Content uploaded by Agneta H Fischer
Author content
All content in this area was uploaded by Agneta H Fischer on Nov 19, 2024
Content may be subject to copyright.
Available via license: CC BY 4.0
Content may be subject to copyright.
OliverGenschow
EmielCraccoEditors
Automatic
Imitation
Automatic Imitation
Oliver Genschow • Emiel Cracco
Editors
Automatic Imitation
ISBN 978-3-031-62633-3 ISBN 978-3-031-62634-0 (eBook)
https://doi.org/10.1007/978-3-031-62634-0
© The Editor(s) (if applicable) and The Author(s) 2025
This book is an open access publication.
Open Access This book is licensed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), 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 license and indicate if
changes were made.
The images or other third party material in this book are included in the book’s Creative Commons
license, unless indicated otherwise in a credit line to the material. If material is not included in the book’s
Creative Commons license 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.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specic statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the
editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional afliations.
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
If disposing of this product, please recycle the paper.
Editors
Oliver Genschow
School of Management and Technology
Leuphana University Lüneburg
Lüneburg, Germany
Emiel Cracco
Department of Experimental Clinical
and Health Psychology
Ghent University
Ghent, Belgium
v
“Genschow and Cracco have compiled a remarkable collection of state-of-the-art research
that spans a quarter of a century on how and why we imitate others, often automatically.
Given the crucial role that automatic imitation plays in social interactions and bonding this
book will become essential to researchers across the life and social sciences.”
—Manos Tsakiris, Professor of Psychology, Royal Holloway, University of London, UK
“ChatGPT says that I coined the term ‘automatic imitation’.I am not sure that is true, but I
certainly admire this wide ranging volume.Drawing insights from cognitive, social, and
affective psychology, the chapters give an overview of what is known, and what is yet to be
discovered, about spontaneous mimicry—our tendency to copy the actions of others even
when we do not want or intend to do so.This will make it a valuable resource for anyone
interested in the topic.”
—Cecilia Heyes, Professor of Psychology, All Souls College, University of Oxford, UK
vii
This book was supported by two grants from the German Research Foundation
(DFG; funding numbers: 246329797 and 497678237).
Acknowledgments
ix
1 Introduction to Automatic Imitation . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Oliver Genschow and Emiel Cracco
2 Measuring Movement Imitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Divyush Khemka and Caroline Catmur
3 Emotional Mimicry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Ursula Hess and Agneta Fischer
4 Common Coding of Speech Imitation . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Patti Adank and Hannah Wilt
5 Automatic Imitation and the Correspondence Problem
of Imitation: A Brief Historical Overview of Theoretical
Positions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Marcel Brass
6 The Promise and Pitfalls of Studying the Neurophysiological
Correlates of Automatic Imitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Kohinoor M. Darda and Richard Ramsey
7 Levels of Imitation: Movements, Outcomes, and Goals . . . . . . . . . . . 127
Jochim Hansen
8 Anticipated Imitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Roland Pster, Bence Neszmélyi, and Wilfried Kunde
9 Automatic Imitation in Infants and Children . . . . . . . . . . . . . . . . . . . 177
Sumeet Farwaha and Virginia Slaughter
10 Automatic Imitation of Multiple Agents . . . . . . . . . . . . . . . . . . . . . . . . 199
Emiel Cracco
11 Social Modulation of Imitative Behavior . . . . . . . . . . . . . . . . . . . . . . . 219
Oliver Genschow and Emiel Cracco
Contents
x
12 Automatic Imitation of Hand Movements in Clinical
and Neurodiverse Populations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
Ellen Poliakoff and Emma Gowen
13 The Benefits—and Costs—of Behavioral Mimicry: Applications
in Marketing, Sales, and Therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
Wojciech Kulesza and Tanya Chartrand
14 Cognitive Mechanisms of Being Imitated . . . . . . . . . . . . . . . . . . . . . . . 275
Paula Wicher, Harry Farmer, and Antonia Hamilton
15 Mimicry in Psychological Disorders and Psychotherapy . . . . . . . . . . 309
Maike Salazar Kämpf and Cornelia Exner
16 Watching Others Mirror: Explaining the Range of Third-Party
Inferences from Imitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333
Lindsey J. Powell and Piotr Winkielman
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
Contents
1© The Author(s) 2025
O. Genschow, E. Cracco (eds.), Automatic Imitation,
https://doi.org/10.1007/978-3-031-62634-0_1
Chapter 1
Introduction toAutomatic Imitation
OliverGenschow andEmielCracco
Individuals have the tendency to automatically imitate a wide range of different
behaviors such as facial expressions (e.g., Dimberg, 1982), characteristics of lan-
guage (e.g., Cappella & Planalp, 1981), emotions (e.g., Hess & Fischer, 2014), pos-
tures (e.g., LaFrance, 1982), gestures (e.g., Cracco et al., 2018b), and simple
movements (e.g., Brass etal., 2000; Genschow etal., 2013). In this respect, auto-
matic imitation refers to the nding that the execution of an action is facilitated
when observing compatible actions and impeded when observing incompatible
actions (Cracco etal., 2018a; Genschow etal., 2017; Heyes, 2011). Research over
the last two decades indicates that such imitative behavior fullls an important
social function as it bonds humans closely together by creating feelings of aflia-
tion and fostering pro-social attitudes (for a review, see Duffy & Chartrand, 2015).
