Visual perspective as a two-dimensional construct in episodic future thought

https://doi.org/10.1016/j.concog.2021.103148Get rights and content

Highlights

  • We measure visual perspective as “first-person predominance” and self-visibility.

  • Both of these variables independently predict subjective ratings of vividness.

  • Self-visibility predicts ratings of emotional intensity.

  • Some participants report acentred imagery while others report switching between perspectives.

  • Trait dissociation is related to the tendency to switch between perspectives.

Abstract

Visual perspective (first-person vs. third-person) is a salient characteristic of memory and mental imagery with important cognitive and behavioural consequences. Most work on visual perspective treats it as a unidimensional construct. However, third-person perspective can have opposite effects on emotion and motivation, sometimes intensifying these and other times acting as a distancing mechanism, as in PTSD. For this reason among others, we propose that visual perspective in memory and mental imagery is best understood as varying along two dimensions: first, the degree to which first-person perspective predominates in the episodic imagery, and second, the degree to which the self is visually salient from a third-person perspective. We show that, in episodic future thinking, these are anticorrelated but non-redundant. These results further our basic understanding of the potent but divergent effects visual perspective has on emotion and motivation, both in everyday life and in psychiatric conditions.

Introduction

One of the earliest works discussing visual perspective in memory is Freud’s 1899 essay on “screen memories”—vivid childhood memories of seemingly anodyne situations which, he argued, were replacements for or sanitized versions of distressing but autobiographically significant events (Freud, 1962). The essay draws on Victor and Catherine Henri’s survey of early childhood experiences (Nicolas, Gounden, & Piolino, 2013) to argue that memories retrieved from the perspective of an outside observer (“third-person” memories), as early memories often are, cannot be faithful re-renderings of the original experiences, which must have been originally experienced from a first-person perspective. Freud argued that these memories must instead have been “worked over”.

The topic seems to have lain dormant for nearly a century until the seminal work of Nigro and Neisser (1983), which not only re-established the existence of third-person memories, but explored memory characteristics predictive of first- or third-person perspective. Among these were self-awareness (predictive of third-person perspective) and a focus on emotions rather than objective details (predictive of first-person perspective). Also, older memories were generally more likely to be recalled from a third-person perspective.

Visual perspective is not an immutable characteristic of memory. Robinson and Swanson (1993) note that people report switching between first- and third-person points of view within the same memory. This switching is easier for more recent and vivid memories and, while the switch from first- to third-person is associated with reduced affect, the reverse switch is not associated with increased affect. They suggest that when the sensory-affective details of a memory are lost, it tends to be retrieved from a third-person perspective; only the cognitive component of the memory trace remains, and forcing a switch to first-person perspective cannot bring forth sensory-affective detail that is no longer accessible. However, forcing a switch away from a naturally-occurring first-person perspective involves suppressing accessible detail, resulting in reduced affect (Robinson & Swanson, 1993).

Subsequent findings have aligned well with this model: first-person perspective in natural recall is associated with greater emotional intensity (McIsaac and Eich, 2002, D’Argembeau et al., 2003, McIsaac and Eich, 2004, Talarico et al., 2004), detail (McIsaac and Eich, 2002, McIsaac and Eich, 2004), vividness (Sutin & Robins, 2010), and sense of reliving (Berntsen & Rubin, 2006). There are also asymmetries in perspective switching (decreases after the first-to-third-person switch but no increase after the reverse switch) for sense of reliving (Berntsen & Rubin, 2006), emotional intensity (Vella and Moulds, 2014, Gu and Tse, 2016), vividness (Williams & Moulds, 2008), and episodic detail (Akhtar, Justice, Loveday, & Conway, 2017). Furthermore, retrieval from a third-person perspective alters memories, reducing vividness (Butler, Rice, Wooldridge, & Rubin, 2016), accuracy (Marcotti & St. Jacques, 2018), emotional intensity (Sekiguchi & Nonaka, 2014) and sensory-affective detail (Bagri & Jones, 2009) at subsequent recall. First-person perspective appears, then, to be closely related to the episodicity of memory. Indeed, the first-person–third-person distinction is closely related (though not identical; Sutton, 2010) to the remember–know distinction (Crawley & French, 2005) and the proportion of memories recalled from a first-person perspective follows a reminiscence bump for events in young adulthood (Piolino et al., 2006).

