Brain and Cognition 94 (2015) 52–59

Contents lists available at ScienceDirect

Brain and Cognition journal homepage: www.elsevier.com/locate/b&c

Lateralized differences in tympanic membrane temperature, but not induced mood, are related to episodic memory Ruth E. Propper a,⇑, Taylor D. Barr a, Tad T. Brunyé b,c a

Psychology Department, Montclair State University, 1 Normal Avenue, 225 Dickson Hall, Montclair, NJ 07043, USA Tufts University, Psychology Department, Medford, MA, USA c US Army Natick Soldier Research, Development and Engineering Center, Natick, MA, USA b

a r t i c l e

i n f o

Article history: Accepted 13 January 2015 Available online 2 February 2015 Keywords: Memory Lateralization Recall Episodic

a b s t r a c t The present research examined the effects of pre-encoding and pre-recall induced mood on episodic memory. It was hypothesized that happy and/or angry mood prior to encoding (increasing left hemisphere activity), in tandem with fearful mood prior to recall (increasing right hemisphere activity) would be associated with superior episodic memory. It was also hypothesized that tympanic membrane measures (TMT), indicative of hemispheric activity, would change as a function of induced mood. Although subjectively-experienced mood induction was successful, pre-encoding and pre-recall mood did not alter memory, and only altered TMT in the pre-encoding fear and pre-recall angry mood induction conditions. Interestingly, baseline absolute difference between left and right TMT, a measure of differential hemispheric activity, regardless of the direction of that activity, was significantly positively related to number of total words written, number of correctly recalled words, and corrected recall score. This same TMT measure pre-encoding, regardless of specific mood, was significantly negatively related to false recall. Results are discussed in terms the HERA model of episodic memory, and in the nature of interhemispheric interaction involved in episodic recall. Ó 2015 Elsevier Inc. All rights reserved.

1. Introduction The left and right frontal lobes are differentially involved in the experiencing of emotional/motivational state, with increased left hemisphere activity associated with positive/approach emotion/ motivation, such as happiness, and increased right hemisphere activity associated with negative/withdrawal emotion/motivation, such as fear (e.g.; Davidson, 2002, 2004; Tomarken, Davidson, Wheeler, & Doss, 1992). Although there are other theoretical accounts of hemispheric lateralization of emotion/motivational state [see Shobe, 2014, for example and for review], examination and detailed discussion of these other accounts is beyond the scope of this manuscript. Instead, the current research assumes left hemisphere-happiness and right-hemisphere anxiety cortical lateralization, a conception that is well supported by other work (see Davidson, 2004; Urry et al., 2004). There is controversy regarding lateralization of anger, with some suggestion that anger is a left hemisphere approach emotion (Carver & Harmon-Jones, 2009) while other work indicating it is a right hemisphere, withdrawal state (Zinner, Brodish, Devine, & Harmon-Jones, 2008). Given this ⇑ Corresponding author. Fax: +1 973 655 5000. E-mail address: [email protected] (R.E. Propper). http://dx.doi.org/10.1016/j.bandc.2015.01.005 0278-2626/Ó 2015 Elsevier Inc. All rights reserved.

controversy, anger is also investigated here, and it is hoped that the current work could help to illuminate the lateralization of anger. Interestingly, increased neuronal activity within one versus the other hemisphere results in a biasing of information processing, such that the more active hemisphere’s mode of ‘experiencing’ dominates the processing of incoming information (e.g.: Goldstein, Revivo, Kreitler, & Metuki, 2010; Harmon-Jones, 2006; Propper, Christman, Brunyé, & Januszewski, 2013; Propper, McGraw, Brunyé, & Weiss, 2013; Propper, Brunyé, Christman, & Januszewski, 2012; Spielberg et al., 2011; Seta, McCormick, Gallagher, McElroy, & Seta, 2010). For example, increased performance on the Remote Associates Test (RAT) was found after left hand contractions, presumed to have activated global processing mechanisms in the right hemisphere, compared to following right hand contractions (Goldstein et al., 2010). Specifically, the RAT, considered a measure of convergent creativity, requires participants to find the commonality between three word roots. For example, individuals may be presented with the word trio of ‘falling’, ‘actor’ and ‘dust’, with the correct response being ‘star’. Superior performance following left, but not right, hand contractions was suggested to be the result of an increased right hemisphere

R.E. Propper et al. / Brain and Cognition 94 (2015) 52–59

neural activation, resulting in an increased spread of activation and superior creativity (Goldstein et al., 2010). Similarly, other research has also demonstrated that sustained unilateral motor activity increases hemispheric activation in the contralateral hemisphere, and that this increased activation is associated with increases in lateralized cognitive processes. For example, Schiffer et al. (2004), using fMRI, reported increased activity in the dorsolateral prefrontal cortex contralaterally to side of sustained unilateral gaze. In support of the notion that such sustained unilateral movements alter not only brain activity, but cognition as well, Propper et al. (2012), using methodologies for gaze identical to that of Schiffer et al. (2004), reported changes in semantic memory abilities as a function of sustained unilateral gaze. Presumably these changes in semantic memory were the result of alteration in hemispheric activity in (at least) dorsolateral prefrontal cortex, as a function of side of gaze. Work has also demonstrated that activity designed to increase lateralized hemispheric activation is associated with changes in affective state. For example, right hand clenching (left hemisphere activation) versus left hand clenching (right hemisphere activation) resulted in increased approach (e.g.: happiness, anger) versus withdrawal (e.g.: sadness, anxiety) emotional states, respectively (e.g.: Harmon-Jones, 2006; Peterson, Shackman, & Harmon-Jones, 2008; Schiff & Lamon, 1989). Additionally, the pattern of results in the unilateral hand clenching literature also supports models (e.g.; Davidson, 2002, 2004; Tomarken et al., 1992) of left hemisphere anger lateralization. Thus, induced unilateral hemispheric activity of one versus the other hemisphere is associated with the experiencing of a particular emotional/motivational orientation. Given that positive/approach and negative/withdrawal affects/ motivations may be left versus right hemisphere (respectively) oriented (e.g.; Davidson, 2002, 2004; Tomarken et al., 1992).), it may be possible to alter lateralized hemispheric activity via changes in mood. Mood induction itself may therefore alter hemispheric activity (Flores-Gutiérrez et al., 2009; Schmidt & Trainor, 2001; Tsang, Trainor, Santesso, Tasker, & Schmidt, 2001). Supporting this notion, Schmidt and Trainor (2001) reported changes in mood-congruent affect, as well as decreased alpha power (increased hemispheric activity) in left frontal areas in response to joyful and happy music, and decreased alpha power (increased hemispheric activity) in right frontal areas in response to sad and anxiety producing music. These results indicate that mood induction alters hemispheric activity and, furthermore, alters such activity in a manner consistent with theories of left lateralization of positive/ approach and right hemisphere negative/withdrawal states (e.g.; Davidson, 2002, 2004; Tomarken et al., 1992). Given that music-induced mood induction alters hemispheric activity (e.g.; Schmidt & Trainor, 2001), and given that hemispheric activity is reflected in cognition, it is proposed that music-induced mood, thereby changing hemispheric activity, will be associated with changes in performance on tasks thought to be lateralized to the cerebral hemispheres. In support of this notion, mood induced changes in cognition may be associated with lateralized processes involving attention (e.g.; Ford et al., 2010). For example, Ford et al. (2010) reported that anger increased visual attending to rewarding information, an orientation that may support anger as an approach, left hemisphere motivational state in some contexts. The Hemispheric Encoding/Retrieval Asymmetry (HERA) model of memory proposes that left prefrontal regions are associated with encoding, and right prefrontal regions with retrieval, of episodic memories (Habib, Nyberg, & Tulving, 2003; Tulving, Kapur, Craik, Moscovitch, & Houle, 1994). Although this model has its detractors (e.g.; Lee, Robbins, Pickard, & Owen, 2000; Owen, 2003), as countered by Habib et al. (2003) the criticisms themselves do not necessarily invalidate the HERA model, which in itself

