A computational neuroethology perspective on body and expression perception

Observers predict actions from facial emotional expressions during real-time social interactions

In face-to-face social interactions, emotional expressions provide insights into the mental state of an interactive partner. This information can be crucial to infer action intentions and react towards another person's actions. Here we investigate how facial emotional expressions impact subjective experience and physiological and behavioral responses to social actions during real-time interactions. Thirty-two participants interacted with virtual agents while fully immersed in Virtual Reality. Agents displayed an angry or happy facial expression before they directed an appetitive (fist bump) or aversive (punch) social action towards the participant. Participants responded to these actions, either by reciprocating the fist bump or by defending the punch. For all interactions, subjective experience was measured using ratings. In addition, physiological responses (electrodermal activity, electrocardiogram) and participants' response times were recorded. Aversive actions were judged to be more arousing and less pleasant relative to appetitive actions. In addition, angry expressions increased heart rate relative to happy expressions. Crucially, interaction effects between facial emotional expression and action were observed. Angry expressions reduced pleasantness stronger for appetitive compared to aversive actions. Furthermore, skin conductance responses to aversive actions were increased for happy compared to angry expressions and reaction times were faster to aversive compared to appetitive actions when agents showed an angry expression. These results indicate that observers used facial emotional expression to generate expectations for particular actions. Consequently, the present study demonstrates that observers integrate information from facial emotional expressions with actions during social interactions.

Three-stage Dynamic Brain-cognitive Model of Understanding Action Intention Displayed by Human Body Movements

The ability to comprehend the intention conveyed through human body movements is crucial for effective interpersonal interactions. If people can't understand the intention behind other individuals' isolated or interactive actions, their actions will become meaningless. Psychologists have investigated the cognitive processes and neural representations involved in understanding action intention, yet a cohesive theoretical explanation remains elusive. Hence, we mainly review existing literature related to neural correlates of action intention, and primarily propose a putative Three-stage Dynamic Brain-cognitive Model of understanding action intention, which involves body perception, action identification and intention understanding. Specifically, at the first stage, body parts/shapes are processed by those brain regions such as extrastriate and fusiform body areas; During the second stage, differentiating observed actions relies on configuring relationships between body parts, facilitated by the activation of the Mirror Neuron System; The last stage involves identifying various intention categories, utilizing the Mentalizing System for recruitment, and different activation patterns concerning the nature of the intentions participants dealing with. Finally, we delves into the clinical practice, like intervention training based on a theoretical model for individuals with autism spectrum disorders who encounter difficulties in interpersonal communication.

Differential contributions of body form, motion, and temporal information to subjective action understanding in naturalistic stimuli

Introduction

We investigated the factors underlying naturalistic action recognition and understanding, as well as the errors occurring during recognition failures.

Methods

Participants saw full-light stimuli of ten different whole-body actions presented in three different conditions: as normal videos, as videos with the temporal order of the frames scrambled, and as single static representative frames. After each stimulus presentation participants completed one of two tasks-a forced choice task where they were given the ten potential action labels as options, or a free description task, where they could describe the action performed in each stimulus in their own words.

Results

While generally, a combination of form, motion, and temporal information led to the highest action understanding, for some actions form information was sufficient and adding motion and temporal information did not increase recognition accuracy. We also analyzed errors in action recognition and found primarily two different types.

Discussion

One type of error was on the semantic level, while the other consisted of reverting to the kinematic level of body part processing without any attribution of semantics. We elaborate on these results in the context of naturalistic action perception.

