders, and through the presence of other persons (Duval & Wicklund, Reference Duval and Wicklund1972).
Whereas the two states of self-awareness posited in the theory of objective self-awareness are induced by situational factors, the pervasive and trait nature of self-reflection is stressed in the concept of self-consciousness, defined as ‘the consistent tendency of persons to direct attention inward or outward’ (Fenigstein et al., Reference Fenigstein, Scheier and Buss1975, p. 522). A number of criteria informed construction of the Self-Consciousness Scale:
(a) preoccupation with past, present, and future behavior; (b) sensitivity to inner feelings; (c) recognition of one's positive and negative attributes; (d) introspective behavior; (e) a tendency to picture or imagine oneself; (f) awareness of one's physical appearance and presentation; and (g) concern over the appraisal of others. (Fenigstein et al., Reference Fenigstein, Scheier and Buss1975, p. 523)
On the basis of factor analysis of an item pool derived from these criteria, three subscales were identified: private self-consciousness (i.e., ‘attending to one's thoughts and feelings’), public self-consciousness (‘a general awareness of the self as a social object that has an effect on others’; Fenigstein et al., Reference Fenigstein, Scheier and Buss1975, p. 523); and social anxiety, which was significantly positively correlated with public self-consciousness, but unrelated to private self-consciousness. While private and public self-consciousness were interpreted as process-oriented, social anxiety was seen as outcome-oriented.
Outcomes of Self-Reflection
Much of the early work on self-consciousness investigated the two theories in the same studies; this mostly took the form of researchers examining the effects of dispositional self-reflection (self-consciousness) and situational self-reflection (self-awareness). The difference between the theories focused upon at this earlier stage was this state-trait distinction. It was only later that debate intensified between proponents of each theory regarding problems with the other in accounting for conceptual questions and empirical findings regarding self-reflection and awareness (see Carver & Scheier, Reference Carver and Scheier1987; Fenigstein, Reference Fenigstein1987; Gollwitzer & Wicklund, Reference Gollwitzer and Wicklund1987; Wicklund & Gollwitzer, Reference Wicklund and Gollwitzer1987). However, it is now generally recognised that overall experimental findings vary little regardless of the particular theoretical underpinning of the studies (Scheier, Carver, & Gibbons, Reference Scheier, Carver and Gibbons1979). Therefore, it is at the explanatory level that the studies differ: one theory (objective self-awareness theory) posits self-discrepancy reduction or consistency concerns (e.g., Pryor, Gibbons, Wicklund, Fazio, & Hood, Reference Pryor, Gibbons, Wicklund, Fazio and Hood1977), while the other (self-consciousness) invokes more self-knowledge oriented needs (e.g., Scheier, Buss, & Buss, Reference Scheier, Buss and Buss1978).
The central finding from the early body of research is that manipulated or dispositional self-focus results in increased concordance between self-reports and objective benchmarks (e.g., academic performance, emotionally-inducing stimuli) and behaviour (e.g., sociability, aggression; Pryor et al., Reference Pryor, Gibbons, Wicklund, Fazio and Hood1977; Scheier et al., Reference Scheier, Buss and Buss1978; Scheier & Carver, Reference Scheier and Carver1977; Turner & Peterson, Reference Turner and Peterson1977), as well as higher levels of self-focus reflected in cognitive tasks (e.g., sentence completion; Carver & Scheier, Reference Carver and Scheier1978), and attributions of heightened self-responsibility (D.M. Buss & Scheier, Reference Buss and Scheier1976). Several studies have examined the importance of trait versus state attention. In one early study, the effect of private self-consciousness was statistically stronger than that of mirror-induced self-awareness (Buss & Scheier, Reference Buss and Scheier1976). There was also an interaction effect where dispositional self-reflection and state self-reflection seemed to have an additive effect in negative situations, while in positive situations high and low private self-consciousness participants were most differentiated in self-attributions when state self-awareness was not manipulated (see Hass, Reference Hass1984, for results in a similar direction for public self-consciousness and non-manipulated state self-awareness). In another study, a small interaction effect (at p < .08) was apparent between high or low private self-consciousness (obtained via median split) and presence or absence of a mirror, such that low self-consciousness participants were more self-focused when exposed to a mirror (Carver & Scheier, Reference Carver and Scheier1978). In this case, a ceiling-type effect was advanced for the lack of effect for high private self-conscious participants when they were exposed to their reflection. In addition, private self-consciousness was correlated with self-focus measured using proportion of participants’ self-focused completion of sentences; when this correlation was calculated separately for the two levels of the manipulation, it was significant only for the no-mirror group. While these studies provide information on the outcomes of self-reflection, it is important to examine specific theoretical and empirical treatment of the influence of self-reflection on perspective taking.
The Relationship Between Self-Reflection and Perspective Taking
The interdependence of self-reflection and perspective taking for the notion of self is the foundation of earlier work in the symbolic interactionist tradition and related schools of thought. For theorists, it was the very process of taking another's perspective that led to a state of self-awareness, rather than self-reflection informing the understanding of another person's perspective. In one of the early conceptions, Cooley (1902/Reference Cooley1967) suggested that:
In a very large and interesting class of cases the social reference takes the form of a somewhat definite imagination of how one's self — that is any idea he appropriates — appears in a particular mind, and the kind of self-feeling one has is determined by the attitude toward this attributed to that other mind. A social self of this sort might be called the reflected or looking-glass self. (pp. 183–184)
Similarly, Mead (Reference Mead1913, 1934/Reference Mead and Morris1950, 1936/Reference Mead and Strauss1964) considered the development of self and self-consciousness to be fundamentally based on taking the perspective of another (both the specific and generalised other) toward the self. The site at which this occurred was the social act, which contains various social positions, as well as specific perspectives that are associated with these positions (Gillespie, Reference Gillespie2006; Mead, 1934/Reference Mead and Morris1950).
Notions of self-awareness are also fundamental to theories of human development. Piaget (1928/Reference Piaget, Cartalis, Escher, Hanhart, Hahnloser, Matthes, Perret and Warden1965) conceived of a child as initially being unaware of their own thought processes, where egocentrism impedes logic and perspective taking is at a minimum. The concrete-operational stage of cognitive development, which is conceptualised as beginning around 7 years of age, entails growth in perspective-taking ability, whereby the child is able to better differentiate between and consider their own and another person's perspective (Piaget & Inhelder, 1966/Reference Piaget, Inhelder and Weaver1969). The development of an understanding regarding self and other perspectives drives the ability to both self-reflect and take the perspective of the other. Selman's (Reference Selman1980) later theory of the development of social perspective taking stresses the importance of self-reflection within the development of this cognitive empathic response. At level 2 (approximately ages 7–12 years), children are able to take an outsider's perspective on their thoughts and behaviours and more generally take the perspective of the other. This self-reflection is built upon at level 3 (approximately ages 10–15 years), and a third-person perspective emerges, whereby consideration of multiple perspectives (i.e., both self and other) and coordination of these perspectives are simultaneously possible.
