Unraveling the linguistic nature of specific autobiographical memories using a computerized classification algorithm

In the present study, we explored the linguistic nature of specific memories generated with the Autobiographical Memory Test (AMT) by developing a computerized classifier that distinguishes between specific and nonspecific memories. The AMT is regarded as one of the most important assessment tools to study memory dysfunctions (e.g., difficulty recalling the specific details of memories) in psychopathology. In Study 1, we utilized the Japanese corpus data of 12,400 cue-recalled memories tagged with observer-rated specificity. We extracted linguistic features of particular relevance to memory specificity, such as past tense, negation, and adverbial words and phrases pertaining to time and location. On the basis of these features, a support vector machine (SVM) was trained to classify the memories into specific and nonspecific categories, which achieved an area under the curve (AUC) of .92 in a performance test. In Study 2, the trained SVM was tested in terms of its robustness in classifying novel memories (n?= 8,478) that were retrieved in response to cue words that were different from those used in Study 1. The SVM showed an AUC of .89 in classifying the new memories. In Study 3, we extended the binary SVM to a five-class classification of the AMT, which achieved 64%–65% classification accuracy, against the chance level (20%) in the performance tests. Our data suggest that memory specificity can be identified with a relatively small number of words, capturing the universal linguistic features of memory specificity across memories in diverse contents.     

Point of view in personal memories

A structural aspect of personal memories was examined in four studies. In some memories, one has the perspective of an observer, seeing oneself “from the outside.” In other memories, one sees the scene from one's own perspective; the field of view in such memories corresponds to that of the original situation. The existence of “observer” and “field” memories was confirmed in Study 1, using a recall questionnaire. In Study 2, the similarity structure of a specified set of eight to-be-recalled situations was established: the significant dimensions were “emotionality” and “self-awareness.” Study 3 related these dimensions to the observer-field distinction; situations involving a high degree of emotion and selfawareness were most likely to be recalled with an observer perspective. Recall set was varied in Study 4: a focus on feelings (as opposed to objective circumstances) produced relatively more field memories. Studies 3 and 4 also showed that events reported as more recent tend to be recalled in the field mode. Thus a qualitative characteristic of personal memories—the perspective from which they are experienced—is apparently related to characteristics of the original event, to the individual's purpose in recalling that event, and to the reported recall interval.     

Resilience and Positive Emotions: Examining the Role of Emotional Memories

Abstract Resilience has been frequently associated with positive emotions, especially when experienced during taxing events. However, the psychological processes that might allow resilient individuals to self-generate those positive emotions have been mostly overlooked. In line with recent advances in memory research, we propose that emotional memories play an important role in the self-generation of positive emotions. The present research examined this hypothesis in two studies. Study 1 provided initial data on the validity and reliability of a measure of emotional memories networks (EMN) and showed that it had a predictive value for broad emotion regulation constructs and outcomes. In addition, Study 1 showed that positive EMN mediated the relationship between psychological resilience and the experience of positive emotions in a context of sadness, even after controlling for pre-experimental positive mood. Study 2 replicated results of Study 1 in a context of anxiety and after controlling for positive affectivity trait.     

Childhood remembered: Reports of both unique and repeated events

To explore the significance of repeated memories for individuals' personal histories, we compared the characteristics of young adults' unique and repeated memories of childhood experiences. Memory type (unique vs. repeated) was a within-participant variable. In Experiment 1, college-age participants generated as many early memories as possible in 4 minutes; in Experiment 2, another sample provided complete reports of five early memories in each condition. In both experiments, participants rated the vividness, biographical importance and personal meaning of each memory and labelled the accompanying emotion. Unique memories were more vivid than repeated memories as well as more likely to include negative emotion, regardless of the method of reporting. Most importantly, college students rated their memories for unique and repeated events as equivalently infused with personal meaning. Analysis of the content of the memories reported in Experiment 2 established that unique and repeated memories did not differ in word count or percentages of perceptual terms or words indicating positive affect, although unique memories contained a greater percentage of negative affect. Additional analyses of content provided evidence for differences in the functions served by unique and repeated memories. The results have implications for the study of autobiographical memory and for identifying over-general memories.     

