Time course of effects of emotion on item memory and source memory for Chinese words

Neurobiology of Learning and Memory 95 (2011) 415–424

Contents lists available at ScienceDirect

Neurobiology of Learning and Memory journal homepage: www.elsevier.com/locate/ynlme

Time course of effects of emotion on item memory and source memory for Chinese words Bo Wang a, Xiaolan Fu b,⇑ a b

Department of Psychology, School of Social Development, Central University of Finance and Economics, Beijing 100081, China State Key Laboratory of Brain and Cognitive Science, Institute of Psychology, Chinese Academy of Sciences, Beijing 100101, China

a r t i c l e

i n f o

Article history: Received 14 September 2010 Revised 23 January 2011 Accepted 1 February 2011 Available online 15 February 2011 Keywords: Time course Emotion Free recall Recognition Recollection Familiarity Source memory

a b s t r a c t Although many studies have investigated the effect of emotion on memory, it is unclear whether the effect of emotion extends to all aspects of an event. In addition, it is poorly understood how effects of emotion on item memory and source memory change over time. This study examined the time course of effects of emotion on item memory and source memory. Participants learned intentionally a list of neutral, positive, and negative Chinese words, which were presented twice, and then took test of free recall, followed by recognition and source memory tests, at one of eight delayed points of time. The main findings are (within the time frame of 2 weeks): (1) Negative emotion enhances free recall, whereas there is only a trend that positive emotion enhances free recall. In addition, negative and positive emotions have different points of time at which their effects on free recall reach the greatest magnitude. (2) Negative emotion reduces recognition, whereas positive emotion has no effect on recognition. (3) Neither positive nor negative emotion has any effect on source memory. The above findings indicate that effect of emotion does not necessarily extend to all aspects of an event and that valence is a critical modulating factor in effect of emotion on item memory. Furthermore, emotion does not affect the time course of item memory and source memory, at least with a time frame of 2 weeks. This study has implications for establishing the theoretical model regarding the effect of emotion on memory. Ó 2011 Elsevier Inc. All rights reserved.

1. Introduction Over the past several decades the relation between emotion and memory has increasingly attracted attention from cognitive psychologists (Uttl, Ohta, & Siegenthaler, 2006). To elucidate the relation between emotion and memory, it is important to understand how emotion affects memory. Memory for events or episodes in particular places at particular times is called episodic memory; most laboratory tasks that psychologists had used over the past century to study memory could be classified as requiring episodic memory (Tulving, 1972). The concept of episodic memory has not only been treated as a psychological construct useful for heuristic and descriptive purposes, but has been used to refer to a specialized mind-brain system (Tulving, 2002). Episodic memory is composed of two elements: item memory and source memory (Slotnick, Moo, Segal, & Hart, 2003). In laboratory studies, item memory refers to recognition or recall of previously presented information itself, whereas source memory refers to recollection or recall of the context from which the fact ⇑ Corresponding author. Address: Institute of Psychology, Chinese Academy of Sciences, 4A Datun Road, Chaoyang District, Beijing 100101, China. Fax: +86 10 64872070. E-mail address: [email protected] (X. Fu). 1074-7427/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.nlm.2011.02.001

or information is acquired. The term source can refer to a variety of characteristics that, collectively, specify the conditions under which a memory is acquired (e.g., the spatial, temporal, and social context of the event; the media and modalities through which it was perceived) (Johnson, Hashtroudi, & Lindsay, 1993). Studies have shown that item memory and source memory are dissociable elements of episodic memory. For example, under certain circumstances, people can correctly recall a piece of information without being able to recollect its source such as who told them or how they came to know about it (e.g., Schacter, Kaszniak, Kihlstrom, & Valdiserri, 1991). Neuropsychological evidence also supports the dissociation between item memory and source memory (e.g., Glisky, Polster, & Routhieaux, 1995). Many studies have investigated the effect of emotion on item memory; however, the findings are contradictory. Some studies have shown that emotion enhances item memory (Aycicegi-dinn & Caldwell-Harris, 2009; Blake, Varnhagen, & Parent, 2001; Bradley, Greenwald, Petry, & Lang, 1992; Comblain, D’Argembeau, Van Der Linden, & Aldenhoff, 2004; Danion, Kauffmann-Muller, Grange, Zimmermann, & Greth, 1995; Grider & Malmberg, 2008; Guya & Cahill, 1999; Hertel & Parks, 2002; Kensinger & Corkin, 2003; Mathews & Barch, 2006; Shigemune et al., 2010); some studies have shown that emotion impairs item memory (Corson & Verrier, 2007; Maratos, Allan, & Rugg, 2000); other studies have