Actions are characterized by 'canonical moments' in a sequence of movements

Understanding what others are doing is an essential aspect of social cognition that depends on our ability to quickly recognize and categorize their actions. To effectively study action recognition we need to understand how actions are bounded, where they start and where they end. Here we borrow a conceptual approach - the notion of 'canonicality' - introduced by Palmer and colleagues in their study of object recognition and apply it to the study of action recognition. Using a set of 50 video clips sourced from stock photography sites, we show that many everyday actions - transitive and intransitive, social and non-social, communicative - are characterized by 'canonical moments' in a sequence of movements that are agreed by participants to 'best represent' a named action, as indicated in a forced choice (Exp 1, n = 142) and a free choice (Exp 2, n = 125) paradigm. In Exp 3 (n = 102) we confirm that canonical moments from action sequences are more readily named as depicting specific actions and, mirroring research in object recognition, that such canonical moments are privileged in memory (Exp 4, n = 95). We suggest that 'canonical moments', being those that convey maximal information about human actions, are integral to the representation of human action.1.

Modality-specific brain representations during automatic processing of face, voice and body expressions

A central question in affective science and one that is relevant for its clinical applications is how emotions provided by different stimuli are experienced and represented in the brain. Following the traditional view emotional signals are recognized with the help of emotion concepts that are typically used in descriptions of mental states and emotional experiences, irrespective of the sensory modality. This perspective motivated the search for abstract representations of emotions in the brain, shared across variations in stimulus type (face, body, voice) and sensory origin (visual, auditory). On the other hand, emotion signals like for example an aggressive gesture, trigger rapid automatic behavioral responses and this may take place before or independently of full abstract representation of the emotion. This pleads in favor specific emotion signals that may trigger rapid adaptative behavior only by mobilizing modality and stimulus specific brain representations without relying on higher order abstract emotion categories. To test this hypothesis, we presented participants with naturalistic dynamic emotion expressions of the face, the whole body, or the voice in a functional magnetic resonance (fMRI) study. To focus on automatic emotion processing and sidestep explicit concept-based emotion recognition, participants performed an unrelated target detection task presented in a different sensory modality than the stimulus. By using multivariate analyses to assess neural activity patterns in response to the different stimulus types, we reveal a stimulus category and modality specific brain organization of affective signals. Our findings are consistent with the notion that under ecological conditions emotion expressions of the face, body and voice may have different functional roles in triggering rapid adaptive behavior, even if when viewed from an abstract conceptual vantage point, they may all exemplify the same emotion. This has implications for a neuroethologically grounded emotion research program that should start from detailed behavioral observations of how face, body, and voice expressions function in naturalistic contexts.

Multi-view emotional expressions dataset using 2D pose estimation

Human body expressions convey emotional shifts and intentions of action and, in some cases, are even more effective than other emotion models. Despite many datasets of body expressions incorporating motion capture available, there is a lack of more widely distributed datasets regarding naturalized body expressions based on the 2D video. In this paper, therefore, we report the multi-view emotional expressions dataset (MEED) using 2D pose estimation. Twenty-two actors presented six emotional (anger, disgust, fear, happiness, sadness, surprise) and neutral body movements from three viewpoints (left, front, right). A total of 4102 videos were captured. The MEED consists of the corresponding pose estimation results (i.e., 397,809 PNG files and 397,809 JSON files). The size of MEED exceeds 150 GB. We believe this dataset will benefit the research in various fields, including affective computing, human-computer interaction, social neuroscience, and psychiatry.

Stage 2: Visual information and communication context as modulators of interpersonal coordination in face-to-face and videoconference-based interactions

Interpersonal coordination of body movement-or similarity in patterning and timing of body movement between interaction partners-is well documented in face-to-face (FTF) conversation. Here, we investigated the degree to which interpersonal coordination is impacted by the amount of visual information available and the type of interaction conversation partners are having. To do so within a naturalistic context, we took advantage of the increased familiarity with videoconferencing (VC) platforms and with limited visual information in FTF conversation due to the COVID-19 pandemic. Pairs of participants communicated in one of three ways: FTF in a laboratory setting while socially distanced and wearing face masks; VC in a laboratory setting with a view of one another's full movements; or VC in a remote setting with a view of one another's face and shoulders. Each pair held three conversations: affiliative, argumentative, and cooperative task-based. We quantified interpersonal coordination as the relationship between the two participants' overall body movement using nonlinear time series analyses. Coordination changed as a function of the contextual constraints, and these constraints interacted with coordination patterns to affect subjective conversation outcomes. Importantly, we found patterns of results that were distinct from previous research; we hypothesize that these differences may be due to changes in the broader social context from COVID-19. Taken together, our results are consistent with a dynamical systems view of social phenomena, with interpersonal coordination emerging from the interaction between components, constraints, and history of the system.