In studies stemming from self-awareness theory and the self-consciousness perspective that directly investigated the self-reflection and perspective-taking relationship, self-awareness was often induced first, examining the impact that reflection had on taking others’ perspectives. These studies suggest that when participants are made self-aware, they take an outsider's perspective on their own thoughts and behaviours, which is consonant with theories of self-reflection that propose the importance of another perspective on one's own behaviour (Diener & Srull, Reference Diener and Srull1979; Hass, Reference Hass1984). In one study involving self-reinforcement tasks, participants adhered more to socially derived standards than to personal standards when made self-aware (with a television screen and an audio recording), and adhered more to personal standards than to social standards when not made self-aware (Diener & Srull, Reference Diener and Srull1979). In a series of studies, participants who were made self-aware (through the presence of a video camera or audio recorder) prior to beginning the task more often outlined the letter E on their forehead backwards, making it a correct orientation for an external person (Hass, Reference Hass1984). However, dispositional private or public self-consciousness had little effect on the behaviour of participants. These findings involving perceptual perspective taking have much in common with earlier work, where participants who watched themselves on videotape provided relatively more dispositional attributions for behaviour (akin to the perspective of an observer), in comparison to the usual situational focus inherent in the actor-observer effect (Storms, Reference Storms1973; see also Jones & Nisbett, Reference Jones and Nisbett1971).
The relationship between dispositional rather than situationally induced self-reflection and perspective taking has been examined. Two experiments by Bernstein and Davis (Reference Bernstein and Davis1982) investigated the influences of a participant's trait perspective taking (using the Interpersonal Reactivity Index; Davis, Reference Davis1980) and a real target's private self-consciousness, on the accuracy of the participants’ matching of targets to self-descriptions. In the first study, participants high in perspective taking (determined via a median split) were more accurate in matching self-descriptions to 10 targets who they had watched on video discussing a problem. In the second study, participants found it easier to predict targets that were higher in private self-consciousness. Interactions were found first for perspective taking by tape length, and second for private self-consciousness by tape length, indicating that the effects of both variables were more prevalent for longer videotapes. These results were interpreted as reflecting the greater opportunity for participants to use their perspective-taking ability with more information (thus avoiding problems such as projection), as well as increased tape length providing more information on a target who had provided more adequate self-descriptions. Both studies point to the similar functions of perspective taking and self-reflection in understanding and interpreting either another's or one's own emotions, cognitions, and behaviours. In one of the few studies that has investigated private self-consciousness and perspective taking within intimate relationships (Franzoi, Davis, & Young, Reference Franzoi, Davis and Young1985), private self-consciousness and perspective taking both predicted relationship satisfaction in heterosexual dating couples, with the private self-consciousness and satisfaction association mediated by degree of self-disclosure. Unfortunately, the extent to which self-reflection (or self-disclosure) is implicated in perspective taking and relationship satisfaction was not a focus of this investigation.
Collectively, these results suggest that improvement in perspective-taking ability may be facilitated through awareness of differences between self and other perspectives via self-focused attention. At the same time, specific concerns such as time-pressures (Darley & Batson, Reference Darley and Batson1973) or self-validational needs (Gollwitzer & Wicklund, Reference Gollwitzer and Wicklund1985) have been found to compromise perspective taking and empathic responses. Across three experiments (Stephenson & Wicklund, Reference Stephenson and Wicklund1983), participants who were made self-aware (through audio or visual means) made fewer perspective-taking errors such as attributing privileged knowledge to a target or exhibited increased perspective taking (where this was measured) than both a control and a self-aware group who were also exposed to a personal concern condition (e.g., being told that their questionnaire responses would be assessed for neuroticism). These results suggest that it is imperative to consider the nature or ‘types’ of self-focus that individuals undertake, particularly the affective and motivational aspects underlying such self-reflection.
Types of Reflection
Self-reflection has been hypothesised and empirically shown as being related to perspective taking, as well as to increases in self-knowledge, self-attribution, and more general self-focus. However, these findings do not provide much information on the nature of this self-reflection. It could be argued that the type of self-reflection that individuals undertake (e.g., a ruminative-type focus) would determine the relationship to both investigated outcomes of reflection, as well as perspective taking. In addition, self-reflection, like perspective taking, is used in the service of some goal (e.g., better understanding of a relevant aspect of the self; the reduction in a discrepancy or self-defensive reasons—see Franzoi, Davis, & Markwiese, Reference Franzoi, Davis and Markwiese1990), so that the motivation for such reflection becomes important to also address.
Psychometric Investigation of Types of Self-Reflection
Many of these issues have actually been addressed within psychometric work on the private self-consciousness subscale of the Self-Consciousness Scale (Fenigstein et al., Reference Fenigstein, Scheier and Buss1975) described above, which measures reflection on one's own thoughts and feelings. This work indicates that there are at least two distinctive types of private self-consciousness tapped by the scale (e.g., Abrams, Reference Abrams1988; Anderson, Bohon, & Berrigan, Reference Anderson, Bohon and Berrigan1996; K.M. Cramer, Reference Cramer2000; Kingree & Ruback, Reference Kingree and Ruback1996; A.J. Martin & Debus, Reference Martin and Debus1999; Mittal & Balasubramanian, Reference Mittal and Balasubramanian1987; Piliavin & Charng, Reference Piliavin and Charng1988; cf. Bissonnette & Bernstein, Reference Bissonnette and Bernstein1990; Britt, Reference Britt1992), with differing relationships between these types of reflection and various intrapersonal and interpersonal outcomes.
Drawing on the work of A.H. Buss (Reference Buss1980), Burnkrant and Page (Reference Burnkrant and Page1984) divided private self-consciousness items into two factors: internal state awareness (i.e., of affect, mood, and so on) and self-reflectiveness (e.g., heightened self-focus, self-related inquiry). In two studies using confirmatory factor analysis, there was greater support for the appropriateness of a two-factor model of private self-consciousness over a one-dimensional model. Based on their own factor analytic and correlational work, Anderson et al. (Reference Anderson, Bohon and Berrigan1996) suggested that these subscales be differently conceptualised, with self-reflectiveness considered ‘a negative and oppressive style of private self-consciousness, a style characterized by self-mistrust and ruminative self-preoccupation’ (named self-oppression), and internal state awareness ‘a more neutral and perhaps mildly positive style of private self-consciousness, one characterized by interest and attention’ (named balanced self-awareness; p. 150). Similarly, it has been suggested that items can be grouped into two categories: those items involving a negative ruminative-type focus that is fairly overarching in its scope; and other items that relate more to being attuned to aspects of the self (e.g., feelings, problem-solving techniques), which is related to self-regulation (A.J. Martin & Debus, Reference Martin and Debus1999). In studies in which a two-factor structure was adopted, self-reflectiveness items were negatively related to self-esteem and positively associated with depression and anxiety, with the inverse apparent for internal state awareness items (Anderson et al., Reference Anderson, Bohon and Berrigan1996; Höping, de Jong-Meyer, & Abrams, Reference Höping, de Jong-Meyer and Abrams2006; Kingree & Ruback, Reference Kingree and Ruback1996; Piliavin & Charng, Reference Piliavin and Charng1988).
Trapnell and Campbell (Reference Trapnell and Campbell1999) suggested that self-consciousness, and in particular private self-consciousness, should be further theorised as involving motives and values that determine the nature of an individual's self-focus. Based on a series of studies utilising several thousand participants, measures of rumination and reflection were developed (Rumination-Reflection Questionnaire [RRQ]). Whereas the conceptualisation of rumination stressed ‘a general, neurotic category of self attentiveness . . . prompted by threats, losses, or injustices to the self’, reflection was considered more intellective, motivated by ‘pleasurable, intrinsic interest in abstract or philosophical thinking’ (p. 292) rather than distress. The subsequent scales were psychometrically distinct, and both were significantly correlated with the original private self-consciousness measure. Associations were found between rumination and neuroticism-related concepts (e.g., anxiety, depression, negative affect), and between reflection and openness to experience-related concepts (e.g., need for cognition, personal identity-related measures), as well as a relationship between volunteering for psychology studies and reflection, suggested to indicate differences between those high and low on this trait in knowledge-related needs.
Few studies have examined the relationship between rumination and reflection using the RRQ and perspective taking, as well as the relationship between these types of self-reflection and past experience. In one of the few studies, relationships between the measures and perspective taking (using the Interpersonal Reactivity Index [IRI]; Davis, Reference Davis1980) were relatively straightforward, with reflection significantly positively associated with perspective taking and rumination significantly negatively associated with perspective taking (Joireman, Parrott, & Hammersla, Reference Joireman, Parrott and Hammersla2002b). Rumination was significantly positively associated with two affective components of empathy: empathic concern, which involves compassionate and sympathetic feelings for another person; and personal distress, which involves self-focused discomfort and unease when exposed to another person's troublesome experiences (Batson et al., Reference Batson, Early and Salvarani1997). Reflection was also significantly positively associated with empathic concern, although the inclusion of perspective taking and personal distress on a second step in regression analysis rendered this relationship non-significant (relationships between reflection and rumination and perspective taking and personal distress remained significant when other empathy measures were included).
In a follow-up study (Joireman, Reference Joireman2004), reflection and rumination mediated relationships between perspective taking and personal distress and guilt and shame. Most important to the present discussion regarding self-reflection and perspective taking, reflection partially mediated the relationship between guilt (predictor variable) and perspective taking (outcome variable). However, a hypothesised empathic concern model mirroring this perspective taking model was not supported. In addition, models where rumination mediated the relationship between shame (predictor) and personal distress (outcome) or shame mediated the relationship between rumination (predictor) and personal distress were supported.
The relationships between reflection and rumination and autobiographical memory were investigated in a study whereby participants generated memories related to a set of neutral cue words (Teasdale & Green, Reference Teasdale and Green2004). Rumination was negatively related to feelings of happiness and ‘at-oneness’ at the time of the event, and was positively related to feelings of unhappiness at the time of the event; only ‘at-oneness’ remained a significant predictor in regression analyses including all three types of affect. Reflection was not significantly related to the nature of the event. Support for the role of higher order constructs in the rumination and memory relationship was provided when a measure of neuroticism was included with rumination in regression analysis predicting ‘at-oneness’. In this analysis, neuroticism but not rumination was a significant predictor.
Another measure of self-reflection, the Self-Reflection and Insight Scale (SRIS; Grant, Franklin, & Langford, Reference Grant, Franklin and Langford2002), which consists of two scales, examines not only dispositional self-reflection, but also the nature of that reflection. The Self-Reflection scale (SRIS-SR) measures an individual's propensity to engage in reflection on their thoughts, feelings, and behaviours, together with the extent to which the individual feels the need to engage in this process. The Insight scale (SRIS-IN) measures an individual's understanding of their thoughts, feelings, and behaviours.
In three validation studies, the SRIS-SR was significantly positively correlated with anxiety and stress measured using the Depression Anxiety Stress Scales (Lovibond & Lovibond, Reference Lovibond and Lovibond1995); the SRIS-IN was significantly negatively correlated with these measures (and depression on that scale). Both measures significantly correlated with the original measure of private self-consciousness (Fenigstein et al., Reference Fenigstein, Scheier and Buss1975) — the SRIS-SR positively and the SRIS-IN negatively — and insight correlated significantly and negatively with the 20-item version of the Toronto Alexithymia Scale (Bagby, Parker, & Taylor, Reference Bagby, Parker and Taylor1994) and significantly and positively with the Cognitive Flexibility Scale (M.M. Martin & Rubin, Reference Martin and Rubin1995) and the Self-Control Schedule (Rosenbaum, Reference Rosenbaum1980). The two scales were not significantly associated in one of the validation studies and significantly negatively associated in another of the studies.
Finally, associations between both self-reflection and perspective taking and other psychological processes suggest that individuals disposed, encouraged, or experimentally manipulated to self-reflect exhibit more valid (accurate) or insightful processing of experiences and taking the perspectives of others. For example, one study (Beitel, Ferrer, & Cecero, Reference Beitel, Ferrer and Cecero2005) found that private self-consciousness, perspective taking, and empathic concern positively predicted psychological mindedness in regression analysis.
This work on reflection, rumination, and insight raises important concerns for the present discussion and the investigation of self-reflection and perspective taking. While definitions of reflection denote an understanding and search regarding aspects of the self that is conducive to perspective taking (if, as the current examination suggests, reflection on past experiences and its accompanying thoughts and feelings leads to an application to the other's situation), the definition of rumination has a particular focus on past experience and problematic experiences. Indeed, past experiences — at least those investigated in previous studies (e.g., relationship problems, bereavement; see Batson, Reference Batson2011) — would likely contain an aspect of ruminative self-focus by the very nature of their affective tone and impact on the individual's life. Research in the area of self-reflection and perspective taking would suggest that self-reflection can only be effective as a process of switching between one's own perspective and that of another when more self-oriented concerns are absent. As such, what needs to be accounted for is how past experience reflection can lead to beneficial (e.g., accurate, relationship-enhancing, or easy) outcomes, when these experiences are often negative. In the final section, clinical theory and research with a particular focus on past experience is used to provide a potential answer.
Self-Reflection in Clinical Theory and Practice
Self-Reflection and Psychological Disorders
There is a large body of research examining the relationship between self-reflection and psychological disorders, particularly depression (e.g., Nolen-Hoeksema, Reference Nolen-Hoeksema1991; Pyszczynski, Greenberg, Hamilton, & Nix, Reference Pyszczynski, Greenberg, Hamilton and Nix1991). In a review of this literature, Ingram (Reference Ingram1990) suggested that self-focused attention is a common feature of several psychopathologies (e.g., depression, anxiety), is implicated in the underlying features of other disorders (e.g., schizophrenia, psychopathy), and may serve both an initiation and intensification function in relation to negative affect (cf. Pyszczynski et al., Reference Pyszczynski, Greenberg, Hamilton and Nix1991).
In Mor and Winquist's (Reference Mor and Winquist2002) meta-analysis investigating the relationship between self-focused attention and negative affect, weighted mean effect sizes for 149 correlational studies and 72 experimental studies involving a self-focus manipulation were 0.51 and 0.44, respectively. Stronger positive associations between self-focus and negative affect were present for: clinical and subclinical samples versus non-clinical samples; rumination versus non-ruminative self-focus; public self-consciousness versus private self-consciousness; attention to negative versus positive aspects of the self; the aftermath of a negative versus positive event; and depression versus anxiety. The direction of the relationship between self-focused attention and negative affect was also examined, with moderate effect sizes found that did not differ significantly in magnitude between studies employing a self-focus manipulation, studies manipulating negative affect, and correlational studies that did not involve a manipulation. The authors concluded that ‘these results seem to support the reciprocity of the relationship between self-focus and NA [negative affect]’ (p. 650).
In Nolen-Hoeksema's (Reference Nolen-Hoeksema1991) influential response styles theory, rumination is implicated in depression through retrieval of more negative memories and past experiences, increased focus on current problems, and pessimistic expectations for the future; interference with problem solving and instrumental behaviour; and reduced social support (for a review of studies, see Nolen-Hoeksema, Wisco, & Lyubomirsky, Reference Nolen-Hoeksema, Wisco and Lyubomirsky2008). Indeed, in a study of the psychometric properties of Nolen-Hoeksema's Ruminative Responses Scale (see Nolen-Hoeksema & Morrow, Reference Nolen-Hoeksema and Morrow1991), a two-factor model similar to that of Trapnell and Campbell's (Reference Trapnell and Campbell1999) Reflection and Rumination scales was supported. The factors were reflection, ‘a purposeful turning inward to engage in cognitive problem solving to alleviate one's depressive symptoms’; and brooding, ‘a passive comparison of one's current situation with some unachieved standard’ (Treynor, Gonzalez, & Nolen-Hoeksema, Reference Treynor, Gonzalez and Nolen-Hoeksema2003, p. 256).
