Authors: Fabrizia Gallo, Alberto González-Villar, Laurent Ott, Adriana Sampaio, Jean-Louis Nandrino, Angela Bartolo
Categories: Article, Human behaviour, Psychology
Source: Scientific Reports
Authors: Fabrizia Gallo, Alberto González-Villar, Laurent Ott, Adriana Sampaio, Jean-Louis Nandrino, Angela Bartolo
Intransitive gestures are expressive and symbolic, whereas pantomimes are object-related actions. These gestures convey different meanings depending on whether they are directed toward (TB) or away from the body (AB). TB gestures express mental states (intransitive) or hygiene/nutritional activities (pantomime), while AB gestures modify the behaviour of the observer (intransitive) or demonstrate tool use with an object (pantomime). A substantial body of literature suggests that females exhibit stronger social cue processing compared to males. Considering the social significance of gestures, this study aimed to explore the physiological gender differences in the observation of AB and TB gestures. Pupil dilation and High Frequency Heart Rate Variability (HF-HRV) were measured in 54 participants (27 female) while observing TB and AB gestures. The Interpersonal Reactivity Index (IRI) and the Vicarious Distress Questionnaire (VDQ) were used to assess social-emotional processes. Results showed greater pupil dilation in females for TB gestures, but no significant gender differences for HF-HRV. Males showed a significant correlation between increased pupil dilation to both TB and AB gestures and empathy levels (IRI). The support scale of the VDQ correlated significantly with TB gestures in males. These findings provide insight into the neurobiological basis of gender differences in perceiving social gestures.
Cognitive neuropsychology distinguishes between meaningless and meaningful gestures. The former are gestures that have no meaning to the observer (e.g. hand under the chin), whereas meaningful gestures are divided into object-related and non-object-related actions. Object-related actions include pantomime gestures, which are gestures that mimic the use of an object without holding the object in the hand. Non-object-related actions include intransitive gestures, which are communicative, expressive gestures^1^. Recent studies further distinguish gestures according to where the action is directed, i.e. toward the body (TB) or away from the body (AB)^2–6^. This distinction is based on previous studies conducted with meaningless gestures (i.e. gestures that have no meaning) that showed differences between hand posture and finger configurations^7,8^. Hand postures are performed by placing the hand on a specific part of the body, for example, the palm on the head. This is a body-related gesture directed toward the performer’s personal space (TB). In contrast, finger configurations involve extending the hand in a visible way to show the arrangement of the fingers, such as folding the thumb and ring finger. These gestures are performed away from the body (AB), in the peripersonal or action space of the person performing the gesture^9^.
In Bartolo et al. (2019), we claimed that in the case of meaningful gestures (i.e., pantomime and intransitive), there may be a relationship between gesture direction and gesture intention^2^. In the case of intransitive gestures, gestures TB, with a few exceptions, tend to convey information about the personal state of the person performing the action, making them primarily mental state gestures. For example, the gesture of shivering (which is a gesture TB) indicates that the person performing the action is cold. Although this gesture is not typically used for direct interaction with others, observing someone making this gesture can trigger a response from the observer. For instance, seeing someone shivering may prompt the observer to offer something to help warm the person up. Such gestures have the potential to elicit prosocial behaviours. On the other hand, gestures AB are often used to facilitate two-way interaction. For example, the ‘come here’ gesture is performed by extending the hand towards the viewer and folding the hand or index finger, palm up. Furthermore, the observation of this gesture requires a response from the observer. Interestingly, these gestures often involve reducing the physical distance between the sender and receiver in order to create a better and more effective interaction between them. In other words, this distinction between meaningful gestures TB and AB corresponds to the categorisation proposed by Gallagher and Frith (2004)^10^. These authors distinguish between inner gestures that evoke the mental state of the person performing the action (e.g. “I'm hungry”) and instrumental gestures that modify the behaviour of the observers (e.g. “Go there”). Bartolo et al. considered that most of the inner gestures are TB, whereas instrumental gestures are AB^2^. Gallagher & Frith also found a neuroanatomical dissociation between these two types of intransitive gesture^10^. Specifically, whereas instrumental/AB gestures activated brain regions associated with language processing, such as the left inferior and middle frontal cortices, inner/TB gestures activated regions of the “social brain”^10^, including the anterior paracingulate cortex, the right superior temporal sulcus (STS), and the bilateral temporal lobes.
