==== Front Sci Rep Sci Rep Scientific Reports 2045-2322 Nature Publishing Group UK London 77951 10.1038/s41598-020-77951-w Article Augmenting aesthetic chills using a wearable prosthesis improves their downstream effects on reward and social cognition Haar A. J. H. adamjhh@mit.edu 1 Jain A. 1 Schoeller F. 12 Maes P. 1 1 grid.116068.80000 0001 2341 2786Fluid Interfaces Group, Media Lab, Massachusetts Institute of Technology, Cambridge, USA 2 Centre de Recherches Interdisciplinaires, Paris, France 10 12 2020 10 12 2020 2020 10 216033 11 2019 16 11 2020 © The Author(s) 2020Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.Previous studies on aesthetic chills (i.e., psychogenic shivers) demonstrate their positive effects on stress, pleasure, and social cognition. We tested whether we could artificially enhance this emotion and its downstream effects by intervening on its somatic markers using wearable technology. We built a device generating cold and vibrotactile sensations down the spine of subjects in temporal conjunction with a chill-eliciting audiovisual stimulus, enhancing the somatosensation of cold underlying aesthetic chills. Results suggest that participants wearing the device experienced significantly more chills, and chills of greater intensity. Further, these subjects reported sharing the feelings expressed in the stimulus to a greater degree, and felt more pleasure during the experience. These preliminary results demonstrate that emotion prosthetics and somatosensory interfaces offer new possibilities of modulating human emotions from the bottom-up (body to mind). Future challenges will include testing the device on a larger sample and diversifying the type of stimuli to account for negatively valenced chills and intercultural differences. Interoceptive technologies offer a new paradigm for affective neuroscience, allowing controlled intervention on conscious feelings and their downstream effects on higher-order cognition. Subject terms EmotionRewardSomatosensory systemHuman behaviourissue-copyright-statement© The Author(s) 2020 ==== Body Introduction Body perception plays a critical role in making emotional signals accessible to consciousness through feelings and their associated embodied effects1–3. Internal representations of the external world are actively mediated by physical feelings from within the body (i.e., interoception). The theory of embodied predictive coding suggests that feelings indicate a process of continually updating self-generated predictions about the probable causes of sensory input4–6. This suggests new opportunities for intervening on human affect through the controlled stimulation of interoceptive signals and somatosensation. The actuation of human emotion via bodily stimulation has been suggested as an ideal case for investigating interoception and designing clinical interventions7,8. Here, we investigated embodied predictive coding by modulating bodily sensations underlying aesthetic chills (i.e., goosebumps, psychogenic shivers), using a device modulating cold and shiver sensations down the spine characteristic of this peak emotion. Our study is based on the paradigm of interoceptive illusions or misattribution of arousal9, wherein external stimulation, such as temperature change or increased heart rate, modulate the interoceptive inferences underlying decision-making to trigger predictable behavioral change10,11. Interoceptive illusions have attracted some attention in recent years as experimental tools for interoceptive neuroscience11. In a seminal study, Valins et al. demonstrated that increasing heart rate via mere physical exercise facilitates romantic attraction to confederates9. This is attributed to the misinterpretation of increased heart rate as due to attraction, as opposed to exercise12. Other experiments have manipulated various forms of physiological emotional feedback, including facial muscles13–15. The modulation of interoceptive inferences through the controlled simulation of emotional somatic markers has shown promising results in reducing stress16 and fear17. Recent evidence in favor of the model of embodied predictive coding suggests a causal relation between altering interoception and corresponding changes in emotion10,11. Modulating somatic markers and altering interoceptive inference during the phase of emotion experience may allow us to disentangle historical key questions in the field of affective neuroscience, namely the problem of causality and dynamics in emotional networks, and in turn suggest new interventions for somatic disorders18. However, interoceptive illusion studies are generally based on physiological markers lacking in universality. For example, a smile conveys various meanings across cultures and communities19. In such cases most studies failed to replicate and thereby scientific paradigms for bottom-up influences of the body on higher cognitive processes remain opaque20. The aim of this study is to test the possibility of modulating cognition from the bottom-up (from body to mind), through actuation of a device designed to simulate and