Several recent studies have found that visual perspective in memory recall is a reliable individual-differences variable (Siedlecki, 2015, Verhaeghen et al., 2018), and it may be one with important connections to psychological health. Third-person perspectives are argued to be constructed from “contextual” memory representations that are dissociable from the low-level sensory representations underlying vivid intrusions (Brewin, Gregory, Lipton, & Burgess, 2010). As such, individuals with PTSD often deliberately adopt a third-person perspective when recalling trauma memories to avoid reliving them (McIsaac & Eich, 2004). Similarly, more avoidant survivors of trauma are more likely to adopt third-person perspectives on their trauma memories (Kenny & Bryant, 2007). Third-person perspective memories are also more common among depressed individuals (Kuyken & Howell, 2006, but cf. McFadden & Siedlecki, 2020), possibly for positive events in particular (Lemogne et al., 2006, Nelis et al., 2013) or for individuals with a tendency toward cognitive avoidance of intrusions (Williams and Moulds, 2007, Kuyken and Moulds, 2009). Finally, trait dissociation is linked to increased third-person perspective in memory recall (Williams and Moulds, 2007, Sutin and Robins, 2010, Radvansky and Svob, 2019). The fact that third-person perspective can dampen affect apparently causes it to function as a psychological distancing mechanism across psychiatric conditions.

The personal future can also be imagined from a first- or third-person perspective through a process known as episodic future thinking (EFT; Atance & O’Neill, 2001). Just as visual perspective in memory recall has important implications for affect, visual perspective in EFT has important implications for motivation and decision making. First-person perspective tends to emphasize sensory-affective information while third-person perspective tends to highlight more abstract information that contextualizes events, and this difference has behavioural consequences (Libby & Eibach, 2011). For example, 90% of registered voters who visualized voting in the 2004 American presidential election from a third-person perspective actually followed through with their voting intentions, compared to 72% who visualized voting from a first-person perspective (Libby, Shaeffer, Eibach, & Slemmer, 2007). Similarly, visualizing an academic achievement from a third-person perspective increases motivation toward it more than visualization from a first-person perspective, an effect mediated by high construal level, which casts academic achievement within the larger context of more abstract goals (Vasquez and Buehler, 2007, Trope and Liberman, 2010). Finally, visualizing the future self from a third-person perspective increases hypothetical retirement saving, an effect mediated by the visual salience of the future self (Macrae et al., 2017).

There is an apparent contradiction in the divergent effects of visual perspectives on motivation and emotion: why should third-person perspective be associated with increased motivation if it is associated with dampened affect? To resolve this, McCarroll (2019) argues that third-person perspective does not always function as an emotional distancing mechanism. Instead, it can sometimes enhance self-conscious emotions such as pride and thereby motivate future-oriented decisions. In the memory literature, Sutin and Robins (2008) have introduced a model in which third-person perspective can either heighten or dampen affect: for positive memories that are consistent with the current self-concept, third-person perspective will heighten affect by bringing the remembered self into the visual foreground of a memory, thereby increasing self-focus (the “salient self” effect). Meanwhile, for negative memories that are inconsistent with the current self-concept, third-person perspective will dampen affect by creating psychological distance between the remembering and remembered self (the “dispassionate observer”; see Rice (2010), for a discussion of this model’s limitations).

Extending this model to EFT, third-person perspective should increase or decrease motivation toward future goals depending on whether the “salient self” or the “dispassionate observer” effect prevails, which in turn should depend on whether the goal in question is consistent with the current self-concept. In line with this, pursuit of health goals is hindered by third-person visualization if they are peripheral to the self-concept (Stornelli, Pereira, & Vann, 2020).