In the case of pantomime gestures, when performed as TB, they are usually associated with hygiene-related actions (e.g., mimicking toothbrushing) or eating behaviours (e.g., mimicking the use of a spoon)^1,11^ and are directed toward a specific part of the body. Pantomime AB are gestures that are directed toward another object, thus they typically involve the representation of both a tool (e.g. the hammer) and an object (i.e. the nail).
Bartolo et al. (2019) studied the physiological responses associated with the observation of meaningless, intransitive, and pantomime gestures TB and AB in a group of healthy individuals^2^. In particular, they measured pupil dilation and High Frequency Heart Rate Variability (HF-HRV), as indicators of psychological processes at work during gesture observation. The authors have reported increased pupil dilation during the observation of intransitive and pantomime gestures TB compared to AB. For meaningless gestures, no significant difference in pupil dilation between TB and AB was observed^2^. Since pupil dilation has been shown to be sensitive to emotional stimuli^12–17^, the increase in pupil dilation when observing intransitive gestures TB has been explained by the fact that that these gestures are mental state gestures, i.e. gestures that express feelings, moods, and emotions. In the case of pantomime TB, as they belong to the personal sphere of the person performing the action; consequently, observing someone performing these gestures can be compared to entering their private sphere, which provokes a feeling of discomfort and thus evokes a stronger emotional response^2^. Furthermore, an increased HF-HRV (i.e. increased vagal suppression) was also observed for intransitive gestures AB compared to TB. Given that HRV reflects the organism’s ability to adapt to the demands of the environment and generate appropriate behaviour^18^, being recognized to be important in social communication^19^, the increase of vagal suppression observed in response to intransitive gestures AB has been explained by proposing that these gestures likely elicit a response from the receiver that alters her or his homeostasis. For instance, a gesture such as “come here” requires the receiver to approach, necessitating the receiver to adapt and generate an appropriate response^2^.
However, the findings reported by Bartolo et al. (2019) were obtained from a sample consisting of predominantly healthy individuals, the majority of whom were female (26 females out of 33 participants). The potential implications of this gender disparity within the study sample should be considered, as the preponderance of female subjects may lead to gender biases that could impact the generalizability of the findings. Previous research has extensively investigated gender differences in various aspects of socio-affective and socio-cognitive tasks. In particular, females have been found to be more interested in social information, to have superior facial expression recognition skills, and to be better at understanding gestures and body language^20^. These gender differences include analysis of social interactions, empathic responses to others’ pain, interest in social cues and information, interpretation of gestures and actions, perception of biological motion, and responsiveness to emotional stimuli^20–26^. In particular, females exhibit greater emotional responsiveness and better emotion recognition abilities, and they tend to engage more emotional brain areas during tasks related to social cognition^20^ and empathy^22,24–26^. This enhanced affective empathy is reflected in their more accurate and faster recognition of facial expressions and their increased involvement in prosocial and altruistic behavior^22,27^ and different engagement of the mirror system^28^. Interestingly, a percentage of mirror neurons were found to activate to respond when affiliative communicative gestures are observed (e.g.,^22^), highlighting the link between social-cognitive and affective skills in the understanding and processing of social and communicative gestures.
The primary aim of this study was to investigate potential differences in physiological responses (pupil dilation and Heart Rate Variability—HRV) between females and males when observing meaningless, pantomime and intransitive gestures directed TB or AB. Pupil dilation responds to stimuli that carry emotional content^12,15,16^ and HRV responses to different types of gestures have been previously documented^2^.