416

B. Wang, X. Fu / Neurobiology of Learning and Memory 95 (2011) 415–424

demonstrated that emotion has no effect on item memory (Doerksen & Shimamura, 2001; Dougal & Rotello, 2007; Johansson, Mecklinger, & Treese, 2004; Kapucu, Rotello, Ready, & Seidl, 2008; Windmann & Kutas, 2001). With regard to the scenario of the effect of emotion on source memory, the findings are likewise mixed. Although a number of studies have demonstrated an enhancement effect of emotion (Anderson & Shimamura, 2005; D’Argembea and Van der Linden, 2005a, 2005b; Doerksen & Shimamura, 2001; Guillet & Arndt, 2009; Kensinger & Corkin, 2003; Kensinger & Schacter, 2006; Mather & Nesmith, 2008), some studies have found the impairment effect on source memory (e.g., Cook, Hicks, & Marsh, 2007; Maddock & Frein, 2009) and one study have found that emotion has no effect on source memory (Sharot & Yonelinas, 2008). The current controversy may be because the effect of emotion on episodic memory can be modulated by a variety of factors, such as type of memory task (e.g., Ramponi, Handelsman, & Barnard, 2010) and the time at which memory is tested (e.g., Sharot & Yonelinas, 2008). Although many studies have examined the effect of emotion on both item memory and source memory, to the best of our knowledge very few studies have concerned themselves with how emotion affects item memory and source memory over time. In a study by Kleinsmith and Kaplan (1963), for learning participants were presented paired associates either of high arousal or low arousal, and were tested at various time intervals: 2 min, 20 min, 45 min, 1 day, and 1 week. Low arousal paired associates led to high immediate recall value and rapid forgetting, whereas high arousal paired associates resulted in low immediate recall and high permanent memory, indicating that emotional arousal is conducive to memory retention over time. In a recent study by Sharot and Phelps (2004), recognition of neutral and arousing words were examined at two time points. Recognition of neutral words became worse over time, whereas recognition of arousing words remained the same and was better than neutral word recognition at a 24-h delay. This study indicates that emotion serves to maintain item memory over time. Sharot and Yonelinas (2008) examined the time course of both item memory and source memory for emotional and neutral pictures at two retention intervals. Consistent with the finding from Sharot and Phelps (2004), recollection was enhanced for emotional compared to neutral photos after a 24-h delay, but not immediately after encoding. However, at both time points, no effect of emotion on source memory was observed. The above studies demonstrate that the time course of the effect of emotion on item memory is different from that on source memory. Although previous studies provided important insights into the time course of effect of emotion, they have the limitation that only two time points were used, which makes it difficult to have a comprehensive understanding of the time course of the effect of emotion. To better capture the pattern for the time course, in this study we used the following eight time points: immediate, 19-min delay, 63-min delay, 4.9-h delay, 8.8-h delay, 24-h delay, 6-day delay, and 2-week delay (see Ebbinghaus, 1885). During learning, participants memorized intentionally both the words themselves and their associated font colors. All words were presented twice. Then they were randomly assigned to eight delay conditions in which to take test of free recall, followed by recognition and source memory tests, at corresponding time points. There has been a number of research investigating on the difference between free recall and recognition. According to some theorists (e.g., Anderson & Bower, 1972; Kintsch, 1970), the process of free recall consists of two major sub-processes: a search process and a recognition process. First, information in the long-term store must be searched. Once a piece of information has been located, a recognition process than starts to ascertain whether that piece of information has indeed occurred. If the judgment is that it has

not, the search process continues until a next piece of information is located. In a recognition test, however, there is no need for the search process because all the information is presented and the participant only needs to decide which piece of information had indeed occurred during initial learning. Given the above theories on the difference between free recall and recognition, it is not surprising that some studies have shown the effect of emotion over time can be modulated by whether memory is tested by free recall or by recognition. It was only at a delay that emotion enhanced recognition (Sharot & Phelps, 2004; Sharot & Yonelinas, 2008), and such enhancement effect remained at 2-week delay tests (Comblain, D’Argembeau, Van Der Linden, & Aldenhoff, 2004; Hamann, Monarch, & Goldstein, 2000). However, emotion enhanced free recall either at a 5-min delay (Hertel & Parks, 2002) or at 1-week delay test (Guya & Cahill, 1999). Based on the above findings as well as the theory regarding the difference regarding free recall and recognition (Anderson & Bower, 1972; Kintsch, 1970), we hypothesized that emotion would enhance recognition only at delayed tests and that emotion would enhance free recall both at immediate and delayed tests. It has been suggested that emotion affects memory consolidation via the modulation of the amygdala, specifically the basolateral (BLA) region (McGaugh, 2002). In fact, studies have shown that lesions of the BLA can lock the induction of long-term potentiation (LTP) in the dentate gyrus of the hippocampus (e.g., Ikegaya, Nakanishi, Saito, & Abe, 1997). Memory consolidation takes time, and the slow consolidation serves an adaptive function by enabling neurohormonal processes triggered by an arousing stimulus to modulate memory strength (McGaugh, 2000). In fact, studies have shown that the effect of emotion on memory would be apparent following a delay (e.g., Sharot & Phelps, 2004). Therefore, we hypothesized that the magnitude of effect of emotion would become more marked over time. Based on the finding from Sharot and Yonelinas (2008), we hypothesized that emotion would not have any effect on source memory whether at immediate or at 24-h delay test or other shorter delay tests (i.e., 19-min delay, 63-min delay, 4.9-h delay and 8.8h delay tests). However, if consolidation of source memory takes longer than 24 h, emotion would enhance source memory in other longer delayed tests. 2. Methods 2.1. Participants One hundred and thirty-six undergraduates and graduate students (86 female and 50 male, mean age = 22.91 years) from several universities attended this experiment. All participants reported themselves to be non-smoking and free from any emotional disorders. Participants were paid 20 Yuan (RMB) an hour. Data from four participants were not included for final analysis because they failed to follow the instructions. This study was approved by the Institutional Review Board of the Institute of Psychology, Chinese Academy of Sciences. 2.2. Material 2.2.1. Material for memory ability test In the memory ability test 16 abstract pictures were used. These pictures were downloaded from http://www2.bc.edu/~slotnics/ scripts.htm. 2.2.2. Material for primary test A total of 333 Chinese words, selected from Dictionary of Word Frequency of Modern Chinese (Compiled by Beijing Institute of

417

4.495(0.182) 4.691(.197)

f c a

5.771(0.164) 5.786(.133) 2.440(0.090) 2.417(.071)

p

Negative words

f

0.0263(0.0139) .0075(.0019)

Languages), were divided into four sets, each of which were rated on a Likert-type scale (ranging from 1 to 9) respectively by 28 participants (16 male, 12 female), 23 participants (11 male, 12 female), 28 participants (13 male, 15 female), and 24 participants (12 male, 12 female). All the words were rated with regard to pleasantness, arousal, and concreteness; the rating instructions, which are presented in Supplement, were adapted from Bradley and Lang (1999). From the 333 rated words a total of 120 words were chosen to be used in the primary experiment. The 120 words, including 40 neutral (4.5 < pleasantness < 5.5, arousal < 5), 40 positive (pleasantness > 6, arousal > 5) and 40 negative (pleasantness < 4, arousal > 5) words, were evenly divided into set 1 and set 2 such that each set consists of 20 neutral words, 20 positive words, and 20 negative words. The two sets of words, which were matched in pleasantness, arousal, concreteness, and word frequency (see Table 1), were used in a counterbalanced manner across participants, as target items in the learning session and distracters in the testing session. In each set positive words have greater pleasantness than both neutral and negative words, and neutral words have greater pleasantness than negative words (all ps < .05). Both positive and negative words have greater arousal than neutral words (all ps < .05), but arousal is comparable between positive and negative words (p > .05). In addition, in each set, the three types of words do not significantly differ in abstractness and word frequency (all ps > .05).