Ingram (Reference Ingram1990) developed a conceptualisation of self-reflection or focus relevant to psychopathology. This stressed three important variables or parameters: degree of focused attention on the self (particularly in comparison to focus external to the self; these two constructs are formulated as existing along a continuum); duration of this focused attention; and flexibility in attention change. It was suggested that maladaptive self-focused attention, or self-absorption, is characteristic of an excessive degree of self-focus for an extended period of time and with limited change in direction of attention. The content of the self-absorption is related to the activation of a particular schema, which, according to Ingram, explains differences between disorders with regard to self-related thought.
Clinical Research and Psychological Models of Change
While the work described above did not examine self-reflection and perspective taking directly, it suggests that thinking about one's own past experiences in the service of perspective taking could be seen as involving a temporary shift away from the person with whom one is empathising toward a self-focus, but to a degree whereby thinking has not become solely self-focused, ruminative, or brooding (Joormann, Dkane, & Gotlib, Reference Joormann, Dkane and Gotlib2006; Takano & Tanno, Reference Takano and Tanno2009).
The actual content of self-related thought would also need to be of a certain quality. Much like perspective taking only being sufficient to the extent that the individual is accurate about the cognitions and emotions of the other person, the self-reflection most needed would be one where the individual has veridical assessment of their cognitions, emotions, and behaviours in the previous situation, so that these can be applied to a new situation and another person. The notion of insight into, or accurate self-knowledge of, oneself, which was introduced earlier, is particularly important here. Insight has actually been studied alongside empathy. For example, in a series of studies on empathy, Dymond investigated both insight into the relationship patterns one has with others (Dymond, Reference Dymond1948) and insight into one's own personality traits (Dymond, Reference Dymond1949, Reference Dymond1950). Finding that insight and empathy were positively related, Dymond suggested that ‘empathy may be one of the underlying mechanisms on which insight is based’ (1948, p. 233).
While the notion of or term insight is somewhat more specific to certain therapies than others, most cognitive-related psychotherapies address inaccurate cognitions and attempt to generate more accurate self-knowledge (Beck, 1976/Reference Beck1989; Meichenbaum, Reference Meichenbaum1977). Indeed, introspection and thinking about past experiences through imagery are described as ways to examine cognitions, as well as to uncover dominant errors in thinking and underlying attitudes and beliefs (Shaw & Beck, Reference Shaw, Beck, Ellis and Grieger1977; Young, Rygh, Weinberger, & Beck, Reference Young, Rygh, Weinberger, Beck and Barlow2008). Similarly, rational-emotive therapy involves the detection, examination, and active challenge of one's irrational beliefs regarding events (the A-B-C theory where Belief mediates between an Activating event and a Consequence), resulting in the development of new beliefs (or effect; Ellis, Reference Ellis, Ellis and Grieger1977, Reference Ellis and Bernard1991). In insight-oriented psychotherapies (e.g., Freudian psychoanalysis, Jungian analysis, person-centered psychotherapy), it is the process of self-exploration that is said to lead to an accurate or truthful uncovering of information (Jopling, Reference Jopling2001).
The idea of working toward understanding of and insight into past experiences (and their underlying cognitive, emotional, and behavioural aspects) seems vital in the case of the present examination. Indeed, processing or thinking deeply about a specific past experience is unlikely to happen solely in the moment when one is empathising with another person in a similar situation. Instead, it is likely that the use of past experience involves applying past experience to the new situation in the moment, but also consideration and processing of one's own situation outside of the specific perspective-taking episode. Models of psychotherapeutic stages of change stress the process of working toward such an understanding. According to the assimilation model (Stiles, Reference Stiles2001; Stiles et al., Reference Stiles, Elliott, Llewelyn, Firth-Cozens, Margison, Shapiro and Hardy1990), change involves assimilation and accommodation of problematic experiences into schema. This process occurs in a number of stages, beginning with limited awareness of the problem to awareness and clarification of the problem and related insight, working through the problem, and moving toward problem solution and mastery (see Day, Bryan, Davey, & Casey, Reference Day, Bryan, Davey and Casey2006; Honos-Webb, Stiles, & Greenberg, Reference Honos-Webb, Stiles and Greenberg2003; Stiles, Reference Stiles2001). As Stiles (Reference Stiles2001) suggests:
Another way to formulate the change process is using the metaphor of voice . . . This metaphor expresses the theoretical suggestion that the traces of past experiences are active agents within people and are capable of communication. The traces can act and speak. Dissociated (unassimilated) voices tend to be problems, whereas assimilated voices can be resources, available when circumstances call for their capacities and talents. The interlinked traces of experiences that have been assimilated previously can be considered as a community of voices within the person. (pp. 462–463)
Ways to Think About Experiences
In this final section, we examine social psychological, neuroscientific, and clinical research that has investigated the perspectives an individual may take toward the experiences of others and their own experiences. What emerges from this research is that reflecting on thoughts, feelings, and behaviours from a distanced perspective or moving away from one's own point of view may be beneficial for development of the insight and assimilation described above.
There have been several studies that have investigated outcomes of asking participants to take another person's perspective in one of two ways (e.g., Batson et al., Reference Batson, Lishner, Carpenter, Dulin, Harjusola-Webb, Stocks and Sampat2003; Batson et al., Reference Batson, Early and Salvarani1997; Davis et al., Reference Davis, Soderlund, Cole, Gadol, Kute, Myers and Weihing2004; Stotland, Reference Stotland and Berkowitz1969). In these studies, which often focused on empathic emotion or altruism, participants were asked to examine another person's situation from either an imagine-self perspective, where participants imagine themselves in the other person's place, or an imagine-other perspective, where participants focus on how the other person feels about their experiences (Batson et al., Reference Batson, Early and Salvarani1997). A third control or comparison condition tends to discourage the participant from taking the perspective of the other person, instead asking them to remain objective.
In these studies, differences between participants in imagine-self and imagine-other groups have been observed on physiological, self-report, and behavioural measures. In Stotland's (Reference Stotland and Berkowitz1969) seminal study, participants received one of the three perspective-taking instructions and watched while an experimental confederate ostensibly received a diathermy treatment in front of them. Participants in an imagine-him pain condition were found to experience vasoconstriction reactions that were temporally related to when the diathermy machine was in operation (suggesting, with participant self-report, that participant reactions were tied to the situation of the other), while those in the imagine-self pain condition exhibited palmar sweating and self-reported in ways that were less temporally related to the operation of the machine.
In a study that investigated instructional sets from a self-report perspective, Batson et al. (Reference Batson, Early and Salvarani1997) asked undergraduate students to listen to an ostensible broadcast of a student describing her experiences after her parents had been killed in an automobile accident. Participants in either imagine condition reported greater (and not significantly different) empathic concern for the target than those in the objective condition, while those in the imagine-self condition reported greater personal in comparison to the two other groups (which did not significantly differ). It is well documented that personal distress inhibits other-oriented responses (e.g., helping), while empathic concern increases such responses (Batson, Reference Batson2011).
In two later studies reported by Batson et al. (Reference Batson, Lishner, Carpenter, Dulin, Harjusola-Webb, Stocks and Sampat2003), participants in the imagine-other condition were significantly more likely to assign a positive task to another individual (opposed to themselves) than those participants the imagine-self condition and a condition where no perspective-taking instructions were given; in addition, participants who had imagined their own perspective were significantly more likely to choose to assign one raffle ticket each to themselves and a target rather than keep both tickets, in comparison to participants in the objective condition (the imagine-other perspective was not used). These findings suggest that while participants asked to perspective take using either instructional set report increased empathic emotion and behaviour, those who focus on their own point of view may not be as effective in formulating an empathic response.