stimulate aesthetic chills. Aesthetic chills are a somatic marker for individual emotional peaks21 described as a highly pleasurable tingling sensation down the spine22. These are most often elicited by music23,24, but also poetry25, scientific insights26 or social rituals26. These events have deep significance to individuals, often related to life changing experience such as the overview effect27 and transformative experiences28,29, and are sometimes accompanied by tears29. In terms of physiology, chills have a specific cardiac signature30 and are primarily accompanied by an increase of phasic electrodermal activity and respiration depth31. Chills have been described as a physiological marker of salience26, a self signalling mechanism allowing the cognitive system to orient its attentional resources toward evolutionary relevant stimuli26,32. Furthermore, negatively valenced aversive chills have also been observed in relation to aversive auditory stimulation or traumatic stimuli33,34. Both positive and negative chills engage neural populations coding for salience such as amygdaloid complex, ventromedial prefrontal cortex and the nucleus accumbens22,34. Chills have been related to various complex emotional states such as awe35,36, insight37, prosocial emotions38, and being moved25. Their downstream effects include positive effects on social cognition37,39 and promotion of altruistic behavior38,40. Schoeller and Perlovsky have put forward a theory of chills as a satiation of an internal drive for knowledge26,41. Chills would thus correspond to a sudden acceleration of learning42–45 described formally in terms of an event when the rate of change of a learning function tends towards zero26,37,39,41. This account is coherent with current accounts of emotional valence in terms of error dynamics46,47. Crucially, and even though their prevalence across human populations is still an open question, psychogenic shivers seem to present a high degree of universality, making them a useful somatic marker for affective neuroscience in light of their myriad emotional links38,43,44,48,49. As a salient interoceptive inference and somatic marker of a peak emotion, chills are an ideal case for disentangling the role of conscious and unconscious evaluations of bodily signals during emotional processes. Here, we tested whether we could artificially induce peak emotion by stimulating chills with spatial and temporal precision using a wearable prosthesis designed for this purpose. The prosthesis delivers cold and vibration stimulus at critical reference points of an audiovisual stimulus of awe-provoking images of the planet Earth from above accompanied with a chills-eliciting speech and music (see “Methods” section and “Supplementary Information”). Through this controlled stimulation, we aimed to amplify the sensory signals serving as a basis of the feeling of chills, and replicate some of the downstream effects. We therefore predicted that the device would increase the frequency and intensity of chills, and trigger changes in terms of pleasure and social cognition, which are known downstream effects for aesthetic chills40,41. To get at the underlying psychology driving the positive relationship between aesthetic chills and altruism, we chose to distinguish between the two broad categories of emotional and cognitive social processes which are likely factors: emotional contagion (i.e., the degree to which one party shares the feelings of another, separate from the ability to accurately intuit what another person is feeling) and cognitive empathy (i.e., the degree to which one share’s somebody’s viewpoint)50–52. Results Subjects (n = 21) reported number of chills on a range from 0–10. Participants experienced a higher number of chills while wearing the device (M = 4.00; STD = 2.90) than without the device (M = 2.76; SD = 2.86) (Fig. 1). A pairwise one-tailed t-test revealed that the device significantly increases the frequency of chills (t = 3.2274, df = 20, p = 0.0021) and Shapiro–Wilk normality test (W = 0.96964, p = 0.7251) does not reject the null hypothesis of data normality. Participants also reported a higher intensity of chills while wearing the device (M = 5.62; SD = 2.44) than without the device (M = 4.33, SD = 2.57) (Fig. 1). Since reported intensity is a likert scale non parametric data, we used a pairwise one-tailed Wilcoxon test to assess the difference and found a significant difference in chills intensity across groups (V = 104, p = 0.0327) (Table 1). Hence, at any significance level, this test rejects the null hypothesis of independence between conditions and chills frequency and intensity.Figure 1 Number and Intensity of chills per conditions: participants wearing the device reported experiencing significantly greater number and intensity of chills during the experience with the device than during the experience without the device. Table 1 Comparison of the intensity of chills, sharing of the speaker’s feelings, sharing of the speaker’s viewpoint, and perceived pleasure with and without the device. Measure With device Without device