0.007(0.002) .0024(.0004)

B. Wang, X. Fu / Neurobiology of Learning and Memory 95 (2011) 415–424

4.781(0.244) 4.691(.197) Note: p, a, c and f respectively stands for pleasantness, arousal, concreteness and word frequency.

a

6.016(0.083) 6.010(.105) 6.307(0.068) 6.418(.0662)

p f

0.0091(0.0038) .0087(.0045) 4.854(0.149) 4.884(.177)

c a

4.947(0.045) 4.947(.0513)

p

5.059(0.050) 5.052(.0578) Set 1 Set 2

Positive words Neutral words

2.3.2. Design and procedure of primary test About one minute after the memory ability test, the primary test was conducted. A mixed design was used, with emotion (negative, positive, and neutral) representing the within-subject factor and delay condition (immediate, 19-min delay, 63-min delay, 4.9-h delay, 8.75-h delay, 1-day delay, 6-day delay, and 14-day delay) representing the between-subject factor. Dependent variables include performances of item memory and source memory. The above eight time points of delay were chosen largely based on the study of Ebbinghaus (1885), in which the following eight time points were used: immediate, 19-min, 63-min delay, 8.75-h, 1-day, 6-day, and 14-day and 30-day. We did not use a 30-day delay

Table 1 Descriptive statistics for the two sets of words (standard errors in parentheses).

2.3.1. Design and procedure of memory ability test Memory ability test was conducted in order to ascertain whether participants who had been assigned to the eight delay conditions had comparable ability of item memory and source memory. During the learning phase, participants sat about 50 cm in front of computer screens, ready to memorize 16 abstract pictures. In each trial a crosshair first appeared at the center of a screen for 1 s. Then an abstract picture stayed at the center of the screen for 3 s, with the color of screen background being either blue or red. Participants were asked to remember each picture and its associated background color. To avoid possible floor effect, participants were asked to conduct two blocks of learning, in each of which the stimuli were randomly presented. Immediately after learning, a memory test was arranged in which the 16 pictures that had been learned were mixed with another set of 16 new pictures. In each trial, a crosshair first appeared at the center of a screen, followed by a picture on a white background. For the item memory task, participants were asked to judge whether a picture had been learned previously by clicking a corresponding button. If they responded in the affirmative, they were further asked to conduct the source memory task of recalling the background color in which the picture was initially presented during the learning. If they gave a negative response, then a crosshair appeared at the center of the screen, followed by the next picture.

c

2.3. Design and procedure

418

B. Wang, X. Fu / Neurobiology of Learning and Memory 95 (2011) 415–424

During the memory test, participants were first given 5 min to recall the words they had learned. They were encouraged to write down any words that they felt they had learned. Following the free recall, the 60 old words and 60 new words were mixed and randomly presented at the center of screen in the font color of black (Courier New, font size = 40), still on a white screen background. By clicking a corresponding button on a screen, participants were required to choose from: (1) I ‘‘remember’’ this word; (2) I ‘‘know’’ this word; (3) I did not learn this word. The exact definitions of ‘‘remember’’ and ‘‘know’’ had been previously explained to participants: To ‘‘remember’’ a word means to consciously recollect its details, whereas to ‘‘know’’ a word means to be only familiar with it but unable to have any recollection of its details. After making either a ‘‘remember’’ or ‘‘know’’ judgement, they were required to further judge the color of the word. Their responses were selfpaced. The details of instructions, adapted from Geraci and Mccabe (2006), are given below in italics.

Table 2 Multiple comparisons for free recall (immediate condition versus other conditions). Delay condition (I) Delay Mean condition (J) difference (I Immediate

19-min 63-min 4.9-h 8.75-h 1-day 6-day 2-week

.045 .048 .090 .105 .109 .146 .150

Standard error p value J) .031 .032 .024 .023 .025 .022 .022

.816 .809 .013 .002 .003 .000 .000

considering that, with the dependent measures we used, a floor effect is very likely to occur, especially in terms of source memory performance. However, to better capture the effect of time, we used a time point of 4.9-h delay (the mid-point between 63-min delay and 8.75-h delay) which was not used in the study of Ebbinghaus (1885). In the immediate and 1-day delay conditions, there were respectively 22 participants (16 female and 6 male) and 18 participants (10 female and 8 male). In each of the remaining six delay conditions, there were 16 participants (10 female and 6 male). During the learning phase, participants sat about 50 cm in front of the computer screen, preparing themselves to memorize 60 words. In each trial a crosshair first appeared at the center of screen for 1 s, followed by a word in either red or blue font (Courier New, font size = 40) appearing for 2 s, on a white screen background. In each category of words (neutral, positive and negative), half of the words were in red font and the remaining half were in blue font. Participants were asked to remember both the words and their font colors. To buffer primacy and recency effects, two neutral words were placed at the beginning and two neutral words were placed at the end of the word list. Participants conducted two blocks of the same list of 60 words so as to avoid floor effect, which is especially likely to occur in long delay conditions. Words were randomized in each of the two blocks.

2.3.3. The instructions for ‘‘remember’’ and ‘‘know’’ judgements You should click the ‘‘remember’’ button if you can consciously recollect a word’s prior occurrence. That is, you can mentally travel back in time to the specific moment that you saw the word. For example, you may recall the specific font color of the word, or you may recall something that happened in the room when that word was presented (a noise), or you may recall a specific thought you had about that particular word at the moment it was presented. You may even remember the words that came before or after it. The important point is that you should indicate that you remember the word if you can become consciously aware again of some particular details of the word’s occurrence. You should click the ‘‘know’’ button if you feel that a word was presented in the study phase but you cannot consciously recollect any details of the word’s actual occurrence. For participants in the immediate condition, the above-mentioned memory tests were conducted immediately after the learning. For participants in the other delay conditions, some mathematical tasks (lasting for 19 min) were arranged right after

0.25

correct percentage

0.20 neutral words positive words

0.15

negative words 0.10 0.05 0.00 immediate

19-min

63-min

-0.05

4.9-h

8.8-h

24-h

6-day

2-week

delay conditions Fig. 1. Time course of effect of emotion on free recall.