Construction and validation of the Dalian emotional movement open-source set (DEMOS)

Human body movements are important for emotion recognition and social communication and have received extensive attention from researchers. In this field, emotional biological motion stimuli, as depicted by point-light displays, are widely used. However, the number of stimuli in the existing material library is small, and there is a lack of standardized indicators, which subsequently limits experimental design and conduction. Therefore, based on our prior kinematic dataset, we constructed the Dalian Emotional Movement Open-source Set (DEMOS) using computational modeling. The DEMOS has three views (i.e., frontal 0°, left 45°, and left 90°) and in total comprises 2664 high-quality videos of emotional biological motion, each displaying happiness, sadness, anger, fear, disgust, and neutral. All stimuli were validated in terms of recognition accuracy, emotional intensity, and subjective movement. The objective movement for each expression was also calculated. The DEMOS can be downloaded for free from https://osf.io/83fst/ . To our knowledge, this is the largest multi-view emotional biological motion set based on the whole body. The DEMOS can be applied in many fields, including affective computing, social cognition, and psychiatry.

Voxelwise encoding models of body stimuli reveal a representational gradient from low-level visual features to postural features in occipitotemporal cortex

Research on body representation in the brain has focused on category-specific representation, using fMRI to investigate the response pattern to body stimuli in occipitotemporal cortex without so far addressing the issue of the specific computations involved in body selective regions, only defined by higher order category selectivity. This study used ultra-high field fMRI and banded ridge regression to investigate the coding of body images, by comparing the performance of three encoding models in predicting brain activity in occipitotemporal cortex and specifically the extrastriate body area (EBA). Our results suggest that bodies are encoded in occipitotemporal cortex and in the EBA according to a combination of low-level visual features and postural features.

To switch or not? Effects of spokes-character urgency during the social app loading process and app type on user switching intention

Users of mobile phone applications (apps) often have to wait for the pages of apps to load, a process that substantially affects user experience. Based on the Attentional Gate Model and Emotional Contagion Theory, this paper explores the effects of the urgency expressed by a spokes-character's movement in the loading page of a social app the app type on users' switching intention through two studies. In Study 1 (N = 173), the results demonstrated that for a hedonic-orientated app, a high-urgency (vs. low-urgency) spokes-character resulted in a lower switching intention, whereas the opposite occurred for a utilitarian-orientated app. We adopted a similar methodology in Study 2 (N = 182) and the results showed that perceived waiting time mediated the interaction effect demonstrated in Study 1. Specifically, for the hedonic-orientated (vs. utilitarian-orientated) social app, the high-urgency (vs. low-urgency) spokes-character made participants estimate a shorter perceived waiting time, which induces a lower user switching intention. This paper contributes to the literature on emotion, spokes-characters, and human-computer interaction, which extends an enhanced understanding of users' perception during loading process and informs the design of spokes-characters for the loading pages of apps.

A large-scale brain network of species-specific dynamic human body perception

This ultrahigh field 7 T fMRI study addressed the question of whether there exists a core network of brain areas at the service of different aspects of body perception. Participants viewed naturalistic videos of monkey and human faces, bodies, and objects along with mosaic-scrambled videos for control of low-level features. Independent component analysis (ICA) based network analysis was conducted to find body and species modulations at both the voxel and the network levels. Among the body areas, the highest species selectivity was found in the middle frontal gyrus and amygdala. Two large-scale networks were highly selective to bodies, dominated by the lateral occipital cortex and right superior temporal sulcus (STS) respectively. The right STS network showed high species selectivity, and its significant human body-induced node connectivity was focused around the extrastriate body area (EBA), STS, temporoparietal junction (TPJ), premotor cortex, and inferior frontal gyrus (IFG). The human body-specific network discovered here may serve as a brain-wide internal model of the human body serving as an entry point for a variety of processes relying on body descriptions as part of their more specific categorization, action, or expression recognition functions.