Much of the memory and EFT literature on visual perspective treats it as a unidimensional construct, with first- and third-person perspectives at opposite poles (e.g., Berntsen & Rubin, 2006). However, the fact that third-person perspective can have divergent effects on emotion/motivation suggests that visual perspective is not unidimensional. Here we make a case against a unidimensional model of visual perspective (see also Rice & Rubin, 2009).

The definition of a third-person perspective usually contains 2 criteria: a point of view different from first-person and the visibility of the remembered or imagined self (Nigro & Neisser, 1983). These two criteria appear to correspond to the two divergent effects of third-person perspective posited by Sutin and Robins (2008): a shift in perspective away from first-person should amplify the “dispassionate observer” effect, while the vivid visibility of the self should amplify the “salient self” effect.

Moreover, these two criteria seem logically separable: the fact that a point of view is not first-person does not necessarily mean that the remembered or imagined self will be visible (Sutton, 2010, McCarroll, 2019). This proposal derives from the concept of “acentred” memories, in which the point of view is not the same as the original, but neither is the self visible in the memory scene, having been “edited out” (Wollheim, 1984). However, as far as we are aware, there are presently no empirical data on the prevalence of this phenomenon.

Additionally, there is an infinite number of spatial locations and angles from which a third-person perspective can be constructed (Callow and Roberts, 2010, Morris and Spittle, 2012), and the particular third-person vantage point from which an observer memory is retrieved relates to that memory’s content (e.g., when remembering giving a presentation, one is more likely to recall the scene from in front of oneself; Rice & Rubin, 2011). Clearly, then, third-person perspective is not just one end of a unidimensional construct but is instead a category of perspectives within which meaningful variation exists.

Finally, it is possible to adopt multiple perspectives when retrieving a single memory. For example, Berntsen and Rubin (2006) provide the following description of a memory in which the perspective switches:

I see myself dancing at a party at the university. I remember my clothes and my legs (the way they moved). Suddenly, I am “inside my own body” looking out. A guy I know a little walks by me and says as he passes: “You look good today”.

Thus a single memory or imagined future episode can be described as involving both first- and third-person perspectives—i.e., as simultaneously occupying both opposite poles of visual perspective understood as a unidimensional construct. Switching between perspectives is also observed in sports imagery, where it has been shown to benefit performance (Epstein, 1980, Gordon et al., 1994, Smith et al., 1998). It has even been suggested that multiple perspectives can coexist simultaneously within a mental image in a way that is not possible in perception (Sutton, 2012). This is based in part on the observation that spontaneous verbal descriptions of space often employ route (first-person) and survey (third-person) descriptions within a single clause (Taylor & Tversky, 1992). Similarly, Sartre’s view that mental images reflect rather than precede knowledge implies that multiple visual perspectives can be adopted simultaneously (Sartre (1972), as cited in McCarroll, 2018, pp. 144–145).

In light of the fact that multiple visual perspectives can co-occur within a single memory retrieval, Rice and Rubin (2009) argue that first- and third-person perspectives are independent dimensions that characterize memory, and show that the two correlate differentially with ratings of vividness. However, as discussed above, meaningful variation exists between third-person perspectives, which may not be fully captured when describing a memory or imagined future episode as involving simply more or less third-person perspective. Here we offer an alternative two-dimensional conceptualization of visual perspective.

We argue that visual perspective, in both memory and episodic future thinking, is best understood as comprising (1) the predominance of first-person perspective (henceforth “first-person predominance”) and (2) the visibility of the imagined self when a third-person perspective is adopted (henceforth “third-person self-visibility”). As noted above, people can adopt multiple visual perspectives on the same event, meaning that first-person perspective can predominate to varying degrees. Furthermore, from a third-person perspective, the imagined self may be more or less visually salient.

These two dimensions articulate the space of possibilities for visual perspective: when first-person perspective predominates and third-person self-visibility is low, the overall perspective is best described as first-person. When these are both reversed, the result is a third-person perspective as conceived by Nigro and Neisser (1983). When third-person perspective predominates but self-visibility is low, the result is acentred imagery. Finally, when both third-person self-visibility and the predominance of first-person perspective are moderate to high, this describes mental imagery in which the visual perspective switches between first- and third-person.

The present work aimed to evaluate the utility of measuring visual perspective in terms of the two dimensions proposed above. In order to measure effects on motivation and decision making, we used an episodic future thinking paradigm. We also aimed to identify perceptual and affective correlates of the two proposed dimensions.

First (prediction 1), we expected that the two dimensions would be separable, allowing the measurement of both acentred imagery (low first-person predominance and low third-person self-visibility) and perspective switching (high first-person predominance and high third-person self-visibility). Second (prediction 2), in line with previous research, we expected first-person predominance to be positively correlated with subjective ratings of vividness and negatively correlated with temporal distance (i.e. lower first-person predominance with increasing temporal distance). Third (prediction 3), based on the model of Sutin and Robins (2008), we expected emotional intensity to be positively correlated with both first-person predominance and third-person self-visibility: first-person predominance should reduce the “dispassionate observer” effect, while third-person self-visibility should increase the “salient self” effect.

We also aimed to measure trait predictors of each of these two dimensions. One likely candidate is trait dissociation, which manifests as feelings of detachment from one’s experiences and surroundings (Lyssenko et al., 2018). As noted earlier, trait dissociation is correlated with increased rates of third-person perspective in memory recall (Williams and Moulds, 2007, Sutin and Robins, 2010, Radvansky and Svob, 2019), an effect interpreted in terms of emotional distancing (i.e., the “dispassionate observer” effect). Extending this idea to EFT, we predicted that trait dissociation would be associated with low ratings of both first-person predominance and third-person self-visibility (prediction 4).

Another relevant variable is trait imagery vividness, which we expected to be positively associated with first-person predominance. Mental imagery is generated by projecting representations in long-term memory into imagery areas (Byrne et al., 2007, Pearson, 2019). This activates modality-specific sensory cortices in a top-down manner, and the strength of this activation is correlated with subjective vividness ratings (Belardinelli et al., 2009). The ability to rapidly construct vivid mental images differs between individuals (Isaac & Marks, 1994), and these differences can be understood in terms of the information-theoretic concept of channel capacity: in high trait imagery individuals, there is greater information flow from long-term memory to imagery areas (Hishitani, Miyazaki, & Motoyama, 2011). We therefore expected that individuals with greater trait imagery vividness would construct more perceptually detailed mental images of the future and, given the inherently egocentric/body-centred nature of perception, would report more first-person images of the future (prediction 5).

Finally, we explored the role of visual perspective in EFT-driven reductions of impulsivity. EFT shifts preferences away from smaller immediate rewards and toward larger delayed rewards (Peters & Büchel, 2010), and is thought to do so by simulating the value of future rewards in the present (Benoit, Gilbert, & Burgess, 2011). If this is the case, first-person perspective in EFT should increase the vividness of these simulations and thereby amplify the effect on impulsivity. However, as noted earlier, visualizing the far future from a third-person perspective increases hypothetical saving behaviour (Macrae et al., 2017). Thus we expected (prediction 6) that the effect of EFT on impulsivity is amplified by both first-person predominance (through increased vividness) and third-person self-visibility (through the “salient self” effect).

To summarize, the present work aimed to examine visual perspective in episodic future thought as a two-dimensional construct, comprising first-person predominance and third-person self-visibility. We expected that first-person predominance would be associated with increased state and trait vividness, while third-person self-visibility would be associated with positive affect and negatively correlated with trait dissociation. Finally, we expected that both first-person predominance and third-person self-visibility would be independently correlated with the effect of EFT on impulsivity.