Table 3 Descriptive statistics of hit rates and false alarm rates (FAR) for the three types of words (standard errors in parentheses). Delay condition

Immediate 19-min 63-min 4.9-h 8.75-h 1-day 6-day 2-week

Neutral words

Positive words

Negative words

Hit rate

FAR

Hit rate

FAR

Hit rate

FAR

.823(.040) .722(.047) .750(.048) .738(.047) .675(.047) .708(.044) .569(.047) .593(.048)

.191(.048) .172(.057) .200(.059) .300(.057) .278(.057) .367(.054) .338(.057) .377(.059)

.848(.037) .756(.043) .793(.045) .803(043) .706(.043) .817(.041) .700(.043) .687(.045)

.239(.049) .159(.057) .247(.059) .438(.057) .294(.057) .458(.054) .453(.057) .507(.059)

.891(.033) .781(.038) .820(.039) .844(.038) .831(.038) .864(.036) .756(.038) .727(.039)

.320(.049) .234(.057) .343(.059) .513(.057) .453(.057) .594(.054) .559(.057) .550(.059)

419

.321(.058) .328(.059) .353(.058) .330(.043) .297(.060) .550(.074) .438(.054) .343(.049) .096(.025) .109(.027) .137(.044) .177(.040) .222(.055) .092(.030) .141(.043) .318(.068) .570(.062) .484(.068) .467(.061) .507(.053) .534(.070) .311(.073) .300(.065) .386(.065) .196(.038) .141(.053) .173(.045) .357(.054) .194(.044) .381(.057) .356(.047) .289(.038) .322(.055) .359(.069) .323(.068) .323(.045) .347(.068) .575(.055) .444(.050) .357(.057) .043(.014) .022(.010) .073(.047) .057(.016) .100(.034) .078(.027) .097(.030) .221(.044) .525(.066) .441(.080) .470(.083) .473(.059) .359(.070) .228(.049) .250(.059) .311(.063) .164(.034) .156(.057) .143(.043) .240(.051) .191(.041) .333(.067) .294(.057) .246(.037) .300(.051) .325(.063) .290(.062) .343(.036) .319(.063) .511(.065) .369(.038) .282(.046)

FAR ‘‘Know’’

FAR FAR

.027(.011) .016(.008) .057(.028) .043(.019) .088(.035) .050(.023) .069(.028) .157(.042) .522(.066) .419(.077) .460(.077) .387(.062) .356(.063) .177(.047) .203(.048) .282(.064)

Hit rate

‘‘Remember’’ ‘‘Know’’

Hit rate

FAR ‘‘Remember’’ ‘‘Know’’

Hit rate

‘‘Remember’’

Hit rate

Positive words

2.5.2.2. Effect of emotion on the time course of recognition. Statistics for hit rates and false alarm rates for the three types of words were presented in Tables 3 and 4. With recognition accuracy (hit rates minus false alarm rates) being the dependent measure, the main effect of emotion is significant (F (2, 254) = 6.527, p = .002, partial g2 = .049). Recognition for negative words was significantly worse than for neutral words (p = .003) and for positive words (p = .002). No significant difference was found between recognition for neutral and for positive words (p = .736). The main effect of delay condition was significant (F (14, 254) = .528, p = .916, partial g2 = .028). The results of

Neutral words

2.5.2. Results of primary test 2.5.2.1. Time course of effect of emotion on free recall. The main effect of emotion was significant (F (2, 254) = 21.060, p < .001, partial g2 = .142). Further analyses indicated that free recall for negative words was significantly better than for neutral words (p < .001) and for positive words (p < .001). Free recall for positive words was marginally significantly better than for neutral words (p = .087). The main effect of delay condition was significant (F (7, 127) = 10.077, p < .001, partial g2 = .357). The results of multiple comparisons were presented in Table 2. The interaction between emotion and delay condition was non-significant (F (14, 254) = 1.000, p = .453, partial g2 = .052), meaning that the pattern for the effect of emotion remained the same across all the delay conditions (see Fig. 1).

Delay condition

2.5.1. Results of memory ability test Item memory was determined by subtracting false alarm rates from hit rates, according to the two-high threshold model (Snodgrass and Corwin, 1988). Source memory was calculated as the percentage of correctly recalled colors of screen for pictures that were correctly judged as having been presented in the learning phase. Data from four participants were not collected because they failed to follow the instructions. Analysis of variance (ANOVA) indicated that neither the main effect of group on item memory nor on source memory was significant (F (7,124) = 1.634, p = .132, and F (7,124) = 1.492, p = .176). Therefore, participants who had been randomly assigned to the eight delay conditions had comparable ability of both item memory and source memory.

Table 4 Descriptive statistics of hit rates and false alarm rates (FAR) for ‘‘remember’’ and ‘‘know’’ responses (standard errors in parentheses).

2.5. Results

Hit rate

FAR

Negative words

FAR

Repeated-measures analyses were conducted with emotion (positive and negative and neutral) being the within-subject factor. The dependent measure of free recall was the percentage of words correctly recalled. Recognition was determined by subtracting false alarm rates from hit rates. Source memory was calculated as the percentage of correctly recalled font colors for words that were correctly judged as having been presented in the learning phase. Repeated-measures analyses were also conducted on recollection and familiarity derived respectively from ‘‘remember’’ (R) and ‘‘know’’ (K) responses. Recollection was calculated by subtracting false alarm rates of R responses from hit rates of R responses; familiarity was calculated by subtracting false alarm rates of K responses from hit rates of K responses.

Hit rate

2.4. Data analysis

Immediate 19-min 63-min 4.9-h 8.75-h 1-day 6-day 2-week

learning to avoid rehearsal. Participants, except for those in the immediate, 19-min delay and 63-min delay conditions, went back home after finishing the mathematical tasks; they were instructed to avoid discussing anything about the experiment and to return to the same laboratory at appropriate times according to the delay conditions to which they were assigned.