Emotion is perceived accurately from isolated body parts, especially hands

Body posture and configuration provide important visual cues about the emotion states of other people. We know that bodily form is processed holistically, however, emotion recognition may depend on different mechanisms; certain body parts, such as the hands, may be especially important for perceiving emotion. This study therefore compared participants' emotion recognition performance when shown images of full bodies, or of isolated hands, arms, heads and torsos. Across three experiments, emotion recognition accuracy was above chance for all body parts. While emotions were recognized most accurately from full bodies, recognition performance from the hands was more accurate than for other body parts. Representational similarity analysis further showed that the pattern of errors for the hands was related to that for full bodies. Performance was reduced when stimuli were inverted, showing a clear body inversion effect. The high performance for hands was not due only to the fact that there are two hands, as performance remained well above chance even when just one hand was shown. These results demonstrate that emotions can be decoded from body parts. Furthermore, certain features, such as the hands, are more important to emotion perception than others. STATEMENT OF RELEVANCE: Successful social interaction relies on accurately perceiving emotional information from others. Bodies provide an abundance of emotion cues; however, the way in which emotional bodies and body parts are perceived is unclear. We investigated this perceptual process by comparing emotion recognition for body parts with that for full bodies. Crucially, we found that while emotions were most accurately recognized from full bodies, emotions were also classified accurately when images of isolated hands, arms, heads and torsos were seen. Of the body parts shown, emotion recognition from the hands was most accurate. Furthermore, shared patterns of emotion classification for hands and full bodies suggested that emotion recognition mechanisms are shared for full bodies and body parts. That the hands are key to emotion perception is important evidence in its own right. It could also be applied to interventions for individuals who find it difficult to read emotions from faces and bodies.

Emotional body expressions facilitate working memory: Evidence from the n‐back task

In daily life, individuals need to recognize and update emotional information from others' changing body expressions. However, whether emotional bodies can enhance working memory (WM) remains unknown. In the present study, participants completed a modified n-back task, in which they were required to indicate whether a presented image of an emotional body matched that of an item displayed before each block (0-back) or two positions previously (2-back). Each block comprised only fear, happiness, or neutral. We found that in the 0-back trials, when compared with neutral body expressions, the participants took less time and showed comparable ceiling effects for accuracy in happy bodies followed by fearful bodies. When WM load increased to 2-back, both fearful and happy bodies significantly facilitated WM performance (i.e., faster reaction time and higher accuracy) relative to neutral conditions. In summary, the current findings reveal the enhancement effect of emotional body expressions on WM and highlight the importance of emotional action information in WM.

A functional magnetic resonance imaging examination of audiovisual observation of a point-light string quartet using intersubject correlation and physical feature analysis

We use functional Magnetic Resonance Imaging (fMRI) to explore synchronized neural responses between observers of audiovisual presentation of a string quartet performance during free viewing. Audio presentation was accompanied by visual presentation of the string quartet as stick figures observed from a static viewpoint. Brain data from 18 musical novices were obtained during audiovisual presentation of a 116 s performance of the allegro of String Quartet, No. 14 in D minor by Schubert played by the 'Quartetto di Cremona.' These data were analyzed using intersubject correlation (ISC). Results showed extensive ISC in auditory and visual areas as well as parietal cortex, frontal cortex and subcortical areas including the medial geniculate and basal ganglia (putamen). These results from a single fixed viewpoint of multiple musicians are greater than previous reports of ISC from unstructured group activity but are broadly consistent with related research that used ISC to explore listening to music or watching solo dance. A feature analysis examining the relationship between brain activity and physical features of the auditory and visual signals yielded findings of a large proportion of activity related to auditory and visual processing, particularly in the superior temporal gyrus (STG) as well as midbrain areas. Motor areas were also involved, potentially as a result of watching motion from the stick figure display of musicians in the string quartet. These results reveal involvement of areas such as the putamen in processing complex musical performance and highlight the potential of using brief naturalistic stimuli to localize distinct brain areas and elucidate potential mechanisms underlying multisensory integration.

