Anya Bergstrom and Karina Agadzhanyan
Do you ever wonder why some information is remembered better than others? Why do we remember what we remember and forget what we forget? On a daily bases, we are bombarded with hundreds of images, faces, news headlines but yet remember only a handful of the encountered information. Research has shown that human memory operates in a selective manner differentiating between significant experiences and information of less importance (Castel et al., 2011). One factor that may also be critical in this process is the emotionality of information. Generally, emotional information is remembered better than neutral or non-emotional information, a phenomenon known as the emotion salience effect (Dolan, 2002; Fung & Carstensen, 2008; see Murphy & Isaacowitz, 2008 for a review). Given these findings, researchers have recently begun to explore the underlying reasons for why people tend to remember emotional information better than neutral information and how metamemory, assessments of our memory performance, play a role in this effect.
To illustrate what it means to recall emotional information better than neutral information, many studies have been conducted to identify the differences in memory performance between positively and negatively stimuli in comparison to neutrally valenced stimuli. By valence, researchers refer to the subjective emotional evaluation of a given item, with lower numbers indicating negativity, higher numbers indicating positivity, and numbers in-between indicating neutrality. Studies have assessed the memorability of emotional information using a variety of stimuli, such as words, pictures, videos, and narrated slideshows (Kensinger et al., 2002). For example, when given a list of words of mixed valences, subjects are more likely to recall the word “murder” than “flower,” due to the former word eliciting a negative valence, therefore, being more salient than the latter. In an experiment by Donald MacKay and colleagues (2004), participants were presented with a mix of taboo and neutral words and were later asked to recall as many words as they could. Results showed that participants remembered a greater number of the taboo words, which are known to elicit an emotional response. Thus, emotionally valenced stimuli are known to be more salient, thus more memorable, than neutral words
In order to understand why emotional information has a higher memorability, we must identify the factors that influence and constitute metacognition, the awareness of one’s memory processes, which consists of both monitoring and control (Nelson & Narens, 1994; see also Dunlosky et al., 2016). Previous findings have shown that emotional information has an impact on our predictions of future memory performance, a phenomenon known as metacognitive monitoring (Tauber & Dunlosky, 2012). Metacognitive monitoring refers to self-assessments of one’s learning and memory, and in order to measure it, researchers have widely used judgments of learning (JOLs). To illustrate, after studying a list of words, participants are asked to rate how likely they are to remember the just studied words on a later test by making judgments of learning on a scale from 0 to 100. On the other hand, metacognitive control refers to the self-regulation of learning based on monitoring. To illustrate, if given an opportunity to self-regulate the study time of each item within a given study list, one may spend more time studying words they judge to be more challenging or harder to remember (Nelson, 1996). Hence, according to the discrepancy reduction theory, when learning new information, learners seek to reduce the discrepancy between their current and optimal state of learning by allocating more study time to items they perceive as more challenging (Thiede & Dunlosky, 1999; Ariel & Dunlosky, 2013). Simply put, metacognitive monitoring plays a crucial role in control processes, leading to strategic learning.
Judgments of Learning are a valuable tool in the study of metacognition and memory, and to better understand how different cues influence JOLs, we must consider Koriat’s cue utilization framework (1997). According to the framework, three types of cues influence JOLs: intrinsic cues, which directly affect the study item’s ease of learning (e.g., whether the word is concrete or abstract, font size, emotional valence, etc.), extrinsic cues, such as conditions of learning applied by the learner (e.g., such as repetition during word presentation) and mnemonic cues, the experience of ease during item recall (e.g., feeling of fluency during encoding). Various experiments have been conducted to examine the effect of different cues on metacognition. For example, studies have looked at the effects of perception processing fluency on metacognition. Perception processing fluency refers to the ease with which an item is processed and encoded. Some perceptual characteristics that have been isolated and studied include font size (Rhodes & Castel, 2008), audio volume (Rhodes & Castel, 2009), and repeated vocal responses (Castel, Rhodes, & Friedman, 2013). In addition to the perceptual characteristics of words, the value of words along with perceptual processing fluency has also been shown to influence our metacognition and memory (Murphy et al., 2022). Specifically, while JOLs are sensitive to both perceptual processing fluency and value, the magnitude of the effect of value is greater than that of font size. Conclusively, we find that depending on the encoding conditions, cues can become less impactful when other more salient cues are available for encoding.