Interest in research on imitative behavior dates back to at least the eighteenth
century when Adam Smith (1759) put forward the idea that spontaneous imitation
can be regarded as a form of sympathy. Despite the longstanding theoretical inter-
est, systematic investigations on imitative behavior did not start before the twentieth
century when Hull (1933) discovered that participants unintentionally copied the
body movements of an experimenter during a series of psychological tests. Later,
Bandura (1962) linked imitative behavior to learning from others as part of his well-
known social learning theory. It then took until the end of the 1990s and the begin-
ning of the 2000s for research into automatic imitation to be carried out on a large
scale. In this regard, a landmark nding was published by Chartrand and Bargh
(1999). With reference to the so-called chameleon effect, the authors established a
O. Genschow (*)
School of Management and Technology, Leuphana University Lüneburg, Lüneburg, Germany
e-mail: oliver.genschow@leuphana.de
E. Cracco
Department of Experimental Clinical and Health Psychology, Ghent University,
Ghent, Belgium
e-mail: emiel.cracco@ugent.be
2
social psychological paradigm on motor mimicry in which participants interacted
with different confederates. The authors found that participants in this task touched
their heads more often when a confederate touched their head as compared to when
another confederate moved their foot. Conversely, participants moved their foot
more often, when the confederate moved their foot, compared to when the confeder-
ate touched their head.
At a similar time point, Brass etal. (2000; see also Stürmer etal., 2000) applied
the logic of stimulus-response compatibility (SRC) tasks—which are commonly
used in cognitive psychological research—to investigate automatic imitation. In this
task, which was later called the imitation-inhibition task, participants respond to
one out of two imperative cues on the computer screen (e.g., number 1 or 2) with
two different movements (e.g., lifting the index or middle nger). At the same time,
participants either see another person lifting the same (congruent) or different
(incongruent) nger. The typical nding in such a task is that participants respond
faster and with fewer errors to congruent movements as compared to incongruent
movements (Westfal et al., 2024). An adapted version of the imitation-inhibition
task (Brass et al., 2000) has been recently developed and validated to investigate
automatic imitation in online settings (Westfal et al., 2024). The following link pro-
vides resources for implementing the online task, analyzing its data, as well as a
demo version of the task: https://www.automatic-imitation.com.
Over the last two decades, research on the chameleon effect and the imitation-
inhibition task inspired many different disciplines including social and cognitive
psychology, developmental psychology, clinical psychology, and neuroscience. The
research in these disciplines furthered the understanding of the imitation phenome-
non and fueled several debates. Despite comprehensive investigations, several
important questions remain. In this book, well-known and esteemed experts across
different research disciplines give an overview of the latest research on automatic
imitation and review current debates in the literature. The book is divided into four
parts. In the rst part, the book provides an overview and comparison of different
types of imitation. The second part then sheds light on the processes underlying
automatic imitation. The third part reviews research investigating modulators of
automatic imitation and shows under which conditions people tend to imitate others
more (or less) strongly. Finally, in the fourth part of the book, the consequences of
automatic imitation are reviewed and discussed.
Types ofImitation
In Chap. 2, Khemka and Catmur review a range of tasks for measuring the imitation
of movements. The authors review and compare the advantages and disadvantages
of passive action observation, kinematic measures of imitation, stimulus-response
compatibility tasks, and naturalistic measures of mimicry and action synchrony.
Chapter 3 focuses on emotional mimicry—the imitation of nonverbal behaviors
that signal emotions. In this chapter, Hess and Fischer rst differentiate mimicry
O. Genschow and E. Cracco
3
from other related phenomena, then give a historical overview of the research on
emotional mimicry, and nally review different theories of emotional mimicry.
The rst section on different types of imitation closes with Chap. 4 in which
Adank and Wilt review research on speech imitation. In line with current theories of
speech and language processing (Fadiga etal., 2002; Watkins & Paus, 2004), the
authors incorporate research that connects speech perception and speech production
to explain the underlying mechanisms of speech imitation. Moreover, the authors
explain how SRC tasks can be used to study automatic imitation and then discuss
which theories can explain the results obtained with such SRC tasks.
Processes Underlying Automatic Imitation
In the second part of this book, ve chapters shed light on the processes of auto-
matic imitation. In Chap. 5, Brass reviews functional theoretical accounts of auto-
matic imitation by giving an overview of the historical context of the research on
automatic imitation. In particular, he reviews theories that have been put forward to
explain the “correspondence problem of imitation” (Brass & Heyes, 2005; Heyes,
2001), which refers to the question of how a perceptual representation of a move-
ment can be transformed into a corresponding motor program.