53

also has heuristic value (e.g.; Tulving et al., 1994). Additionally, more recent work has provided results that have been interpreted as supportive of the HERA model (Babiloni et al., 2004; Griessenberger et al., 2012; Okamoto et al., 2011). For example, Babiloni et al. (2004), in a re-analysis of data from a 2004 EEG study, found results supportive of the HERA model using nonverbal stimuli. Okamoto et al. (2011), using functional near-infrared spectroscopy, reported increased right hemisphere activity during recall for a taste, a result they interpreted as supportive of HERA. As suggested by others (e.g.; Cabeza, 2002; Habib et al., 2003; Propper et al., 2012), the left hemisphere encoding/right hemisphere retrieval of episodic memory proposed by HERA may be influenced by a variety of factors, including stimuli material (i.e.; language or spatial-based; e.g.; Propper et al., 2012) and age (e.g.; Cabeza, 2002). The materials in the present study were language-based, and the participants were young adults; both factors which are associated with the patterns of brain activity predicted by HERA. Because increased activity of a given hemisphere is associated with domination of information processing by that hemisphere, increasing one versus the other hemisphere’s neuronal activity immediately prior to encoding, and immediately prior to recalling information, may influence recall ability. Specifically, increased left hemisphere activity during/prior to encoding, and increased right hemisphere activity during/prior to retrieval would be predicted by the HERA model to result in superior recall for episodic information. Previous work has supported this notion. Propper, Christman, et al. (2013) and Propper, McGraw, et al. (2013) reported that presumed increased left hemisphere activity in response to right hand clenching prior to the encoding of list words, and presumed increased right hemisphere activity in response to left hand clenching prior to recall, resulted in superior memory for list words relative to other hand clench conditions. These results suggest (a) differential hemispheric activity in frontal areas may bias cognitive processing and (b) episodic memory may be benefitted by increased left hemisphere activity during encoding and increased right hemisphere activity during retrieval. One purpose of the present research was to examine the effects of mood induction on hemispheric activity and on episodic memory. Happy and fearful mood induction were chosen as differential activators of the left versus right hemisphere, respectively. It was hypothesized that induced moods of happiness prior to encoding, and induced fearfulness prior to recall, would result in superior episodic memory, relative to other mood-induction encoding and retrieval combinations. An anger mood induction was also examined. As mentioned earlier, controversy in the literature regarding anger makes predictions regarding this emotion difficult (e.g.; Carver & Harmon-Jones, 2009; Zinner et al., 2008), and another purpose here was to investigate anger lateralization. If induced anger in the present research causes a pattern of results similar to that found for induced happiness, anger may be similar to left hemisphere lateralized/approach motivational states. Conversely, if anger causes a pattern of results similar to that found for induced fear, then anger might be more similar to right hemisphere lateralized/withdrawal motivational states. Hemispheric activity was assessed via lateralized differences in tympanic membrane temperature (TMT). TMT is a relatively novel measure of lateralized hemispheric activation, and may be a simple, fast way of measuring general cortical activation non-invasively and without the expenses of imaging (e.g.; Helton, 2010; Helton & Carter, 2011; Helton, Harynen, & Schaeffer, 2009; Helton, Kern, & Walker, 2009; Propper & Brunyé, 2013; Propper, Brunyé, Christman, & Bologna, 2010; Propper, Christman, et al., 2013; Propper, Januszewski, Christman, & Brunyé, 2011; Propper, McGraw, et al., 2013). TMT reflects hemispheric activity in frontal and temporal areas, and may be indicative of performance on