Pairwise one-sided Wilcoxon Intensity of chills 5.62 ± 2.44 4.33 ± 2.57 V = 104, p* = 0.0327 Pleasure 7.66 ± 1.28 6.95 ± 1.66 V = 38.5, p* = 0.0321 Cognitive empathy 8.33 ± 1.15 8.09 ± 1.79 V = 18.5, p = 0.5 Emotional contagion 8.00 ± 1.73 7.19 ± 2.40 V = 76, p* = 0.0147 We then considered the physiological data from the camera and cardiac sensors. Some of the physiological data was lost after collection and the remaining data did not reveal any significant difference across groups. We first examined the camera data for the remaining sample (N = 11), and found a higher count of smiles in the recordings for the participants wearing the device (M = 1.18; STD = 1.78) than participants without (M = 0.27; STD = 0.65). A pairwise two-sided Wilcoxon test revealed that this difference is not significant (V = 51, p = 0.3804). We then analysed the heart rate data for the remaining sample (N = 12), and found an increase in the LF/HF ratio for participants with the device (M = 2.23; STD = 4.18) than without (M = 1.75; STD = 2.98). A pairwise two-sided Wilcoxon test revealed this difference is not significant (V = 12.5, p = 0.2228). Hence, these marginal differences do not allow us to reject the null hypothesis for physiological differences in facial expression or heart rate frequency data across groups. We then analyzed the downstream effects of the device on pleasure and empathy as reported by participants (see Table 1 and Fig. 2). Participants reported greater emotional contagion while wearing the device (M = 8.00 SD = 1.73) than without the device (M = 7.19; SD = 2.40). A Shapiro–Wilk normality test rejected the null hypothesis of of data normality (W = 0.79563, p = 0.0006), therefore we used a non-parametric pairwise one-sided Wilcoxon to test the difference across group, which confirmed a significant difference in empathy across groups (V = 76, p = 0.0147). We also analyzed reports of cognitive empathy (i.e., shared viewpoints). Participants wearing the device reported sharing the speaker’s viewpoint (M = 8.33; STD = 1.15) more than when viewing the stimulus without the device (M = 8.09; STD = 1.79). However, a pairwise one-sided Wilcoxon test revealed the difference to be not significant (V = 18.5, p = 0.5). Participants with the device also reported a greater amount of pleasure in the viewing experience (M = 7.66; SD = 1.28) than those without the device (M = 6.95; SD = 1.66). As this data does not seem to follow a normal distribution (W = 0.81652, p = 0.0012), we used a pairwise one-sided Wilcoxon test to test for differences across groups, which revealed a significant difference (V = 38.5, p = 0.032). Hence, we can reject the null hypothesis of independence between conditions and stimulus-elicited pleasure and emotional contagion, but not for cognitive empathy across groups.Figure 2 Empathy and pleasure of chills per conditions: participants wearing the device reported experiencing significantly greater emotional contagion and more pleasure during the experience with the device than during the experience without the device. To understand the relationship of the body-worn device to reported emotions, we further investigated participant’s subjective bodily localization of emotion and perceived point of generation of the felt emotion (see Table 2). Participants reported whether their emotions were felt in the body or mind (on a 0–10 likert scale where 0 corresponded to feeling in the body and 10 to in the mind, Fig. 3) and also whether the emotion was generated internally or externally (where 0 corresponded to reports of emotion generated inside the body and 10 to generated outside the body, Fig. 4). We found no significant differences across conditions. Lastly, we asked participants to rate their confidence regarding their understanding of the speech (on a likert scale from 0 to 10 where 1 corresponded to not confident at all and 10 corresponded to extremely confident). We found no significant difference across conditions in reported confidence.Table 2 Comparison of the viewer’s confidence regarding understanding of the stimulus, perception of emotion generation inside or outside the body, and perception of emotion felt in the body or the mind across device conditions. Measure With device Without device Pairwise two-sided Wilcoxon Confidence 8.00 ± 1.73 8.33 ± 1.62 V = 14, p = 0.1738 Emotion generated in the body or the mind 5.95 ± 2.27 6.14 ± 2.41 V = 48, p = 0.5059 Emotion generated inside or outside the body 4.57 ± 2.80 4.05 ± 2.03 V = 64.5, p = 0.4682 Figure 3 Perception of the felt emotion (in the body or in the mind) reported by the participants. Figure 4 Perception of emotion generation (inside the body or outside the body) reported by the participants. Discussion This study investigated the artificial induction of aesthetic chills using a wearable prosthesis. We found that stimulation from the Frisson device significantly increased the frequency of chills experienced by the participants, as well as ratings of chill intensity. When examining the downstream effects of these augmented chills, we found that participants also