.225(.041) .141(.052) .207(.042) .313(.039) .231(.051) .492(.056) .413(.032) .282(.047)

B. Wang, X. Fu / Neurobiology of Learning and Memory 95 (2011) 415–424

420

B. Wang, X. Fu / Neurobiology of Learning and Memory 95 (2011) 415–424

significant (for the main effect of delay: F (7, 127) = 1.620, p = .136, partial g2 = .082; for the interaction: F (14, 254) = 1.257, p = .235, partial g2 = .065) (see Fig. 4).

Table 5 Multiple comparisons for recognition (immediate condition versus other conditions). Delay condition (I) Delay Mean condition (J) difference (I Immediate

19-min 63-min 4.9-h 8.75-h 1-day 6-day 2-week

.039 .083 .226 .208 .281 .379 .413

Standard error p value J) .064 .064 .064 .064 .061 .064 .065

.999 .895 .012 .029 .000 .000 .000

2.5.2.3. Time course of effect of emotion on source memory. The data of one participant who failed to attend the second session was excluded from the analysis. The main effect of delay condition was significant (F (7, 124) = 5.711, p < .001, partial g2 = .244). The results of multiple comparisons were presented in Table 6. Neither the main effect of emotion nor the interaction between emotion and delay condition was significant (for the main effect of emotion: F (2, 254) = .513, p = .599, partial g2 = .004; for the interaction: F (14, 254) = .964, p = .491, partial g2 = .051) (see Fig. 5).

multiple comparisons were presented in Table 5. The interaction between emotion and delay condition was non-significant (F (14, 254) = .528, p = .916, partial g2 = .028), indicating that the pattern of the effect of emotion remained the same across the eight delay conditions (see Fig. 2). With accuracy of ‘‘remember’’ responses being the dependent variable, the main effect of delay condition was significant (F (7, 127) = 6.551, p < .001, partial g2 = .265). Neither the main effect of emotion nor the interaction between emotion and delay condition was significant (for the main effect of emotion: F (2, 254) = .05, p = .951, partial g2 < . 001; for the interaction: F (14, 254) = 1.109, p = .35, partial g2 = .058) (see Fig. 3). With accuracy of ‘‘know’’ responses being the dependent variable, the main effect of emotion was significant (F (2, 254) = 4.803, p = .009, partial g2 = .034). Further analyses showed that accuracy of ‘‘know’’ responses for negative words was significantly lower than for neutral words (p = .009) and for positive words (p = .007). However, no significant difference existed between accuracy of ‘‘know’’ responses for neutral and for positive words (p = .892). Neither the main effect of delay condition nor the interaction between emotion and delay condition was

3. Discussion This study aimed to examine the time course of effects of emotion on item memory and source memory. Participants learned intentionally a list of neutral, positive, and negative Chinese words, which were presented twice (via two blocks), and then took test of free recall, followed by recognition and source memory tests, at one of eight delayed points of time. The main findings are (within the time frame of 2 weeks): (1) Negative emotion enhances free recall, whereas there is only a trend that positive emotion enhances free recall; in addition, the enhancement effect of emotion on free recall becomes more marked only within a certain frame of time. (2) In the immediately following recognition test, however, the effect was reversed: Negative emotion reduces recognition, whereas positive emotion has no effect on recognition, with this effect being captured by negative words judged to be known rather than negative words judged to be remembered. (3) For the words correctly recognized, source monitoring was proportionally similar for the

0.80 0.70

recognition

0.60 0.50

neutral words positive words

0.40

negative words

0.30 0.20 0.10 0.00

immediate

19-min

63-min

4.9-h

8.8-h

24-h

6-day

2-week

delay conditions Fig. 2. Time course of effect of emotion on recognition.

0.60

recollection

0.50 0.40

neutral words positive words

0.30

negative words

0.20 0.10 0.00

immediate

19-min

63-min

4.9-h

8.8-h

24-h

delay conditions Fig. 3. The time course of effect of emotion on recollection.

6-day

2-week

421

B. Wang, X. Fu / Neurobiology of Learning and Memory 95 (2011) 415–424

Fig. 4. Time course of effect of emotion on familiarity.

Table 6 Multiple comparisons for source memory (immediate condition versus other conditions). Delay condition (I)

Delay condition (J)

Mean difference (I

Immediate

19-min 63-min 4.9-h 8.75-h 1-day 6-day 2-week

.058 .060 .100 .100 .149 .174 .200

Standard error

p value

.045 .046 .041 .033 .038 .033 .030

.897 .891 .250 .076 .008 .000 .000

J)

neutral and emotional words, indicating that neither positive nor negative emotion has any effect on source memory. 4. Time course of the effect of emotion on free recall

source memory

Consistent with some previous studies (e.g., Blake et al., 2001; Davidson, McFarland, & Glisky, 2006; Doerksen & Shimamura, 2001; Hertel & Parks, 2002), this study reveals that emotion (especially negative emotion) enhances free recall. However, in many previous studies, memory tests were conducted at a certain time point. This study provides an extension by showing that under the various delay conditions within the time frame of 2 weeks, the enhancement effect of negative emotion on free recall remains robust. It is important to note that there is only a trend that positive emotion enhances free recall although negative and positive stimuli were matched with regard to arousal, indicating that valence also plays an important role in the effect of emotion on free recall. A novel finding of this study is the time-dependency of the size of effect of emotion on free recall. Specifically, we have found an optimal time point at which the size of effect of emotion is the

greatest. More importantly, the optimal time points for negative and positive emotions are different. The greatest effect size of negative emotion occurs at an 8.8-h delay, whereas that of positive emotion occurs at a 6-day delay. The above-mentioned new finding is theoretically important, because it modifies and refines the theory that the effect of emotion is enhanced following a delay (Baddeley, 1982; LaBar & Phelps, 1998). The results of this study, for the first time, demonstrate that the effect of emotion on free recall becomes more marked only within a certain frame of time. Therefore, the passage of time does not necessarily lead to a more marked effect of emotion and there actually exists a critical point time at which emotion displays its optimal effect. While we are not in the very position to provide any precise mechanism to account for such a phenomenon, we have the following speculation. Effects of emotion on memory have been hypothesized to follow the Yerkes–Dodson law (Yerkes & Dodson, 1908), where moderate levels of emotion enhance encoding processes and subsequent memory performance, but more extreme levels of emotion impair performance (Dougal, 2003). Therefore, the relationship between intensity of emotion and memory is not linear (following an inverted U pattern). Likewise, we suppose that the relationship between length of time and the size of effect of emotion, although not following an inverted U pattern, is not linear either; the optimal size of effect only occurs at a certain point of time. It is critically important to keep this in mind in establishing the theory concerning the time course of effect of emotion on memory.

