Effects of Aging on Functional Connectivity of the Amygdala for Subsequent Memory of Negative Pictures: A Network Analysis of Functional Magnetic Resonance Imaging Data

ABSTRACT— Aging is associated with preserved enhancement of emotional memory, as well as with age-related reductions in memory for negative stimuli, but the neural networks underlying such alterations are not clear. We used a subsequent-memory paradigm to identify brain activity predicting enhanced emotional memory in young and older adults. Activity in the amygdala predicted enhanced emotional memory, with subsequent-memory activity greater for negative stimuli than for neutral stimuli, across age groups, a finding consistent with an overall enhancement of emotional memory. However, older adults recruited greater activity in anterior regions and less activity in posterior regions in general for negative stimuli that were subsequently remembered. Functional connectivity of the amygdala with the rest of the brain was consistent with age-related reductions in memory for negative stimuli: Older adults showed decreased functional connectivity between the amygdala and the hippocampus, but increased functional connectivity between the amygdala and dorsolateral prefrontal cortices. These findings suggest that age-related differences in the enhancement of emotional memory might reflect decreased connectivity between the amygdala and typical subsequent-memory regions, as well as the engagement of regulatory processes that inhibit emotional responses.     

Anhedonia and the experience of emotion in individuals with schizophrenia

Research shows that individuals with schizophrenia report symptoms of anhedonia when assessed by interview or questionnaire. However, when presented with emotional stimuli, they report emotional experiences that are similar to those of control participants. The authors hypothesized that deficits in working memory and episodic memory contribute to such discrepancies. They administered measures of working and episodic memory, self-report anhedonia questionnaires, and several types of emotional stimuli to 49 individuals with schizophrenia and 47 control participants. All participants reported experiencing similar amounts of pleasant-unpleasant emotion (valence) in response to stimuli, but individuals with schizophrenia reported experiencing less arousal for negative stimuli. Individuals with schizophrenia also reported greater social and physical anhedonia on a traditional anhedonia questionnaire. Disturbances in working memory moderated the relationship between physical anhedonia and participants' emotional experience of positive stimuli.     

Arousal,valence, and memory for detail

A common finding in the emotion‐memory literature is that memory is enhanced for positively arousing stimuli and negatively arousing stimuli relative to neutral stimuli. We tested the notion that post‐stimulus elaboration is responsible for these effects. Post‐stimulus elaboration refers to the process of thinking about an event (after its offset) more frequently or more in‐depth. We tested the hypothesis by presenting participants with 36 slides depicting events that varied in arousal (low and high) and valence (positive and negative). The opportunity for elaboration was manipulated by requiring participants, during the inter‐slide interval, to complete addition problems, simply view the addition problems, or view a blank slide. Cued recall memory was tested for central and background details. Based on the post‐stimulus elaboration hypothesis it was expected that the greatest memory decline would occur for the central details of negatively and positively arousing slides when participants were required to complete addition problems (i.e., a distractor task × arousal × detail interaction). Contrary to the hypothesis, we found that filling the inter‐slide interval with a distractor task decreased memory for negative stimuli compared to positive stimuli. This effect was independent of arousal. We also found that arousal increased central detail memory for positive and negative stimuli and background detail memory for positive stimuli but not for negative stimuli. This interaction was explained on the basis of pre‐attentive encoding and cue utilisation. It was concluded that in order to understand the complex relationship between emotion and memory, future studies should include, as a minimum, the variables of valence, arousal, and detail.     

Read-out of emotional information from iconic memory: the longevity of threatening stimuli

Previous research has shown that emotional stimuli are more likely than neutral stimuli to be selected by attention, indicating that the processing of emotional information is prioritized. In this study, we examined whether the emotional significance of stimuli influences visual processing already at the level of transient storage of incoming information in iconic memory, before attentional selection takes place. We used a typical iconic memory task in which the delay of a poststimulus cue, indicating which of several visual stimuli has to be reported, was varied. Performance decreased rapidly with increasing cue delay, reflecting the fast decay of information stored in iconic memory. However, although neutral stimulus information and emotional stimulus information were initially equally likely to enter iconic memory, the subsequent decay of the initially stored information was slowed for threatening stimuli, a result indicating that fear-relevant information has prolonged availability for read-out from iconic memory. This finding provides the first evidence that emotional significance already facilitates stimulus processing at the stage of iconic memory.     