Closure of autobiographical memories: The effects of written recounting from first- or third-person visual perspective

Autobiographical memories are recalled with varying degrees of psychological closure. Closure is a subjective assessment of how far a remembered experience feels resolved, and it has been suggested that one predictor of closure is the amount of emotional detail in the memory. Study 1 examined which aspect of emotional detail is important for closure, and showed that open and closed negative memories were distinguished by ratings of emotion evoked during recall, not by remembered emotion from the time of the event. The recall of open memories was accompanied by more intense, more negative, and less positive emotion than the recall of closed memories. Biased retelling of memories has been shown to influence closure and on the basis of evidence that third-person recall serves a distancing function, Study 2 examined whether instructions to repeatedly recount an open memory from a third-person perspective would increase closure compared with a single or repeated recounting from a first-person perspective. While repeated third-person recounting had the greatest influence on closure, there were also increases in the first-person recounting groups. The results suggest that closure can be increased by reporting memories in written narrative form, particularly if repeatedly expressed from the third-person perspective.     

Autobiographical memory specificity in response to emotion pictorial cues among non‐clinical participants

The objective of the current study was to investigate whether emotion pictorial cues increase memory specificity among non‐clinical participants. Undergraduate university students were presented with emotion word and pictorial cues on a prompted and non‐prompted version of the Autobiographical Memory Test (AMT). In comparison to pictorial cues, participants retrieved significantly fewer specific autobiographical memories in response to word cues on the prompted AMT; however, there was no significant difference on the non‐prompted AMT. Participants also retrieved significantly fewer specific memories in response to both word and pictorial cues on the non‐prompted AMT compared with the prompted AMT. These results provide support for the hypothesis that among non‐clinical participants, visual cues increase memory specificity over and above emotion. Further research is needed to investigate ways in which memory specificity can be increased and the use of imagery may be a promising avenue.     

Emotional valence and the functions

Differences between positive and negative autobiographical memories are often explained with reference to hypothesized evolutionary functions. Generally, it has been proposed that autobiographical memory serves directive, self-, and social functions. However, the relationship between emotional valence and the three functions has never been studied. In Study 1, participants generated memories that mapped onto each of the three functions. Directive memories were dominated by negative emotion, whereas self- and social memories were dominated by positive emotion. In Study 2, participants generated their most positive and most negative memories, as well as their most frequent involuntary and most vivid flashbulb memories, and the three functions were measured through rating-scale questions. The directive function had the lowest ratings across all memory classes, but, consistent with the results of Study 1, positive memories were rated higher on the self- and social functions, whereas negative memories were rated higher on the directive function.     

Closure of autobiographical memories: The effects of written recounting from first- or third-person visual perspective

Autobiographical memories are recalled with varying degrees of psychological closure. Closure is a subjective assessment of how far a remembered experience feels resolved, and it has been suggested that one predictor of closure is the amount of emotional detail in the memory. Study 1 examined which aspect of emotional detail is important for closure, and showed that open and closed negative memories were distinguished by ratings of emotion evoked during recall, not by remembered emotion from the time of the event. The recall of open memories was accompanied by more intense, more negative, and less positive emotion than the recall of closed memories. Biased retelling of memories has been shown to influence closure and on the basis of evidence that third-person recall serves a distancing function, Study 2 examined whether instructions to repeatedly recount an open memory from a third-person perspective would increase closure compared with a single or repeated recounting from a first-person perspective. While repeated third-person recounting had the greatest influence on closure, there were also increases in the first-person recounting groups. The results suggest that closure can be increased by reporting memories in written narrative form, particularly if repeatedly expressed from the third-person perspective.     