reported significantly more empathy towards the speaker presented in the audiovisual stimulus (in terms of emotional contagion i.e. sharing the speaker’s feelings to a greater degree), and a significant increase in ratings of how pleasurable the video-watching experience was. We found that across conditions participants did not report significantly different confidence regarding understanding the stimulus, suggesting the downstream effects mentioned above are not tied to the percept of better understanding of the stimulus speech or the pleasure of understanding. Results suggest that artificially inducing chills may facilitate their downstream effects on social cognition40. Results further lend some credence to the predictive processing model of emotion as predicted interoceptive input4–6, as subjects with interoception of chills (synthetic or organic) were more likely to report the emotions associated with organic chills. Interestingly, subjects wearing the Frisson device did not report a significant difference in embodied localization or point of generation of emotions, suggesting the device itself did not play a large role in the reflection on felt emotion, though the chill it created indeed did. This in turn raises opportunities for future research into affective neuroscience and interoceptive inferences associated with somatosensory stimulus. The neuroscience of emotion has identified close correlational links between states of physiological and emotional arousal. These include relationships between pupil size, galvanic skin response, or heart rate with affective arousal53. Yet historically, multiple theories have posited an active, causal as opposed to correlational role for bodily states in the creation of emotion. William James famously hypothesized that felt emotion is “not a primary feeling, directly aroused by the exciting object or thought, but a secondary feeling indirectly aroused; the primary effect being the organic changes in question, which are immediate reflexes following upon the presence of the object”54. The present results motivate further testing for causal rather than correlational links between changes in bodily state and changes in affective state55. The complex interplay of cognitive and physiological factors involved in emotional processes makes order and causality difficult to disentangle in affective neuroscience56. Experiments in misattribution of arousal, wherein experimenters modulate a subject’s physiology in order to alter cognition, offer an initial probe9. Emotion prosthesis and somatosensory interfaces such as our Frisson device open new pathways for affective neuroscience, by allowing the augmentation or modulation of interoceptive inferences. New research tools allowing researchers to actuate embodied emotion as causal probes beyond the study of correlations are necessary for an effective and applied affective neuroscience. This exploratory study raises a number of significant questions for further research, including what neural mechanisms underlie the integration of artificial sensory information and how organic endogenous sensations are processed differently from exogenous synthetic stimulations. However, this study must also be understood in the light of its limitations. First, the sample sizes in this exploratory study are limited in scope and diversity. With a subject n = 21, the current study was likely powered to detect only very large effects and any effect sizes reported may be overestimated. This exploratory study should be replicated on a larger sample in a wide range of ecological settings including culturally relevant stimuli (which may vary cross-culturally) and across heterogenous human populations. Second, a next iteration of the device should allow for a sham back-worn device to limit expectation effects in the control condition to a strict minimum. Third, emotional contagion is only one element in the complex construct of empathy, and one could ‘catch’ feelings from a stimulus even while misinterpreting the emotion displayed therein. Finally, as chills appear to be independent of valence, further study should examine the potential of the device on modulating the downstream effects of negative chills as well. Emotion prosthetics and somatosensory interfaces offer new possibilities for affective neuroscience inducing human emotions from the bottom-up, modulating their associated metacognitive judgments (i.e., feelings) and downstream effects (i.e., action tendency) through interoceptive technology. The constant feedback loop between actual and expected sensations during interoceptive processing allows for intervention on higher cognitive functioning through controlled stimulation of body signals. Surprisingly little research has been devoted to the topic of how technology may be used to enhance or generate emotion from modulating interoceptive inferences and somatic markers (i.e., somatosensory interfaces). If confirmed in further studies, these results suggest that in the long term, emotion prosthetics and somatosensory interfaces may provide novel non-invasive tools for