Considering potential differences in physiological responses between males and females, our secondary aim was to explore the relationship between these physiological responses and scores on two social-cognitive and affective questionnaires. Therefore, we have included the Interpersonal Reactivity Index (IRI,^29^), which measures various aspects of cognitive and affective empathy, and the Vicarious Distress Questionnaire (VDQ,^30^), which measures prosocial behavior, i.e., how individuals react to and cope with witnessing another person’s distress, including supportive behaviors and avoidance tendencies. Pupil dilation responds to stimuli that carry emotional content^12,15,16^ and relationship between empathy and pupil dilation has been already reported in the literature^13,14,17,31,32^, but also HRV^33^, indicating that more empathic individuals tend to show greater pupil dilation in response to others’ emotional expressions. These physiological measures, such as pupil dilation and HRV, provide valuable insights into the mechanisms underlying empathy and prosocial behaviour in response to emotional stimuli.
By examining the potential associations between physiological responses and the scores at the IRI and VDQ, we aim to shed light on the underlying mechanisms that contribute to gender variations in the perception of gestures.
The results for pupil dilation are largely consistent with those reported in the previous study (Bartolo et al., 2019). In particular, the interaction Gesture Type X Gesture Direction, F(2, 104) = 5.98, p = 0.003, ηp^2^ = 0.10 was significant (see Fig. 1). Post hoc analyses showed greater pupil dilation for Intransitive gestures TB (M = 189.88 SEM = 19.28) vs. AB (M = 97.09 SEM = 18.18; p < 0.001), and for Pantomime gestures TB (M = 199.08 SEM = 16.93) vs. AB (M = 152.22 SEM = 18.66; p = 0.002). No difference was observed between Meaningless gestures TB (M = 165.70 SEM = 16.16) vs. AB (M = 145.14 SEM = 18.26; p = 0.20). Furthermore, pupil dilation was smaller for Intransitive gestures AB than for all the other gestures (p < 0.001), whereas pupil dilation was greater for Intransitive gestures TB than for Pantomime gestures AB (p = 0.01) and Meaningless gestures AB (p = 0.003). Pupil dilation was greater for Pantomime TB than for Meaningless AB (p < 0.001) and TB (p = 0.03).Fig. 1Pupil gesture type as a function of gesture direction. Vertical bars represent the SEM. Asterisks indicate (the main) significant differences (p < 0.05).
The effect of Gesture Type was significant, F(2, 104) = 3.25, p = 0.04, ηp^2^ = 0.06, showing that pupil dilation was greater for Pantomime gestures (M = 175.65 SEM = 15.79) than for Intransitive gestures (M = 143.48 SEM = 16.85, p = 0.001). The effect of Gesture Direction, F(1, 52) = 24.51, p < 0.001, ηp^2^ = 0.32, was also significant, with pupil dilation greater for gestures TB (M = 187 SEM = 14.78) than for gestures AB (M = 131.87 SEM = 15.44, p < 0.001).
Interestingly, the interaction Gender X Gesture Direction was also significant, F(1, 52) = 7.25, p < 0.009, ηp^2^ = 0.12 (see Fig. 2). Fisher’s post hoc showed that while pupil dilation in females was greater for stimuli TB (M = 222.40 SEM = 26) than AB (M = 139.95 SEM = 26.54; p < 0.001), this difference was not significant for males (p = 0.12; MTB 151.59 SEMTB = 10.85 vs. MAB = 123.79 SEMAB = 16.22). Pupil dilation in females for stimuli TB was greater than that registered in males for both stimuli TB (p = 0.01) and AB (p = 0.001).Fig. 2Pupil gesture direction as a function of gender. Vertical bars represent the SEM. Asterisks indicate (the main) significant differences (p < 0.05).
There was no effect of Gender, F(1, 52) = 2.77, p = 0.10, ηp^2^ = 0.05, nor interaction Gesture Type X Gender, F (2, 104) = 0.51, p = 0.60, ηp^2^ = 0.009, nor interaction Gender X Gesture Type X Gesture Direction, F(2, 104) = 2.12, p = 0.12, ηp^2^ = 0.039, meaning that overall, both females and males process the three gestures in the same way and performance on the three gesture types was equal for the two groups of participants.
