Influence of outcome valence in the subjective experience of episodic past, future, and counterfactual thinking

Abstract

Recent findings suggest that our capacity to imagine the future depends on our capacity to remember the past. However, the extent to which episodic memory is involved in our capacity to think about what could have happened in our past, yet did not occur (i.e., episodic counterfactual thinking), remains largely unexplored. The current experiments investigate the phenomenological characteristics and the influence of outcome valence on the experience of past, future and counterfactual thoughts. Participants were asked to mentally simulate past, future, and counterfactual events with positive or negative outcomes. Features of their subjective experiences during each type of simulation were measured using questionnaires and autobiographical interviews. The results suggest that clarity and vividness were higher for past than future and counterfactual simulations. Additionally, emotional intensity was lower for counterfactual simulations than past and future simulations. Finally, outcome valence influenced participants’ judgment of probability for future and counterfactual simulations.

Introduction

Memory and imagination are commonly considered different cognitive faculties. This assumption is held in folk psychology and largely shared by philosophers and psychologists (e.g., Russell, 1921, Stout, 1915). Despite arguments from empiricists philosophers such as Thomas Hobbes (1651) and Hume (1739) that memory and imagination are a single faculty—or at a minimum two integrated faculties sharing many operations—the succeeding philosophical tradition remained largely unmoved by their proposals (e.g., Locke, 1971). Recent scientific evidence, however, is starting to lend credence to Hobbes and Hume’s views, inviting us to question our assumptions about the correct way of individuating the faculties of memory and imagination.

Of particular interest is the substantial amount of evidence gathered during the past three decades suggesting that our capacity to imagine our future may depend on our capacity to remember past events—both processes that constitute what is often referred to as “mental time travel” (Tulving, 1985). Neuropsychological studies have shown that people who have deficits in mentally re-experiencing events that happened in their pasts, also exhibit impairments when mentally pre-experiencing events that may happen in their future, a cognitive process that has come to be known as episodic future thinking (Atance and O’Neill, 2001, Szpunar, 2010). Evidence to this effect comes from research on amnesic subjects (Hassabis et al., 2007, Klein et al., 2002, Tulving, 1983), patients with severe depression (Dickson and Bates, 2005, Williams, 1996), individuals with a diagnosis of schizophrenia (D’Argembeau et al., 2008), young children (Busby and Suddendorf, 2005, Perner et al., 2010, Suddendorf and Busby, 2005), older adults (Addis, Wong, & Schacter, 2008), individuals with amnestic mild cognitive impairment (Gamboz et al., 2010), and patients with Alzheimer’s disease (Addis et al., 2009, Addis et al., 2010). Additionally, research on healthy individuals using neuroimaging techniques has revealed a significant overlap in the brain regions engaged during episodic recollection and episodic future thinking, suggesting that both processes may be mediated by similar cognitive mechanisms (Addis and Schacter, 2008, Addis et al., 2007, Hassabis et al., 2007, Okuda et al., 2003, Schacter et al., 2007, Szpunar et al., 2007). Finally, to further examine the connection between these two processes, researchers have explored the phenomenological features of episodic future thinking in relation to the phenomenology of episodic memory (D’Argembeau and Van der Linden, 2004, D’Argembeau and Van der Linden, 2006, Szpunar and McDermott, 2008, Winfield and Kamboj, 2010). These studies have revealed striking parallels between the subjective experience of both cognitive processes, further suggesting the involvement of similar cognitive mechanisms Schacter and Addis (2007).

In an exciting new development in the search for the cognitive mechanisms of mental time travel, Addis and collaborators (2009) showed that a common “core brain network” engaged during episodic memory and future thinking (Buckner & Carroll, 2007) is also involved in the mental construction of imaginary past events. Further research has shown that the involvement of such brain network is greater during mental simulations of alternative past events we think are more plausible to having occurred relative to simulations of imaginary alternatives we consider implausible (De Brigard et al., submitted for publication). Related studies are starting to provide convergent evidence suggesting that the core brain network engaged when thinking about what may happen in our future is also recruited when we consider what could have happen in our past yet did not occur (Addis, Pan, Vu, Laiser & Schacter, 2009; Kurczek et al., 2010, Van Hoeck et al., 2010, Van Hoeck et al., 2012)—a cognitive process we call episodic counterfactual thinking (De Brigard et al., submitted for publication). The idea behind this line of research is that the common mechanisms that allow episodic memory to reconstruct our personal past and to construct thoughts about possible events that could happen in our future, also enable us to construct thoughts about possible events that, although they did not happen, could have happened in our past.

However, the phenomenology of our thoughts about alternative ways in which our past could have occurred remains largely unexplored. The main purpose of the current experiments was to investigate the nature of the subjective experience of episodic counterfactual thinking, and to contrast it with the phenomenology of episodic recollection and episodic future thinking. We conjectured that this direct contrast could yield one of two results. One possibility was that the phenomenological ratings for episodic autobiographical recollections would be higher than episodic counterfactual thoughts, which in turn would be higher than episodic future thoughts since, presumably, counterfactual thoughts may involve less recombination of memory components than future thoughts. A second possibility, however, was that changing the temporal dimension of the mental simulation from past to future would require less mental manipulation than counterfactual thinking, in which case we would expect to see higher phenomenological ratings for future than for counterfactual thoughts, with episodic autobiographical recollections receiving higher ratings overall. By directly contrasting these three kinds of mental simulation the current studies seek to investigate these possibilities.

In addition, previous studies on the phenomenological characteristics of episodic autobiographical memories (D’Argembeau et al., 2003, Schaefer and Philippot, 2005) and future thinking (D’Argembeau and Van der Linden, 2004) have shown that valence influences the quality of the experience, with events yielding positive outcomes normally receiving higher qualitative ratings than events with negative outcomes. However, in such studies, the valence of the outcomes of the simulated events remained fixed, so that memories that yielded an outcome with a certain valence were simulated as future events with the same outcome valence. Since episodic counterfactual thinking allows one to flip the valence of the outcome of an event, it constitutes a suitable way to investigate the influence of outcome valence on the subjective experience of mental simulations about episodic past, future, and counterfactual events. Thus, if the quality of the experience during the simulation of a counterfactual event differs from the quality of the experience when the same event is either remembered or simulated as a future possibility, then it would suggest that the emotional valence of the outcome of an event, and not only the event itself, can affect the phenomenological quality of the mental simulation. Moreover, this manipulation affords the opportunity to investigate whether the reported higher ratings for positive past and future simulations (D’Argembeau and Van der Linden, 2004, D’Argembeau et al., 2003) are dependent upon the valence of the outcome of the event, or rather whether they depend on the quality of the simulated event regardless of its outcome.

To investigate these issues, in the current experiments participants were asked to remember episodic autobiographical events, think of possible future events, and construct alternative ways in which past experienced events could have turned out. Half of the mental simulations they engaged in had positive outcomes, whereas the other half had negative outcomes. The features of their subjective experiences during these three kinds of mental simulations were measured using adapted forms of the phenomenological characteristic questionnaire (Johnson, Foley, Suengas, & Raye, 1988) and the autobiographical memory interview (Levine, Svoboda, Hay, Winocur, & Moscovitch, 2002).