Research article
Yüklə 0,9 Mb. Pdf görüntüsü
|
|
[ 63
69 ]. Facial Mimicry in Parkinson's Disease PLOS ONE | DOI:10.1371/journal.pone.0160329 July 28, 2016 11 / 20
We also evidenced significant differences regarding these variations between the HC and the PD patients. The responses of the zygomaticus and the orbicularis muscles did not vary with the emotion among the PD patients, as the activity of these muscles did not increase in response to expressions of joy. In addition, the corrugator relaxation in response to happy faces tended to be less marked among the PD patients than in the HC. In response to angry faces, we still noticed an increased activity of the corrugator muscle in the PD patients but it was less marked than that seen in the HC. Thus, PD seems to impact facial mimicry in different man- ners, with a relatively preserved facial mimicry of angry faces but a considerable disruption of facial mimicry of happy faces. As this could result in an imbalance in favour of the expression of —or reaction to—negative emotions, this phenomenon could contribute to the fact that peo- ple suffering from PD are often described by others (including health professionals) as with- drawn, bored or passive, moody, anxious, unhappy or suspicious [ 70 –
]. It is important to note here that the significantly weaker facial reactions to emotions observed among the PD patients could arise from higher tonic muscle activations at baseline. Indeed, analyses on mean EMG amplitudes measured during the last second before stimulus onset showed that zygomati- cus muscle exhibited higher activity at baseline in the PD patients compared to the HC. How- ever, no group difference emerged for the corrugator and the orbicularis. As well, we only recorded muscle activity on the left side on the participants ’ face and we could wonder whether the laterality of the motor symptoms could play a role for the diminished EMG activity in response to emotion highlighted among the 17 patients with left-side predominant motor symptoms. Nevertheless, analyses focusing on the relationship between clinical characteristics and facial reactions did not highlight any significant effect of the disease laterality. In this study, the relative preservation of facial mimicry in response to expressions of anger in PD patients could participate in their abilities to recognize anger as accurately as the healthy participants. This would fit with the assumption of embodied simulation theory asserting that mimicry fosters emotion recognition. According to this assumption, the patients ’ performances in recognizing joy were expected to collapse because their ability to mimic happy faces was almost inexistent. However, our results did not support this expectation since the patients' decoding accuracy scores for joy remained relatively high despite the negative impact of PD in recognizing joy expressions. The relationship between facial reactions and joy decoding accu- racy shown here could provide some elements for discussion. Indeed, our results suggest that corrugator relaxation in response to expressions of joy after 500 ms from stimulus onset fosters the emotion recognition process among both the HC and the PD patients. Among the HC, the information from proprioceptive feedback induced by zygomaticus contraction in the first 500 ms of the perception of the expression also might contribute to the joy recognition. After these first 500 ms, although they still increased, zygomatic contractions did not further boost accu- racy in recognizing joy in the HC. This suggests that among the HC, it could be first the infor- mation from the early reactions of the zygomaticus muscle and then, the feedback from corrugator relaxation —requiring a longer time frame—that contribute to the recognition of joy. Among the PD patients, the information coming from the zygomaticus muscle seemed to not foster the joy recognition anymore, but —even if corrugator relaxation tended to be weaker than normal —the feedback from corrugator activity might have been still efficient in support- ing the recognition of joy. Thus, the findings about joy recognition and joy mimicry are still in favour of the embodiment simulation theory. Finally, we cannot exclude that the motor com- mand related to the mimicry phenomenon might have also an impact on recognition accuracy. Actually, it has been shown that transcranial magnetic stimulation (TMS) above the somato- sensory cortices (S1) and motor region (M1) had an impact on mimicry but only TMS on M1 had a behavioural impact on smile detection [ 73 ].