In Chap. 6, Darda and Ramsey review neuroscientic research to explain the
neurophysiological correlates that account for automatic imitation. When reviewing
this literature, the authors take a critical view of dominant theories in the literature
that explain the inhibition of imitation in terms of self-other distinction mechanisms
that are tied to the theory-of-mind network (e.g., Brass etal., 2009). The authors
argue that SRC tasks of automatic imitation engage in domain-general forms of
control that are underpinned by the multiple-demand network.
While it is widely agreed that people have the automatic tendency to imitate oth-
ers, the question of whether this imitative tendency is based on a goal- or movement-
driven mechanism is part of a longstanding debate in the literature (e.g., Avikainen
etal., 2003; Genschow etal., 2019; Wohlschläger etal., 2003). In Chap. 7, Hansen
reviews the evidence for both mechanisms and then discusses potential processes
that modulate the degree to which individuals engage in goal-based imitation versus
movement-based imitation.
In Chap. 8, Pster, Neszmélyi, and Kunde argue from an ideomotor perspective
that automatic imitation is strongly inuenced by anticipative processes. In essence,
the authors argue that the social consequences of one’s own behavior are readily
integrated into human action representations, suggesting that imitation is strongly
inuenced by anticipative processes.
An interesting question is whether people’s tendency to automatically imitate
others is innate or learned. In Chap. 9, Farwaha and Slaughter review research on
automatic imitation across the lifespan. This review reveals a signicant disconnect
between child and adult research on automatic imitation effects, which complicates
developmental conclusions. To solve this issue, the authors put forward several
promising avenues for future research.
1 Introduction toAutomatic Imitation
4
Modulators ofAutomatic Imitation
The previous sections of this book may give the impression that in any situation and
context, individuals automatically imitate whatever behavior they perceive.
However, there is a rich literature suggesting that automatic imitation is a highly
exible behavior that can be modulated by different psychological factors. In Chap.
10, Cracco gives an overview of the research on imitation in the context of multiple
agents and shows that automatic imitation varies as a function of group size. The
reviewed literature demonstrates that automatic imitation is a complex process that
takes into account regulatory processes to adjust cognitive control parameters as a
function of both group size (Cracco & Brass, 2018) and the topographical relation
between different observed actions (Cracco etal., 2022).
An often put-forward claim in the literature on automatic imitation is that imita-
tive behavior as a social phenomenon should be modulated by social factors. In
Chap. 11, Genschow and Cracco give an overview of theories that argue in favor of
social modulation of automatic imitation and then critically reect upon the idea of
social modulation by reviewing social variables that have been repeatedly found to
modulate automatic imitation and variables that seem not to inuence automatic
imitation. The authors conclude that the evidence for social modulation is rather
mixed. Reasons for the mixed ndings in the literature may especially be due to
methodological shortcomings and imprecise theories.
In Chap. 12, Poliakoff and Gowen discuss whether and how different psycho-
logical pathologies facilitate or inhibit people’s automatic tendency to imitate. Their
review reveals that depending on the psychological condition, automatic imitation
can either be increased, reduced, or intact. At the same time, the authors stress that
these ndings should not be overinterpreted as the reviewed literature includes
rather small numbers of studies and participants.
Consequences ofAutomatic Imitation
The last part of the book sheds light on the consequences of being imitated. Chapters
13 and 14 review literature on the consequences of being behaviorally mimicked. In
Chap. 13, Kulesza and Chartrand review mainly social psychological research nd-
ings showing that being mimicked by others has positive social consequences.
Going one step further, the authors also discuss an often-neglected part of the motor
mimicry literature by showing that under certain conditions, being mimicked can
have negative social consequences as well.
While Kulesza and Chartrand review the social consequences of being mim-
icked, in Chap. 14, Wicher, Farmer, and Hamilton review different theories that
explain these seminal ndings. Their chapter highlights that the cognitive mecha-
nisms underlying the effects of being mimicked are still unknown. The authors then
discuss different possible neurocognitive models. Based on current evidence, they
conclude that a domain-general model involving cognitive predictability and social
learning is the most promising explanation for the effects of being mimicked.
O. Genschow and E. Cracco
5
Since imitating others has been found to have positive social consequences,
scholars assumed that imitation could be a useful tool for psychotherapy.
Interestingly, evidence for this claim is sparse as Salazar Kämpf and Exner conclude
in Chap. 15. To deal with this issue, the authors present different theoretical
approaches from which they derive new ideas on how imitative behavior might
affect different psychological disorders and therapeutic processes.
The previous chapters in this book mainly focus on automatic imitation in dyadic
interactions. Such a view neglects the fact that in many social situations, people are
not interacting with another person in isolation, but are often witnessed by third-
party observers. In Chap. 16, Powell and Winkielman review research about the
inferences both children and adult observers draw from seeing other people imitate.