understanding interoception-related disorders as well18. Methods We used a within-subject design to investigate whether one can artificially modulate interoceptive inferences underlying aesthetic chills, their felt frequency and intensity. The hypothesis was that both psychological and physiological responses would be different between with and without the Frisson stimulation device. Experiments were conducted using self-report, a muscle-bend sensor allowing for silent report of chills experience, image capture for facial expression analysis, physiological sensors (heart rate and skin conductance) for measuring physiological changes concurrent with frisson, and both quantitative and qualitative surveys. Protocol The participant entered the laboratory between 12:00 and 4:00 p.m., sat in front of a computer monitor, and was provided with a consent form. The study and consent form provided were approved by the Institutional Review Board, the Committee on the Use of Humans as Experimental Subjects at MIT. Participants were told that the study examined the relationship between temperature and attention, and were specified the definition of aesthetic chills as psychogenic waves of cold as opposed to external cold stimulations from device, and asked to report only the former. The stimulation device was placed on their back and sensors were wrapped around the middle phalanges of their right hand’s index and middle finger. Participants were told to clench their hand if they experienced aesthetic chills at any time during the film, such that the handworn sensor could collect a count of chills. Next, each participant was exposed to a calming film of a cold landscape (ice mountains and cascade) for 90 s to control for stress baseline and prime the subject with the concept of cold. Next, the chill-eliciting video stimulus began, entailing a 213 s long speech and introductory message about the film content. Artificial chills were delivered at timecodes 2:43, 3:52 and 4:03. The same video viewing procedure was done with or without the Frisson back-worn stimulation device, depending on condition, but participants wore hand-worn physiological sensors regardless of condition. We chose to remove the back-worn device in the control condition in order to avoid any transient cold temperature from the device. To mitigate expectation effects differing across conditions, we presented the chills device as both the sensor on the back and actuator on the wrist, allowing us to remove the back-worn part of the device in the control condition but maintain that ‘the device’ remained on their body. We collected subjective data in the form of surveys after each condition which asked questions such as the frequency of chills experienced by the participant, the intensity of the chills experienced, degree to which subjects shared the speaker's viewpoint, and the degree to which subjects shared the speaker's feelings. The subjective questions were asked on a likert scale of 0–10 (see “Supplementary Information” for further details about the procedure, stimulus and questionnaire). Once the experiment was finished, the experimenter disconnected the sensors and provided the subject with a questionnaire. Finally, the subjects were thanked for their participation and fully debriefed. Each session lasted about 20 min. Participants A total of 21 students participated in the experiment (N females = 7, N males = 14, M age = 27, STD = 6.5). Sample size was determined based on a compiled exhaustive database of aesthetic chills research (see “Supplementary Information”). We determined that an acceptable sample size for an exploratory study of this kind ranged from 20 to 50. On their arrival at the laboratory waiting room, they were randomly assigned to one of the experimental conditions. Following a within subject design, each participant took part in both experimental conditions: once with and once without the device, each time exposed to the same audiovisual stimulus. The order was counterbalanced to account for learning effects. Materials The materials used in the experiment are in the following section. Stimulus The audiovisual stimulus was presented using a standard computer screen and headphones with a fixed volume. Following on a preliminary study using responses to a survey inquiring into the properties of chill-eliciting situations (see41) and a software for searching YouTube videos in terms of their density of chills-related comments, we designed the stimulus combining two modalities (audio and visual) likely to trigger chills in the studied population. The content for each modality was identified using a tool specifically developed for this purpose42, a software to automatically retrieve online content triggering chills on YouTube (as measured by the prevalence of chills related comments by users in the platform). Our search yielded a number of videos and we combined audio (speech and music) and visual components (images of earth) from the content most likely to trigger psychogenic shivers in the target population (the software sorts through networks of videos