Self-Defining Memories: Structure, Function and Measurement

Abstract

Self-defining memories are a small subset of autobiographical memories that are vivid, affectively intense, repeatedly recalled and linked to enduring goals or unresolved conflicts. The present review asks what distinguishes these memories structurally, what functions they serve, and how reliably they can be measured. We examine the original formulation by Singer and Salovey (1993), the four-dimension coding system of Blagov and Singer (2004), and the subsequent literature relating specificity, integrative meaning, content and affect to adjustment, defensiveness and narrative identity. We situate self-defining memories within the Self-Memory System, in which correspondence and coherence are negotiated by the working self, and review evidence for a directive function of specific episodes. Clinical studies in schizophrenia and depression indicate that impairments concern meaning-making and affective phenomenology more than memory availability. The main conclusion is that a self-defining memory is not reducible to its event content: its affective charge is inseparable from the goal structure and interpretive frame in which it is embedded. A programme aiming to preserve emotional memory must therefore capture episode, phenomenology and meaning jointly, and treat the resulting record as a snapshot of an evolving representation.

Keywords: self-defining memories; autobiographical memory; narrative identity; integrative meaning; memory specificity; emotion

1. Introduction

Autobiographical memory research has traditionally focused on the accuracy and accessibility of remembered events. A complementary tradition, rooted in personality psychology, asks which memories matter to the rememberer and why. Singer and Salovey (1993) proposed that a limited number of memories occupy a privileged relationship with identity. They termed these self-defining memories and characterised them by five properties: vividness, affective intensity, high rehearsal, linkage to similar memories, and a focus on an enduring concern or unresolved conflict of the personality. On this account, affective charge is the signal that a memory bears on goals that remain active.

Three decades of work have turned this clinical intuition into a measurable construct with a standardised elicitation task, a validated coding manual, and a body of correlational and clinical findings (Singer et al., 2013). The literature nevertheless remains predominantly cross-sectional, reliant on written narratives, and concentrated in Western undergraduate samples. The present review has three aims: to summarise the dimensions along which self-defining memories are coded and their psychometric status; to evaluate the evidence for the directive and identity functions attributed to them; and to extract from clinical comparisons what is dissociable in the system. The final section derives constraints for efforts to preserve human emotional memory.

2. Structure and measurement

2.1 The elicitation task

The Self-Defining Memory Task asks participants to describe, usually in writing, a set of memories (ten in Blagov & Singer, 2004; three in Thorne et al., 2004) satisfying the Singer and Salovey criteria, which are presented in the instructions. Participants then rate each memory on affect, importance and vividness. The task is straightforward to administer and produces broadly comparable distributions of content and specificity across countries (Lardi et al., 2010). Its main limitation is that it elicits what participants are willing and able to write, confounding repression, self-presentation and narrative skill with the underlying representation.

2.2 Four coding dimensions

Blagov and Singer (2004) formalised a coding scheme along four dimensions, summarised in Table 1. Specificity distinguishes single-event memories located in a particular time and place from summarised or generic accounts. Integrative meaning codes whether the narrative moves beyond description to state a lesson, an insight about the self, or a connection between the event and the person’s life. Content classifies the thematic domain (relationships, achievement, life-threatening events, leisure, guilt or shame). Affect is scored from participants’ valence and intensity ratings and, in some studies, from coder ratings of narrative tone. Inter-rater reliabilities for specificity and integrative meaning are consistently high, which is one reason these two dimensions dominate subsequent research.

Table 1. Coding dimensions of self-defining memories (after Blagov & Singer, 2004).

In a sample of 103 undergraduates, Blagov and Singer (2004) reported that the proportion of memories containing integrative meaning was positively associated with self-restraint, a composite of impulse control, responsibility and consideration of others on the Weinberger Adjustment Inventory. Memory specificity was inversely related to repressive defensiveness: highly defensive participants produced fewer specific and more summarised memories, which the authors interpreted as an avoidant retrieval style rather than an encoding deficit. Affect was the dimension most closely related to self-reported distress. These associations were modest and correlational, so directionality cannot be inferred; the pattern has nonetheless been replicated in its essentials (Singer et al., 2013).

2.3 Phenomenology and stability

Coding captures structure but not experience. Sutin and Robins (2007) developed the Memory Experiences Questionnaire (MEQ) to measure the phenomenology of a recalled memory along dimensions including vividness, coherence, accessibility, sensory detail, emotional intensity, visual perspective, sharing, distancing and valence. Applied to self-defining memories, the MEQ makes it possible to test whether the properties Singer and Salovey attributed to this class are elevated relative to ordinary memories, and to separate vividness from intensity. In an earlier longitudinal study, Sutin and Robins (2005) had shown that the emotions and motives expressed in self-defining memories are moderately stable across several years and related to personality traits, supporting the claim that these memories index enduring concerns rather than transient states.