To explore the interplay between metacognition and memory, one must note that there exists a discrepancy between our predicted and actual memory, referred to as metacognitive illusions (Rhodes & Castel, 2008; Rhodes & Castel, 2009). To illustrate, when presented with words varying in font size, participants often predict that their memory for large font size words would be better than their memory for small font size words, when in fact, font size has been shown to elicit little or no effect on participants’ actual recall performance. When it comes to emotional stimuli, the following question arises: how does the valence of the to-be-learned information influence JOLs and actual memory performance? Previous work has shown that people give significantly higher JOLs to negative words (e.g., coffin) than to neutral words (e.g., chair), which is known as the emotional salience effect on JOLs (Zimmerman & Kelley, 2010; Tauber et al., 2017). This effect has been replicated with various types of emotional stimuli, including single words, word pairs, and emotional pictures (Witherby et. al., 2021). Generally, research suggests several potential mechanisms that could explain the influence of emotion on JOLs. First, participants might inherently believe that emotional information will be remembered better than neutral information, and therefore, purposely give higher JOLs to emotional words. This approach can be explained by the theory-based account, which requires a conscious cognitive strategy where one intentionally categorizes an item as an emotionally salient one based on prior beliefs (Koriat, 1997; Hourihan et. al 2016). Beliefs include people’s explicit understanding of factors that influence metacognitive processing. For example, students often have the preexisting belief that highlighted words are more easily remembered, so they may award higher JOLs to highlighted words than non-highlighted words. The second possible mechanism, also known as experience based, incorporates unconscious factors, such as perception processing fluency or the feeling-of-fluency effect (Koriat, 1997; Benjamin et al., 1998). According to this mechanism, emotional words are known to elicit physiological responses similar to metamnemonic processes (e.g. a subjective judgment about a studied item), which are also known to be associated with greater activation in the amygdala (a region in the brain responsible for emotional processing) (Kensinger & Corkin, 2004). Therefore, both prior beliefs and experiences combined explain why emotional stimuli have a great effect on our predictions of memory performance.
In an attempt to investigate which of the two mechanisms – conscious or unconscious – explains the effect of emotion on metacognition, researchers have conducted a series of experiments, which resulted in support for the conscious (theory-based) account rather than the unconscious account (Hourihan et. al 2016). To illustrate, when participants were given a list that included both negatively and neutrally valenced words, it was found that subjects gave higher judgments of learning towards words that had a higher emotional valence as opposed to neutral words, and the subjects’ predictions were in line with their actual ability to recall the negatively-valenced words. However, the effect of emotional valence quickly diminishes when the words within a list are all of the same valence: either all positively valenced, all negatively valenced or all neutral. When the saliency of the emotional words was removed, there was no longer an increase in JOLs or actual recall for emotionally valenced words. The contrasts between the results of the above two experiments suggest that it is not solely emotional valence that is influencing the participants’ rating of JOLs, but rather the relative distinctiveness of emotional words, specifically negatively emotional words, that led participants to believe that the words will be remembered better. In other words, it is safe to say that people believe emotional words to be more memorable based on their saliency relative to nonemotional words and make a cognitive choice by giving higher JOLs to emotional words.
In conclusion, it is clear that emotional information is generally remembered differently than neutral information, which is essential to our understanding of memory processes. Despite the fact that people tend to give higher JOLs to emotional items and remember them better than neutral items, these judgments may be made based on the distinctiveness of emotional information in contrast to neutral information, leading to conscious strategic decisions about memory performance rather than relying on unconscious physiological responses (Hourihan et. al 2016). These findings have important implications for how information can best be recalled through manipulating the emotional valence of words and other intrinsic factors, such as font size. Future work could focus on the interplay between font size, a cue that is known to have little to no effect on memory but has an influence on JOLs and emotional valence, a cue that is known to have an effect on both metacognition and actual memory (Agadzhanyan & Castel, in prep). Understanding the underlying mechanism of the effect of emotion on memory can be important when it comes to eyewitness memory of emotional events since predictions of future memory for emotional information may not be as accurate as we might think.
Agadzhanyan, K. & Castel, A. D. (in prep). The effect of font size and emotional valence on metacognition and recall.
Ariel, R., & Dunlosky, J. (2013). When do learners shift from habitual to agenda-based processes when selecting items for study? Memory & Cognition, 41, 416-428.
Benjamin, A. S., Bjork, R. A., & Schwartz, B. L. (1998). The mismeasure of memory: When retrieval fluency is misleading as a metamnemonic index. Journal of Experimental Psychology: General, 127, 55–68.