Repression and the inaccessibility of affective memories

The fundamental assumption that repression involves an inaccessibility to affective memories has not been directly addressed in empirical research. In the present study we examined three groups of subjects (repressors, low anxious, and high anxious) under six conditions of recall (general, happy, sad, anger, fear, and wonder). Subjects were asked to recall personal experiences from childhood and to rate their current mood and the affective intensity of the memories. The results indicated that repressors recalled significantly fewer negative memories than did low-anxious and high-anxious subjects and, furthermore, that they were substantially older at the time of the earliest negative memory recalled. Compared with low-anxious subjects, repressors also recalled fewer positive affective memories as well. This pattern of findings is consistent with the hypothesis that repression involves an inaccessibility to negative emotional memories and indicates further that repression is associated in some way with the suppression or inhibition of emotional experiences in general. The concept of repression as a process involving limited access to negative affective memories appears to be valid.     

The Effect of Imageability and Predicability of Cues in Autobiographical Memory

Four experiments investigated the role of imagery in the recollection of autobiographical memories. The first two experiments examined the effects of word imageability and word frequency on the retrieval of personal memories in a cued autobiographical memory task. They showed that the imageability of cues (but not frequency) mediates specificity in the recall of personal memories. Experiment 2 explored how different imagery modalities (visual, olfactory, tactile, auditory, and motor) influence autobiographical retrieval. Consistent with research on imagery modalities in verbal learning paradigms, visual imageability emerged as the most significant predictor of specificity. Experiments 3 and 4 examined how far a knowledge-based account of imagery effects might account for these effects, using predicability as a measure of semantic richness of a cue. Results found that visual imageability of cues accounted for more variance in specificity of recall than did predicability. The results are explained in terms of the way images represent the most efficient form of summarizing the information that can be used at each stage of the recollection process: setting the retrieval plan, strategic search, and evaluation of candidate episodes.     

Influence of general and specific autobiographical recall on subsequent behavior: The case of cognitive performance

Two studies tested the impact of autobiographical recall of general versus specific academic success or failure on actual task performance. As expected, it was found that general memories of failure and specific memories of success resulted in worse performance than general memories of success and specific memories of failure. In Study 1, this performance pattern was obtained on a standard math test. In Study 2, it was replicated on a test of intellectual ability, and a mediation by fear of failure was documented. The present findings offer direct evidence that autobiographical memories of success and failure impact actual performance and also reveal the role of memory specificity in this influence.     

Social schemata of peripheral changes in emotion

Psychophysiological research failed to establish consistent physiological patterns differentiating emotion. Recent data showed that people verbally report experiencing peripheral changes that differ among emotions. The present studies tested the hypothesis that these reports originate in social schemata. Study 1 showed that Ss' reports of peripheral changes experienced during actual emotion do not differ from those defined in social schemata. Studies 2 and 3 showed that these schemata are similar across cultures. Overall, these data suggest that (a) people can directly access schemata about peripheral changes in emotion, (b) people are likely to do so when they believe to be reporting actual memories of such changes, and (c) the specific patterns revealed by past research may reflect prototypical knowledge of emotion. Finally, the data highlight the various peripheral patterns as they exist in schematic knowledge of emotion.     

Continuity and correlates of emotions and motives in self-defining memories

Two studies examined emotions and motives in self-defining memories. In Study 1, participants recalled five self-defining memories (four recent and one earliest childhood), rated their emotions and motives during each memory, and completed a set of personality measures. A subset of participants provided a second set of memories, as well as emotion and motive ratings, approximately 2 weeks after the initial session. Results suggest that emotions and motives are moderately stable across memories and over time and show theoretically meaningful relations with self-esteem, narcissism, and affective dispositions. Study 2 extended the findings of Study 1 to a longitudinal context. Emotions and motives coded from self-defining memories were associated with changes in personality, well-being, and academic performance over a 4-year period.     