Lardi et al. (2010) extended the characterisation to a Swiss francophone sample of young adults. Consistent with American data, most memories were specific, positively valenced and concerned relationships or achievement, clustering in adolescence and early adulthood. The study also related memory characteristics to self-concept measures: individuals whose memories carried more integrative meaning tended to report clearer and more positive self-views, though effects were small. The cross-cultural consistency of the basic distribution supports the construct; the modest adjustment correlates caution against strong claims.

3. Self-defining memories within the self-memory system

Conway, Singer and Tagini (2004) placed self-defining memories within a hierarchical model in which autobiographical knowledge runs from lifetime periods through general events to episodic detail, and in which retrieval is controlled by the working self, the currently active goal hierarchy. The model’s central claim is that autobiographical memory must satisfy two partly incompatible constraints. Correspondence requires that memories reflect experience with sufficient fidelity for learning and adaptive action. Coherence requires that accessible memories remain consistent with the self’s goals and beliefs. When the two conflict, the working self modulates encoding and access: memories discrepant with current self-views become harder to retrieve, whereas supportive memories are elaborated and rehearsed.

Self-defining memories, on this account, are the episodes the working self has selected because they instantiate its most important goals. They stabilise the system by anchoring abstract self-knowledge in concrete experience, and they are the memories most likely to have been reshaped by coherence pressures precisely because they are so frequently retrieved. A methodological corollary follows: vividness and confidence in a self-defining memory cannot be taken as evidence of accuracy. Experimental support for the model’s dynamic side comes from Çili and Stopa (2015), who found that retrieving a self-defining memory activated self-descriptions matching the memory’s content and shifted current affect accordingly, whereas ordinary memories produced weaker effects, consistent with functional coupling to the working self rather than storage as inert records.

4. Functions

4.1 Directive function

Pillemer (2003) argued that specific episodic memories have a directive function underrepresented in accounts emphasising semantic or narrative memory. A single well-encoded episode can carry a rule together with the contextual cues under which it applies, and can be retrieved rapidly when a similar situation arises. Pillemer reviewed evidence that people report using specific memories to guide decisions, that originating events are often remembered for personal rules, and that instruction anchored in a vivid example is better retained than abstract instruction. Self-defining memories are the paradigmatic case: their rehearsal and goal relevance make them the episodes most available as behavioural precedents. Experimental demonstrations that they causally alter subsequent choices remain few and most evidence is retrospective self-report, so the directive function is well-motivated but incompletely tested.

4.2 Identity and meaning-making

Thorne, McLean and Lawrence (2004) examined when late adolescents elaborate self-defining memories into meaning. Among 168 college students providing three memories each, only a minority of narratives (roughly a quarter) contained an explicit lesson or insight. Meaning-making was not uniformly distributed across content: memories of life-threatening and relationship events were markedly more likely to prompt reflection than memories of achievement or leisure. Memories with meaning were also more likely to have been told to others, and the authors emphasised that meaning is constructed partly through listener response: for certain events, the psychological work lies in interpretation rather than recall.

Singer et al. (2013) integrated these findings into a three-level account of narrative identity: self-defining memories are the basic units; scripts, following Tomkins, are recurrent affective-narrative sequences abstracted from families of similar memories; and the life story organises scripts into an overarching account with a characteristic tone. The review reported that integrative meaning is associated with greater well-being and maturity across multiple studies, whereas the mere presence of negative memories is not, and argued for the explicit use of self-defining memories in psychotherapy to identify maladaptive scripts and revise their meaning without attempting to alter the underlying events. The association between meaning and adjustment is established in direction but not in mechanism; it is compatible with meaning promoting adjustment, with adjustment enabling meaning, and with third variables such as verbal ability.

5. Clinical dissociations

Clinical comparisons indicate which components of the system can fail independently. A robust background finding is that depression and other emotional disorders are characterised by overgeneral autobiographical memory, a reduced tendency to retrieve specific episodes to cue words (Williams et al., 2007). Raffard et al. (2009) administered the task to patients with schizophrenia and matched controls. Patients’ memories were less specific and, more strikingly, contained integrative meaning far less often; they were also disproportionately concentrated around illness onset and hospitalisation. Raffard et al. (2010) further reported that patients drew fewer connections between memories and their current sense of self, and that meaning-making deficits were related to negative symptoms and executive functioning. The pattern suggests that the availability of emotionally charged episodes is relatively preserved while the interpretive operations that integrate them into identity are impaired.

Werner-Seidler and Moulds (2012) compared self-defining memories in formerly depressed, currently dysphoric and never-depressed individuals. Group differences were limited for specificity but emerged in phenomenology and meaning: individuals with a history of depression reported negative self-defining memories that were more central to identity and less often accompanied by redemptive or integrative meaning, while their positive memories were less vivid. The authors interpreted this as a vulnerability marker persisting beyond remission. Because groups were defined retrospectively and samples were modest, the causal status of these differences is uncertain, but the convergence with the schizophrenia findings is notable: in both populations the deficit lies in the coupling between memory and self rather than in the memory record.

6. Implications for emotion preservation

The literature reviewed above yields several constraints for a programme that seeks to preserve human emotional memory, formulated here as requirements and testable propositions.