The mixed ANOVA revealed a significant interaction Gesture Type X Gesture Direction: F(2, 104) = 4.09, p = 0.02, ηp^2^ = 0.08 (Fig. 3). Fisher Post Hoc analyses showed less vagal suppression for Intransitive gestures TB (M = -0.26 SEM = 0.07) vs. AB (M = -0.39 SEM = 0.08; p = 0.002). Then, vagal suppression was also lower for Intransitive gestures TB compared to Pantomime gestures TB (M = -0.36 SEM = 0.09; p = 0.045) and Meaningless gestures AB (M = -0.45 SEM = 0.09; p = 0.001). The results also showed less vagal suppression for Pantomime gestures AB (M = -0.28 SEM = 0.08) compared to Meaningless gestures AB (p = 0.007).Fig. 3ΔHF-HRV: gesture type as a function of gesture direction. Vertical bars represent the SEM. Asterisks indicate (the main) significant differences (p < 0.05).
The other results were not the effect of the Gesture Type, F(2, 104) = 2.11, p = 0.13, ηp^2^ = 0.04, the effect of Gender: F(1, 52) = 0.34, p = 0.60, ηp^2^ = 0.006; the interaction Gender X Gesture Type: F(2, 104) = 0.48, p = 0.62, ηp^2^ = 0.009; the effect of Gesture Direction: F(1, 52) = 1.61, p = 0.21, ηp^2^ = 0.003; the interaction Gender X Gesture Direction: F(1, 52) = 0.07, p = 0.79, ηp^2^ = 0.001; the interaction Gender X Gesture Type X Gesture Direction: F(2, 104) = 0.56, p = 0.57, ηp^2^ = 0.01.
The results of the two questionnaires, the IRI and the VDQ, for females and males are shown in Table 1. No significant difference was found between females and males in the IRI and VDQ.Table 1Range, Mean and (standard deviations) of the results at the questionnaires for females and males. The last column shows the t-tests carried out to compare the results of females and males.QuestionnairesFemalesMalesFemales vs. MalesIRICognitive23.5–37, 31.30 (3.90)19–48, 30 (5.45)t52 = 1.00, p = 0.32Affective20–37, 30.41 (4.01)18–39, 28.26 (4.40)t52 = 1.87, p = 0.07VDQDistress11–29, 20 (4.37)11–27, 17.78 (4.28)t42 = 1.70 p = 0.11Support20–38, 29.14 (4.13)11–36, 28.91 (5.42)t42 = 0.16, p = 0.88Avoidance4–13, 7.24 (2.84)4–20, 7.87 (3.82)t42 = -0.62, p = 0.54
While the ANOVA did not show a significant main effect of gender, we found a significant difference between females and males in pupil dilation during gesture direction processing (Gender X Gesture Direction interaction). Therefore, we analysed the correlation between gesture direction (TB and AB gestures regardless of gesture type) and the scores on the IRI Cognitive and Affective and the three subscales of the VDQ (Distress, Support, and Avoidance), with the hypothesis that females and males might use different social-affective resources when processing gesture direction (Table 2). The results showed significant correlations only for males, in particular, both the Cognitive and Affective IRIs were significantly correlated with the pupil dilation registered in both TB and AB gestures. On the VDQ, only scores on the Support subscale were significantly correlated with pupil dilation to TB gestures for males only (see Table 2).Table 2Correlation coefficients and p-values of the correlations between IRI and VDQ and pupil dilation in gestures TB and AB in females and males. The last column shows the comparisons between the correlations in females and males. Significant results are shown in bold.QuestionnairesPupil dilation for gestures TBFemalesMalesFemales vs. malesIRICognitiver = -0.19r = 0.40****Z = -2.13, p = 0.02p = 0.33p = 0.04Affectiver = 0.22r = 0.41Z = -0.73, p = 0.23p = 0.28p = 0.03VDQDistressr = 0.09r = -0.17Z = 0.81, p = 0.21p = 0.70p = 0.43Supportr = -0.08r = 0.48****Z = -1.86, p = 0.03p = 0.73p = 0.02Avoidancer = 0.052r = -0.30Z = 1.11, p = 0.13p = 0.82p = 0.16QuestionnairesPupil dilation for gestures ABFemalesMalesFemales vs. MalesIRICognitiver = -0.08r = 0.39****Z = -1.7, p = 0.04p = 0.68p = 0.04Affectiver = 0.19r = 0.51Z = -1.28, p = 0.10p = 0.33p = 0.01VDQDistressr = 0.04r = -0.08Z = 0.36, p = 0.36p = 0.87p = 0.69Supportr = 0.07r = 0.22Z = -0.47, p = 0.32p = 0.77p = 0.31Avoidancer = 0.14r = -0.22Z = 1.12, p = 0.13p = 0.54p = 0.31
Fisher r-to-z transformation was used to compare the correlation coefficients registered in females vs. males (see Table 2). The results showed that the correlation coefficients obtained for the correlations between IRI Cognitive and both pupil dilation in TB and AB gestures differed significantly between females and males. For the correlations between VDQ support and pupil dilation in TB gestures, there was also a significant difference between females and males.