54

R.E. Propper et al. / Brain and Cognition 94 (2015) 52–59

cognitive tasks and of affective/motivational state (Cherbuin & Brinkman, 2004; Helton, Harynen, et al., 2009; Helton, Kern, et al., 2009; Helton & Maginnity, 2012; see also Propper & Brunyé, 2013, for review of TMT-affect/motivation literature). The exact physiological mechanism by which the relationship between TMT and hemispheric activity occurs has been elusive (see Propper & Brunyé, 2013). In the most common prevailing perspective, decreased TMT is thought to be associated with increased hemispheric activation on the ipsilateral side of the brain. This finding has been interpreted as a result of a positive correlation between ipsilateral carotid artery and cerebral blood flow, and perfusion of TMT by the same cardiovasculature as that which perfuses the cortex, in conjunction with ipsilateral skull and artery related heat dissipation. That is, ultimately, blood flowing around the TMT is thought to be cooler than blood being actively recruited by the brain during neuronal activity (e.g.; Cherbuin & Brinkman, 2004). According to this model, the difference between left and right TMT is indicative of relative hemispheric activation within an individual (Cherbuin & Brinkman, 2004; Helton, Kern, et al., 2009), with increased hemispheric activity being associated with decreased TMT ipsilaterally. On the other hand, Helton (2010) has suggested also that between-subjects differences in TMT measures may reflect individual differences in skull heat-build up between individuals, and may therefore result in a relationship between TMT and hemispheric activity such that increased TMT is associated with increased hemispheric activation on the ipsilateral side of the brain. Adding to the puzzle, other work has indicated that state versus trait variables may influence the relationship between TMT and hemispheric activity (see Propper & Brunyé, 2013). It should be noted that there is only one (of which we are aware) paper explicitly examining TMT measures and brain activity. Schiffer, Anderson, and Teicher (1999) used EEG and TMT within the same study. Due to the way the data were analyzed however, this latter research unfortunately could not address questions regarding the specific nature of the relationship between TMT and hemispheric activity. Thus, research examining TMT-hemispheric activity relationships have primarily used cognitive and affective measures to infer actual hemispheric activity, which may have added to the confusion in the literature. Additionally, some of the conflict in the literature may arise from the fact that the physiological mechanisms responsible for the relationship between TMT and hemispheric activation are not completely understood, and the association between increased TMT and decreased/increased ipsilateral hemispheric activity is likely modulated, and possibly reversed, by many factors; conflicts within this burgeoning literature await further explanation. Of particular relevance here, TMT measures have been associated with lateralized cognitive and emotional/motivational states (e.g.; Boyce, Higley, Jemerin, Champoux, & Suomi, 1996; Boyce et al., 2002; Gunnar & Donzella, 2004; Helton, 2010). For example, performance on a visuo-spatial task, presumed to rely on right hemisphere processes, resulted in decreased right TMT, while performance on a verbal task, presumed to rely on left hemisphere processes, resulted in decreased left TMT (Cherbuin & Brinkman, 2004). Regarding TMT and emotional/motivational state, Boyce et al. (1996) reported that increasing left side TMT (and therefore decreased left hemisphere activity) was associated with increased negative/withdrawal emotions, and behavior, while Boyce et al. (2002) reported that warmer left side TMT (and therefore increased left hemisphere activity) was associated with increased positive/approach emotions, and warmer right side TMT (and therefore increased right hemisphere activity) associated with increased negative/withdrawal emotions. Therefore, unfortunately, despite these (and other, see Propper & Brunyé, 2013) works, it is

still not clear if an increased or decreased TMT is associated with an increased or decreased ipsilateral hemispheric activity. Despite conflict in the literature, it is clear that TMT measures are predictive of emotional/motivational state and of hemispheric activity, being significantly predictive (sometimes positively and sometimes negatively) with cognition (e.g.; Cherbuin & Brinkman, 2004; Helton & Maginnity, 2012) and with measures of happiness and approach motivational states, as well as with anxiety, withdrawal motivational states (see Propper & Brunyé, 2013). Other work has demonstrated an association between increasing absolute difference between left and right TMT and increasing anger, indicating that increased difference in activity between the cerebral hemispheres, regardless of the direction of that difference, is associated with increased anger (Propper, Christman, et al., 2013; Propper, McGraw, et al., 2013; Propper et al., 2010, 2011). Interestingly, such a finding has been suggested to be related to the variability in lateralization of anger; that is, if anger may be left or right hemisphere lateralized, depending on state and trait individual characteristics, then increasing difference in activity between the hemisphere, regardless of the direction of that difference, would be predicted to be related to the experiencing of anger (e.g.; Propper et al., 2011). Therefore, TMT can be considered an indicator of differential hemispheric activity, and is associated with both cognitive performance and with affective/motivational state. The following hypotheses were made: 1. Induced happiness prior to encoding and induced fear prior to recall will result in superior episodic memory. 2. If anger is an approach, left lateralized emotion, then anger will result in memory performance similar to that of happiness. Conversely, if anger is a withdrawal, right lateralized emotion, then anger will result in memory performance similar to that of fear. 3. If increased TMT is associated with increased ipsilateral hemispheric activity, then induced happiness will be associated with increased left hemisphere TMT, and induced fear with increased right TMT. If decreased TMT is associated with increased ipsilateral hemispheric activity, then the converse will occur. If anger is an approach, left lateralized emotion, then anger will result in TMT similar to that of happiness. Conversely, if anger is a withdrawal, right lateralized emotion, then anger will result in TMT similar to that of fear. 4. Induced anger will be associated with increased absolute difference between left and right TMT. 5. If increased TMT is associated with increased ipsilateral hemispheric activity, then increased left TMT prior to encoding, and increased right TMT prior to recall, will be associated with superior episodic memory. If TMT is associated with decreased ipsilateral hemispheric activity, then the converse will occur. 6. Other work has indicated that episodic memory benefits from increased interhemispheric interaction (e.g.; Christman, Garvey, Propper, & Phaneuf, 2003; Christman, Propper, & Dion, 2004). If so, then relative activity of the hemispheres, measured via TMT, may reflect interhemispheric interaction. For example, there is considerable research indicating that interhemispheric interaction via the corpus callosum may at times be inhibitory in nature (Bloom & Hynd, 2005; Palmer, Schulz, & Larkum, 2013). Callosally-mediated ipsilateral inhibition of motor movements is well established (e.g.: Tazoe & Perez, 2013). Additionally, inhibition of sensory information occurs in cortical areas as well; interestingly, this inhibition may occur in tandem with exitation across multiple cortical levelas (Palmer et al., 2013). Finally, Homae (2014) suggests a developmentally mediated increasing inhibition of homologous cortical areas via the corpus callsoum. If callosal interaction is