PLOS ONE | DOI:10.1371/journal.pone.0160329 July 28, 2016 12 / 20
The relationship between facial mimicry and emotion recognition observed in this study fits previous findings reporting a positive effect of facial mimicry on recognition of emotions. However, studies investigating the role of facial mimicry in this process have shown mixed results. Some have found that facial mimicry could be considered as a functional element among emotion-related abilities enabling us to infer the emotional state of our interlocutor. This is the case with studies reporting impairment (or improvement) of emotion recognition when facial mimicry is blocked (or intensified) [ 9 , 11 , 74 ] as well as among people suffering from the locked-in syndrome which leads to a paralysis of facial movements [ 75 ]. Moreover, in the study by Sato and colleagues [ 76 ], facial EMG activity predicted the recognition of emo- tional valence through its influence on the experience of emotional valence in response to dynamic facial expressions and Korb et al. [ 77 ] showed that facial mimicry predicted authentic- ity judgments of smiles. In the same way, Künecke and collaborators [ 78 ] evidenced a correla- tion between corrugator responses to angry, happy and sad faces and accuracy of the perception of these emotions. Conversely, other authors have suggested that facial mimicry is neither necessary nor linked to the process of recognizing emotion. The study by Bogart and Matsumoto [ 12 ] among people with Moebius syndrome is in line with this view. Likewise, Hess and Blairy [ 16 ] could not con- firm any relationship between mimicry and emotion recognition or emotion contagion while Blairy et al. [ 15 ] showed that neither spontaneous nor voluntary mimicry increased accuracy in decoding emotions. They did not find a negative impact of "blocking" mimicry —whereby par- ticipants were required to show incompatible facial expressions —on decoding accuracy either. These discrepancies could result from methodological differences including the methods for measuring mimicry (facial EMG vs. Ekman ’s Facial Action Coding System) and emotion rec- ognition (categorical accuracy scores vs. ratings of emotional valence, single task vs. multiple tasks approaches), the characteristics of the stimuli (static vs. dynamic, prototypical vs. more ambiguous) and the analyses conducted (correlations vs. path or mediational analyses). This also underlines the importance of dynamic features in relation to facial expressions (stimuli) as well as the importance of taking into account the dynamic aspect of facial reactions (mimicry) in analyses. Psychological and physiological evidences suggest that facial emotions are per- ceived and mimicked differently when the stimuli are dynamic as opposed to static expressions. Indeed, using static expressions not only affects ecological validity but also limits our under- standing of the role of facial mimicry [ 60 ].
—as well as on facial reactions —could conceal possible role of facial feedback in this process. Further investi- gations assessing facial mimicry among unmedicated PD patients could clarify this point. Furthermore, we need to interpret these findings carefully given that compensatory strate- gies could be used by people suffering from a long-lasting motor impairment and not only in temporary experimental manipulations of muscle activity. Indeed, fMRI studies have shown compensatory cortical mechanisms among PD patients [ 79 ] and in Parkin mutation carriers showing a stronger than normal activity in the ventrolateral premotor cortex (part of the mir- ror neurons system) during the execution and the perception of affective facial gestures as well as a slightly reduced ability to recognize facial emotions [ 80 ]. To conclude, in their recent review, Hess and Fischer [ 81 ] claimed that facial mimicry is not necessary to decode emotions but could facilitate the speed