First, the unit of preservation cannot be the emotion alone. In a self-defining memory the affective charge is defined relationally, by the goal the memory serves and the interpretation the person has placed on it (Conway et al., 2004; Singer et al., 2013). A record capturing valence and intensity without the episode’s specific detail, content and stated meaning would lose the properties that make the memory self-defining. We propose as a requirement that any preserved record contain, minimally, the four Blagov and Singer dimensions plus MEQ-type phenomenological ratings, since these variables have demonstrated reliability and predictive value.

Second, self-defining memories are not stable objects. Their emotions and motives are moderately stable over years (Sutin & Robins, 2005), but meaning changes with rehearsal, telling and listener response (Thorne et al., 2004), and coherence pressures reshape content. A preserved record is therefore a dated version. Preservation protocols should be longitudinal by design, re-eliciting the same memories at intervals, with divergence between versions treated as data about identity change rather than as error.

Third, integrative meaning is the component most vulnerable to psychopathology and most closely tied to well-being. A testable proposition follows: records that include the rememberer’s own meaning statement will be judged, by the rememberer and by close others, as more faithful representations of the event’s emotional significance than records containing only event and affect ratings.

Fourth, because retrieving a self-defining memory activates the corresponding working self and shifts affect (Çili & Stopa, 2015), elicitation for preservation is itself an intervention. Protocols should assess whether repeated elicitation of negative memories without a meaning-making component increases distress, and whether structured meaning prompts mitigate this risk.

Finally, meaning belongs to the rememberer. Since it is co-constructed with listeners, a preservation programme should document the interpersonal context of the account and refrain from imposing interpretive frames the person has not endorsed.

7. Conclusion

Self-defining memories constitute a measurable class of autobiographical memory whose distinctive property is functional coupling to the self’s goal structure. Two coding dimensions have proved reliable and informative: specificity indexes retrieval style and defensiveness, whereas integrative meaning indexes the interpretive work relating the past to present identity and is most consistently associated with adjustment. Clinical dissociations in schizophrenia and depression suggest that the record of emotionally charged episodes can survive while their integration into the self fails. The central open questions are causal: whether meaning-making produces adjustment or reflects it, and whether the directive influence of specific episodes can be demonstrated prospectively. Answering them will require longitudinal and experimental designs the field has rarely employed.

References

Blagov, P. S., & Singer, J. A. (2004). Four dimensions of self-defining memories (specificity, meaning, content, and affect) and their relationships to self-restraint, distress, and repressive defensiveness. Journal of Personality, 72(3), 481–511.

Çili, S., & Stopa, L. (2015). The retrieval of self-defining memories is associated with the activation of specific working selves. Memory, 23(2), 233–253.

Conway, M. A., Singer, J. A., & Tagini, A. (2004). The self and autobiographical memory: Correspondence and coherence. Social Cognition, 22(5), 491–529.

Lardi, C., D’Argembeau, A., Chanal, J., Ghisletta, P., & Van der Linden, M. (2010). Further characterisation of self-defining memories in young adults: A study of a Swiss sample. Memory, 18(3), 293–309.

Pillemer, D. B. (2003). Directive functions of autobiographical memory: The guiding power of the specific episode. Memory, 11(2), 193–202.

Raffard, S., D’Argembeau, A., Lardi, C., Bayard, S., Boulenger, J.-P., & Van der Linden, M. (2009). Exploring self-defining memories in schizophrenia. Memory, 17(1), 26–38.

Raffard, S., D’Argembeau, A., Lardi, C., Bayard, S., Boulenger, J.-P., & Van der Linden, M. (2010). Narrative identity in schizophrenia. Consciousness and Cognition, 19(1), 328–340.

Singer, J. A., Blagov, P., Berry, M., & Oost, K. M. (2013). Self-defining memories, scripts, and the life story: Narrative identity in personality and psychotherapy. Journal of Personality, 81(6), 569–582.

Singer, J. A., & Salovey, P. (1993). The remembered self: Emotion and memory in personality. Free Press.

Sutin, A. R., & Robins, R. W. (2005). Continuity and correlates of emotions and motives in self-defining memories. Journal of Personality, 73(3), 793–824.

Sutin, A. R., & Robins, R. W. (2007). Phenomenology of autobiographical memories: The Memory Experiences Questionnaire. Memory, 15(4), 390–411.

Thorne, A., McLean, K. C., & Lawrence, A. M. (2004). When remembering is not enough: Reflecting on self-defining memories in late adolescence. Journal of Personality, 72(3), 513–541.

Werner-Seidler, A., & Moulds, M. L. (2012). Characteristics of self-defining memory in depression vulnerability. Memory, 20(8), 935–948.

Williams, J. M. G., Barnhofer, T., Crane, C., Herman, D., Raes, F., Watkins, E., & Dalgleish, T. (2007). Autobiographical memory specificity and emotional disorder. Psychological Bulletin, 133(1), 122–148.