5. Time course of the effect of emotion on recognition Consistent with some previous studies (e.g., Danion et al., 1995; Maratos et al., 2000), this study shows that negative emotion reduces recognition whereas positive emotion has no effect on it.

0.90 0.80

neutral words positive words

0.70 0.60

negative words

0.50 0.40 0.30 0.20 0.10 0.00

immediate

19-min

63-min

4.9-h

8.8-h

24-h

delay conditions Fig. 5. Time course of effect of emotion on source memory.

6-day

2-week

422

B. Wang, X. Fu / Neurobiology of Learning and Memory 95 (2011) 415–424

One explanation for the impairment effect of negative words may be that they tend to be more semantically related than neutral words (Talmi & Moscovitch, 2004), leading to higher rates of false alarm, thus impairing the accuracy of general recognition. However, positive emotion has no effect on recognition; we speculate that this is because positive emotion broadens the scope of attention, cognition, action, widening the array of percepts, thoughts, and actions presently in mind (Fredrickson, 2001), thus offsetting the impairing effect resulting from greater semantic relatedness. However, because we did not have a specific measurement concerning the semantic relatedness for negative and positive words, it is possible that under our experimental condition the different effect results from negative words having greater semantic relatedness than positive words. It is worth noting that our finding is not congruent with that from some previous studies (e.g., Bradley et al., 1992; Comblain, D’Argembeau, Van Der Linden, & Aldenhoff, 2004; Kensinger & Corkin, 2003; Mathews & Barch, 2006). A reason may lie in the difference of encoding mode (incidental versus intentional). For example, in the study by Kensinger and Corkin (2003), which shows memory enhancement for negative words, participants were asked to make concrete/abstract judgements for words and therefore it is reasonable to assume that such encoding is more incidental than that in our study, where participants were instructed to try their best to memorize words. In fact, the modulating role of encoding mode in the effect of emotion on memory has been demonstrated in some studies (D’Argembeau and Van der Linden, 2004; Kensinger, Piguet, Krendl, & Corkin, 2005). It is worth mentioning that in this study we used a one-step RK procedure (i.e., participants were asked to choose from ‘‘remember’’, ‘‘know’’ and ‘‘new’’). According to Eldridge, Sarfatti, and Knowlton (2002), the one-step procedure is more likely to lead participants to make false recognitions and consequently use the know response as a guess response when they are unsure that they actually recognize a word has having been presented earlier. 6. Time course of effect of emotion on source memory This study demonstrates that under the various delay conditions, emotion has no effect on source memory, which is consistent with some studies (e.g., Davidson et al., 2006; Sharot & Yonelinas, 2008). According to Sharot and Yonelinas (2008), emotion enhances source memory only when the contextual information associated with emotional stimuli provides some adaptive value to an individual or is important for predicating future events. In this study, font colors of words, which may not have any adaptive value, were used as the source information and therefore emotion was not observed to have enhanced this type of source memory. However, Davidson et al. (2006) argued that the enhancement effect of emotion on source memory may only occur by using an experimental paradigm in which attention is simultaneously allocated to items and their related sources. In this study, participants were asked to memorize both words and their font colors and as such it can be assumed that attention was evenly divided to items and their sources. However, we did not observe an enhancement of emotion on source memory, indicating that the view point of Sharot and Yonelinas (2008) may better explain our results. 7. Different time courses of item memory and source memory This study demonstrates that for item memory (i.e., free recall and recognition of words themselves, see Slotnick et al., 2003) the initial point of time at which a statistically significant decrease occurs is 4.9 h from the end of learning, whereas for source memory (i.e., memory for color of words) the initial point of time at

which a statistically significant decrease occurs is 24 h from the end of learning. That is, in comparison to item memory, source memory appears to be more resistant to decay over time. Although we cannot propose a precise mechanism to account for this phenomenon, we have the speculation that it may be because item memory and source memory are two dissociable elements of episodic memory (Glisky et al., 1995) that have different neural substrates. In fact, studies have shown that frontal lobe is primarily responsible for source memory (e.g., Janowsky, Shimamura, & Squire, 1989; Schacter, Harbluk, & McLachlan, 1984), whereas medial temporal lobe is linked to item memory (Shimamura & Squire, 1987). Since source memory and item memory rely on different brain structures, it may not be surprising that they have different patterns of change over time, as observed in this study. However, future neuroscientific studies are needed to elucidate exactly why source memory is more resistant to decay over time. This above novel finding is interesting considering that a source memory task, which involves the binding of an item to its context, is always more difficult than the corresponding item memory task (Glisky et al., 1995); it appears that the difficulty involved in encoding might be inversely related to the strength of retention. The above finding has important implications for learning activities in that different reviewing strategies should be used depending on the properties of materials learned. If the materials that have been learned are information themselves that are to be tested by free recall or recognition, then it can be wise to review them before 4.9 h has elapsed; however, if the materials that have been learned are contexts of information, then there seems to be no need to start reviewing that soon because it is 24 h from the initial learning that source memory starts to decrease significantly. 8. Limitations Our study has some limitations. The first limitation is that recognition and source memory tests followed free recall. Recalling a word would create a new episode which can interfere with the critical episode of when the word was first presented. Therefore, the effect of emotion and delay on recognition and source memory could be substantially altered by the preceding free recall. Another limitation is that the words were presented twice, making it difficult for participants to make ‘‘remember’’ judgements and possibly increasing the number of ‘‘know judgements’’ because the recollection of a certain episode is more difficult. The third limitation is that recognition memory and source memory were combined in a single task, which may lead source memory to be constrained by item memory because participants had the opportunity to judge source information only after they made a ‘‘hit’’ response. 9. Future directions There are some issues that remain to be solved in future studies. Emotion can be represented by the two orthogonal components of valence and arousal (Greenwald, Cook, & Lang, 1989; Russell, 1980). This study demonstrates that emotion (especially negative emotion) enhances free recall. However, because emotional words differ from neutral words not only in terms of valence but also in terms of arousal, it is unclear whether the enhancement effect of emotion is contributed by valence or arousal. This issue is important and needs to be clarified in future studies, because there exists controversy regarding the roles of valence and arousal (Kensinger, 2009; Mather, 2007). The source monitoring framework (Johnson et al., 1993) proposes that there are three types of source memory tasks that respectively examine the abilities to distinguish between internal