R.E. Propper et al. / Brain and Cognition 94 (2015) 52–59

inhibitory (at least in those cortical laminae involved in memory), then increasing difference between the two hemispheres in neuronal activity, measured via TMT, may indicate increased inhibitory interhemispheric interaction, and TMT measures may therefore be predictive of episodic memory. 2. Methods 2.1. Participants 247 students (63 men) at a mid-sized University participated for either research credit for their Psychology course or for $20.00 remuneration. 14 participants failed to follow instructions (e.g.: fill out all the answers on the mood questionnaire), 3 did not have complete TMT recorded, and 1 demonstrated erratic behavior indicative of possible intoxication; these individuals were excluded from analyses (final N = 229, 54 men). Ages ranged from 18 to 38 years (M = 20.44, SD = 2.70). The research was approved by the Montclair State University IRB and the U.S. Army Human Research Protection Office. Participants provided their written informed consent to participate in the study. Participants were randomly assigned to one of nine Mood Induction Conditions (MIC) made by a combination of happy, angry, or fearful mood as a function of Pre-Encoding and Pre-Recall Memory Conditions. Thus, participants experienced a happy, angry or fearful mood prior to memory encoding followed by a happy, angry, or fearful mood prior to memory recall. See Table 1 for ns as a function of Memory Condition and MIC. 2.2. Materials All stimuli were presented on a 21.5 in. iMac computer monitor using Superlab 4.5. Memory stimuli: 72 words randomly taken from (Tulving, Schacter, & Stark, 1982) were used to create two lists of 36 words (two lists were created for counterbalancing purposes). List words were presented at the rate of 5 s each, in upper case, 28 point, Courier New font. Mood induction stimuli: Moods were induced via the method and stimuli used in Mayer, Allen, and Beauregard (1995). Briefly, participants listened to 5 min of mood-congruent music. Musical selections were those previously validated by Mayer et al. (1995) to elicit happy, fearful, and angry moods. One minute after a musical piece began, participants were presented with eight mood-congruent vignettes (see Mayer et al., 1995) that they were asked to imagine occurring. Vignettes were presented one at a time, for 30 s each. Two musical selections per mood were included for (i) counterbalancing and (ii) to prevent participants in a repetitious condition (e.g. happy prior to encoding and happy prior to recall) from hearing the same musical piece twice. Participants did not wear headphones in order to ensure that TMT was not artificially increased; computer sound was set to volume ‘4’ for all participants. Questionnaires: Participants completed the Brief Mood Introspection Scale (BMIS; Mayer & Gaschke, 1988). This scale consists of 12 adjectives that describe affects varying in valence.

Table 1 Ns per pre-encoding and pre-recall mood induction condition. Pre-encoding

Happy Angry Fearful

Pre-recall Happy

Angry

Fearful

22 26 26

25 25 26

27 27 25

55

Participants, using a likert scale from 1 to 4, self-assess how well each adjective describes their current mood. Participants also completed an Affect Grid wherein they rated their mood on combined arousal and valence dimensions, the Edinburgh Handedness Inventory (EHI; Oldfield, 1971), and the Waterloo Handedness Questionnaire (WHQ). Tympanic membrane temperature assessment: TMT was assessed via a Braun ThermoScan Pro 4000 (WelchAllyn) tympanic membrane thermometer, with probe covers used for each ear, each time TMT was recorded. 2.3. Procedure Participants were tested individually. After reading and signing the consent form, left and right TMT were assessed (in that order) while the participant was seated (Baseline; TMT1). Order of TMT assessment was used in order minimize any potential ‘rebound’ effects on TMT occurring with repeated measurements (e.g.; McCarthy & Heusch, 2006), and replicated previous work (Propper, Christman, et al., 2013; Propper, McGraw, et al., 2013; Propper et al., 2010, 2011). It should also be noted that participants were asked to remove any headgear (e.g.; hats, hoods) upon entering the laboratory. Lab greeting/orientation, consenting the participants, and description of the procedure, resulted in an acclimatization period of approximately 5–7 min prior to initial TMT assessment. Participants completed the BMIS (Baseline; BMIS1) and an Affect Grid (the Affect Grid was included at each Time (Baseline, Pre-Encoding, and Pre-Recall), counterbalanced with BMIS in order of presentation. However, because participants tended to misunderstand instructions for this questionnaire, resulting in incorrect completion, the Affect Grid was excluded from analyses). Participants then engaged in the mood induction paradigm (Mayer et al., 1995), listening to the appropriate musical selection and viewing the mood-congruent vignettes. Immediately following mood induction, TMT was assessed (Pre-encoding; TMT2), and BMIS and Affect Grid completed (Pre-encoding; BMIS2). Participants next viewed the list word stimuli, and were told to ‘study the words’ as they ‘may be tested on them later’. Participants next completed the EHI and the WHQ as filler items (order of presentation counterbalanced). Following questionnaire completion, participants again engaged in the assigned mood induction condition, followed immediately by TMT assessment (Pre-recall; TMT3) and BMIS and Affect Grid completion (Prerecall; BMIS3). Participants were told they had as much time as they wanted to write down on a blank piece of paper as many words as they could remember from the list they saw previously. If a participant had not completed recall after five minutes, they were prompted to complete recall. Participants’ recall times were recorded surreptitiously by the experimenter. At the conclusion of the experiment, all participants were offered the opportunity to view either a funny video (Animal Planet’s Einstein, The Talking Parrot; see http://www.youtube.com/watch?v=7rfGEtALHYs) or to engage in the Happy MIC in order to ensure that their mood was positive when leaving the experimental protocol. 2.4. Analyses Recall: A given word was scored as correct if it contained two or fewer of spelling and was phonetically accurate. Analyses were performed on total number of words written down, hits (number of items correctly recalled), false alarms (number of items written that had not been presented), and corrected scores (hits minus false alarms; Graf & Mandler, 1984). 3 (Pre-Encoding MIC; Happy versus Fearful versus Angry)  3 (Pre-Recall MIC: Happy versus Fearful versus Angry) between-subjects ANOVAs were conducted on each of the four recall measures.

56

R.E. Propper et al. / Brain and Cognition 94 (2015) 52–59

TMT: Effects of Time (Baseline, Pre-Encoding, and Pre-Recall), Memory Condition MIC on TMT were examined via 3 (betweensubjects; Pre-Encoding MIC)  3 (between-subjects; Pre-Recall MIC)  3 (within-subjects; Time: Baseline 1 versus Pre-Encoding 2 versus Pre-Recall 3) mixed ANOVAs on left TMT, right TMT, left minus right TMT (l-rTMT; a measure of relative hemispheric activity within an individual), and absolute value of l-rTMT (abl-rTMT; a measure of relative difference in activity between the hemispheres, regardless of the direction of that difference, e.g. Propper et al., 2010). Recall and TMT: Lastly, correlations between TMT measures and recall measures as a function of Time were conducted.

