This study investigates gender differences in emotional memory. We hypothesized that males and females would be able to remember emotional stimuli better than non-emotional stimuli, but this effect would be greater for females. There were 44 participants, 22 men and 22 women. Respondents studied two lists of words, emotional and non emotional, then instructed to record the total number of correct words recalled on response sheets. Our data indicated an overall better recall for emotional than for non emotional words; however, there was no interaction between gender and type of words recalled. This demonstrates how emotional memories are more readily remembered. Emotional enhancement effect in memory implies that relating material to one’s experience and emotion may be an effective learning strategy.
Statement of the Problem
Have you ever wondered why some memories in our lives stay vivid for years while the others fade over a short time? The appropriate answer to this question is emotion as observations in everyday life provides a lot of evidence thus emotional events and facts are better remembered then neutral ones. For example, people reported that flashbulb memories are more vivid and photographic than other autobiographical memories. On the other hand, people who have post-traumatic stress disorder may have persistent memories for traumatic events that they wish to forget. Why and how our memory is influenced by emotion is an interesting question to address. Living in a community, social interaction is a very important part of our lives that will elicit emotional feelings. Memory, on the other hand, is crucial to our daily life activities and plays an important role in learning and intelligence. Thus, understanding the interaction between emotion and memory is central to understanding our motivations, behavior, and well being. Having an understanding of how this mechanism works can contribute to developing effective learning strategies where emotional stimuli could be used to enhance memory. On the other hand, it can contribute to ways of avoiding intrusive emotional recollection to persist in our memory or distort it. To understand better how emotion influence memory, a lot of psychological researches have used emotional stimuli such as words, pictures or faces to study the effect of emotion on memory in the laboratory. Although in a narrower context when compared to studying real life events, these emotional stimuli could be better controlled for and experimental results can have greater implications on the interaction of emotion and memory.
There are a lot of theories accounting for why emotional content are be better remembered than neutral items and the underlying mechanism. It is still a subject of debate as studies try to determine how emotion affects memory. A study by Kensinger and Corkin (2003) demonstrated that negative words were more vividly remembered than neutral words by using remember/know response procedure. 18 participants were asked to study 140 words, with 70 negative and 70 neutral words controlled for word familiarity, word length, and imageability. In a recognition test, 140 new words were added to the original list. Participants were then asked to select whether they vividly remembered the word from the list (measuring recollection), knew they studied the word because it was familiar (measuring familiarity), or thought that the word was new (not studied before). Comparison of the hit and false alarm rates confirmed that participants’ memory was better for negative words, showing the traditional emotional enhancement effect. In addition, more remember than know responses for negative words indicated a recollection enhancement. Overall, the results support that there is a memory enhancement for emotional, as compared with neutral, words.
Another study by Harris and Pashler (2005) challenged the theory of selective rumination, which explained emotional enhancement effect by increased attention, rumination, and rehearsal for emotion contents. One of their studies found that memory for emotional pictures was better for neutral pictures, even when selective rumination is not possible because of fast presentation rate and manipulated retention interval. In this study, 55 participants viewed a very rapidly presented series of 5 pictures for 12 trials at a rate of 4 pictures per second. In total, they viewed 60 pictures, with 50 neutral and 10 emotional (negative). They were told after each trial to describe each picture in a few words either immediately or after a 20 seconds demanding task filled delay. Results indicated that emotional pictures were named and remembered more often than neutral pictures in both immediate and delayed recall conditions. In conclusion, this experiment confirmed the emotional enhancement effect with the use of pictures and without allowing for selective rumination.
Later, another experiment by Kensinger (2008) demonstrated that the elderly showed emotional enhancement effect for only arousing emotional words but not non-arousing emotional words, and also showed a different memory pattern when compared to the young. This study compared the ability of 30 young and 30 old adults to free recall words from different emotional categories. Words that are used in this study differed in valence (how positive or negative) and in arousal (how exciting or calming). Participants studied in total 75 words (divided into 3 lists) with 15 words in each of these categories: neutral, negative non-arousing, negative arousing, positive non-arousing and positive arousing. They were then given a recall test after studying a word list. Results indicated that both age group remembered negative and positive arousing words equally well and more often than neutral words, confirming the emotional enhancement effect. However, young adults remembered more negative than positive non-arousing words, while still showing memory enhancement over neutral words. Old adults, on the other hand, remembered more positive than negative non-arousing words, and showed no memory enhancement for negative non-arousing words over neutral words. The results imply that age is a factor in determining where emotional enhancement effect will be shown or not.
These experiments above provide a background for our study by suggesting that emotion contents were better remembered than neutral contents under different conditions. All of these experiments provide insights for the possible nature of the underlying mechanism for emotional memory; whereas the experiment by Kensinger (2008) provides an insight for possible individual differences in the emotional enhancement effect. The results in this experiment suggested that different age group do not show the same memory enhancement for all emotional information. Just as understanding the underlying process and mechanism is important, understanding individual differences could contribute equally on understanding the interactions between emotion and memory. Although all these studies confirmed the existence of emotional enhancement effect across different conditions and age group, they did not address the possible gender differences in emotional enhancement effect. Many people believed that women are more emotional than men, and experience more emotional experiences. Books like Men Are from Mars, Women Are from Venus (John Gray, 1992), argue that women are emotionally complex and expressive, while men are stoic, better able to suppress their feelings. Some researchers suggest that this sex difference can be explained by the different structures of women and men’s brains. Thus, it is reasonable to suspect that female would have a greater emotional enhancement effect when compared to male.