reflecting the human populations uploading them). Film audio tracks are more powerful than music in eliciting piloerection, a common marker of aesthetic chills31. The visual stimulus was a 4 K satellite view of planet Earth, a stimulus known to trigger awe and the overview effects likely to trigger chills27,35,36. As a voice over, we used a speech excerpt of Charlie Chaplin from the movie ‘The Great Dictator’ accompanied by a Hans Zimmer musical soundtrack. The film was subtitled in English. The introductory message that preceded the film was a block of text presenting the following: “What you are about to hear is the voice of Charlie Chaplin from the middle of World War II. For his entire career, he had been a silent actor. But in 1940, he decided to speak out. This is his message of hope”. For further discussion on contents likely to elicit chills, refer to8,26,37,39,41–45. Actuator A prototype was developed to deliver a thermal and vibrotactile feedback down the spine and imitate the sensation of chills. The device consisted of three peltier elements at different spatial locations: top of the back, one on middle and one on the lower back, and a BLE enabled control circuit board. We tested the device in a series of preliminary experiments, and improved the design based on participant feedback to reproduce the sensation of chills (see “Supplementary Information”). The final device delivered thermal feedback in a manner closely resembling the internal chill, a traversing cold temperature from top to bottom for a period of 3 s and a short burst of tingling vibration at the top of the back for 1 s (Fig. 5). The device was powered by a 2S 7.4 V 500 mAH Lipo battery, and was cast in silicone for easy attachment to the back of the participants. An Android mobile application was developed to activate the device with specific timing, delivering the chills in synchrony with the video stimulus.Figure 5 The Frisson prosthesis: a device delivering thermal feedback in a manner closely resembling to the internal chill, a traversing cold temperature from top to bottom. Sensors and software Physiological data was collected using hand worn sensors from the Dormio device57–59. We collected heart rate using a Sparkfun Pulse Sensor Amped PPG sensor from the middle finger and electrodermal activity using dry electrodes from the wrist. We also attached a flexion sensor on the index finger and asked participants to clench their index finger when experiencing a chill sensation. The data was sampled at 100 Hz. All the data was collected using Bluetooth to a PC. Facial expressions data was recorded using a camera mounted over the screen and stored in a SD card.We used Affdex SDK for detecting facial expressional from recorded video at 30fps. Affex SDK detected 12 relevant metrics from the video. The metrics were ‘smile’, ‘anger’, ‘valence’, ‘browFurrow’, ‘noseWrinkle’, ‘joy’, ‘surprise’, ‘browRaise’, ‘upperLipRaise’, ‘mouthOpen’, ‘eyeClosure’, ‘cheekRaise’. Ethics The experiment is in compliance with the Helsinki Declaration. The study was approved by the Committee on the Use of Humans as Experimental Subjects at MIT. All participants gave their voluntary informed consent and we followed the Ethics Code of the American Psychological Association. All participants were informed about the purpose of the research, about their right to decline to participate and to withdraw from the experiment and about the limits of confidentiality. We also provided them with a contact for any questions concerning the research and with the opportunity to ask any questions regarding the phenomenon under study (aesthetic chills) and receive appropriate answers. All participants reacted positively to the experiment and were thankful for the opportunity to learn about the phenomenon. Following the conclusion of the experiment, we engaged subjects in many discussions on the potential abuses inherent in technologies which are built to alter human emotion. As these devices become more subtle and more advanced, these issues will come clearly to the fore. We build our devices for impossibility of surreptitious use (i.e. subjects could not fail to notice their application) but are aware that continued conversation around ethical use of these devices and public education around potential for influence are necessary and fruitful. Reviewer disclosure Following the standard reviewer disclosure request endorsed by the Center for Open Science60, We confirm to have reported all measures, conditions, data exclusions and how we determined our sample sizes. Supplementary information Supplementary Information. Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: A. J. H. Haar, A. Jain and F. Schoeller. Supplementary information is available for this paper at 10.1038/s41598-020-77951-w. Acknowledgements The authors would like to thank Leonid Perlovsky and Karl Friston for providing valuable suggestions for the experimental design. Author contributions A.J.H. and A.J. conceived the original idea for the device. A.J.H., A.J., and F.S. conceived the experiments. A.J.H. and A.J. conducted the experiments. A.J. performed the analysis. P.M. supervised the project. All authors discussed the results and contributed to the final manuscript. Data availability Supporting data is available at public repository: https://doi.org/10.7910/DVN/E4ZYOT. Competing interests The authors declare no competing interests. ==== Refs References 1. Azzalini D Rebollo I Tallon-Baudry C Visceral signals shape brain dynamics and cognition Trends Cognit. Sci. 2019 23 488 509 10.1016/j.tics.2019.03.007 31047813 2. Garfinkel SN Critchley HD Threat and the body: How the heart supports fear processing Trends Cognit. Sci. 2016 20 34 46 10.1016/j.tics.2015.10.005 26628111 3. Seth AK Critchley HD Extending predictive processing to the body: Emotion as interoceptive inference Behav. Brain Sci. 2013 36 227 228 10.1017/S0140525X12002270 23663284 4. Pezzulo G Why do you fear the bogeyman? An embodied predictive coding model of perceptual inference Cognit. Affect. Behav. Neurosci. 2013 14 902 911 10.3758/s13415-013-0227-x 5. Seth AK Interoceptive inference, emotion, and the embodied self Trends Cogn. Sci. 2013 17 565 573 10.1016/j.tics.2013.09.007 24126130 6. Seth AK Suzuki K Critchley HD An interoceptive predictive coding model of conscious presence Front. Psychol. 2012 10.3389/fpsyg.2011.00395 22416238 7. Costa J Adams AT Jung MF Guimbretière F Choudhury T EmotionCheck: A wearable device to regulate anxiety through false heart rate feedback GetMobile Mobile Comput. Commun. 2017 21 22 25 10.1145/3131214.3131222 8. Schoeller F Haar AJH Jain A Maes P Enhancing human emotions with interoceptive technologies Phys. Life Rev. 2019 31 310 319 10.1016/j.plrev.2019.10.008 31757602 9. Valins S Cognitive effects of false heart-rate feedback J. Pers. Soc. Psychol. 1966 4 400 408 10.1037/h0023791 5969996 10. Iodice P Porciello G Bufalari I Barca L Pezzulo G An interoceptive illusion of effort induced by false heart-rate feedback Proc. Natl. Acad. Sci. 2019 116 13897 13902 10.1073/pnas.1821032116 31235576 11. Pezzulo G Increased heart rate facilitates the processing of fearful but not disgusted faces Sci. Rep. 2018 10.1038/s41598-017-18761-5 29330467 12. White GL Fishbein S Rutsein J Passionate love and the misattribution of arousal J. Pers. Soc. Psychol. 1981 41 56 62 10.1037/0022-3514.41.1.56 13. Flack W Peripheral feedback effects of facial expressions, bodily postures, and vocal expressions on emotional feelings Cogn. Emot. 2006 20 177 195 10.1080/02699930500359617 14. Laird JD Lacasse K Bodily influences on emotional feelings: Accumulating evidence and extensions of William James’s theory of emotion Emot. Rev. 2013 6 27 34 10.1177/1754073913494899 15. Strack F Martin LL Stepper S Inhibiting and facilitating conditions of the human smile: A nonobtrusive test of the facial feedback hypothesis J. Pers. Soc. Psychol. 1988 54 768 777 10.1037/0022-3514.54.5.768 3379579 16. Schulz A Interoception and stress Front. Psychol. 2015 10.3389/fpsyg.2015.00993 26257668 17. Van Diest, I., Davenport, P., Van den Bergh, O., Miller, S., & Robertson, E. Interoceptive fear conditioning to an inspiratory load using 20% CO2 inhalation as an unconditioned stimulus. in Paper presented at the 12th Annual Meeting of the International Society for the Advancement of Respiratory Psychophysiology (ISARP), (2005). 18. Khalsa SS Interoception and mental health: A roadmap Biol. Psychiatry Cognit. Neurosci. Neuroimaging 2018 3 501 513 10.1016/j.bpsc.2017.12.004 29884281 19. Ekman, P. ed. Darwin and facial expression: A century of research in review. Ishk. http://www.ishkbooks.com/mal (2006). 20. Strack F Reflection on the smiling registered replication report Perspect. Psychol. Sci. 2016 11 929 930 10.1177/1745691616674460 27784750 21. Goldstein A Thrills in response to music and other stimuli Physiol. Psychol. 1980 8 126 129 10.3758/BF03326460 22. Blood AJ Zatorre RJ Intensely pleasurable responses to music correlate with activity in brain regions implicated in reward and emotion Proc. Natl. Acad. Sci. 2001 98 11818 11823 10.1073/pnas.191355898 11573015 23. Grewe O Kopiez R Altenmuller E Chills as an indicator of individual emotional peaks Ann. N. Y. Acad. Sci. 2009 1169 351 354 10.1111/j.1749-6632.2009.04783.x 19673806 24. Panksepp J The emotional sources of “chills” induced by music Music Percept. Interdiscip. J. 1995 13 171 207 10.2307/40285693 25. Wassiliwizky E Koelsch S Wagner V Jacobsen T Menninghaus W The emotional power of poetry: Neural circuitry, psychophysiology and compositional principles Social Cognit. Affect. Neurosci. 2017 12 1229 1240 10.1093/scan/nsx069 26. Schoeller F Knowledge, curiosity, and aesthetic chills Front. Psychol. 2015 6 1546 10.3389/fpsyg.2015.01546 26539133 27. Stepanova ER Quesnel D Riecke BE Space—A virtual frontier: How to design and evaluate a virtual reality experience of the overview effect Front. Digital Human. 2019 10.3389/fdigh.2019.00007 28. Kitson, A. et al. Transformative experience design. in Extended Abstracts of the 2019 CHI Conference on Human Factors in Computing Systems (ACM, 2019). 