It is important to note that in both imagine-self and imagine-other conditions, participants report reflecting on both their own thoughts and experiences and those of the other person (e.g., Davis et al., Reference Davis, Soderlund, Cole, Gadol, Kute, Myers and Weihing2004; Gerace et al., Reference Gerace, Day, Casey and Mohr2015). For example, in manipulation checks reported in the Batson et al. (Reference Batson, Early and Salvarani1997) study, there was no significant difference in the extent to which participants in the imagine-self condition reported imagining their feelings and the feelings of the target person. In addition, those in the imagine-other condition reported imagining their own feelings to a greater extent than those in the objective condition (although they reported imagining the others’ feelings to a greater extent than imagining their own).
In a study involving a thought-listing procedure, recipients of imagine-self and imagine-target instructions differed (in the expected directions) in the number of self-reported and target-reported thoughts they listed after watching a talk show vignette (Davis et al., Reference Davis, Soderlund, Cole, Gadol, Kute, Myers and Weihing2004). Interestingly, participants who were not given any instructions regarding perspective taking produced results most consistent with the imagine-target scenario, leading the researchers to suggest that real-life (non-induced) perspective-taking may be most consistent with the type of cognitive work of participants in an imagine-target condition. However, research by Vorauer and Sucharyna (Reference Vorauer and Sucharyna2013) that investigated perspective taking in romantic relationship and close friendship dyads found that, in comparison to participants in a control condition where no specific perspective-taking instructions were given, participants who took an imagine-other (target) perspective overestimated the extent to which their values or preferences would be transparent to their relationship partner or friend; these participants also reported in a thought-listing task an increased focus on the self as an object of the other person's evaluation. In both cases, imagine-self participant responses did not differ from those of control condition participants. Furthermore, a relationship between taking an imagine-other perspective and decreased relationship satisfaction was mediated by perceived transparency of one's negative feelings.
These findings raise an intriguing question: why does inferring the perspective of another involve the use of one's own experiences and perspective? One explanation advanced is that perspective taking leads to increasingly similar mental representations of the self and the target (Davis, Conklin, Smith, & Luce, Reference Davis, Conklin, Smith and Luce1996). In one study, participants generated lists of traits to describe themselves, and then did so for a video vignette target. Instructions required participants take an imagine-self, imagine-other, or watch-target perspective. Results indicated that participants in the imagine-self or imagine-other perspective-taking groups ascribed a higher percentage of traits they had used to characterise themselves to the other person, than did those participants in the objective group. These results thus suggest that perspective taking may lead to a self-other merging or overlap in a way that involves some level of application of what is known about the self to the other (see also Batson et al., Reference Batson, Early and Salvarani1997; Cross, Bacon, & Morris, Reference Cross, Bacon and Morris2000; Cialdini, Brown, Lewis, Luce, & Neuberg, Reference Cialdini, Brown, Lewis, Luce and Neuberg1997). This is not to say that perspective taking leads to more projection or activation of self-knowledge when targets are perceived as more like oneself. In fact, it has been reported that while participants in both an imagine-other and control condition demonstrated greater projection when exposed to a similar versus dissimilar target, imagine-other participants reported lesser projection compared to control participants for similar targets, and more projection for dissimilar targets (Todd, Simpson, & Tamir, Reference Todd, Simpson and Tamir2016). In another study, self-other overlap was associated with more situational attributions for a romantic relationship partner (Aron & Fraley, Reference Aron and Fraley1999); situational attributions, as mentioned, are indicative of taking another person's perspective (see Storms, Reference Storms1973).
Neuroscientific research has investigated shared and unique brain regions and neural activation associated with whether a person is focused on their own point of view and actions or that/those of another person (for a review, see Lamm, Decety, & Singer, Reference Lamm, Decety and Singer2011). At a theoretical level, the perception-action model (Preston & de Waal, Reference Preston and de Waal2002) proposes that empathising with another person's ‘state automatically activates the subject's [empathiser] representations of the state, situation, and object, and that activation of these representations automatically primes or generates the associated autonomic and somatic responses, unless inhibited’ (p. 4). However, as found in a number of studies, while there is ‘a widely shared network of activation between self and other representation’ (Ruby & Decety, Reference Ruby and Decety2004, p. 995), when one focuses on either their own or another person's perspective, there is also unique neural activation in a number of other areas indicative of self-other distinction (e.g., D'Argembeau et al., Reference D'Argembeau, Ruby, Collette, Degueldre, Balteau, Luxen, Maquet and Salmon2007; Lamm, Batson, & Decety, Reference Lamm, Batson and Decety2007).
Lack of self-other distinction and an excessive focus on one's own emotions is likely to be problematic. In a study that examined both brain activation and self-reported empathic responses, Lawrence et al. (Reference Lawrence, Shaw, Giampietro, Surguladze, Brammer and David2006) found that self-other overlap (measured using a procedure similar to that of Davis et al., Reference Davis, Conklin, Smith and Luce1996) was negatively associated with accuracy in judging bodily expressions using the Profile of Nonverbal Sensitivity (PONS; Rosenthal, Hall, DiMatteo, Rogers, & Archer, Reference Rosenthal, Hall, DiMatteo, Rogers and Archer1979), and that higher trait personal distress (measured using the IRI; Davis, Reference Davis1980) was associated with longer latency in labelling mental states. In concluding, the researchers suggested that ‘there is an optimal degree of reliance on first person knowledge in social perception and that a complete blurring of self and other is detrimental’ (p. 1182), and that personal distress may have an inhibitory effect on other-oriented empathic responses.
Studies have also examined the type of perspective one takes to one's own experiences. In particular, studies by Kross and colleages (Kross & Ayduk, Reference Kross and Ayduk2008; for reviews, see Ayduk & Kross, Reference Ayduk and Kross2010a; Kross, Reference Kross2009) present two ways of thinking about past experiences. The first is a self-immersed perspective, where people ‘visualize events happening to them through their own eyes’, and which is likely to involve a focus on what had occurred and was felt. The second is a self-distanced perspective, where examination of experiences occurs ‘from the perspective of an observer or “fly on the wall”’ (Ayduk & Kross, Reference Ayduk and Kross2010a, p. 843). This self-distanced perspective facilitates a focus on understanding the why of the experience, and leads to ‘reconstructing it in ways that promote insight and closure’ (Kross, Reference Kross2009, p. 36). While a self-immersed perspective is likely to lead to rumination and negative affect, a self-distance perspective is seen to involve more adaptive self-reflection.
In one study, participants recalled an anger experience from one of the two perspectives, and with a focus on either the emotions felt (i.e., the ‘what’ of the event) or the reasons for those emotions (i.e., the ‘why’ of the event). Those participants in the self-distanced perspective group who focused on the reasons for their emotions, in comparison to participants in the three other groups, reported less implicit (using a word completion task) and self-reported anger and negative affect. In a follow-up study by the same authors, participants took one of the two perspectives (both groups focused on reasons for emotions), wrote about an anger experience, and reported on the extent to which, during the task, they experienced the emotions associated with the original event and the intensity of that experience. Narratives were analysed for two types of underlying construals: concrete construals, where participants described experienced events, emotions, and behaviours, and attributed blame to another person; and abstract construals, where participants included ‘metacognitive insight statements describing a “realization” about or change in the way the participant understood the causes underlying the event, his or her feelings, or his or her partner’, and ‘metacognitive closure statements . . . taking into account past and current experiences to make sense of the past’ (Kross, Ayduk, & Mischel, Reference Kross, Ayduk and Mischel2005, p. 713). There was support for a proposed mediational path where a self-distanced perspective led to less concrete versus abstract construals, resulting in decreased emotional reactivity. This mediational path has been replicated when self-focus is not manipulated; that is, participants report on how naturally immersed or distant they were when thinking about a negative past experience (Ayduk & Kross, Reference Ayduk and Kross2010b). In an investigation involving in vivo provocations, the effects of a self-distanced versus self-immersed perspective on anger, aggressive thoughts and behaviours were in the expected directions (Mischkowski, Kross, & Bushman, Reference Mischkowski, Kross and Bushman2012).