of the process [ 10 , 82 ] or the recog- nition of emotion when the task is difficult. They further reported that facial mimicry is sensi- tive to the emotional and social context such as the emotional meaning of the facial display, the identity of the sender or the relationship between the observer and the sender. Thus, they sug- gested that mimicry could occur when it reinforces social bonds, enhances social coordination and improves the quality of social interactions. Therefore, in the same way as facial amimia Facial Mimicry in Parkinson's Disease PLOS ONE | DOI:10.1371/journal.pone.0160329 July 28, 2016 13 / 20
could lead to inaccurate impressions and reduce the desire for social interaction [ 70 ], we can wonder whether the reduction in facial expression of emotion and facial mimicry observed in PD could in turn disturb the way others interpret the emotions of patients and affect the quality of their interactions in real social contexts. Conclusions To sum up, this is the first study to focus on facial mimicry in PD using EMG recordings in a facial emotion recognition paradigm. Using analyses of the temporal aspects of facial EMG reactions in response to dynamic avatars, we highlighted disturbances in facial mimicry among PD patients. In addition, regarding the beneficial effect of mimicry on emotion decoding accu- racy evidenced here, reduced facial mimicry could be a new explanatory factor with regard to emotional disturbances associated with PD, notably regarding to the already known deficits in facial expression decoding in PD, once again confirmed in our study. Finally, we provide addi- tional arguments in favour of embodied simulation theory asserting that mimicry could foster the recognition of emotion. Supporting Information S1 Appendix. Detailed description of the procedure and the stimulus material. (DOC)
S2 Appendix. Inter-muscle comparisons. (DOC)
S3 Appendix. Effects of clinical characteristics of the patients on facial reactions to emotion ( α = 0.05). (DOC) S4 Appendix. Effect of facial reactions on emotion decoding accuracy ( α = 0.05). (DOC)
S1 Fig. EMG data management. For each trial, the last second before stimulus onset was con- sidered as baseline. Then, to examine the temporal profiles of facial reactions to emotions, the EMG amplitudes were averaged on sequential 100 ms intervals (x 20) of stimulus exposure (top panel A) and expressed as a relative percentage of the mean amplitude from baseline (bot- tom panel A). To examine the impact of medication therapy and disease severity (disease duration, LEDD, Hoehn and Yahr stages and UPDRS III scores both ON and OFF DRT) on EMG responses and to assess the relationship between emotion recognition and facial reac- tions, facial EMG responses were calculated as previously on sequential 500 ms periods of stimulus exposure. Four periods were thus considered: 0 –500; 500–1000; 1000–1500 and 1500
–2000 ms (B). (TIF)
S1 File. Sociodemographic, neuropsychological and clinical characteristics of the partici- pants: dataset. Group = healthy controls (HC) and PD patients (PD); Subject = subject number; Sex = participant ’s gender (W = woman and M = man); Age = participant’s age at inclusion; STAI_state = state anxiety score on the State-Trait Anxiety Inventory; STAI_trait = trait anxiety score on the State-Trait Anxiety Inventory; Matrix = standard note on the Matrix Reasoning subtest from the Wechsler Adult Intelligence Scale; Benton = standardized score on the on the Benton unfamiliar-face matching test; VOSP_screeningT = score on the shape detection screen- ing subtest from the Visual and Object Space Perception battery; VOSP_posdiscriT = score on the position discrimination VOSP subtest; VOSP_nblocT = score on the number location VOSP Facial Mimicry in Parkinson's Disease PLOS ONE | DOI:10.1371/journal.pone.0160329 July 28, 2016 14 / 20