Decoding spatiotemporal features of emotional body language in social interactions

How are emotions perceived through human body language in social interactions? This study used point-light displays of human interactions portraying emotional scenes (1) to examine quantitative intrapersonal kinematic and postural body configurations, (2) to calculate interaction-specific parameters of these interactions, and (3) to analyze how far both contribute to the perception of an emotion category (i.e. anger, sadness, happiness or affection) as well as to the perception of emotional valence. By using ANOVA and classification trees, we investigated emotion-specific differences in the calculated parameters. We further applied representational similarity analyses to determine how perceptual ratings relate to intra- and interpersonal features of the observed scene. Results showed that within an interaction, intrapersonal kinematic cues corresponded to emotion category ratings, whereas postural cues reflected valence ratings. Perception of emotion category was also driven by interpersonal orientation, proxemics, the time spent in the personal space of the counterpart, and the motion–energy balance between interacting people. Furthermore, motion–energy balance and orientation relate to valence ratings. Thus, features of emotional body language are connected with the emotional content of an observed scene and people make use of the observed emotionally expressive body language and interpersonal coordination to infer emotional content of interactions.

Mirror neurons 30 years later: implications and applications

Mirror neurons (MNs) were first described in a seminal paper in 1992 as a class of monkey premotor cells discharging during both action execution and observation. Despite their debated origin and function, recent studies in several species, from birds to humans, revealed that beyond MNs properly so called, a variety of cell types distributed among multiple motor, sensory, and emotional brain areas form a 'mirror mechanism' more complex and flexible than originally thought, which has an evolutionarily conserved role in social interaction. Here, we trace the current limits and envisage the future trends of this discovery, showing that it inspired translational research and the development of new neurorehabilitation approaches, and constitutes a point of no return in social and affective neuroscience.

More Than the Face: Representations of Bodies in the Inferior Temporal Cortex

Visual representations of bodies, in addition to those of faces, contribute to the recognition of con- and heterospecifics, to action recognition, and to nonverbal communication. Despite its importance, the neural basis of the visual analysis of bodies has been less studied than that of faces. In this article, I review what is known about the neural processing of bodies, focusing on the macaque temporal visual cortex. Early single-unit recording work suggested that the temporal visual cortex contains representations of body parts and bodies, with the dorsal bank of the superior temporal sulcus representing bodily actions. Subsequent functional magnetic resonance imaging studies in both humans and monkeys showed several temporal cortical regions that are strongly activated by bodies. Single-unit recordings in the macaque body patches suggest that these represent mainly body shape features. More anterior patches show a greater viewpoint-tolerant selectivity for body features, which may reflect a processing principle shared with other object categories, including faces. Expected final online publication date for the Annual Review of Vision Science, Volume 8 is September 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Correlated expression of the body, face, and voice during character portrayal in actors

Actors are required to engage in multimodal modulations of their body, face, and voice in order to create a holistic portrayal of a character during performance. We present here the first trimodal analysis, to our knowledge, of the process of character portrayal in professional actors. The actors portrayed a series of stock characters (e.g., king, bully) that were organized according to a predictive scheme based on the two orthogonal personality dimensions of assertiveness and cooperativeness. We used 3D motion capture technology to analyze the relative expansion/contraction of 6 body segments across the head, torso, arms, and hands. We compared this with previous results for these portrayals for 4 segments of facial expression and the vocal parameters of pitch and loudness. The results demonstrated significant cross-modal correlations for character assertiveness (but not cooperativeness), as manifested collectively in a straightening of the head and torso, expansion of the arms and hands, lowering of the jaw, and a rise in vocal pitch and loudness. These results demonstrate what communication theorists refer to as “multichannel reinforcement”. We discuss this reinforcement in light of both acting theories and theories of human communication more generally.