There is also support for the effects of type of self-reflection on emotional and cognitive outcomes in clinical populations. Participants who had a current major depressive disorder were asked to engage in either experiential (focusing on experience of symptoms) or analytical (thinking about meaning of or evaluating symptoms) self-focus regarding their depressive symptoms (Watkins & Teasdale, Reference Watkins and Teasdale2004). Participants in the experiential self-focus condition decreased significantly in their proportion of recalled overgeneral memories, which are implicated in depression. This appears somewhat contradictory to findings regarding the self-immersed and self-distanced perspectives. However, in this conceptualisation, analytical self-focus is of a more ruminative and less mindful self-awareness.
In the last decade, an increasing amount of work has also been conducted on another other-oriented response turned inward: compassion toward the self (for a review, see Barnard & Curry, Reference Barnard and Curry2011). Neff (Reference Neff2003) defined self-compassion in the context of suffering or personal failure with reference to other-oriented compassion, and suggested that it involved:
(a) self-kindness — extending kindness and understanding to oneself rather than harsh judgment and self-criticism, (b) common humanity — seeing one's experiences as part of the larger human experience rather than seeing them as separating and isolating, and (c) mindfulness — holding one's painful thoughts and feelings in balanced awareness rather than over-identifying with them. (p. 89)
These components imply a less ruminative, self-immersed perspective and a move toward insightful processing of one's experiences. It has been reported that brooding (the rumination component of the Ruminative Responses Scale [RRS]; Nolen-Hoeksema & Morrow, Reference Nolen-Hoeksema and Morrow1991) but not reflection (measured using the RRS reflection subscale) or worry partially mediated the relationship of self-compassion to depression, and that rumination and worry (but not reflection) partially mediated the self-compassion and anxiety relationship (Raes, Reference Raes2010; see also Neff, Kirkpatrick, & Rude, Reference Neff, Kirkpatrick and Rude2007).
Unlike research on self-immersed and self-distanced perspectives (Ayduk & Kross, Reference Ayduk and Kross2010a), work has been conducted to examine the relationship between self-compassion and other-oriented empathy. In this work, higher self-compassion was associated with balancing one's own and other people's concerns, as well as more beneficial outcomes in relationship conflicts with family, friends, and partners (Yarnell & Neff, Reference Yarnell and Neff2013; for study on self-perspective and problem solving, see Ayduk & Kross, Reference Ayduk and Kross2010b). Higher self-compassion was also associated with perspective taking, empathic concern, and personal distress (negatively), as well as compassion for humanity, forgiveness, and altruism (Neff & Pommier, Reference Neff and Pommier2013).
Discussion
In this review, we have documented theory and research relevant to an understanding of the role that self-reflection has to play in the perspective-taking process and, in particular, the use of similar past experiences to understand another person. Self-reflection and perspective taking are conceptualised as interdependent (e.g., Piaget & Inhelder, 1966/Reference Piaget, Inhelder and Weaver1969), but there has been limited empirical investigation of the ways in which self-reflection might influence the taking of another person's psychological point of view. Our review of the dominant research streams, namely those related to objective self-awareness (Duval & Wicklund, Reference Duval and Wicklund1972) and private self-consciousness (Fenigstein et al., Reference Fenigstein, Scheier and Buss1975), revealed, however, that the outcomes of self-reflection, such as increased accuracy in understanding thoughts, feelings, and behaviours, are similar to those of perspective taking (Bernstein & Davis, Reference Bernstein and Davis1982), although the focus of the individual is different (i.e., self versus other).
What emerges is the importance of considering the purpose of reflecting on one's thoughts, feelings, behaviours, and, most importantly, past experiences; and whether this involves more negatively toned rumination or more positively framed reflection and insight (Trapnell & Campbell, Reference Trapnell and Campbell1999). Rumination is negatively associated with perspective taking, while reflection that is driven by wanting to understand oneself is positively associated with taking the perspective of others. Clinical research further suggests that reflection and understanding (or ‘insight’) can be conceptualised as a process, in much the same way that perspective taking is considered a process (Dymond, Reference Dymond1950; Gerace et al., Reference Gerace, Day, Casey and Mohr2013), in which the individual becomes temporarily self-focused but is aware and able to switch between a focus on himself or herself to that of the environment and others (Ingram, Reference Ingram1990). Taking an outsider's perspective and moving away from one's own perspective on experienced events is one way to develop such an understanding (Ayduk & Kross, Reference Ayduk and Kross2010a; Batson et al., Reference Batson, Early and Salvarani1997; Kross, Reference Kross2009).
Drawing on this research, we believe that self-reflection on past experiences is a prerequisite for working toward any resolution of problematic experiences (Day et al., Reference Day, Bryan, Davey and Casey2006; Stiles, Reference Stiles2001), and that this understanding can then be applied to interpreting others’ similar situations. While rumination and adverse reactions to problematic experiences may be inevitable, it is through the process of attempting to make sense of, and work through, such experiences (and understanding the associated thoughts, feelings, and behaviours), that self-reflection aids the process of perspective taking. It is therefore likely that similar past experiences are insufficient in seeking to understand another person unless the experience has been reflected upon and assimilated. This helps to explain the mixed research findings regarding whether those with similar past experiences are more accurate in apprehending another's experience (e.g., Hodges et al., Reference Hodges, Kiel, Kramer, Veach and Villanueva2010). It may be that when personal distress is prevalent, or there is not sufficient differentiation between self and other when trying to understand another person, accuracy is likely to be compromised (Lawrence et al., Reference Lawrence, Shaw, Giampietro, Surguladze, Brammer and David2006).
The literature reviewed herein raises a number of questions, both empirical and applied. Primarily, it seems important for future studies to directly examine the potential role of self-reflection and insight on past experiences in understanding others’ perspectives. Previous studies that examined different types of reflection (Ayduk & Kross, Reference Ayduk and Kross2010a) on one's own experiences did not investigate other-oriented perspective taking. Researchers could ask participants to generate a similar experience to a presented target's and measure their rumination, more positively focused self-reflection, and insight into their particular past experiences or, indeed, encourage them to develop some insight and assimilation in vivo through taking a distanced perspective. Examination of whether this resulted in increased ability to take the other's perspective, in terms of ease, accuracy, as well as emotional (e.g., empathic concern, personal distress) and behavioural responses (e.g., intentions to help, relationship-enhancing behaviours, aggression) could then be undertaken. Studies could examine participant retrospective experiences or those performed in experimental manipulations (e.g., Mischkowski et al., Reference Mischkowski, Kross and Bushman2012), as well as using experimental confederates/vignettes and actual existing relationships. For example, a study by Ayduk and Kross (Reference Ayduk and Kross2010b) involved couples reflecting on conflict and engaging in the laboratory, and this paradigm could be extended to other types of relationship. It is likely that, particularly in close relationships, self-other merging and perceptions of similarity to the other person would influence effects.
