Share this post on:

Can further lead to diminished recognition [35, 68]. It can be assumed that the internal representations of facial emotional expressions we use for recognition of such do include both emotion-specific facial features and their temporal characteristics [35]. The latter is not captured in static stimuli and generally not in morphed BasmisanilMedChemExpress RG1662 sequences where the same temporal characteristics are typically applied to all emotional expressions when facial emotion recognition is investigated. With respect to the visible facial features, it is possible that the onsets of facial action units within an emotional expression vary [69]. For example in happiness, at a very low intensity of expression, the wrinkling around the eyes is not visible, but the smile is [44], which wouldPLOS ONE | DOI:10.1371/journal.pone.0147112 January 19,21 /Validation of the ADFES-BIVmean that the mouth corners are pulled up before the wrinkling around the eyes occurs. True (un-manipulated) video recordings capture this, as timings are preserved, whereas morphed sequences present simultaneous appearance changes of the facial features. Another point to consider is that there is inter-individual variability that can be found in the speed of emotional expression [88] in facial expressions. Pollick et al. [88] used 3-D point light displays of posed expressions (anger, happiness, sadness, and surprise) recorded to video and assessed the time of expression from onset to apex of four encoders. They found jir.2012.0140 the speed of facial expressions to range by 466ms between the slowest and fastest encoder across the expressions included. Interestingly, this range was bigger than the one between the slowest and fastest emotion, which was 208ms. Again, those variations are captured by true video recordings, but not in morphed sequences, demonstrating the benefit of using video recordings.LimitationsThe varying durations of emotional content visible between the intensities of the ADFES-BIV could be seen as limitation in terms of standardisation. However, wcs.1183 they could also be interpreted as a further indicator of ecological validity, as the point of onset always precedes the reaching of maximum expression intensity. It is virtually impossible to create stimuli of facial expression at varying intensities and have the emotional content visible for the exact same duration between the intensities without manipulating the expression dynamics. Given the anatomical underpinnings of muscular movements, expressing low intensity facial emotional expressions should take less time than high intensity ones, since a more intense FT011 biological activity muscle contraction should take more time. Except if it was the case that we speed up our facial expression when expressing full intensity and slow down for subtle expressions which seems unlikely, but research on that has yet to be undertaken. Further support comes from the DaFEx video database [56] where professional actors were specifically instructed to express low, intermediate, and high intensity emotional facial expressions leading to the videos in the low condition being shorter in lengths than the intermediate ones and the latter shorter than the high intensity videos, just as in the current study. It can therefore be concluded that even though the current videos were edited, variations in timings between intensities resemble a natural occurrence for posed expressions –the intended purpose of the stimuli. A limitation of the stimuli is that the videos are edited to varying intensitie.Can further lead to diminished recognition [35, 68]. It can be assumed that the internal representations of facial emotional expressions we use for recognition of such do include both emotion-specific facial features and their temporal characteristics [35]. The latter is not captured in static stimuli and generally not in morphed sequences where the same temporal characteristics are typically applied to all emotional expressions when facial emotion recognition is investigated. With respect to the visible facial features, it is possible that the onsets of facial action units within an emotional expression vary [69]. For example in happiness, at a very low intensity of expression, the wrinkling around the eyes is not visible, but the smile is [44], which wouldPLOS ONE | DOI:10.1371/journal.pone.0147112 January 19,21 /Validation of the ADFES-BIVmean that the mouth corners are pulled up before the wrinkling around the eyes occurs. True (un-manipulated) video recordings capture this, as timings are preserved, whereas morphed sequences present simultaneous appearance changes of the facial features. Another point to consider is that there is inter-individual variability that can be found in the speed of emotional expression [88] in facial expressions. Pollick et al. [88] used 3-D point light displays of posed expressions (anger, happiness, sadness, and surprise) recorded to video and assessed the time of expression from onset to apex of four encoders. They found jir.2012.0140 the speed of facial expressions to range by 466ms between the slowest and fastest encoder across the expressions included. Interestingly, this range was bigger than the one between the slowest and fastest emotion, which was 208ms. Again, those variations are captured by true video recordings, but not in morphed sequences, demonstrating the benefit of using video recordings.LimitationsThe varying durations of emotional content visible between the intensities of the ADFES-BIV could be seen as limitation in terms of standardisation. However, wcs.1183 they could also be interpreted as a further indicator of ecological validity, as the point of onset always precedes the reaching of maximum expression intensity. It is virtually impossible to create stimuli of facial expression at varying intensities and have the emotional content visible for the exact same duration between the intensities without manipulating the expression dynamics. Given the anatomical underpinnings of muscular movements, expressing low intensity facial emotional expressions should take less time than high intensity ones, since a more intense muscle contraction should take more time. Except if it was the case that we speed up our facial expression when expressing full intensity and slow down for subtle expressions which seems unlikely, but research on that has yet to be undertaken. Further support comes from the DaFEx video database [56] where professional actors were specifically instructed to express low, intermediate, and high intensity emotional facial expressions leading to the videos in the low condition being shorter in lengths than the intermediate ones and the latter shorter than the high intensity videos, just as in the current study. It can therefore be concluded that even though the current videos were edited, variations in timings between intensities resemble a natural occurrence for posed expressions –the intended purpose of the stimuli. A limitation of the stimuli is that the videos are edited to varying intensitie.

Share this post on: