- Email: [email protected]
Relationship between cognitive activity involved in emotional processing and personality
244
International Journal of Psychophysiology 94 (2014) 120–261
Results: Results showed that the depression group, in comparison with the control group, recognized facial expressions were as accurate as controls in recognizing emotions but not happiness (t(57) = 2.44, p b .05). Furthermore, dwell time on happy expressions was significantly longer in the depression group compared to the control group (F(4, 228) = 4.585, p b .05) while dwell time on other expressions was comparable in both groups. Discussion & conclusions: Deficits in emotion recognition, specifically of happy expressions, may contribute to an overall negative mood state among depressed individuals (Beck et al., 1979). Individuals with MDD seem to spend more time to see to happy stimuli. On the other hand, depressed subjects might not need to be paid greater attention to negative stimuli (i.e., sad, disgust, and angry) since these stimuli were more congruent with negative mood that is prevailing emotion in depression. doi:10.1016/j.ijpsycho.2014.08.935
Trustworthiness modulates facial responses to emotional faces Tomomi Fujimuraa, Kazuo Okanoyab National Institute of Advanced Industrial Science and Technology, Japan b The University of Tokyo, Japan a
People often produce facial movements congruent with the emotional face of the person being observed. These spontaneous and automatic facial movements have been called as facial mimicry. Recent studies have revealed that the occurrence of facial mimicry is considered to depend on social context like the relationship between expresser and observer. This study was aimed to investigate how social context modulated facial mimicry by measuring facial electromyogram (EMG) in response emotional faces. We focused on trustworthiness as social context because trust is an important precursor in the development of cooperation. To manipulate trustworthiness of expresser, participants engaged in the trust game with 4 female partners. In the game, the two players sent point back and forth. Each round began with a neutral face of the partner, and then the participant decided which points she wanted to investigate with the partner, 2 points or 4 points. The investigated points to the partner were tripled, then the partner split it between herself and the participant. There were three types of partner's decision; fair return, just return, and unfair return. The participants played 10 rounds randomly with each of 4 partners who behaved fairly or unfairly in the game. A fair partner returned fairly in 8 rounds and just returned in 2 rounds while an unfair trustee returned unfairly in 8 rounds and just returned in 2 rounds. Before and after the game, the participants viewed emotional faces (happy, sad, and angry faces) displayed by the partners and evaluated trustworthiness of a neutral face of the partner on a 7-point rating scale. Trustworthiness of the partner was successfully manipulated by the trust game. The results showed that corrugator activities in response to sad faces displayed by the untrustworthy partner were more inhibited than those displayed by the trustworthy partner. On the other hand, trustworthiness had no effect on zygomatic activities in response to happy faces. These results suggest that the impact of social context on facial mimicry depends on the nature of emotion. doi:10.1016/j.ijpsycho.2014.08.936
Verification of the three dimensional model of emotions with spectral analysis of independent EEG components Miroslaw Wyczesany, Tomasz S. Ligeza Jagiellonian University, Poland
The question about which specific parts of the human brain “produces” emotions is a fundamental one for human neuroscience. So far, the locationist paradigm, which propose specialized brain centres for each discrete emotion has not found enough support. An alternative, constructionist view suggests that neural correlates of emotions should be considered from a higher-level perspective. It assumes that elicitation of emotions involves the interaction of wider neural networks which mediate more general functions, not specific to a single, discrete emotion. Here, we apply a three-dimensional model (valence, arousal and dominance — VAD) and explore whether it can reflect the neural basis of subjectively experienced affect in a more clear and understandable way than the locationist approach. We assume that an emotional space consisting of few basic factors can better reflect the way neural processes constitute affective states. Fifty-eight volunteers participated in the study, in which spontaneous brain activity was recorded using 64-channel EEG. After preprocessing, the recording was subject to Independent Component Analysis for blind separation of independent sources of activity. Next, their equivalent