Trade-offs in visual attention and the enhancement of memory specificity for positive and negative emotional stimuli

Emotional influences on memory can lead to trade-offs in memory for gist or detail and trade-offs in memory for central or peripheral aspects of an event. Attentional narrowing has often been proposed as a theoretical explanation for this pattern of findings with negative emotion. These trade-offs have been less extensively investigated with positive emotion. In three experiments, we investigate memory for specific visual details of positive and negative stimuli, examine central–peripheral trade-offs in memory, and assess the hypothesis that attentional narrowing contributes to emotional enhancement of memory specificity. We found that memory for details was enhanced by negative and positive emotion. Central–peripheral trade-offs were found in memory for negative emotional stimuli but not in memory for positive emotional stimuli. These trade-offs with negative emotion were associated with attentional narrowing at the time of encoding, as measured by eye movements. There were no attentional effects at the time of encoding found with positive emotional stimuli. Evidence was found for the attentional narrowing hypothesis of memory specificity and central–peripheral trade-offs in memory for negative emotional events. Alternative explanations for the positive emotional enhancement of memory specificity are required.     

Cognitive and neural mechanisms of emotional memory

A highly adaptive aspect of human memory is the enhancement of explicit, consciously accessible memory for emotional stimuli. Recent findings from neuroimaging, neuropsychological, drug and neural stimulation studies indicate that emotional stimuli engage specific cognitive and neural mechanisms that enhance explicit memory. Emotional arousal influences memory via factors that act during memory encoding (attention and elaboration) and factors that modulate memory consolidation. Across studies, the amygdala has been consistently implicated as playing a key role in enhancing explicit memory for both pleasant and unpleasant emotional stimuli through modulation of encoding and consolidation processes.     

Arousal, valence, and memory for detail

A common finding in the emotion-memory literature is that memory is enhanced for positively arousing stimuli and negatively arousing stimuli relative to neutral stimuli. We tested the notion that post-stimulus elaboration is responsible for these effects. Post-stimulus elaboration refers to the process of thinking about an event (after its offset) more frequently or more in-depth. We tested the hypothesis by presenting participants with 36 slides depicting events that varied in arousal (low and high) and valence (positive and negative). The opportunity for elaboration was manipulated by requiring participants, during the interslide interval, to complete addition problems, simply view the addition problems, or view a blank slide. Cued recall memory was tested for central and background details. Based on the post-stimulus elaboration hypothesis it was expected that the greatest memory decline would occur for the central details of negatively and positively arousing slides when participants were required to complete addition problems (i.e., a distractor task x arousal x detail interaction). Contrary to the hypothesis, we found that filling the inter-slide interval with a distractor task decreased memory for negative stimuli compared to positive stimuli. This effect was independent of arousal. We also found that arousal increased central detail memory for positive and negative stimuli and background detail memory for positive stimuli but not for negative stimuli. This interaction was explained on the basis of pre-attentive encoding and cue utilisation. It was concluded that in order to understand the complex relationship between emotion and memory, future studies should include, as a minimum, the variables of valence, arousal, and detail.     

Short-term episodic memory for visual textures: a roving probe gathers some memory

Cognition is shaped by the way that past experiences are represented in memory. To examine the representation of recent visual experiences, we devised a novel procedure that measures episodic recognition memory for synthetic textures. On each trial, two brief study stimuli were followed by a probe, which either replicated one of the study stimuli or differed in spatial frequency from both. The probe's spatial frequency roved from trial to trial, testing recognition with a range of differences between probe and study items. Repeated testing of recognition generated mnemometric functions, snapshots of memory strength's distribution. The distributional characteristics of the mnemometric functions rule out several hypotheses about memory representation, including the hypothesis that representations are prototypes constructed from previously seen stimuli; instead, stimuli are represented in memory as noisy exemplars.     

Effect of negative emotional content on working memory and long-term memory

In long-term memory, negative information is better remembered than neutral information. Differences in processes important to working memory may contribute to this emotional memory enhancement. To examine the effect that the emotional content of stimuli has on working memory performance, the authors asked participants to perform working memory tasks with negative and neutral stimuli. Task accuracy was unaffected by the emotional content of the stimuli. Reaction times also did not differ for negative relative to neutral words, but on an n-back task using faces, participants were slower to respond to fearful faces than to neutral faces. These results suggest that although emotional content does not have a robust effect on working memory, in some instances emotional salience can impede working memory performance.     

Verbal recoding of visual stimuli impairs mentalimagetransformations

Two experiments were carried out to test the hypothesis that verbal recoding of visual stimuli in short-term memory influences long-term memory encoding and impairs subsequent mental image operations. Easy and difficult-to-name stimuli were used. When rotated 90 degrees counterclockwise, each stimulus revealed a new pattern consisting of two capital letters joined together. In both experiments, subjects first learned a short series of stimuli and were then asked to rotate mental images of the stimuli in order to detect the hidden letters. In Experiment 1, articulatory suppression was used to prevent subjects from subvocal rehearsal when learning the stimuli, whereas in Experiment 2, verbal labels were presented with each stimulus during learning to encourage a reliance on the verbal code. As predicted, performance in the imagery task was significantly improved by suppression when the stimuli were easy to name (Experiment 1) but was severely disrupted by labeling when the stimuli were difficult to name (Experiment 2). We concluded that verbal recoding of stimuli in short-term memory during learning disrupts the ability to generate veridical mental images from long-term memory.     