The reviewed literature indicates that young observers typically draw positive infer-
ences from imitative behaviors. However, as observers become more mature and
more aware of social dynamics, they start taking into account more complex factors
such as intention, mutual knowledge, social skills, theory of mind, and social strate-
gies. As a consequence, adults not only form positive inferences but also negative
inferences about people who imitate, depending on the situation.
Summary
Taken together, this book gives a comprehensive overview of the ubiquitous phe-
nomenon of automatic imitation, by showing what and how people imitate. The
chapters are written by esteemed experts from different psychological research
elds who take a critical view of the research that has been carried out within the
last two and a half decades. The reviewed research indicates that while there is
strong evidence for several claims made in the literature, there is still an ongoing
and lively debate for other research questions. We hope the chapters included in this
book will be helpful for both scholars and students alike to get deep insights into an
interesting social phenomenon and to develop new investigations to further the
understanding of research questions that are currently part of ongoing debates.
References
Avikainen, S., Wohlschläger, A., Liuhanen, S., Hänninen, R., & Hari, R. (2003). Impaired mirror-
image imitation in Asperger and high-functioning autistic subjects. Current Biology, 13,
339–341. https://doi.org/10.1016/s0960-9822(03)00087-3
Bandura, A. (1962). Social learning through imitation. In J.R. Marshall (Ed.), Nebraska sympo-
sium on motivation (pp.211–274). University Press.
Brass, M., & Heyes, C. (2005). Imitation: Is cognitive neuroscience solving the correspondence
problem? Trends in Cognitive Sciences, 9, 489–495. https://doi.org/10.1016/j.tics.2005.08.007
Brass, M., Bekkering, H., Wohlschläger, A., & Prinz, W. (2000). Compatibility between observed
and executed nger movements: Comparing symbolic, spatial, and imitative cues. Brain and
Cognition, 44, 124–143. https://doi.org/10.1006/brcg.2000.1225
1 Introduction toAutomatic Imitation
6
Brass, M., Ruby, P., & Spengler, S. (2009). Inhibition of imitative behaviour and social cogni-
tion. Philosophical Transactions of the Royal Society B: Biological Sciences, 364, 2359–2367.
https://doi.org/10.1098%2Frstb.2009.0066
Cappella, J. N., & Planalp, S. (1981). Talk and silence sequences in informal conversations
III: Interspeaker inuence. Human Communication Research, 7, 117–132. https://doi.
org/10.1111/j.1468-2958.1981.tb00564.x
Chartrand, T. L., & Bargh, J. A. (1999). The chameleon effect: The perception-behavior link
and social interaction. Journal of Personality & Social Psychology, 76, 893–910. https://doi.
org/10.1037//0022-3514.76.6.893
Cracco, E., & Brass, M. (2018). The role of sensorimotor processes in social group contagion.
Cognitive Psychology, 103, 23–41. https://doi.org/10.1016/j.cogpsych.2018.02.001
Cracco, E., Bardi, L., Desmet, C., Genschow, O., Rigoni, D., De Coster, L., … Brass, M. (2018a).
Automatic imitation: A meta-analysis. Psychological Bulletin, 144, 453–500. https://doi.
org/10.1037/bul0000143
Cracco, E., Genschow, O., Radkova, I., & Brass, M. (2018b). Automatic imitation of pro- and
antisocial gestures: Is implicit social behavior censored? Cognition, 170, 179–189. https://doi.
org/10.1016/j.cognition.2017.09.019
Cracco, E., Braem, S., & Brass, M. (2022). Observing conicting actions elicits conict adapta-
tion. Journal of Experimental Psychology: General, 151, 493–505. https://doi.org/10.1037/
xge0001089
Dimberg, U. (1982). Facial reactions to facial expressions. Psychophysiology, 19, 643–647.
Duffy, K.A., & Chartrand, T.L. (2015). Mimicry: Causes and consequences. Current Opinion in
Behavioral Sciences, 3, 112–116. https://doi.org/10.1016/j.cobeha.2015.03.002
Fadiga, L., Craighero, L., Buccino, G., & Rizzolatti, G. (2002). Speech listening specically mod-
ulates the excitability of tongue muscles: A TMS study. European Journal of Neuroscience, 15,
399–402. https://doi.org/10.1046/j.0953-816x.2001.01874.x
Genschow, O., Florack, A., & Wänke, M. (2013). The power of movement: Evidence for context-
independent movement imitation. Journal of Experimental Psychology: General, 142,
763–773. https://doi.org/10.1037/a0029795
Genschow, O., van Den Bossche, S., Cracco, E., Bardi, L., Rigoni, D., & Brass, M. (2017). Mimicry
and automatic imitation are not correlated. PLoS One, 12, e0183784. https://doi.org/10.1371/
journal.pone.0183784
Genschow, O., Hansen, J., Wänke, M., & Trope, Y. (2019). Psychological distance modulates
goal-based versus movement-based imitation. Journal of Experimental Psychology: Human
Perception and Performance, 45, 1031–1048. https://doi.org/10.1037/xhp0000654
Hess, U., & Fischer, A. (2014). Emotional mimicry: Why and when we mimic emotions. Social
and Personality Psychology Compass, 8, 45–57. https://doi.org/10.1111/spc3.12083
Heyes, C. (2001). Causes and consequences of imitation. Trends in Cognitive Sciences, 5, 253–261.
https://doi.org/10.1016/s1364-6613(00)01661-2
Heyes, C. (2011). Automatic imitation. Psychological Bulletin, 137, 463–483. https://doi.
org/10.1037/a0022288
LaFrance, M. (1982). Posture mirroring and rapport. In M. Davis (Ed.), Interaction rhythms:
Periodicity in communicative behavior (pp.279–298). Human Sciences Press.