Castel, A. D., Humphreys, K. L., Lee, S. S., Galván, A., Balota, D. A., & McCabe, D. P. (2011). The development of memory efficiency and value-directed remembering across the life span: a cross-sectional study of memory and selectivity. Developmental Psychology, 47, 1553.
Castel, A. D., Rhodes, M. G., & Friedman, M. C. (2013). Predicting memory benefits in the production effect: the use and misuse of self-generated distinctive cues when making judgments of learning. Memory & Cognition, 41, 28–35.
Dolan, R. J. (2002). Emotion, cognition, and behavior. Science, 298, 1191–1194.
Dunlosky, J., Mueller, M. L., & Thiede, K. W. (2016). Methodology for investigating human metamemory: Problems and pitfalls. In J. Dunlosky & S. K. Tauber (Eds.), The Oxford handbook of metamemory (pp. 23–37). Oxford University Press.
Fung, H. H., & Carstensen, L. L. (2003). Sending memorable messages to the old: Age differences in preferences and memory for advertisements. Journal of Personality and Social Psychology, 85, 163–178.
Hourihan, K. L., Fraundorf, S. H. & Benjamin, A. S (2016). The influences of valence and arousal on judgments of learning and on recall. Memory & Cognition, 45, 121–136.
Kensinger, E. A., Brierley, B., Medford, N., Growdon, J. H., & Corkin, S. (2002). Effects of normal aging and Alzheimer’s disease on emotional memory. Emotion, 2, 118.
Kensinger, E. A., & Corkin, S. (2004). Two routes to emotional memory: distinct neural processes for valence and arousal. Proceedings of the National Academy of Sciences of the United States of America, 101, 3310–3315.
Koriat, A. (1997). Monitoring one’s own knowledge during study: A cue-utilization approach to judgments of learning. Journal of Experimental Psychology: General, 126, 349.
Mackay, D.G., Shafto, M., Taylor, J.K., Marian, D.E., Abrams, L and Dyer, J.R. (2004). Relations between emotion, memory, and attention: Evidence from taboo Stroop, lexical decision, and immediate memory tasks. Memory & Cognition, 32, 474-488.
Mueller, M. L., & Dunlosky, J. (2017). How beliefs can impact judgments of learning: Evaluating analytic processing theory with beliefs about fluency. Journal of Memory and Language, 93, 245–258.
Murphy, N. A., & Isaacowitz, D. M. (2008). Preferences for emotional information in older and younger adults: a meta-analysis of memory and attention tasks. Psychology & Aging, 23, 263.
Murphy, D. H., Huckins, S. C., Rhodes, M. G., & Castel, A. D. (2022). The effect of perceptual processing fluency and value on metacognition and remembering. Psychonomic Bulletin & Review, 29, 910-921.
Nelson, T. O., & Narens, L. (1994). Why investigate metacognition. Metacognition: Knowing about Knowing, 13, 1-25.
Nelson, T. O. (1996). Consciousness and metacognition. American Psychologist, 51, 102.
Rhodes, M. G., & Castel, A.D. (2008). Memory predictions are influenced by perceptual information: evidence for metacognitive illusions. Journal of Experimental Psychology: General, 137, 615–625.
Rhodes, M. G., Castel, A. D. (2009) Metacognitive illusions for auditory information: Effects on monitoring and control. Psychonomic Bulletin & Review, 16, 550–554.
Tauber, S. K., & Dunlosky, J. (2012). Can older adults accurately judge their learning of emotional information?. Psychology and Aging, 27, 924.
Tauber, S. K., Dunlosky, J., Urry, H. L., & Opitz, P. C. (2017). The effects of emotion on younger and older adults’ monitoring of learning. Aging, Neuropsychology, and Cognition, 24, 555–574.
Thiede, K. W., & Dunlosky, J. (1999). Toward a general model of self-regulated study: An analysis of selection of items for study and self-paced study time. Journal of Experimental Psychology: Learning, Memory, and Cognition, 25, 1024.
Witherby, A.E., Tauber, S.K., Dunlosky, J. (2021). Why Do Emotional Stimuli Influence Judgments of Learning? Theory, Evidence, and Future Directions. In: Moraitou, D., Metallidou, P. (eds) Trends and Prospects in Metacognition Research across the Life Span. Springer, Cham.
Zimmerman, C. A., & Kelley, C. M. (2010). “I’ll remember this!”. Effects of emotionality on memory predictions versus memory performance. Journal of Memory and Language, 62, 240–253.