Section snippets

Materials and methods

McMaster undergraduate students were recruited to complete an episodic future thinking task, a delay discounting task, and several questionnaires. Participants (N = 92, 77 women, 1 other/unspecified, ages 18 – 24, median = 18) completed the experiment for introductory psychology course credit, giving written consent as approved by the local research ethics board. The experiment was programmed using jsPsych (De Leeuw, 2015) and hosted on a web server that was accessed from a browser on a

Joint distribution of visual perspective dimensions

We first aimed to characterize the relationship between the two proposed dimensions of visual perspective and determine whether they are indeed separable. To this end, Fig. 1a illustrates the bivariate distribution along both dimensions of all imagined future events pooled from all participants. There was a strong negative association between the two dimensions, as indicated by a linear mixed effects model (dependent variable: third-person self-visibility; fixed effect: first-person

Discussion

The present work tested a novel conceptualization of visual perspective as comprising the two dimensions of first-person predominance and third-person self-visibility. Several lines of evidence support this model. First, the two dimensions, while anticorrelated, were separable. That is, we found evidence for both acentred imagery (low scores on both dimensions) and perspective switching (high scores on both dimensions). Second, the two dimensions were non-redundant, with both independently

Conclusions

The present work provides evidence for a novel conceptualization of visual perspective in episodic imagery as a two-dimensional construct, comprising first-person predominance and third-person self-visibility. While not all of our specific predictions regarding trait and state correlates of these were borne out, a two-dimensional understanding of visual perspective is nonetheless useful: both dimensions independently predicted vividness ratings, and third-person self-visibility was uniquely

CRediT authorship contribution statement

Isaac Kinley: Conceptualization, Data curation, Formal analysis, Writing - original draft, Writing - review & editing. Morgan Porteous: Conceptualization, Data curation, Writing - review & editing. Yarden Levy: Conceptualization, Writing - review & editing. Suzanna Becker: Conceptualization, Resources, Supervision, Writing - original draft, Writing - review & editing.

Acknowledgements

This research was supported by an Ontario Graduate Scholarship to IK and by Grant No. RGPIN-2019-07276 from the Natural Sciences and Engineering Research Council of Canada to SB.

References (101)

  • C. Lemogne et al.

    Episodic autobiographical memory in depression: Specificity, autonoetic consciousness, and self-perspective

    Consciousness and Cognition

    (2006)
  • L.K. Libby et al.

    Visual perspective in mental imagery: A representational tool that functions in judgment, emotion, and self-insight

  • J. Linke et al.

    Mental imagery training increases wanting of rewards and reward sensitivity and reduces depressive symptoms

    Behavior Therapy

    (2017)
  • G. Nigro et al.

    Point of view in personal memories

    Cognitive Psychology

    (1983)
  • G. Northoff et al.

    Cortical midline structures and the self

    Trends in Cognitive Sciences

    (2004)
  • S. O’Donnell et al.

    Does goal relevant episodic future thinking amplify the effect on delay discounting?

    Consciousness and Cognition

    (2017)
  • J. Peters et al.

    Episodic future thinking reduces reward delay discounting through an enhancement of prefrontal-mediotemporal interactions

    Neuron

    (2010)
  • H.J. Rice et al.

    I can see it both ways: First-and third-person visual perspectives at retrieval

    Consciousness and Cognition

    (2009)
  • H.J. Rice et al.

    Remembering from any angle: The flexibility of visual perspective during retrieval

    Consciousness and Cognition

    (2011)
  • A.R. Sutin et al.

    When the “i” looks at the “me”: Autobiographical memory, visual perspective, and the self

    Consciousness and Cognition

    (2008)
  • M. Vannucci et al.

    Individual differences in object and spatial imagery: Personality correlates

    Personality and Individual Differences

    (2009)
  • A.D. Williams et al.

    Cognitive avoidance of intrusive memories: Recall vantage perspective and associations with depression

    Behaviour Research and Therapy

    (2007)
  • M. Amlung et al.