In a previous study, Bartolo et al. (2019) found significant physiological responses associated with the observation of intransitive and pantomime gestures. However, this study included predominantly female participants^2^. As the predominance of female subjects may introduce gender biases that could affect the generalizability of the findings, the primary aim of our work was to investigate whether female participants exhibit an enhanced physiological response to meaningful gestures compared to males. This investigation is based on the understanding that intransitive and pantomime gestures inherently contain a social component and that females and males may display varying levels of proficiency in social-emotional skills^22,34^.
In addition to exploring the physiological responses, the study incorporates questionnaires that assess cognitive and affective empathic tendencies (IRI^29^) and prosocial behaviour (VDQ^30^). We aimed to gain insights into the association between these measures and the physiological reactions registered in our participants during gesture observation. Therefore, the secondary aim of this study was to explore whether higher scores on measures of empathy and prosocial behavior are correlated with increased physiological responses to meaningful gestures and if they differed between females and males.
The results showed that the main effect of gender was not significant, indicating that there are no gender differences in physiological modulation during gesture observation. Thus, the results largely replicated previous research findings^2^. Consistent with previous findings, we found that observing both intransitive and pantomime gestures TB elicited a significant increase in pupil dilation compared to gestures AB. In line with previous research, these findings may be attributed to the nature of intransitive and pantomime gestures directed towards the body, which contain social-emotional content, as reflected in the observed increase in pupil dilation^2^.
Regarding ΔHF-HRV, the results showed significantly higher vagal suppression for intransitive gestures AB compared to intransitive gestures TB, confirming previous findings^2^. As found by Bartolo et al. (2019), the activity of the parasympathetic system was influenced by the observation of intransitive gestures AB, leading to increased vagal suppression. Indeed, intransitive gestures AB require a response from the receiver (e.g., the gesture “come here” requires the receiver to approach), thus requiring the organism to adapt to produce the appropriate response^2^.
An interesting and novel finding of this study was that females exhibited a greater increase in pupil dilation compared to males during the observation of gestures TB, regardless of the type of gesture. This means that the influence of gesture direction on pupil dilation varies between genders, with females displaying a more pronounced response. This finding aligns with previous research indicating that females generally demonstrate a heightened interest in social information and possess enhanced abilities in understanding gestures and body language^20^. The observed gender difference in pupil dilation may reflect these underlying socio-cognitive differences, with females being more attuned and responsive to social cues than males.
No significant difference was found between females and males in their responses to the questionnaires. Given that the gender difference only appeared in pupil dilation for gesture direction regardless of gesture type, correlation analyses were performed to examine the association between the results at the cognitive and affective scales of the Interpersonal Reactivity Index (IRI) and the subscales of the Vicarious Distress Questionnaire (VDQ) with pupil dilation registered in TB and AB gestures for each gender group separately. Correlation coefficients were compared to examine gender differences. Due to the small number of participants in each group (21 women^35^), these results are considered to be exploratory.