Smith, A. (1759). The theory of moral sentiments. A.Miller.
Stürmer, B., Aschersleben, G., & Prinz, W. (2000). Correspondence effects with manual gestures
and postures: A study of imitation. Journal of Experimental Psychology: Human Perception
and Performance, 26, 1746–1759. https://doi.org/10.1037//0096-1523.26.6.1746
Watkins, K., & Paus, T. (2004). Modulation of motor excitability during speech perception:
The role of Broca’s area. Journal of Cognitive Neuroscience, 16, 978–987. https://doi.
org/10.1162/0898929041502616
Westfal, M., Cracco, E., Crusius, J., & Genschow, O. (2024). Validation of an online imitation-
inhibition task. Unpublished Manuscript.
Wohlschläger, A., Gattis, M., & Bekkering, H. (2003). Action generation and action perception
in imitation: An instance of the ideomotor principle. Philosophical Transactions of the Royal
Society B: Biological Sciences, 358, 501–515. https://doi.org/10.1098%2Frstb.2002.1257
O. Genschow and E. Cracco
7
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), 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 license and
indicate if changes were made.
The images or other third party material in this chapter are included in the chapter’s Creative
Commons license, unless indicated otherwise in a credit line to the material. If material is not
included in the chapter’s Creative Commons license 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.
1 Introduction toAutomatic Imitation
9© The Author(s) 2025
O. Genschow, E. Cracco (eds.), Automatic Imitation,
https://doi.org/10.1007/978-3-031-62634-0_2
Chapter 2
Measuring Movement Imitation
DivyushKhemka andCarolineCatmur
Introduction
In this chapter, we present a range of techniques for measuring the speed, accuracy,
and extent of imitation of others’ movements. In general, we dene movement imi-
tation as the production of a congural body movement that matches the movement
performed by another. This denition comprises two components which are of par-
ticular importance when deciding whether a given technique can be said to be mea-
suring imitation. The rst is the focus on congural body movements. This states
that it is the conguration of body parts with respect to other body parts that are of
importance in deciding whether a response is imitative (cf. Heyes, 2021). The
emphasis on body part conguration is important because it allows us to distinguish
imitation from other social learning processes such as stimulus enhancement (where
watching another’s action focuses the observer’s attention on a particular body part,
increasing the likelihood to engage in movements with this body part) or effector
matching (where the observer performs an action using the same effector—hand,
foot, etc.—as that used by the actor; see Whiten etal., 2004, for further denitions
of social learning). For example, when attempting to imitate a movement such as a
swimming stroke, the observer must produce a movement that not only uses the
same effectors as those used by the actor (e.g. the arm and hand) but also moves
those effectors in the same way with respect to the rest of the body. Moving the arm
and hand in a different conguration would be classied as effector matching rather
D. Khemka
Department of Psychiatry, University of Cambridge, Cambridge, UK
e-mail: dk707@cam.ac.uk
C. Catmur (*)
Department of Psychology, Institute of Psychiatry, Psychology and Neuroscience, Kings
College London, London, UK
e-mail: caroline.catmur@kcl.ac.uk
10
than successful imitation: the observer has identied which part of the body to
move, but not how to move it.
The second part of the denition focuses on the matching nature of the relation-
ship between the actor’s movements and the observer’s movements. It is not enough
for the observer to perform a congural body movement in response to the actor’s
movements: the observer’s body movement must match that of the actor. This ele-
ment of the denition of imitation may seem superuous, but in fact it is not always
trivial to determine whether two movements do indeed ‘match’. For example, a
child imitating an adult will produce movements that differ in size and velocity to
those of the actor; in the swimming stroke example above, a full-body movement
does not look the same, visually, when the observer watches the actor as when the
observer produces it themselves. In general, then, matching refers to the visual simi-
larity between the actor’s and observer’s movements when viewed from a third-
person perspective. The visual similarity primarily refers to the form—i.e. the
conguration—of the movement, but in some cases its size and velocity may also
be relevant. We note that it could be argued that the matching component of the de-
nition alone is enough to dene imitation; however, as ‘matching’ can be dened in
a variety of ways, the additional emphasis on congural body movements ensures
that matching at the level of the effector alone is not sufcient for a response to be
considered imitative.