In our study, we were interested in investigating gender differences in emotional memory. Because female are believed to be more emotional in general, we suggest that it may be possible for women to selectively attend to emotional stimuli more than men. We hypothesized that both male and female would be able to remember emotional stimuli better than non emotional stimuli, but this effect would be greater for female. Our study added to previous studies by investigating gender differences of emotional memory. It also increased external validity by testing whether the emotional enhancement effect can be generalized to another setting with different operational definitions.
There were 44 participants in our experiment, 22 men and 22 women. They were undergraduate students at a large Midwestern university, and ranged in age from 18 to 23 years. The participants were not financially compensated but received credit in a psychology course for their participation. Two separate sessions were needed in total as a session included a group of 22 participants, with 11 males and 11 females. We randomly assigned participants to a group and at the same time controlled the number of males and females in each session. Thus, the number of males and females in each group was the same for easy comparison of gender differences.
Two lists with 10 words in total were prepared for participants to study and recall: a list of emotional words with 5 positive and 5 negative words (see Appendix A) and a list of non emotional concrete words (see Appendix A). All words were controlled for word frequency, number of syllables, and length according to the R.A Vasa et al. Personality and Individual Differences (2006) database. Each word list was placed in a single presentation order determined at random. Two recall sheets were prepared for each participant to identify their gender, recall the word lists, and record the total number of words recalled correctly. We would then use the data (score) from the recall sheets as a measure of memory.
Microsoft PowerPoint 2007 was used to display the words to the participants. SPSS 17.0 was used to present and analyze the data.
Our experiment was a two by two mixed factorial design with word type as a within subject factor and gender as a between subject factor. All participants were in both conditions in the experiment: Emotional and Non emotional. Half of the participants (Group 1) were assigned to participate in the Emotional condition (Condition 1) first, and the Non Emotional condition (Condition 2) second. For the other half of the participants (Group 2), the order of the two word lists was reversed. The number of males and females in both groups was the same. All participants saw the same materials in the same presentation order within each condition. Our definition of emotional words referred to words that can or were intended to spontaneously elicit emotional feelings either positively or negatively as accompanied by physiological changes such as a birthday; while non emotional concrete words were neutral words that were not intended to create any feelings for example a wall.
Before running the experiment, we divided participants into two groups, with 22 participants (11 males and 11 females) in each group. At the beginning of the experiment, an experimenter called out names of participants in Group 2, and they were told to step out the room and wait quietly in the hallway. The participants (in Group 1) remaining in the room were instructed that they would learn and later recall two list of words. They were not told, however, that the first list was all emotional words and the second list was all non emotional words. Recall sheets were passed out and participants were instructed to fill in their gender information on the appropriate blank. First, participants studied the emotional word list presented with a PowerPoint slideshow on a projector screen, for 1 second per word. After the slideshow finished, participants were instructed to write down as many words as they could recall on the first sheet labeled Condition 1. The same procedure was repeated for the non emotional word list, and participants were instructed to write down as many words as they could recall on the second sheet labeled Condition 2. Then, the experimenter showed the first (emotional) list again. Participants were instructed to count the number of words correctly recalled for the list, and write the total in the appropriate blank on the sheet labeled Condition 1. The same procedure was repeated for the non emotional word list except that the sheet labeled Condition 2 was used. Afterwards, participants were told to place the recall sheets in a pile on a front desk in the experiment room, and go out to the hallway. Participants in Group 2 were summoned in the experiment room. The procedures and instructions for participants in group 2 were the same as in group 1 except the order of the two word lists (conditions) was reversed. Thus, participants in group 2 studied the non emotional (Condition 2) list first, and then the emotional (Condition 1) list.
The number of words correctly recalled by each participant was recorded for both the emotional and non emotional conditions. The highest possible number of correctly recalled words was ten for both conditions. The mean number of words recalled for emotional versus non-emotional words was compared across participants, and the mean number of words recalled for emotional words was compared between male and female participants.
Results and Recommendations
We hypothesized that the mean number of words correctly recalled for the emotional condition would be significantly higher than that of non-emotional condition for both males and females. This hypothesis was supported: Participants recalled more words from the emotional word list when compared to the non emotional word list (see Table 1). A paired-samples t-test for the emotional and non emotional conditions showed this difference was significant, t (43) = 0.51, p <.05. A repeated measures ANOVA also showed a significant main effect of word type (emotional versus non emotional) was significant, F (1, 42) = 5.36, p <.05.