dipoles were localized using the DIPFIT procedure and then clustered across subjects. The average spectral power in the beta range was calculated for each cluster. The emotional state of participants was measured using the PANAS-X questionnaire. The resulting eleven discrete scales were transformed into the three-dimensional space using data from another 168 subjects, who ranked all of the PANAS-X items on the VAD dimensions. Finally, stepwise regression analyses were performed to determine the self-report VAD factors, which significantly predicted activity of the brain clusters. Four of ten clusters revealed consistent associations between their activity and emotional self-report expressed in the VAD space. The activity of the anterior cingulate cluster was associated with emotions of negative valence and low arousal. The cluster located in the middle cingulate cortex was related to low dominance emotions, irrespective of the valence. The anterior orbitofrontal cluster turned out to be associated with negative emotions of high arousal. Finally, the left temporal cluster was related to negative emotions of both low and high arousal. The considered results support the dimensional approach to emotions and show that the VAD space is reflected in patterns of brain activity. doi:10.1016/j.ijpsycho.2014.08.937
Relationship between cognitive activity involved in emotional processing and personality Midori Motoia,b, Yuka Egashiraa, Takayuki Nishimurac, Damee Choia, Matsumoto Rikoa, Shigeki Watanukid a Graduate School of Integrated Frontier Sciences, Kyushu University, Japan b Japan Society for the Promotion of Sciences, Japan c Graduate School of Biomedical Sciences, Nagasaki University, Japan d Faculty of Design, Kyushu University, Japan Background: Previous studies have shown that the processing of unpleasant stimuli occurs early (0–300 ms); however, it is not clear how cognitive processing related to pleasant/unpleasant emotions occur at later time windows (≥300 ms). On the other hand, as evident from the previous reports, the Behavioral Inhibition System (BIS)/Behavioral Activation System (BAS) personality traits are strongly associated with unpleasant and pleasant responses, respectively. Therefore, in the present study, we aim to identify the time window involved in human pleasant/unpleasant emotional processing by investigating event-related potentials (ERP) components correlated with BIS/BAS personality traits.
International Journal of Psychophysiology 94 (2014) 120–261
Methods: Twenty-nine men without hearing impairments (21– 24 years old) took part in the study. In each session, three types of car door closing sounds were randomly presented 60 times each at 1.5–2 second intervals. Three sessions were conducted with 5minute breaks in between. Brainwaves were collected with a 64channel net. For the remainder of the trials, stimulus presentation was set as 0 ms, and an arithmetic mean of the − 100 to 800 ms range was taken to obtain the ERP waveform. The number of additions was set as ≥100. The mean of the potentials of the electrodes in close proximity was taken, with the frontal region as the region of interest. With 80–160 ms as N1 and 450–800 ms as LPP, the mean value of these intervals was set as the amplitude for each ERP component. BIS/BAS score was calculated using the Japanese edition of BIS/BAS questionnaire. Results: A significant and positive correlation was observed between N100 amplitude and BIS score (r = .33, p b 0.05). A positive correlation was found between BAS fun seeking subscale score and LPP amplitude (r = .40, p b 0.05). Discussion: Our results suggest that the processing of unpleasant emotions takes place early on, since N100 response was larger in high BIS subjects who are known to be sensitive to unpleasant emotions. Our findings did not contradict previous study results. LPP was larger in high BAS subjects who are known to be sensitive to pleasant emotions. The LPP was considered to be augmented because of the activity level in the ACC. Conclusions: Our results suggest that the processing of unpleasant emotions takes place early on. Additionally, the association of pleasant emotions was suggested as a late component response. doi:10.1016/j.ijpsycho.2014.08.938