3. Results Given known differences between right- and non-right handers in memory (e.g.; Prichard, Propper, & Christman, 2013), in TMT measures (e.g. Propper et al., 2010), and in lateralization of emotion (e.g. Hellige, 2001), analyses were conducted to determine whether each Memory Condition (e.g.: Pre-Encoding and PreRecall MIC) was similar in EHI score. A 3 (Pre-Encoding MIC)  3 (Pre-Recall MIC) between-subjects ANOVA revealed no main effects or interaction in EHI as a function of Memory MIC (p > .05 for all comparisons). Therefore, the remaining analyses collapsed across participant handedness. Manipulation Check: Next it was determined whether the mood inductions were successful overall. Happy mood was examined via the combined ‘happy’ and ‘lively’ scores on the BMIS, per Time (e.g.: Baseline versus Pre-Encoding versus Pre-Recall); fearful was ‘nervous’ and ‘jittery’, and angry was ‘fed up’ and ‘grouchy’. For each MIC (i.e.: happy, fearful, and angry), paired t-tests were conducted between BMIS1 and BMIS2 and between BMIS1 and BMIS3. (Note that Baseline BMIS was compared within participants, therefore for the Pre-Encoding and Pre-Recall MICs, there are different Baseline ns). The Pre-Encoding Happy MIC ( x ¼ 7:01, SD = .10, n = 74) successfully altered mood relative to Baseline ( x ¼ 6:03, SD = 1.29, n = 74) in the anticipated direction (t(73) = 9.60, p < .01), as did Pre-Encoding Fear MIC ( x ¼ 4:90, SD = 1.70, n = 77, t(76) = 8.66, p < .01: Baseline  x ¼ 3:39, SD = 1.34), and Pre-Encoding Anger MIC ( x ¼ 5:33, SD = 1.58, n = 78, t(77) = 13.15, p < .01: Baseline  x ¼ 2:80, SD = 1.04). Relative to Baseline ( x ¼ 6:26, SD = 1.17, n = 74), Pre-Recall Happy MIC did not alter mood ( x ¼ 6:38, SD = 1.28, t(73) = .84), p > .05). Pre-Recall Fear did alter mood in the predicted direction ( x ¼ 5:06, SD = 1.58, n = 79, t(78) = 6.94, p < .01: Baseline  x ¼ 3:57, SD = 1.28), as did Pre-Recall Anger ( x ¼ 4:66, SD = 1.61, n = 76, t(75) = 7.63, p < .01: Baseline  x ¼ 23:00, SD = 1.22). Recall: Given that the MIC conditions were successful (with the exception of the Pre-Recall Happy condition; see Section 4), we examined recall as a function of Pre-Encoding and Pre-Recall MIC conditions. There were no main effects or interactions between Pre-Encoding and Pre-Recall MIC on any measure of recall (p > .05 for all comparisons). TMT: Note all measures are in degrees Celsius. A main effect of Time (Baseline versus Pre-Encoding versus Pre-Recall) on Left (F(2, 440) = 25.67, p < .01) and Right TMT (F(2, 440) = 51.48, p < .01) indicated increasing TMT across Time, regardless of PreEncoding or Pre-Recall MIC. There were no main effects of MIC on l-rTMT, however there was an interaction between Time, PreEncoding and Pre-Recall MIC on l-rTMT (F(8, 440) = 2.05, p < 05). Analyses of simple effects indicated this interaction reflected significant differences between l-rTMT1 (Baseline, X = .04, SD = .29) and l-rTMT2 (Pre-Encoding MIC, X = .06, SD = .33, t(25) = 2.23, p < .05) and between l-rTMT1 and l-rTMT3 (Pre-Recall MIC, X = .07, SD = 2.53, t(25) = 2.43, p < .05) in the Fear Pre-Encoding/

Anger Pre-Recall MIC group only, indicating decreased l-rTMT in this group at Times 2 and 3 relative to Baseline (Time 1). There were no main effects or interaction on abl-rTMT as a function of MIC. Recall and TMT: Among the entire sample (N = 229), there were no significant correlations between Left or Right TMT, or l-rTMT, as a function of Time and any measure of recall (p > .05 for all comparisons). There were significant correlations between abl-rTMT1 (Baseline) and number of total words written (r = .22, p < .01), hits (r = .24, p < .01), and corrected scores (r = .23, p < .01). There were also significant correlations between abl-rTMT2 (Pre-Encoding) and number of hits (r = .16, p = .015), corrected scores (r = .19, p < .01), and false alarms (r = .18, p < .01), and between abl-rTMT3 (Pre-Recall) and number of hits (r = .15, p < .05), corrected scores (r = .17, p. < 01), and false alarms (r = .13, p < .05). Because abl-rTMT across the three different Times was significantly positively correlated (abl-rTMT1 and abl-rTMT2 r = .42, p < .01; abl-rTMT1 and abl-rTMT3 r = .30, p < .01; abl-rTMT2 and abl-rTMT3 r = .43, p < .01), and in order to determine whether and how much each abl-rTMT contributed to the recall measures, four multiple regressions were conducted with each abl-rTMT as a function of Time as the independent measures, and each recall measure (total words written, hits, corrected scores, and false alarms) as dependent measures. The regression for number of total items written was significant, accounting for 3.6% of the variance in total words written (R2 = .05, F(3, 225) = 3.86, p = .01), with abl-rTMT1 the only significant contributor to the regression (b = .20, t = 2.80, p < .01). Similarly, the regression for hits was also significant, accounting for 5.0% of the variance (R2 = .06, F(3, 225) = 5.11, p < .01), with ablrTMT1 the only significant variable (b = .19, t = 2.69, p < .01). The regression for corrected scores was also significant, with 5.7% of the variance accounted for (R2 = .07, F(3, 225) = 5.62, p < .01), with abl-rTMT1 again the only significant variable in the regression (b = .17, t = 2.34, p < .05). The regression for false alarms was also significant, accounting for 2.2% of the variance (R2 = .04, F(3, 225) = 2.74), with abl-rTMT2 negatively correlated with false alarms (b = .15, t = 1.98, p < .05). See Fig. 1a–d.

4. Discussion There are three main findings of the present work; (i) despite the fact that participants’ affects were altered as a function of mood induction condition, recall and TMT measures were not influenced by MIC; (ii) left and right TMT increased as a function of Time, regardless of MIC, and; (iii) recall was primarily predicted by the absolute difference between left and right TMT at baseline. First, it is not certain why recall was not influenced by MIC, even though subjectively mood induction was successful, with the exception of Pre-Recall Happy MIC. The Pre-Recall Happy exception may reflect the fact that higher levels of happiness are difficult to induce given the relatively high happiness people maintain generally (e.g.: Lucas, Diener, & Suh, 1996). Other research has demonstrated effects of mood on memory, particularly mood during encoding, such that happiness is associated with increased, and sadness with decreased, false memory (e.g.: Storbeck & Clore, 2005, 2011). Other work has supported the mood congruency effect, whereby positive mood at learning and recall facilitate memory (e.g.: Loeffler, Myrtek, & Peper, 2013). However, research on mood and false memory focuses primarily on semantically related words within the Deese– Roediger–McDermott paradigm (1995). The present work, in contrast, examined recall for semantically unrelated words, which ultimately may decrease the likelihood of false recall generally.