Non-monotonic relationships between emotional arousal and memory for color and location

Recent research points to the decreased diagnostic value of subjective retrieval experience for memory accuracy for emotional stimuli. While for neutral stimuli rich recollective experiences are associated with better context memory than merely familiar memories this association appears questionable for emotional stimuli. The present research tested the implicit assumption that the effect of emotional arousal on memory is monotonic, that is, steadily increasing (or decreasing) with increasing arousal. In two experiments emotional arousal was manipulated in three steps using emotional pictures and subjective retrieval experience as well as context memory were assessed. The results show an inverted U-shape relationship between arousal and recognition memory but for context memory and retrieval experience the relationship was more complex. For frame colour, context memory decreased linearly while for spatial location it followed the inverted U-shape function. The complex, non-monotonic relationships between arousal and memory are discussed as possible explanations for earlier divergent findings.     

Reduced memory for the spatial and temporal context of unpleasant words

Emotional stimuli are consistently better remembered than neutral stimuli. However, the reported effects of emotional stimuli on source memory are less consistent. In four experiments, we examined spatial and temporal source memory and free recall for emotional words previously studied in an fMRI experiment. In the fMRI experiment, the unpleasant but not the pleasant words were shown to activate the amygdala. In the experiments reported here, spatial and temporal source memory were reduced for the unpleasant words compared to pleasant and neutral words. Reduced spatial memory for unpleasant words was observed across studies that varied the specific stimuli, task conditions, and incidental versus intentional encoding instructions. Free recall was enhanced for the pleasant and unpleasant words compared to neutral words. The results demonstrate reduced source memory for unpleasant words even when item recall is enhanced. Our findings are consistent with prior reports of amygdala-mediated amnestic effects evoked by unpleasant stimuli.     

Impaired emotional declarative memory following unilateral amygdala damage

Case studies of patients with bilateral amygdala damage and functional imaging studies of normal individuals have demonstrated that the amygdala plays a critical role in encoding emotionally arousing stimuli into long-term declarative memory. However, several issues remain poorly understood: the separate roles of left and right amygdala, the time course over which the amygdala participates in memory consolidation, and the type of knowledge structures it helps consolidate. We investigated these questions in eight subjects with unilateral amygdala damage, using several different measures. For comparison, our main task used stimuli identical to those used previously to investigate emotional declarative memory in patients with bilateral amygdala damage. Contrasts with both brain-damaged and normal control groups showed that subjects with left amygdala damage were impaired in their memory for emotional stimuli, despite entirely normal memory for neutral stimuli (because of a number of caveats, the findings from subjects with right amygdala damage were less clear). Follow-up experiments suggested that the normal facilitation of memory for emotional stimuli may develop over an extended time course (>30 min), consistent with prior findings, and that the specific impairment we report may depend in part on the lexical nature of the task used (written questionnaire). We stress the complex and temporally extended nature of memory consolidation and suggest that the amygdala may influence specific components of this process.     

Auditory and visual memory in musicians and nonmusicians

Numerous studies have shown that musicians outperform nonmusicians on a variety of tasks. Here we provide the first evidence that musicians have superior auditory recognition memory for both musical and nonmusical stimuli, compared to nonmusicians. However, this advantage did not generalize to the visual domain. Previously, we showed that auditory recognition memory is inferior to visual recognition memory. Would this be true even for trained musicians? We compared auditory and visual memory in musicians and nonmusicians using familiar music, spoken English, and visual objects. For both groups, memory for the auditory stimuli was inferior to memory for the visual objects. Thus, although considerable musical training is associated with better musical and nonmusical auditory memory, it does not increase the ability to remember sounds to the levels found with visual stimuli. This suggests a fundamental capacity difference between auditory and visual recognition memory, with a persistent advantage for the visual domain.     