B. Wang, X. Fu / Neurobiology of Learning and Memory 95 (2011) 415–424

sources from external sources (e.g., between what one imagines from what one sees), to distinguish between several internal sources (e.g., between the memory of what one thought and the memory of what one said), and to distinguish between several external sources (between what one sees and what one hears). In this study source memory task only examines the ability to distinguish two external sources (font colors of words). Future studies should examine the time course of effect of emotion on the other two types of source memory based on internal monitoring and reality monitoring. Acknowledgments This study is supported partially by Grant from 973 Program (No. 2011CB302201) and the National Natural Science Foundation of China (No. 61075042). We thank Judy Fleiter and anonymous reviewers for their helpful comments on a prior draft of the paper. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.nlm.2011.02.001. References Anderson, J. R., & Bower, G. H. (1972). Recognition and retrieval processes in freerecall. Psychological Review, 79, 97–123. Anderson, L., & Shimamura, A. P. (2005). Influences of emotion on context memory while viewing film clips. American Journal of Psychology, 118, 323–337. Aycicegi-dinn, A., & Caldwell-Harris, C. (2009). Emotion-memory effects in bilingual speakers: A levels-of-processing approach. Bilingualism: Language and Cognition, 12, 291–303. Baddeley, A. D. (1982). Implications of neuropsychological evidence for theories of normal memory. Philosophical Transactions of the Royal Society of London: Series B, 298, 59–72. Blake, T. M., Varnhagen, C. K., & Parent, M. B. (2001). Emotionally arousing pictures increase blood glucose Levels and Enhance Recall. Neurobiology of Learning and Memory, 75, 262–273. Bradley, M. M., & Lang, P. J. (1999). Affective norms for english words (ANEW): Instruction manual and affective ratings. Technical report C-1, The Center for Research in Psychophysiology, University of Florida. Bradley, M. M., Greenwald, M. K., Petry, M. C., & Lang, P. J. (1992). Remembering pictures: Pleasure and arousal in memory. Journal of Experimental Psychology: Learning, Memory and Cognition, 18, 379–390. Comblain, C., D’Argembeau, A., Van Der Linden, M., & Aldenhoff, L. (2004). The effect of ageing on the recollection of emotional and neutral pictures. Memory, 12, 673–684. Cook, G. I., Hicks, J. L., & Marsh, R. L. (2007). Source monitoring is not always enhanced for valenced material. Memory and Cognition, 35, 222–230. Corson, Y., & Verrier, N. (2007). Emotions and false memories: Valence or arousal? Psychological Science, 18, 208–211. Danion, J.-M., Kauffmann-Muller, F., Grange, D., Zimmermann, M.-A., & Greth, P. (1995). Affective valence of words, explicit and implicit memory in clinical depression. Journal of Affective Disorders, 34, 227–234. D’Argembea, A., & Van der Linden, M. (2005a). Influence of emotion on memory for temporal information. Emotion, 5(4), 503–507. D’Argembea, A., & Van der Linden, M. (2005b). Influence of emotion on memory for temporal information. Emotion, 5, 503–507. D’Argembeau, A., & Van der Linden, M. (2004). Influence of affective meaning on memory for contextual information. Emotion, 4, 173–188. Davidson, P. S. R., McFarland, C. P., & Glisky, E. L. (2006). Effects of emotion on item and source memory in young and older adults. Cognitive, Affective and Behavioral Neuroscience, 6(4), 306–322. Doerksen, S., & Shimamura, A. (2001). Source memory enhancement for emotional words. Emotion, 1, 5–11. Dougal, S. (2003). A dual process approach to emotional memory: Effects of emotion on familiarity and retrieval processes in recognition. Unpublished Doctoral Dissertation, University of Pittsburgh. Dougal, S., & Rotello, C. M. (2007). ‘‘Remembering’’ emotional words is based on response bias, not recollection. Psychonomic Bulletin & Review, 14(3), 423–429. Ebbinghaus, M. (1885). Memory: A contribution to experimental psychology. Originally published in New York in 1913 by Teachers College, Columbia University. Eldridge, L. L., Sarfatti, S., & Knowlton, B. J. (2002). The effect of testing procedure on remember-know judgments. Psychonomic Bulletin & Review, 9, 139–145. Fredrickson, B. L. (2001). The role of positive emotions in positive psychology: The broaden-and-build theory of positive emotions. American Psychologists, 56, 218–226.