subtest; PD_duration = year of diagnosis; PD_laterality = worst affected side (R = right and L = left); UPDRS3_ON and UPDRS3_OFF = scores on the Unified Parkinson ’s Disease Rating Scale motor part under dopamine replacement therapy (ON) and during a temporary with- drawal from treatment (OFF); HY_ON and HY_OFF = stages on the Hoehn and Yahr disability scale under dopamine replacement therapy (ON) and during a temporary withdrawal from treatment (OFF); LARS = score on the Lille Apathy Rating Scale; LEDD = levodopa-equivalent daily dose (mg/day). (TXT) S2 File. Performances on the facial emotion recognition task: dataset. Group = healthy con- trols (HC) and PD patients (PD); Subject = subject number; Trial = trial number (e01 —e36); Emotion = emotion displayed (Angry, Happy, Neutral); Avatar = identification code of the avatar; Decoding_Accuracy = accurately identified expressions were coded as 1 and misidenti- fied expressions were coded as 0; Response = categorical judgements (emotion recognized by the participants). (TXT)
S3 File. EMG responses to emotion displayed: dataset. Group = healthy controls (HC) and PD patients (PD); Subject = subject number; Trial = trial number (e01 —e36); Emotion = emotion displayed (Angry, Happy, Neutral); Avatar = identification code of the avatar; Muscle = recorded muscle (Corru = corrugator supercilii, Zygo = zygomaticus major and Orbi = orbicularis oculi); Interval = sequential recording 100 ms interval (i01 —i20); EMG_response = EMG amplitudes averaged across the sequential 100 ms intervals of stimulus exposure and expressed as a relative percentage of the mean amplitude for baseline (%). (TXT) S1 Table. Characteristics of the patients ’ medication. Type: L = patients under L-dopa medi- cation only (levodopa + carbidopa and/or levodopa + benserazide and/or levodopa + carbi- dopa + entacapone), A = under dopamine agonists only, or L+A = under a combination of L-dopa and dopamine agonists; MAO/COMT: Some patients also took monoamine oxidase (MAO) B and/or catechol-O-methytransferase (COMT) inhibitors; Other(s) = Medication in addition of their dopamine replacement therapy. Specificities: 1 Patient under rotigotine trans- dermal patches (2 x 8 mg/24 hours) and receiving under-cutaneous injection of apomorphine (1 x 5 mg/24 hours), 2 & 5 receiving under-cutaneous injections of apomorphine (22 x 3 mg/24 hours and 53 x 4 mg/24 hours), 3 taking 0.25 mg of alprazolam /24 hours, 4 12.5 mg of cloza- pine + 7 drops of clonazepane (2.5 mg/ml)/24 hours. (DOC)
S2 Table. Inter-emotions comparisons of EMG responses recorded on sequential 100 ms intervals of stimulus exposure in the healthy controls. Test statistics ( χ²) are shown in brack- ets. Figures in bold denote statistically significant differences (p value
cally significant = p value >0.1. (DOC)
S3 Table. Inter-emotions comparisons of EMG responses recorded on sequential 100 ms intervals of stimulus exposure in the PD patients. Test statistics ( χ²) are shown in brackets. Figures in bold denote statistically significant differences (p value
significant = p value >0.1. (DOC)
S4 Table. Inter-muscles comparisons of the EMG responses recorded on sequential 100 ms intervals of stimulus exposure in the healthy controls. CORRU = corrugator supercilii; Facial Mimicry in Parkinson's Disease PLOS ONE | DOI:10.1371/journal.pone.0160329 July 28, 2016 15 / 20
ZYGO = zygomaticus major; ORBI = orbicularis oculi. Test statistics ( χ²) are shown in brack- ets. Figures in bold denote significant differences (p value
p value
>0.1. (DOC)
S5 Table. Inter-muscles comparisons of the EMG responses recorded on sequential 100 ms intervals of stimulus exposure in the PD patients. CORRU = corrugator supercilii; ZYGO = zygomaticus major; ORBI = orbicularis oculi. Test statistics ( χ²) are shown in brackets. Figures in bold denote significant differences (p value <0.05). ns = non significant = p value>0.1. (DOC)