In practice, for many of the techniques discussed in this chapter, the matching
nature of the imitative response is indexed indirectly: by measuring the response in
the muscle(s) that would be involved in performing the observed movement. In such
cases (in particular for the techniques discussed in section “Measuring Movement
Imitation Without Moving”), a matching response is indicated by a particular pat-
tern of activity, such that the muscle that would be involved in performing the
observed movement responds more when observing that movement than when
observing another movement. However, such a pattern on its own does not demon-
strate a matching relationship: it could be that there is a general, non-specic,
increase in motor response when watching particular movements. To control for
such a possibility, it is necessary to measure responses in at least two muscles while
watching a variety of movements: some of which involve one of the measured mus-
cles and some of which involve the other muscle. A muscle-specic pattern of
responses must be observed in at least two muscles (for each muscle, greater
response when observing movements involving that muscle than when observing
movements involving the other muscle) in order to rule out the possibility that a
general increase in motor response is producing the observed effects.
The majority of the techniques described in this chapter therefore index imitation
via the presence of muscle-specic responses or other similar designs. Perhaps the
clearest exception to this approach is in the nal section where we discuss measures
of mimicry. We dene mimicry as a type of imitation that tends to occur in more
naturalistic settings, often without awareness on the observer’s part that they are
imitating the actor (Chartrand & Bargh, 2002; although note that this denition does
not preclude that some of the other measures discussed in this chapter may also
index behaviours that take place without awareness). Studies of mimicry vary in the
D. Khemka and C. Catmur
11
extent to which there is a matching relationship between the movements performed
by the actor and the observer, and in many cases, these studies measure effector
matching or sometimes purely temporal characteristics of the movement. We dis-
cuss the extent to which these measures can be considered truly imitative in section
“Measuring Mimicry” of this chapter.
Finally, as this chapter is in the context of a volume on automatic imitation, we
should note that we are deliberately excluding some techniques from this overview:
we are not discussing measures of intentional or voluntary imitation nor of over-
imitation (see, e.g., Keupp etal., 2018; Marsh etal., 2019).
Measuring Movement Imitation Without Moving
We start by considering methods that can be used to measure imitative activity dur-
ing passive action observation. Since the discovery of ‘mirror’ neurons, neurons in
sensorimotor brain areas that re not only during action performance but also dur-
ing observation of similar actions (di Pellegrino etal., 1992), a range of techniques
have been developed that allow researchers to monitor motor activity while the
observer is not themselves moving. Although such activity is not strictly imitative,
in the sense of producing a motor output that matches the observed movement, it is
generally considered that this motor activity reects subthreshold motor responses
(e.g. Maslovat etal., 2013). In this section, we focus on two techniques that, by
virtue of being measured in the peripheral musculature, can show high muscle spec-
icity compared to cortical measures of activity, allowing the researcher to verify
that the response is produced in the muscle of the observer, which matches the
muscle in the actor that is producing the observed movement.
Electromyographic Measures ofMuscle Activity During
Action Observation
Electromyography (EMG) involves measuring electrical activity in the muscle. In
automatic imitation research, this is carried out non-invasively using surface elec-
trodes. Electrodes are typically placed in a belly-tendon montage and the voltage
difference between the two electrodes is displayed as the EMG signal, such that the
electrical activity in the muscle itself can be isolated from surrounding electrical
noise. The EMG signal therefore provides a relatively clean measure of the electri-
cal activity created by the motor unit action potentials in the targeted muscle and as
such can be considered a measure of the level of activation of that muscle. The EMG
signal contains both positive and negative components and is therefore usually recti-
ed (transforming all negative values to positive) before further data analysis is
performed, allowing peak, total, and/or mean response values to be calculated (see
Fig.2.1).
2 Measuring Movement Imitation
12
Fig. 2.1 (a) Indicative placement of surface electrodes for measurement of facial EMG from the
corrugator (brow) muscle and zygomaticus (cheek) muscle. (b) Example of raw EMG signal (time
runs left to right along the horizontal axis for panels B-E). (c) Rectied EMG signal. (d) Close-up
from panel C illustrating (i) latency of rst ‘peak’ in the EMG response; (ii) amplitude of rst
peak. (e)Illustration of ‘epoch’ approach, with signal averaged over adjacent time ‘bins’ and mea-
sured during certain time windows corresponding to different conditions, illustrated by the grey/
white bars
During action observation, the focus of this section, electromyography is per-
formed in the muscle at rest. In many skeletal muscles, little electrical activity is
recorded at rest, and action observation does not increase muscle activity to the
point where motor unit action potentials can be detected with surface electrodes.
However, in certain populations (e.g. infants and young children; Cattaneo et al.,
2007; Turati etal., 2013) and for certain muscles (in particular those of the face and
neck; e.g. Dimberg, 1982; Ruggiero & Catmur, 2018), it is possible to detect EMG
signals during passive action observation. This permits the researcher to measure
the level of activity in the recorded muscle(s) during various action observation
conditions.