R.E. Propper et al. / Brain and Cognition 94 (2015) 52–59

57

Fig. 1. (a) Correlation between total number of items written and baseline abl-rTMT. (b) Correlation between total number of hits and baseline abl-rTMT. (c) Correlation between corrected score and baseline abl-rTMT. (d) Correlation between number of false alarms and pre-encoding abl-rTMT.

Supporting this suggestion, the overall mean number of words falsely recalled was only .78. (SD = 1.12). Regarding the mood congruency effect, it is important to note that Loeffler et al. (2013) reported increased recall for emotionally valanced words. The words in the present study were not in and of themselves emotionally evocative in a systematic way. Thus, although it is not clear why mood induction here did not produce alterations in recall, one possibility may be that the stimuli used, being un-emotional and without semantic relatedness, involved processes different from those researched previously. Another possibility may be that mood induction was in reality not successful, and that participants responded in the expected manner on the BMIS due to demand characteristics. This possibility is supported by the fact that the mood induction instructions used language encouraging participants to experience the given emotion, and was immediately followed by a questionnaire (the BMIS) rating emotional state, though the lack of mood induction in the Pre-Recall Happy condition contradicts this possibility (i.e.: Why wouldn’t people in this condition also experience demand characteristics?). Still another possibility is that the emotions experienced by participants were not strong enough to alter brain activity in a manner that would ultimately demonstrate more than subtle changes in behavior. Future work could examine these issues by directly assessing brain activity (e.g.: via EEG) during or following mood induction. Second, tympanic membrane temperature was not influenced by mood induction condition, with the exception of the Fear PreEncoding/Anger Pre-Recall MIC group only. It is possible that the decreased l-rTMT from baseline to Times 2 and 3 were spurious in this condition, given that the 3  3  2 ANOVA demonstrated no main effects, and that there are 27 comparisons between the combined Memory Condition MIC groups. The lack of an effect of MIC on TMT measures essentially replicates that of Propper, Christman, et al. (2013) and Propper, McGraw, et al. (2013) who,

although finding significant correlations between mood and TMT measures, nevertheless found no change in TMT from baseline to post-mood induction. Interestingly, and similar to the present work, in the former study right TMT increased significantly from baseline to following mood induction, regardless of mood induction condition. Left TMT did as well, though not significantly so. That left and right TMT increase as a function of Time may reflect an underlying property of TMT measurement; that is, the possibility that the measuring of TMT alters later measurements via a rebound vasoconstriction or vasodilation. Repeated TMT measurements may not reflect actual TMT, but rather properties of the ear canal, vasculature, or other physical aspects of the body (e.g.; McCarthy & Heusch, 2006). Thus, TMT measures subsequent to any baseline may be less likely to be reflective of cortical activity than of vascular or ear canal properties, for example. Third, given i and ii above, the fact that a measure of TMT, ablrTMT at baseline was a significant predictor of recall (total items written, hits, and corrected scores), suggests that this measure may indeed reflect hemispheric activity. Specifically, the results indicate that increased difference between the cerebral hemispheres-regardless of the direction of that difference- is associated with superior episodic memory. Interestingly, previous work implicates a role for corpus calosum-mediated interhemispheric interaction in episodic recall (e.g.: Christman & Propper, 2010; Propper & Christman, 2008). It may be that this hypothesized interhemispheric interaction is inhibitory (Bloom & Hynd, 2005). Such a possibility could result in increased difference in hemispheric activity between the two cerebral hemispheres being associated with increased physiological interhemispheric interaction. Additionally, false recall was negatively associated with abl-rTMT; that is, as the absolute difference in TMT between the left and right side decreased, the number of false recalls increased. This finding supports the notion that decreased differential hemispheric

58

R.E. Propper et al. / Brain and Cognition 94 (2015) 52–59

activity is associated with decreased accuracy in episodic recall, though the fact that this relationship occurred at Pre-Encoding Time 2 deserves further investigation. Ultimately, the present results offer some support for the HERA model, in that greater difference between left and right hemisphere activity was associated with superior episodic memory. Because this relationship occurred at baseline, it might have been anticipated that increased left relative to right hemispheric activity would have been associated with superior episodic memory, due to the proximity of baseline to encoding, and the HERA model proposition that the left hemisphere is particularly involved in episodic encoding. However, stable individual differences in differential hemispheric activity as measured via TMT may obscure specific directional relationships between TMT and memory (e.g.: Propper & Brunyé, 2013). It may be that the result of state changes in TMT overlayed on individual differences in trait TMT would demonstrate that increased difference between the two hemispheres in cortical activity would be associated with increased episodic memory, which is the result reported here. There is an additional factor may have contributed to the lack of findings of inter-relationships between MIC, memory, and TMT. Individual differences in handedness may have contributed to the effects reported here. Specifically, the inclusion of non-right-handers in the present study would be expected to increase variability, and thereby decrease the likelihood of finding significant effects in the relationship between MIC, memory and TMT, because nonright-handers are more variable in their lateralization of cortical functions (Hellige, 2001). Future research could control for participant handedness, and exclusively examine strongly-right-handed individuals, for example. Such a design was not conducted here due to the pragmatics of participant recruitment with a design with such a large number of cells. With regard to the specific predictions proposed in the Introduction: The results here cannot be used to support hypotheses 1–5. However, hypothesis 6, that episodic memory may benefit from interhemispheric interaction, and that such interhemispheric interaction may be inhibitory, has received support. Additionally, results also indicate that this interhemispheric interaction may be reflected in relative differences between left and right TMT, with this measure being predictive of episodic memory. Because this is the first study to demonstrate such a relationship, additional research is warranted. In sum, mood induction did not alter either recall or TMT measures. However, the absolute difference between left and right TMT, a measure of the degree to which the two halves of the brain differ in cortical activity, was significantly related to recall. Future research could examine this relationship in the absence of mood induction, and using non-verbal stimuli. Acknowledgments Part of this work was supported by U.S. Army contract #W911QY-12-C-0046 to author R.E.P. The opinions expressed herein are those of the authors and not necessarily of the US Army. The authors thank Karly H. Hrank and Anthony Molloy for participant recruitment and scheduling, Michael Weiss for study design and computer programming, and Emmy Fonorow, Cara Struble, Jessica Manieri, David Braun, Kyle Dodd, and Connor Jones for data scoring and/or entry. References Babiloni, C., Babiloni, F., Carducci, F., Cappa, S., Cincotti, F., Del Percio, C., et al. (2004). Human cortical EEG rhythms during long-term episodic memory task. A high-resolution EEG study of the HERA model. Neuroimage, 21, 1576–1584. Boyce, W. T., Essex, M. J., Alkon, A., Smider, N. A., Pickrell, T., & Kagan, J. (2002). Temperment, tympanum, and temperature: Four provisional studies of the bio