The results showed significant correlations only in males. In particular, the IRI (both cognitive and affective) correlated significantly with pupil dilation registered in both TB and AB gestures. In males, there was also a significant correlation between the support scale of the VDQ and the pupil dilation in TB. No significant correlations were found in the female participants. Furthermore, the correlation coefficients registered in the correlations between the cognitive IRI and both the gestures TB and AB, as well as the correlation between the VDQ support and the TB gestures, differed significantly between female and males. Taken together, these results suggest that males who showed greater use of emotion regulation strategies, a greater tendency to offer support in response to others’ distress, and higher levels of affective empathy showed greater pupil dilation when observing TB and AB gestures. In summary, in contrast to females, the degree of pupil dilation in response to TB gestures in males may depend on their level of emotional connectedness, empathy, and supportive behaviour^22,36^. This line of reasoning is consistent with neuroimaging data supporting that the neural networks underlying empathy are differentially modulated by gender. More specifically, Schulte-Ruther et al. found increased recruitment of the human mirror neuron system in females relative to males during empathy-related face-to-face interactions^24^. In contrast, males show greater recruitment of areas associated with ToM, corresponding to a more cognitive strategy. Furthermore, the study by Luo et al. using ERP measures also supports the idea of an enhanced ability to recognise and share the emotions of others in females, whereas in males larger early ERP amplitudes (P2) are only observed during self-focused tasks^37^.
Overall, the results suggest that the physiological differences observed between females and males during gesture observation may be related to the distinct disposition of the two genders to process social cues and use socio-emotional resources.
In summary, the results of the study showed an effect of gesture direction on gender. Females exhibited greater pupil dilation than males, specifically in response to gestures TB, irrespective of gesture type.
Furthermore, the correlation analyses indicate a relationship between physiological responses of males to gestures TB and AB and their level of empathy and prosocial skills, while no significant relationship was observed in females. This suggests that males’ physiological reactions to these gestures are more closely tied to their empathic abilities compared to females. However, due to the small sample size, the results should be interpreted with caution. Furthermore, another important consideration is the gender of the individual performing the gestures. In this study, all participants observed a female actor, which raises the question of whether the same effects would be observed if a male actor performed the actions. Further investigation into this would help determine if the observed effect is influenced by gender similarity between the participant and the performer.
These findings have important implications for understanding the neurobiological basis of gender differences in the observation of social gestures. They provide valuable insights into the underlying mechanisms of action understanding and shed light on the gender-specific cognitive processes that contribute to social behaviour. Further research in this area could deepen our understanding of these gender-specific processes and their impact on social interactions.
We calculated that a total sample size of 54 participants would be sufficient in our study to detect a significant effect with a power of 0.80, an alpha of 0.05, and an effect size of 0.25 with both pupil dilation and HF-HRV measures (G*Power Version 3.1.9.7, ANOVA within-between interaction).
Fifty-four participants (age range 19–25, 27 females), were included in the study. All the participants gave written informed consent to participate in the study. The experimental protocol was approved by the local ethical committee in behavioural sciences of the University of Lille (Ref. number 329-S69) and conducted in accordance with the Declaration of Helsinki.