423

Geraci, L., & Mccabe, D. P. (2006). Examining the basis for illusory recollection: The role of remember/know instructions. Psychonomic Bulletin & Review, 13, 466–473. Glisky, E. L., Polster, M. R., & Routhieaux, B. C. (1995). Double dissociation between item and source memory. Neuropsychology, 9, 229–235. Greenwald, M. K., Cook, E. W., & Lang, P. J. (1989). Affective judgment and psychophysiological response: Dimensional covariation in the evaluation of pictorial stimuli. Journal of Psychophysiology, 3, 51–64. Grider, R. C., & Malmberg, K. J. (2008). Discriminating between changes in bias and changes in accuracy for recognition memory. Memory & Cognition, 36, 933–946. Guillet, R., & Arndt, J. (2009). Taboo words: The effect of emotion on memory for peripheral information. Memory & Cognition, 37, 866–879. Guya, S. C., & Cahill, L. (1999). The role of overt rehearsal in enhanced conscious memory for emotional events. Consciousness and Cognition, 8, 114–122. Hamann, S. B., Monarch, E. S., & Goldstein, F. C. (2000). Memory enhancement for emotional stimuli Is impaired in early Alzheimer’s disease. Neuropsychology, 14(1), 82–92. Hertel, R. T., & Parks, C. (2002). Emotional episodes facilitate word recall. Cognition and Emotion, 16, 685–694. Ikegaya, Y., Nakanishi, K., Saito, H., & Abe, K. (1997). Amygdala beta-noradrenergic influence on hippocampal long-term potentiation in vivo. NeuroReport, 8, 3143–3146. Janowsky, J. S., Shimamura, A. P., & Squire, L. R. (1989). Source memory impairment in patients with frontal lobe lesions. Neuropsychologia, 27, 1043–1056. Johansson, M., Mecklinger, A., & Treese, A.-C. (2004). Recognition memory for emotional and neutral faces: An event-related potential study. Journal of Cognitive Neurosciences, 16(10), 1840–1853. Johnson, M. K., Hashtroudi, S., & Lindsay, D. S. (1993). Source monitoring. Psychological Bulletin, 114, 3–28. Kapucu, A., Rotello, C. M., Ready, R. E., & Seidl, K. N. (2008). Response bias in ‘‘remembering’’ emotional stimuli: A new perspective on age differences. Journal of Experimental Psychology: Learning, Memory, and Cognition, 34, 703–711. Kensinger, E. A. (2009). Remembering the details: Effects of emotion. Emotion Review, 1, 99–113. Kensinger, E. A., & Corkin, S. (2003). Memory enhancement for emotional words: Are emotional words more vividly remembered than neutral words? Memory & Cognition, 31, 1169–1180. Kensinger, E. A., Piguet, O., Krendl, A. C., & Corkin, S. (2005). Memory for contextual details: Effects of emotion and aging. Psychology and Aging, 20, 241–250. Kensinger, E. A., & Schacter, D. L. (2006). Reality monitoring and memory distortion: Effects of negative, arousing content. Memory and Cognition, 34, 251–260. Kintsch, W. (1970). Models for free recall and recognition. In D. A. Norman (Ed.), Models of human memory. New York: Academic Press. Kleinsmith, L. J., & Kaplan, S. (1963). Paired-associate learning as a function of arousal and interpolated interval. Journal of Experimental Psychology, 65, 190–193. LaBar, K. S., & Phelps, E. A. (1998). Arousal-mediated memory consolidation: Role of the medial temporal lobe in humans. Psychological Science, 9, 490–493. Maddock, R. J., & Frein, S. T. (2009). Reduced memory for the spatial and temporal context of unpleasant words. Cognition & Emotion, 23, 96–117. Maratos, E. J., Allan, K., & Rugg, M. D. (2000). Recognition memory for emotionally negative and neutral words: An ERP study. Neuropsychologia, 38, 1452–1465. Mather, M. (2007). Emotional arousal and memory binding: An object-based framework. Perspectives on Psychological Science, 2, 233–252. Mather, M., & Nesmith, K. (2008). Arousal-enhanced location memory for pictures. Journal of Memory and Language, 58, 449–464. Mathews, J. R., & Barch, D. M. (2006). Episodic memory for emotional and nonemotional words in individuals with anhedonia. Psychiatry Research, 143, 121–133. McGaugh, J. L. (2000). Memory: A century of consolidation. Science, 14, 248–251. McGaugh, J. (2002). Memory consolidation and the amygdala: A systems perspective. Trends in Neurosciences, 25, 456–461. Ramponi, C., Handelsman, G., & Barnard, P. J. (2010). The memory enhancement effect of emotion is absent in conceptual implicit memory. Emotion, 10, 294–299. Russell, J. A. (1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39, 1161–1178. Schacter, D. L., Harbluk, J. L., & McLachlan, D. R. (1984). Retrieve without recollection: An experimental analysis of source amnesia. Journal of Verbal Learning and Verbal Behavior, 23, 593–611. Schacter, D. L., Kaszniak, A. W., Kihlstrom, J. F., & Valdiserri, M. (1991). The relation between source memory and aging. Psychology and Aging, 6, 559–568. Sharot, T., & Phelps, E. A. (2004). How arousal modulates memory: Disentangling the effects of attention and retention. Cognitive, Affective, & Behavioral Neuroscience, 4, 294–306. Sharot, T., & Yonelinas, A. P. (2008). Differential time-dependent effects of emotion on recollective experience and memory for contextual information. Cognition, 106, 538–547. Shigemune, Y., Abe, N., Suzuki, M., Ueno, A., Mori, E., Tashiro, M., et al. (2010). Effects of emotion and reward motivation on neural correlates of episodic memory encoding: A PET study. Neuroscience Research, 67, 72–79. Shimamura, A. P., & Squire, L. R. (1987). A neuropsychological study of fact memory and source amnesia. Journal of Experimental Psychology: Learning, Memory, and Cognition, 13, 464–473.

424

B. Wang, X. Fu / Neurobiology of Learning and Memory 95 (2011) 415–424

Slotnick, S. D., Moo, L. R., Segal, J. B., & Hart, J. Jr., (2003). Distinct prefrontal cortex activity associated with item memory and source memory for visual shapes. Cognitive Brain Research, 17, 75–82. Snodgrass, J. G., & Corwin, J. (1988). Pragmatics of measuring recognition memory: Applications to dementia and amnesia. Journal of Experimental Psychology General, 117, 34–50. Talmi, D., & Moscovitch, M. (2004). Can semantic relatedness explain the enhancement of memory for emotional words? Memory & Cognition, 32, 742–751. Tulving, E. (1972). Episodic and semantic memory. In E. Tulving & W. Donaldson (Eds.), Organization of memory (pp. 381–403). New York: Academic Press.

Tulving, E. (2002). Episodic memory: From mind to brain. Annual Review of Psychology, 54, 1–25. Uttl, B., Ohta, N., & Siegenthaler, A. L. (2006). Memory and emotion. Blackwell Publishing: Victoria. pp. 2–3. Windmann, S., & Kutas, M. (2001). Electrophysiological correlates of emotioninduced recognition bias. Journal of Cognitive Neuroscience, 13(5), 577–592. Yerkes, R. M., & Dodson, J. D. (1908). The relation of strength of stimulus to rapidity of habit-formation. Journal of Comparative and Neurological Psychology, 18, 459–482.