Memory bias for threatening information related to anxiety: an updated meta-analytic review

The evidence for an anxiety-related memory bias is contradictory. We compiled 171 articles published until October 2016 including a group with clinical or subclinical anxiety and a control group in tasks involving implicit or explicit memory using threatening stimuli. There was an anxiety-related memory bias in free recall tasks, but it was not observed in another memory task. The between-groups differences showed that the anxious group recalled more threatening stimuli than the control group (d?=?0.321). When we compared the group differences (anxious vs. control participants) in the within-groups effect (threatening vs. neutral stimuli), a moderate effect size emerged (d_bw?=?0.714). This anxiety-related memory bias was observed with shallow processing, that is consistent with attentional biases related to anxiety. There was also evidence that high-anxious persons recall fewer positive stimuli. Future research is needed to investigate whether this result is a memory or encoding bias and explore other moderator variables.     

Tactile order memory: evidence for sequence learning phenomena found with other stimulus types

We examine serial order memory for sequences of tactile stimuli and investigate whether established characteristics of order memory, namely serial position effects, error distributions, and Hebb repetition learning, are observed with tactile memory. Visually obscured participants received six tactile stimulations: one to each of six fingers. At test, participants lifted the six fingers in the order of stimulation. For every third trial participants received the same order of stimulation (i.e. the Hebb sequence). Serial recall accuracy produced the canonical bowed serial position function found for immediate serial recall. In addition, recall for the Hebb sequence improved relative to the filler sequences, providing the first demonstration of the Hebb repetition effect with tactile stimuli. Analysis of errors revealed close similarities to that reported with verbal and visual stimuli. This experiment further generalises established features of order memory to tactile memory, supporting the utilisation of an analogous order memory mechanism across stimuli.     

Emotion and memory: A recognition advantage for positive and negative words independent of arousal

Much evidence indicates that emotion enhances memory, but the precise effects of the two primary factors of arousal and valence remain at issue. Moreover, the current knowledge of emotional memory enhancement is based mostly on small samples of extremely emotive stimuli presented in unnaturally high proportions without adequate affective, lexical, and semantic controls. To investigate how emotion affects memory under conditions of natural variation, we tested whether arousal and valence predicted recognition memory for over 2500 words that were not sampled for their emotionality, and we controlled a large variety of lexical and semantic factors. Both negative and positive stimuli were remembered better than neutral stimuli, whether arousing or calming. Arousal failed to predict recognition memory, either independently or interactively with valence. Results support models that posit a facilitative role of valence in memory. This study also highlights the importance of stimulus controls and experimental designs in research on emotional memory.     

Event-related brain potentials reflect traces of echoic memory in humans

In sequences of identical auditory stimuli, infrequent deviant stimuli elicit an event-related brain potential component called mismatch negativity (MMN). MMN is presumed to reflect the existence of a memory trace of the frequent stimulus at the moment of presentation of the infrequent stimulus. This hypothesis was tested by applying the recognition-masking paradigm of cognitive psychology. In this paradigm, a masking sound presented shortly before or after a test stimulus diminishes the recognition memory of this stimulus, the more so the shorter the interval between the test and masking stimuli. This interval was varied in the present study. It was found that the MMN amplitude strongly correlated with the subject’s ability to discriminate between frequent and infrequent stimuli. This result strongly suggests that MMN providesa measure for a trace of sensory memory, and further, that with MMN, this memory can be studied without performance-related distortions.     

Memory retrieval by 18--30-month-olds: age-related changes in representational flexibility

Deferred imitation was used to trace changes in memory retrieval by 18-30-month-olds. In all experiments, an adult demonstrated 2 sets of actions using 2 different sets of stimuli. In Experiments 1A and 1B, independent groups of infants were tested immediately or after a 24-hr delay. Each infant was tested with 1 set of stimuli from the original demonstration and 1 set of stimuli that was different. Recall of the target actions when tested with different stimuli increased as a function of age, particularly after a delay. In Experiment 2, infants were provided with a unique verbal label for the stimuli during the demonstration and the test. The verbal label facilitated performance by 24-month-olds tested with different stimuli but had no effect on performance by 18-month-olds. One hallmark of memory development appears to be an age-related increase in the range of effective retrieval cues for a particular memory.     

Target preexposure eliminates the effect of distraction on event-based prospective memory

Prospective memory is critical to everyday functioning and can be vulnerable to distraction. We conducted an experiment to explore whether we could buffer prospective memory against distraction. For half the participants, we preexposed stimuli that were later designated as prospective memory targets. Then, all participants performed an ongoing task (in which the prospective memory task was embedded) under standard and high attentional demand (i.e., under full and divided attention). Target preexposure improved prospective memory and eliminated the significant divided attention effect. Thus, target preexposure seems to buffer prospective memory against the disruptive effect of dividing attention. Moreover, target preexposure seemed to help participants to respond with the correct intended action. This result implies that preexposure to the target stimuli facilitated the encoding of an association between the target stimuli and the intended action, perhaps promoting relatively more reflexive retrieval and thereby buffering prospective memory against distraction.