We also hypothesized that the mean number of words correctly recalled for the emotional condition would be significantly higher for female when compared to male, but there would be no significant gender difference for the non emotional condition. This hypothesis was rejected: Female participants recalled approximately same number of words for both emotional and non emotional conditions when compared to male participants (see Table 2 & 3). An independent samples t-test for male and female showed that the difference in recall for emotional words was not significant, t (42) = 0.53, p>.05. A repeated measures ANOVA also showed that the predicted interaction between gender and word type was not significant, F (1, 42) = 0.40, p >.05.
To conclude, our results suggested that there was a significant main effect for word type as expected. However, the main effect of gender and interaction between gender and word type was not found.
Discussion and Conclusion
The basic finding of this experiment was that emotional words were better recalled compared to non emotional words, showing the expected emotional enhancement effect. This was found by comparing the means of emotional condition versus non-emotional condition. However, gender differences were not significant: the mean number of words correctly recalled for the emotional condition were not significantly higher for females when compared to males. We conclude that emotion affected participants’ memory. Words that elicit emotions such as fear, happiness, or sadness may be more distinct and memorable because they could be and associated to personal past experiences. We also conclude that this enhancement effect is not affected by gender.
One puzzling factor is that females tend to remember more negative words among the emotional word list, such as “rejected” and “lonely”. This effect was not seen in males, however. This may be explained by social and cultural factors that it is more acceptable for females to express emotions, especially negative emotions. Thus, it may be possible that males have a self defense mechanism to repress negative emotions out of consciousness when placed into a situation eliciting negative emotions. One reason why we fail to find a gender difference between memories for emotional words may be the fact that the word lists were too short to find a significant result as we did not control for the timeliness and primacy effect.
One limitation of the present study is the small sample size used. The results may not be representative to the general population as to the fact that only undergraduate students are involved in the study. The results may then not be able to generalize across other participants. When different participants or settings are used, the results of our research may not be replicated. In addition, the respondents were well aware of what we were doing this could have behaved differently compared to real life situation in their everyday lives. The low ecological validity of the study may therefore provide results that are not true to the real life situations.
Another limitation of this experiment is lack of manipulation checks after the experiment to ensure that the emotional word list did elicit emotions as we expected. Many factors will affect the degree to which participants were aroused. For example, some participants would have more personal experience related to the word birthday than others. Although we controlled for word frequency, number of syllables, and length, it is also possible that the emotional words in our research were simply more memorable than the non-emotional words used. This may be due to the fact that the emotional words chosen were just more semantically related than the non-emotional words, and could be associated with each other more easily. Further research is needed on the alternative explanation. In addition, participants’ mood during recall will also affect their memory for emotional words according to the mood congruent effect. Thus, participants in a negative mood were more likely to remember negative words due to a biased search through memory. Future studies are needed to better understand the effect of mood on memory for emotional words.
Comparing the results in this study with the literature review, our results successfully replicate the emotional enhancement effect where emotional stimuli were better remembered. These results imply that there is an interaction between emotion and memory and more specifically, emotion may enhance memory because of distinctiveness and uniqueness. Our results provide certain insight about applications and understanding of emotional enhancement in real world situation. In particular, parenting style may be important for childhood development of autobiographical memories. For example, discussing the past with children in a more elaborative and emotional way may lead to better memory than a more practical way. This also implies that teachers or lecturers who try to relate material with real life experiences, that tend to elicit emotions, will enhance students’ learning and memory. Students can also use related strategies such as the to-be-remembered material on a personal and emotional level to enhance memory. Our results also differ with a lot of real life phenomenon for example, people reported having flashbulb memories that are more memorable than other autobiographical memories (Schacter, 1996). Our results suggest that this is possible due to more emotional involvement for events that form flashbulb memories. This study can also account for why post-traumatic stress disorder occurs where people have persistent memories for traumatic events that they would wish to forget. Lack of interaction between gender and emotional memory may also suggest why gender will not affect the formation of flashbulb memories or PTSD (Brown, R., & Kulik, J. 1977). However, there are cases where emotional events were forgotten. For example, there are some cases where people forget about either committing, suffering from, or witnessing a violent crime that is emotionally arousing. These cases are referred to as situation specific amnesia. Emotion can also impair memory under certain conditions like prolonged stress, task irrelevant emotional distraction and mood disorders but further research would be adequate to fully understand how emotion interacts with memory.
Brown, R. & Kulik, J. (1977). Flash bulb memories. Cognition, 5, 73–99.
Gray, J. (1992). Men are from Mars, Women are from Venus. USA: Prentice Hall.
Harris, C. R. and Pashler, H. (2005). Enhanced memory for negatively emotionally charged pictures without selective rumination. Emotion, 5, 191-199.
Kensinger, E. A. (2006). Age differences in memory for arousing and non-arousing emotional words. Journal of Gerontology: Psychological Sciences, 63, 13-18.
Kensinger, E. A., & Corkin, S. (2003). Memory enhancement for emotional words: is emotional words more vividly remembered than neutral words?. Memory & Cognition, 31, 1169–1180.
Schacter, D. (1996). Searching for Memory: The brain, the mind and the past. New York: Basic Books.
Vasa, R. A., Carlino, A. R., London, K., & Min, C. (2006). Valence ratings of emotional and Non-emotional words in children. Personality and Individual Differences, 6, 1169–1180.