The relationship between self-reflection and perspective taking is likely not to be straightforward and to be influenced by several factors. For example, there are likely specific situations where constructive self-reflection on and insight into experiences are detrimental to reaching an understanding of the other. This may become apparent when the individual has reflected and assimilated experiences, perhaps almost forgetting what it may be like to be at a different stage of working through the problem. It is also possible that ruminative involvement may actually make apprehending another's experience more vivid and easier to imagine. Of course, due to the nature of rumination, insight into the experience is unlikely to occur, which would result in deficits in perspective-taking abilities and negative emotional reactions. Indeed, a degree of ambiguity is apparent in studies that have investigated the relationship between different types of self-reflection and empathic emotion, particularly empathic concern (Joireman et al., Reference Joireman, Parrott and Hammersla2002b). The relationship of empathic concern to measures of emotional reactivity (Davis, Reference Davis1983) could explain some of the ambivalence, but further research is needed to unpack the component of empathic concern and other responses such as personal distress. That self-compassion is related to resolution of conflict suggests that the nature of the focus of this emotion (i.e., self versus other) is also important to consider.
Turning to applications based on previous and future work, one applied question is whether past experience reflection can be fostered through intervention. Although reflection makes it easier to take another's perspective, it is not always undertaken — even when the individual has similar experience to draw upon (Gerace et al., Reference Gerace, Day, Casey and Mohr2015). Based on results from studies that induce self-awareness, as well as review of psychotherapeutic practices where inaccurate perceptions, cognitions, and notions are examined (e.g., Beck, 1976/Reference Beck1989; Ellis, Reference Ellis, Ellis and Grieger1977), it is nonetheless likely that this is possible. Indeed, engaging in positive abstract forms of self-reflection may be a more deliberate process than rumination and require intervention. Rumination, in particular, involves an almost automatic focus on the self, with situations or dispositions triggering a negative style of focus. This type of focus is inherently excessive or egocentric, fixed upon the self in a way that renders the other-focus required for perspective taking impossible (Joireman et al., Reference Joireman, Parrott and Hammersla2002b). In comparison, conceptualisations of positive reflection imply self-awareness motivated by either a need for self-awareness or similar interests (Franzoi et al., Reference Franzoi, Davis and Markwiese1990; Trapnell & Campbell, Reference Trapnell and Campbell1999). Thus, any focused attempts to increase reflection of the positive or adaptive type should serve to highlight the need and importance of accurate self-related knowledge for understanding others, as well as eradicating problematic areas where self-reflection is avoided. The result would be heightened insight or self-awareness, with direct benefits for perspective taking.
It can be argued that not focusing on problematic situations and associated thoughts and feelings is recommended for people who have experienced particularly negative situations. In the short-term, a self-distanced perspective or distraction task are similarly effective in reducing negative affect and thoughts about an experience (Kross & Ayduk, Reference Kross and Ayduk2008), supporting proposals for the use of short-term distraction in facilitating problem solving and distress reduction in the presence of negative affect (Nolen-Hoeksema et al., Reference Nolen-Hoeksema, Wisco and Lyubomirsky2008; Watkins & Teasdale, Reference Watkins and Teasdale2001). However, a self-distanced perspective is more effective than a self-immersed focus or distraction task over time, suggesting that, eventually, some insightful processing of experiences is needed.
A second question is whether the tendency to self-reflect is a disposition. This is important from an intervention point of view, where attempts are made to influence dominant or trait-like styles of thinking, feeling, and behaving. Since similar outcomes are evidenced in studies whether self-focused attention is manipulated or measured as a disposition, finding ways to increase self-reflection that occurs through environmental stimuli or related means to such an extent that it becomes a trait-like propensity (or tendency) could result in positive change (see Davis & Franzoi, Reference Davis and Franzoi1991). In addition, since regular self-focus and attention — when positively motivated and framed — could lead to a greater complexity and understanding of experiences, this seems particularly useful to the process of taking another's perspective. However, as most of the more recent research has employed dispositional measures of self-reflection, further investigation with methods to increase self-reflection (e.g., mirrors) is warranted.
One clinical intervention concordant with the notion that self- and other-understanding are interdependent is chair-work. Although there are variants of this tool, Greenberg, Elliott, and Lietaer (Reference Greenberg, Elliott, Lietaer, Stricker and Widiger2003) describe the method (with two chairs) as follows:
The therapist initiates the two-chair dialogue by suggesting that the client move back and forth between two chairs, each representing one self-aspect, in order to enact the internal conversation between the two parts . . . In the case of an attributional split, the client is asked to enact the other or external situation. (p. 318)
The focus of chair-work on perspective taking, self-reflection, and self-insight may be particularly useful for people where problems in any of these areas are apparent (Day, Howells, Mohr, Schall, & Gerace, Reference Day, Howells, Mohr, Schall and Gerace2008).
Fostering self-reflection in clinicians is also likely to help them to better take the perspective of their clients, to understand the therapeutic relationship, and to manage their own responses. The importance of self-reflection in clinical practice, including awareness of biases, boundaries, and emotional reactions to clients, is advocated in a number of disciplines (Day, Reference Day2012; Eng & Pai, Reference Eng and Pai2015), and within psychology, this is reflected in the scientist-practitioner model (Blair, Reference Blair2010). For example, clinical supervision and peer debriefing are interpersonal methods that are commonly used to facilitate such reflection (Peternelj-Taylor & Yonge, Reference Peternelj-Taylor and Yonge2003). Other methods, such as a clinician visualising a previous interaction as a play on a stage (Somerville & Keeling, Reference Somerville and Keeling2004), also encourage a self-distanced perspective (Kross, Reference Kross2009).
Therapists are also encouraged to reflect on their own past experiences in order to improve their taking of clients’ perspectives. However, this needs to be done prudently, particularly given that professional distance and lack of self-disclosure is advocated in the helping disciplines (Gerace, Oster, O'Kane, Hayman, & Muir-Cochrane, Reference Gerace, Oster, O'Kane, Hayman and Muir-Cochrane2016; Peternelj-Taylor & Yonge, Reference Peternelj-Taylor and Yonge2003; Rogers, Reference Rogers1957). Hermansson (Reference Hermansson1997), for example, discusses the nature of boundaries in counselling and psychotherapy, which is similar to an understanding of perspective taking as involving both self- and other-information (Gerace et al., Reference Gerace, Day, Casey and Mohr2013). According to Hermansson, ‘to be empathic, a counsellor has to move across a boundary into the life space of the client. . . . To maintain therapeutic potency, however, it must be a qualified boundary cross, with the counsellor never totally leaving his or her own personal territory’ (p. 140).
Implications for Close Relationships
While we anticipate that the main tenets of the proposed relationship between self-reflection and perspective taking investigated in this article apply to interactions between people with different degrees of acquaintance, more research has focused on interactions between new acquaintances or even unseen strangers. There is comparatively less research investigating self-reflection and perspective taking with close relationship partners. However, in close relationships, partners may be particularly motivated to understand and communicate with one another due to factors such as the ongoing nature of the relationship, interdependence, and commitment (Goodwin, Fiske, Rosen, & Rosenthal, Reference Goodwin, Fiske, Rosen and Rosenthal2002; A.M. Gordon & Chen, Reference Gordon and Chen2013).