Investigating the reliability and validity of candidate psychophysiological endophenotypes of affective processing David A. Camfielda, Rodney Crofta, Keith Wesnesb, Sarah Boyallc, Monique Taylorc, Rachael Tascac, Emma Kornfieldc a School of Psychology, Illawarra Health & Medical Research Institute, University of Wollongong, Australia b Wesnes Cognition, Swinburne University of Technology, Northumbria University, United Kingdom c School of Psychology, University of Wollongong, Australia Background: A number of studies have been conducted in recent years documenting electrophysiological and autonomic correlates of affective processing in healthy, as well as clinical, populations. Of particular interest to researchers in this area is the development of endophenotypes/biomarkers that may be used as objective metrics for the diagnosis and selection of treatment alternatives in mood disorders. Methods: Five computerized experimental tasks are to be assessed for their reliability and validity in eliciting psychophysiological measures in a healthy adult population (combined N = 120): (i) An Emotional Sternberg task using positive, negative and neutrally valenced word triplets will be used to investigate the affect of emotionally distracting material on event-related synchronization (ERS) during working memory performance, (ii) loudness dependence of the Auditory Evoked Potential will be investigated in relation to individual differences on the Positive-Negative Affect schedule (PANAS), (iii) a visual Object Pattern Separation test will be used to investigate differences in ERP amplitudes associated with memory retrieval for pictures which are highly similar (‘lure’ images) in comparison to previously presented images. (iv) A differential fear conditioning paradigm using neutral faces paired with aversive images from the International Affective Picture System
245
(IAPS) and startle probes will be used to investigate the affect of fear conditioning on early face-related ERP components, including the N170. (v) Using a battery of 300 IAPS images, in combination with intermittent startle probes, individual differences in skin conductance response (SCR), zygomator/corrugator EMG, and startle P3 amplitude will be investigated in relation to positive and negativelyvalenced highly arousing imagery. Results/conclusions: Data collection was commenced in April 2014, with preliminary data available by August 2014. On the basis of these results in a healthy population, the most reliable endophenotypes will be subsequently selected for investigating deficits in affective processing in patients with Major Depressive Disorder (MDD). doi:10.1016/j.ijpsycho.2014.08.939
Structural equation modelling to assess relationships between event-related potential components, heart rate and skin conductance in the context of emotional stimuli Susan J. Thomasa, Peter R.C. Leesonb, Craig J. Gonsalvezc, Stuart J. Johnstoned a Graduate School of Medicine and Brain & Behaviour Research Institute, University of Wollongong, Australia b School of Psychology, University of Wollongong, Australia c School of Social Sciences and Psychology, University of Western Sydney, Australia d Centre for Psychophysics and Psychopharmacology, and School of Psychology, University of Wollongong, Australia Background: Emotions are associated with central (CNS) and autonomic (ANS) nervous system arousal, which can be indexed through psychophysiological changes in event-related potentials (ERPs), heart rate and skin conductance to emotional compared to neutral stimuli. Relationships between these variables are complex and inadequately understood. ERP changes to emotional stimuli, for example, may reflect the activity of cortical regulation of limbic emotional areas or, conversely, increased attention to emotionally salient stimuli. The current study aimed to provide a stronger assessment of the relationship among these variables than is provided by analysing multiple levels of data separately or using traditional correlational approaches. Method: We employed structural equation modelling to assess pathways between event-related potential component amplitudes and latencies, heart rate and skin conductance, and to statistically assess their relationship to a latent variable of emotional arousal. Twenty healthy volunteers responded to 500 trials of randomly mixed threat and neutral words with simple task requirements performed using a button press device. The psychophysiological data (ERP component amplitudes and latencies, heart rate and skin conductance) to threat and neutral stimuli were included in the analysis. An initial correlational analysis was performed, followed by structural equation modelling using AMOS software (Version 19) in order to clarify interactions among the variables. Fit indices for uniand multi-dimensional models of emotional arousal were compared. Results: We present a schematic of the best-fitting parsimonious model to account for pathways between the variables. The model outlines both relationships and fractionation between these ANS and CNS psychophysiological measures and the hypothesised underlying concept of emotional arousal. Discussion: The results suggest that the structural equation modelling and latent variable analysis is a useful approach to better understanding complex relationships amongst these different levels of CNS and ANS psychophysiological measurement. Conclusion: Structural equation modelling helped to clarify relationships between multiple psychophysiological measures of ANS
