ویژگی های سیگنال از عبارات خود انگیخته صورت: حرکت اتوماتیک در لبخند انفرادی و اجتماعی

Abstract The assumption that the smile is an evolved facial display suggests that there may be universal features of smiling in addition to the basic facial configuration. We show that smiles include not only a stable configuration of features, but also temporally consistent movement patterns. In spontaneous smiles from two social contexts, duration of lip corner movement during the onset phase was independent of social context and the presence of other facial movements, including dampening. These additional movements produced variation in both peak and offset duration. Both onsets and offsets had dynamic properties similar to automatically controlled movements, with a consistent relation between maximum velocity and amplitude of lip corner movement in smiles from two distinct contexts. Despite the effects of individual and social factors on facial expression timing overall, consistency in onset and offset phases suggests that portions of the smile display are relatively stereotyped and may be automatically produced.

. Results 2.1. Preliminary analyses The inclusion of individuals with a history (although no current diagnosis) of depression or anxiety in the solitary sample introduced the possibility that some variation in smiling parameters may be attributable to a history of psychiatric diagnosis, or familial relationship with someone previously diagnosed. Potential effects were compared by assessing the relation between previously diagnosed (n=6), sibling of individual previously diagnosed (n=8), or control participant status (n=35) on the following temporal parameters within the solitary sample: onset, peak, and offset duration, and onset and offset amplitude. None of these parameters were significantly affected by psychiatric status (all P-values>0.05), and data from all solitary participants was considered in further analyses. 2.2. Characteristics of smile onsets, peaks, and offsets Smile onset duration averaged less than 1 s in both social (Full-size image (<1 K)=0.67 s, S.D.=0.24) and solitary contexts (Full-size image (<1 K)=0.52 s, S.D.=0.31) and was not significantly different in length in these two contexts (t(61)=−1.75, P=0.09). The amplitude of lip corner movement, measured as change in values of radius (lip corner movement) between beginning and end of longest continuous increase, was significantly different between contexts (Full-size image (<1 K)=0.050, S.D.=0.07 in the solitary context and Full-size image (<1 K)=0.150, S.D.=0.09 in the social context; t(62)=−3.97, P=0.001). Smile peak duration also differed between solitary and social smiles, with the peaks of laboratory solitary smiles lasting significantly longer (t(61)=3.80, P=0.001). Offset duration did not differ between contexts (t(61)=−0.0002, P=0.99). Offset amplitude also did not differ significantly between contexts (t(61)=−2.78, P=0.02). Significance was set at P<0.01, using the Bonferroni correction for multiple tests (N=5) of context effects on smile parameters. This result, however, represents a strong trend toward larger amplitude offsets in the social context (Full-size image (<1 K)=0.04, S.D.=0.05 in the solitary context; Full-size image (<1 K)=0.10, S.D.=0.08 in the social context). 2.3. Effects of other facial movements Visually observable Orbicularis oculi activity, previously associated with spontaneous smiles of enjoyment ( Frank et al., 1993) occurred in 23 smiles (n=18 (37%) solitary condition, n=5 (33%) social condition). There was not a significant difference in the presence of this marker in the two different contexts (χ2(1)=0.058, P=0.81), therefore, spontaneous smiles from the two contexts were analyzed together for effects of Orbicularis oculi. The presence of Orbicularis oculi activity was not significantly related to either amplitude or duration of smile onset (t(61)=1.10, P=0.28 for amplitude; t(61)=1.21, P=0.23 for duration). Orbicularis oculi activity also did not significantly affect duration of smile peak or amplitude of smile offset. The duration of smile offset, however, was related to the presence of Orbicularis oculi activity, with longer offsets in these smiles (Full-size image (<1 K)=0.67 s, S.D.=0.38 s for smiles with Orbicularis oculi activity; Full-size image (<1 K)=0.39 s, S.D.=0.19 s for smiles without Orbicularis oculi activity; t(61)=3.30, P=0.003). The Bonferroni correction was applied for multiple tests of significance, with significance set at P<0.01. Dampening movements that have been described as potential modifiers of Zygomaticus major activity (and lip corner appearance during smiles) were relatively rare and began following the initial frame of smile onset. Only 7 participants displayed Mentalis (n=4, solitary), or Depressor anguli oris (n=3, solitary) activity during the smile onset. Of those 7 participants with dampened smiles, 5 were control participants with no history of depressive symptoms. Furthermore, the percentage of control participants with dampened smiles (14%, or 5 of 35 participants) was similar to the percentage of participants with either a history of psychiatric illness or siblings of those with a history of psychiatric illness (15%, or 2 of 13 participants). These dampening actions appeared after the visually coded onset of Zygomaticus major activity, and had no apparent effect on amplitude or on duration of onset (t(61)=−1.78, P=0.08 for amplitude, t(61)=0.17, P=0.87 for duration). Effects of dampening movements approached significance for duration of peak, where peaks of smiles with dampening movements lasted 1.87 s on average and peaks of smiles without dampening movements lasted 0.86 s on average (t(61)=2.54, P=0.01) and were significant for duration of offset, where offsets of smiles with dampening movements lasted 0.79 s on average, while offsets of smiles without dampening movements lasted 0.45 s on average (t(61)=2.9, P=0.005). There was no significant effect of dampening movements, however, on amplitude of smile offset (t(61)=0.73, P=0.47). The Bonferroni correction applied for multiple tests of significance, with significance set at P<0.01. 2.4. Relation between velocity and amplitude of lip corner movement Despite differences in smile onset amplitude across contexts (solitary smiles showed less lip movement on average), maximum velocity of lip corner movement was closely related to the amplitude of the movement (see Fig. 4). For smile onsets, this relation was best fit by the power curve y=50.5x1.4 (R2=0.87). Individual regression analyses of social and solitary smile onset data sets showed similar fit, with curves y=79.3x1.5 (R2=0.82) and y=26.5x1.3 (R2=0.82) as the best fit curves for solitary (laboratory elicited) and social (interview) smile onsets, respectively. Relation between maximum velocity and amplitude of spontaneous smile onsets. Fig. 4. Relation between maximum velocity and amplitude of spontaneous smile onsets. Figure options Amplitude of lip corner movement was also related to maximum velocity of lip corner movement during smile offsets. The relation between maximum velocity and offset amplitude was best fit by the power curve y=11.8x1.1 (R2=0.82). Individual regression analyses of offsets in solitary and social smile offsets were best fit by curves y=22.8x1.3 (R2=0.77) and y=7.4x1.0 (R2=0.83), respectively.