behavioral correlates of tympanic membrane temperature asymmetries. Child Development, 73, 718–733. Boyce, W. T., Higley, D., Jemerin, J. J., Champoux, M., & Suomi, S. J. (1996). Tympanic temperature asymmetry and stress behavior in Rhesus Macaques and children. Archives of Pediatric and Adolescent Medicine, 150, 518–523. Cabeza, R. (2002). Hemispheric asymmetry reduction in older adults: The HAROLD model. Psychological Aging, 17, 85–100. Carver, C. S., & Harmon-Jones, E. (2009). Anger is an approach-related affect: Evidence and implications. Psychological Bulletin, 135, 183–204. Cherbuin, N., & Brinkman, C. (2004). Cognition is cool: Can hemispheric activation be assessed by tympanic membrane thermometry. Brain and Cognition, 54, 228–231. Christman, S., Garvey, K. J., Propper, R. E., & Phaneuf, K. (2003). Bilateral eye movements enhance the retrieval of episodic memories. Neuropsychology, 17, 221–229. Christman, S. D., & Propper, R. E. (2010). The interhemispheric basis for the retrieval of episodic memories: Effects of handedness and bilateral eye movements. In G. Davies & D. Wright (Eds.), Current issues in applied memory. London: Psychology Press. Christman, S. D., Propper, R. E., & Dion, A. (2004). Increased interhemispheric interaction is associated with decreased false memories in a verbal converging semantic associates paradigm. Brain and Cognition, 56, 313–319. Davidson, R. J. (2002). Anxiety and affective style: Role of prefrontal cortex and amygdala. Biological Psychiatry, 51, 68–80. Davidson, R. J. (2004). Well-being and affective style: Neural substrates and biobehavioral correlates. Philosophical Transactions of the Royal Society B: Biological Sciences, 359, 1395–1411. Flores-Gutiérrez, E. O., Díaz, J.-L., Barrios, F. A., Guevara, A., Del Río-Portilla, Y., CorsiCabrera, M., et al. (2009). Differential alpha coherence hemispheric patterns in men and women during pleasant and unpleasant musical emotions. International Journal of Psychophysiology, 71, 43–49. Ford, B. Q., Tamir, M., Brunye, T. T., Shirer, W. R., Mahoney, C. R., & Taylor, H. A. (2010). Keeping your eyes on the prize: Anger and visual attention to threats and rewards. Psychological Science, 21, 1098–1105. Goldstein, A., Revivo, K., Kreitler, M., & Metuki, N. (2010). Unilateral muscle contractions enhance creative thinking. Psychological Bulletin and Review, 17, 895–899. Graf, P., & Mandler, G. (1984). Activation makes words more accessible, but not necessarily more retrievable. Journal of Verbal Learning and Behavior, 23, 553–568. Griessenberger, H., Hoedlmoser, K., Heib, D. P. J., Lechinger, J., Klimesch, W., & Schabus, M. (2012). Consolidation of temporal order in episodic memories. Biological Psychology, 91(1), 150–155. Gunnar, M. R., & Donzella, B. (2004). Tympanic membrane temperature and emotional dispositions in preschool-aged children: A methodological study. Child Development, 75, 497–504. Habib, R., Nyberg, L., & Tulving, E. (2003). Hemispheric asymmetries of memory: The HERA model revisited. Trends in Cognitive Science, 7, 241–245. Harmon-Jones, E. (2006). Unilateral right-hand contractions cause contralateral alpha power suppression and approach motivational affective experience. Psychophysiology, 43, 598–603. Hellige, J. B. (2001). Hemispheric asymmetry: What’s right and what’s left. Cambridge, MA: Harvard University Press. Helton, W. S. (2010). The relationship between lateral differences in tympanic membrane temperature and behavioral impulsivity. Brain and Cognition, 74, 75–78. Helton, W. S., & Carter, J. R. (2011). The effect of investigator gender on lateral tympanic membrane temperature. Laterality, 16, 156–163. Helton, W. S., Harynen, L., & Schaeffer, D. (2009). Sustained attention to local and global target features is different: Performance and tympanic membrane temperature. Brain and Cognition, 71, 9–13. Helton, W. S., Kern, R. P., & Walker, D. R. (2009). Tympanic membrane temperature, exposure to emotional stimuli and the sustained attention to response task. Journal of Clinical and Experimental Neuropsychology, 31, 611–616. Helton, W. S., & Maginnity, M. (2012). Increased attentiveness is associated with hemispheric asymmetry measured with lateral tympanic membrane temperature in humans and dogs. Experimental Brain Research, 219, 321–326. Homae, F. (2014). A brain of two halves: Insights into interhemispheric organization provided by near-infrared spectroscopy. Neuroimage, 85, 354–362. Lee, A. C., Robbins, T. W., Pickard, J. D., & Owen, A. M. (2000). Asymmetric frontal activation during episodic memory: The effects of stimulus type on encoding and retrieval. Neuropsychologia, 38, 677–692. Loeffler, S. N., Myrtek, M., & Peper, M. (2013). Mood-congruent memory in daily life: Evidence from interactive ambulatory monitoring. Biological Psychology, 93, 308–315. Lucas, R. E., Diener, E., & Suh, E. (1996). Discriminant validity of well-being measures. Journal of Personality and Social Psychology, 71, 616–628. Mayer, J. D., Allen, J. P., & Beauregard, K. (1995). Mood inductions for four specific moods: A procedure employing guided imagery vignettes with music. Journal of Mental Imagery, 19, 133–150. Mayer, J. D., & Gaschke, Y. (1988). The experience and meta-experience of mood. Journal of Personality and Social Psychology, 55, 102–111. McCarthy, P. W., & Heusch, A. I. (2006). The vagaries of ear temperature assessment. Journal of Medical Engineering & Technology, 30, 242–251. Okamoto, M., Wada, Y., Yamaguchi, Y., Kyutoku, Y., Clowney, L., Singh, A. K., et al. (2011). Process-specific prefrontal contributions to episodic encoding and retrieval of tastes: A functional NIRS study. Neuroimage, 54, 1578–1588.