International Journal of Psychophysiology 94 (2014) 120–261
Results: Results showed that the depression group, in comparison with the control group, recognized facial expressions were as accurate as controls in recognizing emotions but not happiness (t(57) = 2.44, p b .05). Furthermore, dwell time on happy expressions was significantly longer in the depression group compared to the control group (F(4, 228) = 4.585, p b .05) while dwell time on other expressions was comparable in both groups. Discussion & conclusions: Deficits in emotion recognition, specifically of happy expressions, may contribute to an overall negative mood state among depressed individuals (Beck et al., 1979). Individuals with MDD seem to spend more time to see to happy stimuli. On the other hand, depressed subjects might not need to be paid greater attention to negative stimuli (i.e., sad, disgust, and angry) since these stimuli were more congruent with negative mood that is prevailing emotion in depression. doi:10.1016/j.ijpsycho.2014.08.935
Trustworthiness modulates facial responses to emotional faces Tomomi Fujimuraa, Kazuo Okanoyab National Institute of Advanced Industrial Science and Technology, Japan b The University of Tokyo, Japan a
People often produce facial movements congruent with the emotional face of the person being observed. These spontaneous and automatic facial movements have been called as facial mimicry. Recent studies have revealed that the occurrence of facial mimicry is considered to depend on social context like the relationship between expresser and observer. This study was aimed to investigate how social context modulated facial mimicry by measuring facial electromyogram (EMG) in response emotional faces. We focused on trustworthiness as social context because trust is an important precursor in the development of cooperation. To manipulate trustworthiness of expresser, participants engaged in the trust game with 4 female partners. In the game, the two players sent point back and forth. Each round began with a neutral face of the partner, and then the participant decided which points she wanted to investigate with the partner, 2 points or 4 points. The investigated points to the partner were tripled, then the partner split it between herself and the participant. There were three types of partner's decision; fair return, just return, and unfair return. The participants played 10 rounds randomly with each of 4 partners who behaved fairly or unfairly in the game. A fair partner returned fairly in 8 rounds and just returned in 2 rounds while an unfair trustee returned unfairly in 8 rounds and just returned in 2 rounds. Before and after the game, the participants viewed emotional faces (happy, sad, and angry faces) displayed by the partners and evaluated trustworthiness of a neutral face of the partner on a 7-point rating scale. Trustworthiness of the partner was successfully manipulated by the trust game. The results showed that corrugator activities in response to sad faces displayed by the untrustworthy partner were more inhibited than those displayed by the trustworthy partner. On the other hand, trustworthiness had no effect on zygomatic activities in response to happy faces. These results suggest that the impact of social context on facial mimicry depends on the nature of emotion. doi:10.1016/j.ijpsycho.2014.08.936
Verification of the three dimensional model of emotions with spectral analysis of independent EEG components Miroslaw Wyczesany, Tomasz S. Ligeza Jagiellonian University, Poland
The question about which specific parts of the human brain “produces” emotions is a fundamental one for human neuroscience. So far, the locationist paradigm, which propose specialized brain centres for each discrete emotion has not found enough support. An alternative, constructionist view suggests that neural correlates of emotions should be considered from a higher-level perspective. It assumes that elicitation of emotions involves the interaction of wider neural networks which mediate more general functions, not specific to a single, discrete emotion. Here, we apply a three-dimensional model (valence, arousal and dominance — VAD) and explore whether it can reflect the neural basis of subjectively experienced affect in a more clear and understandable way than the locationist approach. We assume that an emotional space consisting of few basic factors can better reflect the way neural processes constitute affective states. Fifty-eight volunteers participated in the study, in which spontaneous brain activity was recorded using 64-channel EEG. After preprocessing, the recording was subject to Independent Component Analysis for blind separation of independent sources of activity. Next, their equivalent dipoles were localized using the DIPFIT procedure and