Acknowledgments We would like to warmly thank all the participants for volunteering to take part in this study; the neurologists who helped recruit the PD patients (Drs Sophie Drapier, Rennes; Morgane Frouard, Rennes; Pierre Hinault, Rennes; François Lallement, Saint Brieuc; Marc Merienne, Saint Malo; Marie-Christine Minot-Myhie, Rennes; Isabelle Rivier, Rennes; Anne Salmon, Rennes; Claudie Schleich, Rennes; Marc Vérin, Rennes); Mrs Yves Bocou, Bruno Favier and all the members of the Association des Parkinsoniens d ’Ille-et-Vilaine (APIV) and France Parkin- son; and the pharmaceutical company UCB Pharma. We are very grateful to Benjamin Louis for his valuable support in the statistical analyses. We would also like to thank the NCCR in Affective Sciences, University of Geneva supporting Sylvain Delplanque and Didier Grandjean; as well as Angela Swaine Verdier and Sarah Verdier Leyshon for revising the English. Author Contributions Conceived and designed the experiments: SA SD DG PS. Performed the experiments: SA JFH MA JD. Analyzed the data: SA JD DG. Contributed reagents/materials/analysis tools: SA SD JD MV PS. Wrote the paper: SA SD JFH MA JD MV DG PS. References 1. Blair RJR. Facial expressions, their communicatory functions and neuro-cognitive substrates. Philos Trans R Soc Lond B Biol Sci 2003; 358:561 –72. doi: 10.1098/rstb.2002.1220 PMID:
12689381 2. Darwin C. The Expression of the Emotions in Man and Animals: 2nd ed. Cambridge. Payot & Rivages (2001); 1889. 3. Frith C. Role of facial expressions in social interactions. Philos Trans R Soc Lond B Biol Sci 2009; 364:3453 –8. doi:
10.1098/rstb.2009.0142 PMID:
19884140 4. Hess U. The communication of emotion. Emotions, Qualia and Consciousness 2001:397 –409. 5. Adolphs R. Recognizing emotion from facial expressions: psychological and neurological mechanisms. Behav Cogn Neurosci Rev 2002; 1:21 –62. PMID: 17715585 6. Gallese V. Embodied simulation: From neurons to phenomenal experience. Phenomenology and the Cognitive Sciences 2005; 4:23 –48.
7. Goldman AI, Sripada CS. Simulationist models of face-based emotion recognition. Cognition 2005; 94:193 –213. doi: 10.1016/j.cognition.2004.01.005 PMID:
15617671 8. Bogart KR, Cole J, Briegel W. On the Consequences of living without facial expression. Muller C., Cienki A., Fricke E., Ladewig S., McNeil D., Teßendorf S.; Handbook of Body —Language—Communi- cation: An International Handbook on Multimodality in Human Interaction, Berlin: Mouton de Gruyter: 2014, p. 1969 –82. 9.
ulates emotion perception accuracy. Social Psychological and Personality Science 2011; 2:673 –678.
10. Niedenthal PM, Brauer M, Halberstadt JB, Innes-Ker \AAse H. When did her smile drop? Facial mimicry and the influences of emotional state on the detection of change in emotional expression. Cognition & Emotion 2001; 15:853 –864. Facial Mimicry in Parkinson's Disease PLOS ONE | DOI:10.1371/journal.pone.0160329 July 28, 2016 16 / 20
11. Oberman LM, Winkielman P, Ramachandran VS. Face to face: blocking facial mimicry can selectively impair recognition of emotional expressions. Soc Neurosci 2007; 2:167 –78. doi: 10.1080/ 17470910701391943 PMID: 18633815
12. Bogart KR, Matsumoto D. Facial mimicry is not necessary to recognize emotion: Facial expression rec- ognition by people with Moebius syndrome. Soc Neurosci 2010; 5:241 –51. doi: 10.1080/ 17470910903395692 PMID: 19882440
13. Bogart KR, Tickle-Degnen L, Ambady N. Compensatory expressive behavior for facial paralysis: adap- tation to congenital or acquired disability. Rehabil Psychol 2012; 57:43 –51. doi: 10.1037/a0026904 PMID:
22369116 14.
Bailey PE, Henry JD, Nangle MR. Electromyographic evidence for age-related differences in the mim- icry of anger. Psychol Aging 2009; 24:224 –9. doi: 10.1037/a0014112 PMID: 19290755
15. Blairy S, Herrera P, Hess U. Mimicry and the judgment of emotional facial expressions. Journal of Non- verbal Behavior 1999; 23:5 Yüklə 0,9 Mb. Dostları ilə paylaş:
|