Recollection, belief and metacognition: a reality check

ABSTRACT

Non-believed autobiographical memories [e.g. Mazzoni, G., Scoboria, A., & Harvey, L. (2010). Nonbelieved memories. Psychological Science, 21, 1334–1340] are striking examples of divergences between recollective experiences and beliefs in their correspondence to real events. After reviewing a broader range of similar phenomena, I argue that recollection–belief divergences can arise from normal, “healthy” metacognitive monitoring and control processes that balance memory recollections and reality constraints. Such validating “reality checks” draw on general world knowledge and external/social information. Importantly, changes in (perceived) reality constraints can lead to changes in memory beliefs. More generally, both recollection and (external) reality are keys to the past. In many cases, more or less automatic (System 1-type) reliance on recollection is sufficient (or memory would be useless as a system), but sometimes more elaborate (System 2-type) reality checks are needed. I conclude with some ideas about memory-driven and reality-driven recollection–belief divergences.

KEYWORDS:

• Memory
• recollection
• belief
• metacognition
• reality

Acknowledgements

I am grateful to James Ost for helpful comments on a draft.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1 Beliefs related to autobiographical events – the example I started out with – are typically general, pertaining to whether or not the whole event ever truly happened. In the context of this special issue, it is important to point out that Scoboria et al. (2014) distinguish such general “beliefs in occurrence” (a concept first introduced by Mazzoni & Kirsch, 2002) from what they variously call “beliefs in recollection” or “beliefs in accuracy”, that is, beliefs not just in the occurrence of an event but more specifically in the accuracy of the recollected event details (see also Rubin, 2006; Scoboria, Talarico, & Pascal, 2015). I doubt that these two concepts can be sharply separated, mainly because there is no universally agreed-upon definition of “event”, and consequently what is an event associated with a belief in occurrence relative to one description (e.g., “the attacker pulled a knife”) may be a detail, related to a belief in accuracy, relative to a different description (e.g., “there was an assault”) – it is essentially a matter of event “grain size”. For the present purposes, it is more important to highlight what both types of beliefs have in common, namely that recollections are assigned a (subjective) truth value reflecting their correspondence to the actual past. This also applies to my independently developed notion of memory beliefs (Blank, 2001, 2009), which encompasses both understandings of “belief” discussed so far (i.e., occurrence and accuracy) and which is also more inclusive in that it refers to any episodic memories (not only autobiographical ones). In the remainder of this article, I will variously speak of beliefs or memory beliefs, but always in the most adequate sense for the present purposes, that is, beliefs at any level of generality/specificity that come with a claim to past reality.

2 A feeling of recognition (as featured in déjà vu) would translate into familiarity rather than recollection in some contexts (see e.g., Yonelinas, 2002). This distinction is not important for the argument I am developing here. What is crucial is the conceptual distinction between any initial output of the memory system – whether a feeling of familiarity or a detailed recollection of an event – and the truth value (in the sense of corresponding to past reality) assigned to this output. In keeping with the theme of this special issue, I will stick to the notion of recollection, but use it more broadly to refer to any type of recollective experience, however vague or detailed.

3 As an exception to this rule, the divergence may also be resolved in favour of the initial recollection if details or actions are so clearly stereotype-inconsistent that they had already been encoded (“tagged”) as deviations from the stereotype (Stangor & McMillan, 1992).

4 I am slightly simplifying things here; in many cases, there is a (substantial) decrease but not a complete lack of belief in the memory. See Mazzoni et al. (2010) and Scoboria et al. (2014) for details related to autobiographical event memories; the same holds respectively for original event memories in misinformation studies and recollections of childhood sexual abuse. The important point, for our purposes, is that belief is substantially lower than recollection along a clarity/strength dimension.

5 By external I mean in this context “external to the memory recollection”. For instance, world knowledge may be individually acquired and internally stored, but it is separate from episodic or autobiographical recollections and in this sense considered external evidence.

6 In fact, Mazzoni and Kirsch’s (2002) discussion of autobiographical beliefs (introducing the concept of “belief in occurrence”, see Note 1) draws in part on Koriat and Goldsmith’s work.

7 It is a logical possibility that in these examples not only the memory beliefs but also the recollections themselves varied along with reality constraints. However, in line with research showing higher inertia of recollections than beliefs (Clark et al., 2012; Otgaar et al., 2013), theoretical reviews focussing on beliefs as the locus of memory suggestibility effects (Blank, 2001, 2009; Blank et al., 2016; Nash et al., 2015), as well as considerations of theoretical parsimony, this logical possibility is not very compelling.

8 I use the System 1 and System 2 terminology rather loosely, as a shorthand for distinguishing between fast, automatic and unconscious processes on the one hand (System 1) and slow, deliberative and conscious processes on the other hand (System 2). System 1 processes do not require plenty of cognitive resources, but System 2 processes do. These characterisations represent the endpoints of a continuum of processing, and in any given situation there may be a variable mixture of both types of processing. I do not commit to any specific dual-processing model, nor to the idea that these systems physically exist as such (e.g., somewhere in the brain); see Evans' (2008) review for a critical discussion of these issues.