then clustered across subjects. The average spectral power in the beta range was calculated for each cluster. The emotional state of participants was measured using the PANAS-X questionnaire. The resulting eleven discrete scales were transformed into the three-dimensional space using data from another 168 subjects, who ranked all of the PANAS-X items on the VAD dimensions. Finally, stepwise regression analyses were performed to determine the self-report VAD factors, which significantly predicted activity of the brain clusters. Four of ten clusters revealed consistent associations between their activity and emotional self-report expressed in the VAD space. The activity of the anterior cingulate cluster was associated with emotions of negative valence and low arousal. The cluster located in the middle cingulate cortex was related to low dominance emotions, irrespective of the valence. The anterior orbitofrontal cluster turned out to be associated with negative emotions of high arousal. Finally, the left temporal cluster was related to negative emotions of both low and high arousal. The considered results support the dimensional approach to emotions and show that the VAD space is reflected in patterns of brain activity. doi:10.1016/j.ijpsycho.2014.08.937
Relationship between cognitive activity involved in emotional processing and personality Midori Motoia,b, Yuka Egashiraa, Takayuki Nishimurac, Damee Choia, Matsumoto Rikoa, Shigeki Watanukid a Graduate School of Integrated Frontier Sciences, Kyushu University, Japan b Japan Society for the Promotion of Sciences, Japan c Graduate School of Biomedical Sciences, Nagasaki University, Japan d Faculty of Design, Kyushu University, Japan Background: Previous studies have shown that the processing of unpleasant stimuli occurs early (0–300 ms); however, it is not clear how cognitive processing related to pleasant/unpleasant emotions occur at later time windows (≥300 ms). On the other hand, as evident from the previous reports, the Behavioral Inhibition System (BIS)/Behavioral Activation System (BAS) personality traits are strongly associated with unpleasant and pleasant responses, respectively. Therefore, in the present study, we aim to identify the time window involved in human pleasant/unpleasant emotional processing by investigating event-related potentials (ERP) components correlated with BIS/BAS personality traits.
International Journal of Psychophysiology 94 (2014) 120–261
Methods: Twenty-nine men without hearing impairments (21– 24 years old) took part in the study. In each session, three types of car door closing sounds were randomly presented 60 times each at 1.5–2 second intervals. Three sessions were conducted with 5minute breaks in between. Brainwaves were collected with a 64channel net. For the remainder of the trials, stimulus presentation was set as 0 ms, and an arithmetic mean of the − 100 to 800 ms range was taken to obtain the ERP waveform. The number of additions was set as ≥100. The mean of the potentials of the electrodes in close proximity was taken, with the frontal region as the region of interest. With 80–160 ms as N1 and 450–800 ms as LPP, the mean value of these intervals was set as the amplitude for each ERP component. BIS/BAS score was calculated using the Japanese edition of BIS/BAS questionnaire. Results: A significant and positive correlation was observed between N100 amplitude and BIS score (r = .33, p b 0.05). A positive correlation was found between BAS fun seeking subscale score and LPP amplitude (r = .40, p b 0.05). Discussion: Our results suggest that the processing of unpleasant emotions takes place early on, since N100 response was larger in high BIS subjects who are known to be sensitive to unpleasant emotions. Our findings did not contradict previous study results. LPP was larger in high BAS subjects who are known to be sensitive to pleasant emotions. The LPP was considered to be augmented because of the activity level in the ACC. Conclusions: Our results suggest that the processing of unpleasant emotions takes place early on. Additionally, the association of pleasant emotions was suggested as a late component response. doi:10.1016/j.ijpsycho.2014.08.938