In principle, it is possible to measure a range of dependent variables from EMG
during action observation, including latency of response and its magnitude. In prac-
tice, measuring latency is complicated by the need to dene a threshold above which
it can be considered that an EMG response is not just ‘noise’, with latency of onset
then dened as the timepoint at which the EMG signal exceeds that threshold; how-
ever, during passive action observation the signal-to-noise ratio is low, and as such,
latency measures are not commonly used in this literature. Another drawback of
using latency measures in EMG recorded during action observation is that the
observed movements unfold over time and as such—unlike with more punctate
stimuli—there is not one clear timepoint from which latency can be measured.
Instead, an ‘epoch’ approach is often used, with the signal during certain time win-
dows (e.g. before or after a certain timepoint in the observed movement) being
compared (Fig.2.1). Typically, this signal will comprise the magnitude of the EMG
D. Khemka and C. Catmur
13
response, which can be measured as either peak amplitude, or, more commonly,
area under the curve—in effect a sum of activity—during a certain time window.
Individual differences in a variety of factors (e.g. subcutaneous fat, which impedes
signal detection from surface electrodes) can lead to wide variation in raw EMG
signal magnitude, meaning it is important to control for this in group-level analyses,
for example, by normalising each participant’s EMG signal to their own mean
(Halaki & Ginn, 2012).
As noted above, the use of EMG to measure muscle activity during passive
action observation has on the whole been conned to head muscles. Studies using
this technique to measure imitation have typically focused on imitation of emotional
facial expressions, or of eating/swallowing movements (see also Chap. 3; this vol-
ume). Dimberg (1982) measured EMG responses in facial muscles during the
observation of happy and angry facial expressions. Responses in the zygomaticus
cheek muscle (involved in smiling) and in the corrugator brow muscle (involved in
frowning) were each greater during the observation of the expression that used those
muscles in the actor (i.e. during observation of happy and angry expressions, respec-
tively), consistent with a muscle-specic imitative response to the observed expres-
sions. Subsequent studies have used this technique to compare these imitative
responses across different participant groups (e.g. McIntosh et al., 2006; Kaiser
etal., 2017; Künecke etal., 2018; Scarpazza etal., 2018) and across different types
of actors (e.g. human vs android; Hofree etal., 2014).
EMG responses during observation of eating movements are generally measured
instead from the suprahyoid muscles just under the chin, which are involved in
swallowing. Studies of responses to action observation in these muscles have
focused on whether observers show an anticipatory imitative response when observ-
ing an object being grasped in order to be brought to the actor’s mouth (grasp to eat)
compared to when an object is being grasped in order to be placed at another loca-
tion (grasp to place). Cattaneo et al. (2007) reported differential mylohyoid
responses in neurotypical, compared to autistic, children when observing grasp-to-
eat versus grasp-to-place movements; however, in this study the observed move-
ment was confounded with the nature of the object being grasped, which may have
driven the differential responses (Ruggiero & Catmur, 2018). Studies using the
suprahyoid muscles also tend to record from one muscle only, making it hard to
draw any conclusions regarding the muscle specicity of the response to the
observed movements.
TMS-Evoked Measures ofMotor Responses During
Action Observation
The other key method to measure automatic imitation when participants are not
moving utilises transcranial magnetic stimulation (TMS). TMS is well known in the
cognitive neurosciences as a method for temporarily disrupting brain function (as a
2 Measuring Movement Imitation
14
so-called virtual lesion technique; Walsh & Cowey, 1998; Pitcher etal., 2021), but
it also has a long history in neurophysiological research and clinical settings, when
combined with EMG, as a technique for measuring motor function (Chen etal.,
2008); it is in this latter context that TMS is most relevant to the measurement of
automatic imitation.
TMS is a method for inducing electrical current in the brain. When a TMS ‘pulse’
is red, a strong and rapidly changing electrical current is passed through a coil
consisting of copper wire wound around an iron core. Due to electromagnetic induc-
tion, the changing current generates a magnetic eld which—when the TMS coil is
placed against the head—passes through the skull and induces an electrical current
in any electrical conductors within its range. In the case of TMS applied to the brain,
the conductors are the axons of the neurons directly underneath the coil. If the TMS
pulse is sufciently strong, it depolarises these neurons, creating action potentials.
In the case of TMS applied to the motor cortex, these action potentials are propa-
gated down the corticospinal tract to the neuromuscular junction of the muscle
which they innervate, creating a motor-evoked potential (MEP) in that muscle. The
MEP can be detected and measured with surface electrodes in a similar fashion to
EMG, as described in section “Electromyographic Measures of Muscle Activity
During Action Observation” (see Fig.2.2). Each individual participant will have a
different ‘resting motor threshold’, the intensity of TMS that is required to induce
MEPs in a particular muscle when the participant is at rest. In order to reliably
induce MEPs on every trial during an experiment, TMS is usually delivered at an
intensity just over (e.g. at 110% of) the participant’s threshold.