As discussed, focusing on the other person's perspective (taking an imagine-other perspective) rather than imagining oneself in their position (imagine-self perspective) results in more other-oriented focus, empathic emotion, and behaviour (Batson et al., Reference Batson, Early and Salvarani1997, Batson et al., Reference Batson, Lishner, Carpenter, Dulin, Harjusola-Webb, Stocks and Sampat2003). Similarly, reflecting on one's experiences from a distanced vantage point akin to taking an imagine-other perspective leads to more insight into one's experiences (Ayduk & Kross, Reference Ayduk and Kross2010a). However, in research examining the effects of these instructional sets in romantic relationships and friendships (Vorauer & Sucharyna, Reference Vorauer and Sucharyna2013), participants in imagine-other conditions erroneously believed that their point of view would be more transparent to a relationship partner than what it was. While based on limited investigation, this suggests that in relationships with increased connection, there may be the tendency to attribute too much knowledge to, or not being able to distance from, a relationship partner. Indeed, as the researchers suggested, perceived transparency of one's own perspective may lead to less communication of one's thoughts, feelings, and past experiences to a partner; in turn, the partner may then form an inaccurate or different perception of the other's point of view and relationship experiences.
Another factor that may influence the extent to which individuals use self-reflection to understand others is perceived similarity (e.g., values, preferences) to a close relationship partner. In studies investigating perspective taking in non-intimate relationships, perceptions of similarity can lead to projection of one's own thoughts, feeling, and beliefs when trying to understand that other person's behaviour (Ames, Reference Ames2004, cf. Todd et al., Reference Todd, Simpson and Tamir2016). While these studies show that in less close relationships this can lead to inaccurate perceptions and, in the case of self-reflection, inappropriate use of past experiences to understand others, in close relationships similarity perceptions can influence relationship outcomes such as satisfaction. Within romantic relationships, the picture is complex, with studies finding that people in unhappy marriages more often erroneously assumed that they were similar to their partners and exhibited less empathic accuracy than did people in happy marriages (Dymond, Reference Dymond1954), while other studies have highlighted the positive effects of inaccurate perceptions of similarity (Murray, Holmes, Bellavia, Griffin, & Dolderman, Reference Murray, Holmes, Bellavia, Griffin and Dolderman2002; for a review, see Fletcher & Kerr, Reference Fletcher and Kerr2010). These findings suggest that perceptions of a relationship or interaction partner may help or hinder the empathiser's attempts to use knowledge about the self to understand a relationship partner, and for the other partner to feel understood.
It is also likely that other personality factors beyond trait self-reflection or perspective taking are important in close relationships. For example, in a series of studies by A.M. Gordon and Chen (Reference Gordon and Chen2013), participants who experienced greater power in their relationship reported less consideration of, and were less accurate in understanding of, their romantic relationship partner's perspective when more self-focused (e.g., being individualistic or competitive) than other-focused (e.g., appreciating and seeing a partner and relationship as important to one's own self-construal). Similarly, in an investigation of the association between attachment styles and trait empathy (Joireman, Needham, & Cummings, Reference Joireman, Needham and Cummings2002a), perspective taking and empathic concern were related to measures denoting secure attachment (e.g., depending on relationship partners, being close to partners, including sharing thoughts and feelings) and less avoidant approaches to relationships; personal distress was related to measures indicating anxiety, avoidance, and less secure relationship attachment.
It seems appropriate, therefore, that encouraging use of multiple strategies to consider one's partner's perspective (e.g., imagine-other, imagine-self, and remaining objective) using methods outlined such as chair-work may ameliorate empathic problems stemming from inaccurate perceptions or traits of partners. In addition, encouraging relationship partners to share their thoughts and feelings with their partners is also likely to improve both perspective taking, through the ability to clarify perceptions, and the relationship, through relationship partners demonstrating effort to understand one another. Across several studies, perceptions of one's partner's perspective taking have been found to be a stronger indicator of relationship satisfaction than the partner's self-reported perspective-taking tendency or actual accuracy (Cohen, Schulz, Weiss, & Waldinger, Reference Cohen, Schulz, Weiss and Waldinger2012; D. Cramer & Jowett, Reference Cramer and Jowett2010; Hodges et al., Reference Hodges, Kiel, Kramer, Veach and Villanueva2010; Meeks, Hendrick, & Hendrick, Reference Meeks, Hendrick and Hendrick1998). Similarly, regarding past experiences, disclosure of similar past experience to a stranger target is associated with greater target perceptions of being understood; like other perspective-taking studies, this effect is regardless of actual empathiser accuracy (Hodges et al., Reference Hodges, Kiel, Kramer, Veach and Villanueva2010). This suggests that relationship satisfaction may arise from interactional factors such as sharing one's own perspective, rather than solely accurate perception, though the association between more accurate perception and relationship satisfaction (Dymond, Reference Dymond1954; Vorauer & Sucharyna, Reference Vorauer and Sucharyna2013) highlights the importance of insight and understanding.
To investigate self-reflection and perspective taking in close relationships, dyadic approaches are likely required. A promising and under-utilised model of social interaction developed and investigated by Davis and colleagues (Davis & Kraus, Reference Davis, Kraus, Jones and Perlman1991; Davis & Oathout, Reference Davis and Oathout1987, Reference Davis and Oathout1992) proposes that individual dispositions, most relevant to the present discussion being perspective-taking propensity, lead to specific behaviours in interpersonal relationships that are, subsequently, evaluated by the other individual in the relationship, and lead to outcomes such as satisfaction. Such a model could be used to investigate further the role of self-refection and perspective taking in close relationship processes, as well as the main components of the model being used in interventional work with couples.
In couple counselling and therapy this would likely involve addressing how relationship partners see one another, believe they are perceived, and communicate their thoughts, feelings, and understanding to the other person (see Benson, McGinn, & Christensen, Reference Benson, McGinn and Christensen2012). Increasing the use of specific behaviours (such as empathic listening, responsiveness, and taking time to hear the other's perspective) would also help to enhance both self-and other-awareness (Benson et al., Reference Benson, McGinn and Christensen2012; Finkenauer & Righetti, Reference Finkenauer and Righetti2011). Approaches such as emotion-focused therapy and emotionally focused therapy that examine individuals’ conceptions of self and others, including addressing these conceptions and how individuals process past interactions and experiences, may be particularly helpful in this respect (Busby & Gardner, Reference Busby and Gardner2008; R.N. Goldman & Greenberg, Reference Goldman and Greenberg2013; Johnson, Reference Johnson1996; Makinen & Jonhson, Reference Makinen and Johnson2006); as are approaches that examine how relationship partners conceptualise the relationship in their ideas about the self (Acitelli, Rogers, & Knee, Reference Acitelli, Rogers and Knee1999). While acknowledging the limited empirical research in the area, Block-Lerner, Adair, Plumb, Rhatigan, and Orsillo (Reference Block-Lerner, Adair, Plumb, Rhatigan and Orsillo2007) further advocate for the utility of mindfulness and acceptance-based methods — reflected in psychotherapeutic interventions such as acceptance and commitment therapy, mindfulness-based stress reduction, and mindfulness-based cognitive therapy — to improve empathy between couples, particularly in the context of conflict or rumination.
Conclusion
In this review, we suggest that a useful way to understand others is to self-reflect and focus on one's own past experiences. The most pressing areas for intervention and research are to increase the extent to which individuals draw on appropriate previous experiences in empathic situations and to further examine how these processes play out in different relationships. In addition, encouraging individuals to think about their experiences outside of the specific empathy episode should be fostered (i.e., self-reflection), which would allow for more positive self-reflection and the development of insight, and a move away from rumination and personal distress. The focus of intervention, then, would be on providing individuals with methods to make it easier to perspective take, through making the use of past experiences a more often utilised, and appropriately used, skill.
Financial Support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Conflict of Interest
None.
Ethical Standards
Not applicable (did not involve human and/or animal experimentation).