Investigating the reliability and validity of candidate psychophysiological endophenotypes of affective processing David A. Camfielda, Rodney Crofta, Keith Wesnesb, Sarah Boyallc, Monique Taylorc, Rachael Tascac, Emma Kornfieldc a School of Psychology, Illawarra Health & Medical Research Institute, University of Wollongong, Australia b Wesnes Cognition, Swinburne University of Technology, Northumbria University, United Kingdom c School of Psychology, University of Wollongong, Australia Background: A number of studies have been conducted in recent years documenting electrophysiological and autonomic correlates of affective processing in healthy, as well as clinical, populations. Of particular interest to researchers in this area is the development of endophenotypes/biomarkers that may be used as objective metrics for the diagnosis and selection of treatment alternatives in mood disorders. Methods: Five computerized experimental tasks are to be assessed for their reliability and validity in eliciting psychophysiological measures in a healthy adult population (combined N = 120): (i) An Emotional Sternberg task using positive, negative and neutrally valenced word triplets will be used to investigate the affect of emotionally distracting material on event-related synchronization (ERS) during working memory performance, (ii) loudness dependence of the Auditory Evoked Potential will be investigated in relation to individual differences on the Positive-Negative Affect schedule (PANAS), (iii) a visual Object Pattern Separation test will be used to investigate differences in ERP amplitudes associated with memory retrieval for pictures which are highly similar (‘lure’ images) in comparison to previously presented images. (iv) A differential fear conditioning paradigm using neutral faces paired with aversive images from the International Affective Picture System
245
(IAPS) and startle probes will be used to investigate the affect of fear conditioning on early face-related ERP components, including the N170. (v) Using a battery of 300 IAPS images, in combination with intermittent startle probes, individual differences in skin conductance response (SCR), zygomator/corrugator EMG, and startle P3 amplitude will be investigated in relation to positive and negativelyvalenced highly arousing imagery. Results/conclusions: Data collection was commenced in April 2014, with preliminary data available by August 2014. On the basis of these results in a healthy population, the most reliable endophenotypes will be subsequently selected for investigating deficits in affective processing in patients with Major Depressive Disorder (MDD). doi:10.1016/j.ijpsycho.2014.08.939
Structural equation modelling to assess relationships between event-related potential components, heart rate and skin conductance in the context of emotional stimuli Susan J. Thomasa, Peter R.C. Leesonb, Craig J. Gonsalvezc, Stuart J. Johnstoned a Graduate School of Medicine and Brain & Behaviour Research Institute, University of Wollongong, Australia b School of Psychology, University of Wollongong, Australia c School of Social Sciences and Psychology, University of Western Sydney, Australia d Centre for Psychophysics and Psychopharmacology, and School of Psychology, University of Wollongong, Australia Background: Emotions are associated with central (CNS) and autonomic (ANS) nervous system arousal, which can be indexed through psychophysiological changes in event-related potentials (ERPs), heart rate and skin conductance to emotional compared to neutral stimuli. Relationships between these variables are complex and inadequately understood. ERP changes to emotional stimuli, for example, may reflect the activity of cortical regulation of limbic emotional areas or, conversely, increased attention to emotionally salient stimuli. The current study aimed to provide a stronger assessment of the relationship among these variables than is provided by analysing multiple levels of data separately or using traditional correlational approaches. Method: We employed structural equation modelling to assess pathways between event-related potential component amplitudes and latencies, heart rate and skin conductance, and to statistically assess their relationship to a latent variable of emotional arousal. Twenty healthy volunteers responded to 500 trials of randomly mixed threat and neutral words with simple task requirements performed using a button press device. The psychophysiological data (ERP component amplitudes and latencies, heart rate and skin conductance) to threat and neutral stimuli were included in the analysis. An initial correlational analysis was performed, followed by structural equation modelling using AMOS software (Version 19) in order to clarify interactions among the variables. Fit indices for uniand multi-dimensional models of emotional arousal were compared. Results: We present a schematic of the best-fitting parsimonious model to account for pathways between the variables. The model outlines both relationships and fractionation between these ANS and CNS psychophysiological measures and the hypothesised underlying concept of emotional arousal. Discussion: The results suggest that the structural equation modelling and latent variable analysis is a useful approach to better understanding complex relationships amongst these different levels of CNS and ANS psychophysiological measurement. Conclusion: Structural equation modelling helped to clarify relationships between multiple psychophysiological measures of ANS