    Delay discounting as a transdiagnostic process in psychiatric disorders: a meta-analysis

    JAMA Psychiatry

    (2019)
  • J. Andrade et al.

    Assessing vividness of mental imagery: the plymouth sensory imagery questionnaire

    British Journal of Psychology

    (2014)
  • K.M. Arnold et al.

    Imagining the near and far future: The role of location familiarity

    Memory & Cognition

    (2011)
  • G. Bagri et al.

    Category-specific enhancement of retrieval due to field perspective

    Memory

    (2009)
  • M.O. Belardinelli et al.

    An fmri investigation on image generation in different sensory modalities: the influence of vividness

    Acta Psychologica

    (2009)
  • R.G. Benoit et al.

    A neural mechanism mediating the impact of episodic prospection on farsighted decisions

    Journal of Neuroscience

    (2011)
  • D. Berntsen et al.

    Emotion and vantage point in autobiographical

    Cognition and Emotion

    (2006)
  • C.R. Brewin et al.

    Intrusive images in psychological disorders: characteristics, neural mechanisms, and treatment implications

    Psychological Review

    (2010)
  • A. Bulley et al.

    Deliberating trade-offs with the future

    Nature Human Behaviour

    (2020)
  • P. Byrne et al.

    Remembering the past and imagining the future: a neural model of spatial memory and imagery

    Psychological Review

    (2007)
  • E.B. Carlson et al.

    An update on the dissociative experiences scale

  • J.M. Clark et al.

    Extensions of the paivio, yuille, and madigan (1968) norms

    Behavior Research Methods, Instruments, & Computers

    (2004)
  • S. Cole et al.

    Spontaneous and deliberate future thinking: a dual process account

    Psychological Research Psychologische Forschung

    (2019)
  • Conway, M. A., Justice, L. V., and D’Argembeau, A. (2019). The self-memory system revisited. The organization and...
  • M.A. Conway et al.

    The construction of autobiographical memories in the self-memory system

    Psychological Review

    (2000)
  • S. Crawley et al.

    Field and observer viewpoint in remember-know memories of personal childhood events

    Memory

    (2005)
  • H.F. Crovitz et al.

    Frequency of episodic memories as a function of their age

    Bulletin of the Psychonomic Society

    (1974)
  • A. D’Argembeau et al.

    Phenomenal characteristics of autobiographical memories for positive, negative, and neutral events

    Applied Cognitive Psychology: The Official Journal of the Society for Applied Research in Memory and Cognition

    (2003)
  • A. D’Argembeau et al.

    Tracking the construction of episodic future thoughts

    Journal of experimental psychology: General

    (2011)
  • J.R. De Leeuw

    jspsych: A javascript library for creating behavioral experiments in a web browser

    Behavior Research Methods

    (2015)
  • M.L. Epstein

    The relationship of mental imagery and mental rehearsal to performance of a motor task

    Journal of Sport and Exercise Psychology

    (1980)
  • Freud, S. (1962). Screen memories. In The Standard Edition of the Complete Psychological Works of Sigmund Freud, Volume...
  • F. Galton

    Psychometric experiments

    Brain

    (1879)
  • T. Giesbrecht et al.

    Cognitive processes in dissociation: an analysis of core theoretical assumptions

    Psychological Bulletin

    (2008)
  • S. Gordon et al.

    Effect of internal and external imagery on cricket performance

    Journal of Sport Behavior

    (1994)
  • Hedeker, D. (2005). Generalized linear mixed models. Encyclopedia of Statistics in Behavioral...
  • S. Hishitani et al.

    Some mechanisms responsible for the vividness of mental imagery: Suppressor, closer, and other functions

    Journal of Mental Imagery

    (2011)
  • Hoerl, C. and McCormack, T. (2016). Making decisions about the future. Seeing the future: Theoretical perspectives on...
  • Cited by (0)

    View full text