R.E. Propper et al. / Brain and Cognition 94 (2015) 52–59 Oldfield, R. C. (1971). The assessment and analysis of handedness: The Edinburgh handedness inventory. Neuropsychologia, 9, 97–113. Bloom, J. S., & Hynd, G. W. (2005). The role of the corpus callosum in interheispheric transfer: Excitation or Inhibition? Neuropsychology Review, 15, 59–71. Owen, A. M. (2003). HERA today, gone tomorrow? Trends in Cognitive Sciences, 7, 383–384. Palmer, L. M., Schulz, J. M., & Larkum, M. A. (2013). Layer-specific regulation of cortical neurons by interhemispheric inhibition. Communicative and Integrative Biology. http://dx.doi.org/10.4161/cib.23545. Peterson, C. K., Shackman, A. J., & Harmon-Jones, E. (2008). The role of asymmetrical frontal cortical activity in aggression. Psychophysiology, 45, 86–92. Prichard, E., Propper, R. E., & Christman, S. D. (2013). Degree of Handedness, but Not Direction, is a Systematic Predictor of Cognitive Performance. Frontiers in Psychology, 4. http://dx.doi.org/10.3389/fpsyg.2013.00009. Article 9. Propper, R. E., & Brunyé, T. T. (2013). Lateralized differences in tympanic membrane temperature: Emotion and hemispheric activity. Frontiers in Psychology, 4. http://dx.doi.org/10.3389/fpsyg.2013.00104. Propper, R. E., Brunyé, T. T., Christman, S. D., & Bologna, J. (2010). Negative emotional valence is associated with non-right-handedness and increased imbalance of hemispheric activation as measured by tympanic membrane temperature. Journal of Nervous and Mental Disease, 198, 691–694. Propper, R. E., Brunyé, T. T., Christman, D., & Januszewski, A. (2012). Look over there! Sustained unilateral gaze increases geographical memory for the 50 United States. Brain and Cognition, 78, 59–62. Propper, R. E., & Christman, S. D. (2008). Interhemispheric interaction and saccadic horizontal eye movements: Implications for episodic memory, EMDR, and PTSD. Journal of EMDR Practice and Research, 2, 269–281. Propper, R. E., Christman, D., Brunyé, T. T., & Januszewski, A. (2013). Tympanic membrane temperature, hemispheric activity, and affect: Evidence for a modest relationship. Journal of Neuropsychiatry and Clinical Neurosciences. http:// dx.doi.org/10.1176/appi.neuropsych.12020027. Propper, R. E., Januszewski, A., Christman, D., & Brunyé, T. T. (2011). Increased anger is associated with increased hemispheric asymmetry: Support for anger-tympanic membrane relationships. Journal of Nervous and Mental Disease, 199, 716–720. Propper, R. E., McGraw, S. E., Brunyé, T. T., & Weiss, M. (2013). Getting a grip on memory: Unilateral hand clenching alters episodic recall. PLoS One, 8(4), e62474. http://dx.doi.org/10.1371/journal.pone.0062474. Roediger, H. L., & McDermott, K. B. (1995). Creating false memories: Remembering words not presented in lists. Journal of Experimental Psychology: Learning, Memory and Cognition, 21, 803–814. Schiff, B. B., & Lamon, M. (1989). Inducing emotion by unilateral contraction of facial muscles: A new look at hemispheric specialization and the experience of emotion. Neuropsycholgia, 27, 923–935. Schiffer, F., Anderson, C. M., & Teicher, M. H. (1999). Electroencephalogram, bilateral ear temperature, and affect changes induced by lateral visual field stimulation. Comprehensive Psychiatry, 40, 221–225.

59

Schiffer, F., Mottaghy, F. M., Vimal, R. L. P., Renshaw, P. F., Cowan, R., Pascual-Leone, A., et al. (2004). Lateral visual field stimulation reveals extrastriate cortical activation in the contralateral hemisphere: An fMRI study. Psychiatry Research, 131, 1–9. Schmidt, L. A., & Trainor, L. (2001). Frontal brain electrical activity (EEG) distinguishes valence and intensity of musical emotions. Cognition and Emotion, 15, 487–500. Seta, J. J., McCormick, M., Gallagher, P., McElroy, T., & Seta, C. E. (2010). Voice frequency impacts hemispheric processing of attribute frames. Journal of Experimental Social Psychology, 46, 1089–1192. Shobe, R. (2014). Independent and collaborative contributions of the cerebral hemispheres to emotional processing. Frontiers in Human Neuroscience. http:// dx.doi.org/10.3389/fnhum.2014.00230. Spielberg, J. M., Miller, G. A., Engels, A. S., Herrington, J. D., Sutton, B. P., Banich, M. T., et al. (2011). Trait approach and avoidance motivation: Lateraized neural activity associated with executive function. Neuroimage, 54, 661–670. Storbeck, J., & Clore, G. L. (2005). With sadness comes accuracy; with happiness, false memory: Mood and the false memory effect. Psychological Science, 16(785), 791. Storbeck, J., & Clore, G. L. (2011). Affect influences false memories at encoding: Evidence from recognition data. Emotion, 11, 981–989. Tazoe, T., & Perez, M. A. (2013). Speed-dependent contribution of callosal pathways to ipsilateral movements. Journal of Neuroscience, 33, 16178–16188. Tomarken, A. J., Davidson, R. J., Wheeler, R. E., & Doss, R. C. (1992). Individual differences in anterior brain asymmetry and fundamental dimensions of emotion. Journal of Personality and Social Psychology, 62, 676–687. Tsang, C. D., Trainor, L. J., Santesso, D. L., Tasker, S. L., & Schmidt, L. A. (2001). Frontal EEG responses as a function of affective musical feature. Annals of the New York Academy of Sciences, 930, 439–442. Tulving, E., Kapur, S., Craik, F. L., Moscovitch, M., & Houle, S. (1994). Hemispheric encoding/retrieval asymmetry in episodic memory: Positron emission tomography findings. Proceedings of the National Academies of Science, USA, 91, 2016–2020. Tulving, E., Schacter, D. L., & Stark, H. A. (1982). Priming effects in word-fragment completion are independent of recognition memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 8, 336–342. Urry, H. L., Nitschke, J. B., Dolski, I., Jackson, D. C., Dalton, K. M., Mueller, C. J., et al. (2004). Making a life worth living: Neural correlates of well-being. Psychological Science, 15, 367–372. Zinner, L. R., Brodish, A. B., Devine, P. G., & Harmon-Jones, E. (2008). Anger and asymmetrical frontal cortical activity: Evidence for an anger-withdrawal relationship. Cognition and Emotion, 22(6), 1081–1093.