The characterization of actions at the superordinate, basic and subordinate level

Objects can be categorized at different levels of abstraction, ranging from the superordinate (e.g., fruit) and the basic (e.g., apple) to the subordinate level (e.g., golden delicious). The basic level is assumed to play a key role in categorization, e.g., in terms of the number of features used to describe these actions and the speed of processing. To which degree do these principles also apply to the categorization of observed actions? To address this question, we first selected a range of actions at the superordinate (e.g., locomotion), basic (e.g., to swim) and subordinate level (e.g., to swim breaststroke), using verbal material (Experiments 1-3). Experiments 4-6 aimed to determine the characteristics of these actions across the three taxonomic levels. Using a feature listing paradigm (Experiment 4), we determined the number of features that were provided by at least six out of twenty participants (common features), separately for the three different levels. In addition, we examined the number of shared (i.e., provided for more than one category) and distinct (i.e., provided for one category only) features. Participants produced the highest number of common features for actions at the basic level. Actions at the subordinate level shared more features with other actions at the same level than those at the superordinate level. Actions at the superordinate and basic level were described with more distinct features compared to those provided at the subordinate level. Using an auditory priming paradigm (Experiment 5), we observed that participants responded faster to action images preceded by a matching auditory cue corresponding to the basic and subordinate level, but not for superordinate level cues, suggesting that the basic level is the most abstract level at which verbal cues facilitate the processing of an upcoming action. Using a category verification task (Experiment 6), we found that participants were faster and more accurate to verify action categories (depicted as images) at the basic and subordinate level in comparison to the superordinate level. Together, in line with the object categorization literature, our results suggest that information about action categories is maximized at the basic level.

Decoding the difference between explicit and implicit body expression representation in high level visual, prefrontal and inferior parietal cortex

Recent studies provide an increasing understanding of how visual objects categories like faces or bodies are represented in the brain and also raised the question whether a category based or more dynamic network inspired models are more powerful. Two important and so far sidestepped issues in this debate are, first, how major category attributes like the emotional expression directly influence category representation and second, whether category and attribute representation are sensitive to task demands. This study investigated the impact of a crucial category attribute like emotional expression on category area activity and whether this varies with the participants' task. Using (fMRI) we measured BOLD responses while participants viewed whole body expressions and performed either an explicit (emotion) or an implicit (shape) recognition task. Our results based on multivariate methods show that the type of task is the strongest determinant of brain activity and can be decoded in EBA, VLPFC and IPL. Brain activity was higher for the explicit task condition in VLPFC and was not emotion specific. This pattern suggests that during explicit recognition of the body expression, body category representation may be strengthened, and emotion and action related activity suppressed. Taken together these results stress the importance of the task and of the role of category attributes for understanding the functional organization of high level visual cortex.

Bridging the gap between emotion and joint action

Our daily human life is filled with a myriad of joint action moments, be it children playing, adults working together (i.e., team sports), or strangers navigating through a crowd. Joint action brings individuals (and embodiment of their emotions) together, in space and in time. Yet little is known about how individual emotions propagate through embodied presence in a group, and how joint action changes individual emotion. In fact, the multi-agent component is largely missing from neuroscience-based approaches to emotion, and reversely joint action research has not found a way yet to include emotion as one of the key parameters to model socio-motor interaction. In this review, we first identify the gap and then stockpile evidence showing strong entanglement between emotion and acting together from various branches of sciences. We propose an integrative approach to bridge the gap, highlight five research avenues to do so in behavioral neuroscience and digital sciences, and address some of the key challenges in the area faced by modern societies.