10.1145/3290607.3311762 29. Mori K Iwanaga M Two types of peak emotional responses to music: The psychophysiology of chills and tears Sci. Rep. 2017 10.1038/srep46063 29269734 30. Sumpf M Jentschke S Koelsch S Effects of aesthetic chills on a cardiac signature of emotionality PLoS ONE 2015 10 e0130117 10.1371/journal.pone.0130117 26083383 31. Benedek M Kaernbach C Physiological correlates and emotional specificity of human piloerection Biol. Psychol. 2011 86 320 329 10.1016/j.biopsycho.2010.12.012 21276827 32. Maruskin LA Thrash TM Elliot AJ The chills as a psychological construct: Content universe, factor structure, affective composition, elicitors, trait antecedents, and consequences J. Pers. Soc. Psychol. 2012 103 135 157 10.1037/a0028117 22564010 33. Halpern L Blake R Hilenbrand J Psychoacoustics of a chilling sound Percept. Psychophys. 1986 39 77 80 10.3758/bf03211488 3725541 34. Zald DH Pardo JV The neural correlates of aversive auditory stimulation NeuroImage 2002 16 746 753 10.1006/nimg.2002.1115 12169258 35. Quesnel, D., Di Paola, S., & Riecke, B. E. Deep learning for classification of peak emotions within virtual reality systems. in International SERIES on Information Systems and Management in Creative eMedia (CreMedia)(2017/2), 6–11 (2018). 36. Quesnel D Riecke BE Are you awed yet? How virtual reality gives us awe and goose bumps Front. Psychol. 2018 10.3389/fpsyg.2018.02158 30473673 37. Schoeller F The shivers of knowledge Hum. Social Stud. 2015 4 26 41 10.1515/hssr-2015-0022 38. Zickfeld JH Kama muta: Conceptualizing and measuring the experience often labelled being moved across 19 nations and 15 languages Emotion 2019 19 402 424 10.1037/emo0000450 29888936 39. Schoeller F Perlovsky L Arseniev D Physics of mind: Experimental confirmations of theoretical predictions Phys. Life Rev. 2018 25 45 68 10.1016/j.plrev.2017.11.021 29398558 40. Fukui H Toyoshima K Chill-inducing music enhances altruism in humans Front. Psychol. 2014 5 1215 10.3389/fpsyg.2014.01215 25389411 41. Schoeller F Perlovsky L Aesthetic chills: Knowledge-acquisition, meaning-making, and aesthetic emotions Front. Psychol. 2016 7 1093 10.3389/fpsyg.2016.01093 27540366 42. Schoeller F The satiation of natural curiosity Int. J. Signs Semiotic Syst. 2016 5 27 34 10.4018/ijsss.2016070102 43. Schoeller F Eskinazi M Psychologie du frisson esthétique Psychol. Française 2019 64 305 312 10.1016/j.psfr.2017.11.003 44. Schoeller F Eskinazi M Garreau D Dynamics of the knowledge instinct: Effects of incoherence on the cognitive system Cognit. Syst. Res. 2018 47 85 91 10.1016/j.cogsys.2017.07.005 45. Schoeller F Perlovsky L Great expectations—Narratives and the elicitation of aesthetic chills Psychology 2015 6 2098 2102 10.4236/psych.2015.616205 46. Hesp, C. et al. Deeply felt affect: The emergence of valence in deep active inference. https://psyarxiv.com/62pfd/ (2019) 10.31234/osf.io/62pfd. 47. Kiverstein J Miller M Rietveld E The feeling of grip: Novelty, error dynamics, and the predictive brain Synthese 2019 196 2847 2869 10.1007/s11229-017-1583-9 48. McCrae RR Aesthetic chills as a universal marker of openness to experience Motivation Emot. 2007 31 5 11 10.1007/s11031-007-9053-1 49. Beier EJ Janata P Hulbert JC Ferreira F Do you chill when I chill? A cross-cultural study of strong emotional responses to music Psychol. Aesth. Creativity Arts 2020 10.1037/aca0000310 50. Batson, C. D. These things called empathy: Eight Related but distinct phenomena. in The Social Neuroscience of Empathy (The MIT Press, 2009). 10.7551/mitpress/9780262012973.003.0002. 51. Moore RC Dev SI Jeste DV Dziobek I Eyler LT Distinct neural correlates of emotional and cognitive empathy in older adults Psychiatry Res. Neuroimaging 2015 232 42 50 10.1016/j.pscychresns.2014.10.016 52. Hatfield, E., Rapson, R. L. & Le, Y.-C. L. Emotional contagion and empathy. in The Social Neuroscience of Empathy, 19–30 (The MIT Press, 2009). 10.7551/mitpress/9780262012973.003.0003. 53. Wang C-A Arousal effects on pupil size, heart rate, and skin conductance in an emotional face task Front. Neurol. 2018 10.3389/fneur.2018.01029 30713521 54. James W Discussion: The physical basis of emotion Psychol. Rev. 1894 1 516 10.1037/h0065078 55. Allen M Friston KJ From cognitivism to autopoiesis: Towards a computational framework for the embodied mind Synthese 2016 195 2459 2482 10.1007/s11229-016-1288-5 29887647 56. Pessoa L How do emotion and motivation direct executive control? Trends Cognit. Sci. 2009 13 160 166 10.1016/j.tics.2009.01.006 19285913 57. Horowitz, A.H., Grover, I., Reynolds-Cuéllar, P., Breazeal, C., & Maes, P. Dormio: Interfacing with dreams. in Extended Abstracts of the 2019 CHI Conference on Human Factors in Computing Systems, (ACM, 2019). 10.1145/3170427.3188403 58. Horowitz, A.H., Cunningham, T.J., Maes, P., & Stickgold, R. Dormio: A targeted dream incubation device. Conscious. Cognit.83, 102938 (2020). 59. Carr, M. et al. Dream engineering: Simulating worlds through sensory stimulation. Conscious. Cognit.83, 102955 (2020). 60. Nosek, B. A. et al. Standard reviewer statement for disclosure of sample, conditions, measures, and exclusions. (2017). osf.io/hadz3.