To exclude the presence of mood disorders, participants were administered the French version of the following self-questionnaires: CERQ (Cognitive Emotion Regulation Questionnaire^38^, DERS (Difficulties in Emotion Regulation Scale^39^, and the HADS (Hospital Anxiety and Depression Scale^40^. To rule out differences between females and males on these scales, a series of t-tests were run, which showed no significant results. See Table 3.Table 3Range, Mean (Standard Deviation) on the questionnaires for each gender group. The last column shows the t-tests carried out to compare the results of females and males.QuestionnairesFemalesMalesFemales vs. MalesHADSAnxiety3–231–18t52 = 0.31, p = 0.769.67 (5.20)9.26 (4.59)Depression0–260–22t52 = 1.11, p = 0.277.67 (8.73)5.37 (6.22)CERQAdaptation35–9651–85t52 = 0.63, p = 0.5370.52 (14.03)68.41 (10.37)Non-Adaptation29–6416–84t52 = -0.25, p = 0.8040.85 (8.73)41.63 (13.60)DERS37–7138–65t52 = -0.25, p = 0.8049.48 (8.56)50.04 (7.82)
In the present study, we have used the same method as in Bartolo et al.^2^. The stimuli were videos of gestures performed by an actress with a neutral facial expression. Gestures were meaningless, intransitive and pantomime executed in two directions, either towards the body (TB) or away from the body (AB). Each video was 4 s long and was repeated twice, giving a total of 8 s, so that HRV could be measured for a block of 10 videos. In fact, pupil dilation reflects the sympathetic activity of the autonomic nervous system and is measured for each stimulus in a window between 2 and 5 s after the onset of the stimulus. Conversely, HRV reflects the parasympathetic regulation of heart rate and takes longer to measure and was therefore evaluated at the level of the 10 video block. A total of 60 gesture stimuli were presented, 20 per each gesture type. Half of the 60 gestures were TB and the other half were AB.
To measure HRV, we acquired subjects’ ECG signal using a BIOPAC amplifier (MP35; BIOPAC Systems, Inc., Goleta, CA, USA) together with SS2LB leads and EL500 disposable electrodes placed on the subjects’ wrists. Using custom Matlab software, R peaks locations were first automatically extracted from the ECG signal and then visually inspected by an unbiased experimenter blind to the experimental hypothesis. HRV quantification was performed from the R-R intervals timeseries according to the guidelines in the Kubios software manual^41^. To eliminate very low frequency components, R-R intervals were detrended below 0.04 Hz^41^. Power spectral density analysis was then performed on the baseline rest period (180 s) and each block of gesture types (meaningless, pantomime, intransitive towards the body and intransitive away from the body) for a duration of 130 s. Fast Fourier Transform was used, with a high frequency band set at 0.15–0.4 Hz. The power spectral density within the high frequency band was integrated and plotted using the natural logarithm. Phasic HF-HRV indices were obtained for each condition by subtracting the log-transformed HF-HRV in the gesture direction block from the log-transformed HF-HRV in the baseline. This calculation can be expressed as ΔHF-HRV(gesture direction block) = HF-HRV(gesture direction block)—HF-HRV(baseline).
Pupil diameter was recorded at a 500 Hz samplerate using an SR Research Eyelink 1000 eyetracker. The subject head was located at a distance of 60 cm from the screen and was placed on a chin rest to minimize artefacts due to head movements. Pupil dilation is a time-dependent continuous variable. To examine the changes in pupil dilation based on the stimulus nature^42^, the average pupil area dilation (measured in arbitrary units by Eyelink) relative to the baseline (stimulus onset) was extracted for each video.
Participants were also administered the IRI (Interpersonal Reactivity Index^29^ and the VDQ (Vicarious Distress Questionnaire^30^. They all take the form of items for which participants are asked to position themselves on a 5-point Likert scale. The IRI is a 28-item self-questionnaire designed to measure affective and cognitive components of empathy and measures the level of discomfort and anxiety when individuals witness the negative experiences of others. The four-factor structure of the IRI and the overall structure of empathy are psychometrically invariant, regardless of gender^43,44^. The Vicarious Distress Questionnaire (VDQ) is a self-report questionnaire designed to measure participants’ distress reactions and subsequent outcomes, such as supportive behaviours or avoidance tendencies, in response to witnessing another person’s distress. Distress, supportive responses, and avoidance tendencies registered by the VDQ are associated with an individual’s affective empathy level, anxiety disorders, alexithymic traits, and mood-related conditions^30^. The VDQ questionnaire was included at a later stage, resulting in data availability for the VDQ from 44 (of which 21 females) out of the total 54 participants.