Emotion regulation choice: the role of environmental affordances

Which emotion regulation strategy one uses in a given context can have profound affective, cognitive, and social consequences. It is therefore important to understand the determinants of emotion regulation choice. Many prior studies have examined person-specific, internal determinants of emotion regulation choice. Recently, it has become clear that external variables that are properties of the stimulus can also influence emotion regulation choice. In the present research, we consider whether reappraisal affordances, defined as the opportunities for re-interpretation of a stimulus that are inherent in that stimulus, can shape individuals’ emotion regulation choices. We show that reappraisal affordances have stability across people and across time (Study 1), and are confounded with emotional intensity for a standardised set of picture stimuli (Study 2). Since emotional intensity has been shown to drive emotion regulation choice, we construct a context in which emotional intensity is separable from reappraisal affordances (Study 3) and use this context to show that reappraisal affordances powerfully influence emotion regulation choice even when emotional intensity and discrete emotions are taken into account (Study 4).     

Stress and mood: the moderating role of activity inhibition

The present research tested the hypothesis that activity inhibition (as measured by the picture story exercise) moderates the relationship between stress and mood. Based on prior research that shows that (a) individuals high in activity inhibition restrain emotional and motivational impulses and (b) inhibiting negative emotion may lead to further emotional impairments, we assumed that individuals high in activity inhibition show evidence of increased negative mood when they are confronted with stressful events. Study 1 found evidence of increased negative mood-relevant thought among individuals high in activity inhibition who were confronted with the threat of social rejection. Study 2 found high negative mood among individuals high in activity inhibition who experienced many daily hassles. Also, participants high in activity inhibition reported a disproportionally higher amount of daily hassles in the middle (but not at the beginning) of a university semester. We suggest that this pattern of results can be accounted for by Wegner's (1994) theory of ironic effects of the suppression of thought and emotion.     

The relation between rumination and temporal features of emotion intensity

Intensity profiles of emotional experience over time have been found to differ primarily in explosiveness (i.e. whether the profile has a steep vs. a gentle start) and accumulation (i.e. whether intensity increases over time vs. goes back to baseline). However, the determinants of these temporal features remain poorly understood. In two studies, we examined whether emotion regulation strategies are predictive of the degree of explosiveness and accumulation of negative emotional episodes. Participants were asked to draw profiles reflecting changes in the intensity of emotions elicited either by negative social feedback in the lab (Study 1) or by negative events in daily life (Study 2). In addition, trait (Study 1 & 2), and state (Study 2) usage of a set of emotion regulation strategies was assessed. Multilevel analyses revealed that trait rumination (especially the brooding component) was positively associated with emotion accumulation (Study 1 & 2). State rumination was also positively associated with emotion accumulation and, to a lesser extent, with emotion explosiveness (Study 2). These results provide support for emotion regulation theories, which hypothesise that rumination is a central mechanism underlying the maintenance of negative emotions.     

The effects of thought suppression on autobiographical memory recall

While it is well documented that autobiographical memory (ABM) recall is affected in Posttraumatic Stress Disorders (PTSD), less is known about the cognitive mechanisms that underlie this pattern. This paper presents two research studies which investigated the role of thought suppression in the recall of ABMs. Study 1 assessed the role of thought suppression as a correlate of ABM retrieval in an undergraduate student sample (n = 50). The results showed that higher levels of trait thought suppression were significantly correlated with faster recall of negative episodic memories as well as reduced recall of personal semantic memories. Thought suppression remained as a significant predictor of ABM recall, even when the participants' levels of depression and post-traumatic stress reactions were considered. Study 2 investigated the causal effects of thought suppression on ABM recall. 64 undergraduate students were shown a negative video clip and were asked either to suppress any thought of the video or simply to monitor their thoughts immediately thereafter. Results showed that suppression directly led to significantly enhanced negative episodic ABM recall, as well as a significantly reduced ability to recall personal semantic memories.