The participants were volunteers recruited at the University of Lille (France) through advertisements. They received a letter of information describing the experimental setup and explaining the study. Participants were then asked to come to the experiment and were instructed to abstain from consuming any psychoactive substances such as coffee, tea, tobacco, or alcohol for one hour prior. The experiment took place in the morning (between 10 am and 12 pm) or in the afternoon (between 2 and 4 pm) in the SCALab experimental box. This timing was chosen to minimise the influence of circadian rhythms on the data collected. The temperature in the laboratory was kept constant at 24 °C. Once the participants signed the informed consent form, the experiment could begin.
For the emotional, sympathetic response, pupil dilation was recorded within a short time window of 2 to 5 s after the onset of the stimulus. Heart rate variability (HRV) was measured for the parasympathetic response to assess regulatory changes. This response required more time for activation regulation, so each 4-s gesture video was repeated twice, resulting in an 8-s stimulus duration. Each gesture video was followed by a green screen that matched the background colour in the videos (RGB 86 156 87) and displayed a fixation cross (" + ") for a duration of 4, 5, or 6 s.
Within each Gesture condition (meaningless, pantomime, and intransitive), there were two direction blocks consisting of 10 gesture videos each (10 toward the body and 10 away from the body). Each direction block lasted 130 s. The order of presentation of the blocks was randomised.
The experiment was run on PsychoPy v1.90.2 software and began with on-screen instructions in French. Participants were instructed to watch videos of a person performing gestures ("Des vidéos d’une personne réalisant des gestes vont apparaitre dans l’écran. Regardez simplement chaque geste. Appuyez sur la barre d’espace pour démarrer l’expérience"/ "Videos of a person performing gestures will appear on the screen. Just look at each gesture. Press the spacebar to start the experiment"). Following this, another screen appeared, instructing participants to close their eyes and relax ("Fermez les yeux et détendez-vous" / "Close your eyes and relax"). This relaxation period lasted 3 min and served as a baseline for recording participants’ resting state. After the 3 min, an acoustic signal lasting 0.5 s indicated that participants should open their eyes. On the screen, participants were informed that the task was about to start ("La tâche va commencer" / "The task is ready to start"). A “trial block” was always presented first, and then the Gesture conditions were randomly presented to participants.
Once the task was completed, participants were instructed to close their eyes once again for 3 min ("L'expérience est terminée, fermez les yeux et détendez-vous" / "The experiment is over, close your eyes and relax"). This second resting condition concluded with a 0.5-s acoustic signal, signifying the end of the experiment. Subsequently, participants were administered questionnaires.
Statistica software version 13 was used to perform all analyses, with a significance level established at 0.05. The data’s normal distribution was assessed using Kolmogorov–Smirnov’s test, while Mauchley’s test was used to verify the assumption of sphericity. Results showed that only the pupil dilation data for the Meaningless gestures TB were not normally distributed (K-S d = 0.21, p < 0.05). For the other variables, data normally distributed. Pupil Intransitive AB: d = 0.07, p > 0.20; Intransitive TB: d = 0.14, p > 0.20; Pantomime AB: d = 0.11, p > 0.20; Pantomime TB: d = 0.11, p > 0.20; Meaningless AB: d = 0.10, p > 0.20. For Delta HF-HRV: Intransitive AB: d = 0.08, p > 0.20; Intransitive TB: d = 0.09, p > 0.20; Pantomime AB: d = 0.04, p > 0.20; Pantomime TB: d = 0.08, p > 0.20; Meaningless AB: d = 0.09, p > 0.20; Meaningless TB: d = 0.09, p > 0.20. However, as the results did not change when non-parametric analyses were used (see Supplementary Material), for each physiological measure (pupil and dilation ΔHF-HRV) we run a Mixed ANOVA with 2 independent Gender groups (males and females) X 3 Gesture Types (meaningless, intransitive and pantomime) X 2 Gesture Directions (TB and AB; Notice that when analyses are performed using LMM, the results do not change, see Supplementary Material). Post-hoc Fisher’s analyses have been also carried out for significant ANOVA effects.
Separate correlation analyses (Pearson’s r) for females and males were then carried out between the results at the IRI and the VDQ and the physiological measures that reveal a difference between females and males. Given the size of our sample (see^35^ the results will be considered exploratory.
Supplementary information