When deciding on the dependent variable for MEP studies, some of the same
considerations apply as mentioned above for EMG (see section “Electromyographic
Measures of Muscle Activity During Action Observation”). The main choice is
whether to use the area under the curve of the MEP or the peak-to-peak amplitude
as the dependent variable. However, the MEP tends to be relatively larger and
‘cleaner’ (in terms of signal-to-noise) than the EMG signal; it is also more punctate
as it occurs at a particular timepoint after the TMS pulse is delivered (usually around
20ms for hand muscles, reecting the conduction time between the brain and the
hand musculature). As such, the magnitude of the MEP response is usually indexed
Fig. 2.2 (a) Illustration of TMS coil placement over the primary motor cortex hand area. (b)
Indicative placement of surface electrodes for measurement of motor-evoked potentials from the
rst dorsal interosseus and abductor digiti minimi muscles. (c) Example of a motor-evoked poten-
tial, illustrating (i) peak-to-peak amplitude of MEP response
D. Khemka and C. Catmur
15
by the absolute amplitude (i.e. from the minimum peak to the maximum peak in a
given time window), averaged over multiple trials from a given experimental
condition.
Crucially for the present purposes, for a given strength of TMS pulse, the size of
the MEP reects the relative activation of the motor cortical representation stimu-
lated by the TMS coil. Thus, MEPs generated in a particular muscle when the par-
ticipant is at rest will be smaller than those recorded when the participant is
activating that muscle (e.g. performing a movement or preparing to perform a
movement). Critically, this increase in MEP size is also found when participants are
at rest and merely observing others’ movements. The utility of TMS-MEP measures
for studies of action observation was rst elucidated by Fadiga etal. (1995), who
demonstrated that MEPs in hand muscles while participants were at rest but observ-
ing hand movements were greater than those recorded while the participants were
observing other non-action stimuli such as dimming lights. A clearer demonstration
of muscle specicity was subsequently reported by Strafella and Paus (2000), who
showed greater MEPs in hand muscles while participants were watching hand
movements compared to arm movements and vice versa for MEPs recorded from
arm muscles. Subsequent studies have shown that the increased motor cortical acti-
vation is specic not only to the effector (hand vs. arm) but also to the muscle that
would be involved in the observed movement (e.g. index nger vs. little nger mus-
cle; Romani etal., 2005; Catmur etal., 2007). It should be noted that this specicity
develops during the rst few hundred milliseconds following the onset of the
observed action: a comprehensive review of the TMS-MEP literature on motor
responses to action observation (Naish etal., 2014) concluded that when observing
others’ actions, there is an initial non-specic increase in MEP amplitude, which
occurs in the rst 100ms after the other’s action. This cannot be considered an imi-
tative response since it is not muscle-specic: instead, it suggests that—possibly
due to a general attention-related or alerting mechanism—seeing other people’s
actions produces a general increase in motor activity in the observer. Naish et al.
further reported that muscle-specic MEP responses start to develop around 200ms
after the onset of the observed action. Recall that these muscle-specic responses
are found when participants are at rest, observing another person’s action: as such,
they are considered to be the neurophysiological signature of the tendency to auto-
matically imitate other people’s movements.
TMS-MEP measures have some signicant advantages over other cognitive neu-
roscience methods when measuring motor responses to action observation: the tem-
poral specicity of TMS is very high, allowing the timecourse of motor responses
to be determined (as outlined by Naish etal., 2014); unlike most functional neuro-
imaging methods, the ability to record responses from multiple muscles permits the
researcher to determine whether a response is muscle-specic or merely a reection
of generalised motor activity. It should, however, be noted that carefully designed
behavioural studies of automatic imitation (see section “Stimulus-Response
Compatibility Measures of Automatic Imitation”) share many of these advantages;
andthere are some considerations to be taken into account when using TMS in this
context. As noted above, actions unfold over time and this creates some difculties
2 Measuring Movement Imitation
16
when designing TMS-MEP studies of action observation: notably, if muscle- specic
MEP responses develop from 200 ms after an observed action, one should not
expect such responses to be exactly timelocked to the action that is being observed.
Conversely, when observing ongoing actions (as opposed to single movements) the
observer may be able to predict the outcome of the action, in which case the pattern
of MEP response may be more closely aligned in time to the activity of the muscles
involved in the observed movements. The observation of repeated actions (e.g. in a
blocked design) may induce such predictive responses. Another consideration is
how predictable the TMS pulse becomes during the course of the study: MEP
amplitudes tend to reduce when the TMS pulse is predictable (e.g. occurring at a
certain timepoint after a visual stimulus; Cavallo etal., 2014; Villiger etal., 2011),
making it essential to control for pulse timing and predictability when comparing
experimental conditions, especially when comparing to baseline when pulses may
be less predictable than during stimulus presentation.
Finally, we should note that although single-pulse TMS, when used as outlined
in this section, is a safe procedure, there are some contra-indications to TMS that
make its use on certain populations problematic. The primary safety concern with
TMS is the possibility of inducing a seizure in participants with a low seizure
threshold; as such, TMS should not be used in people with a family or personal his-
tory of epilepsy. It is also not usually considered suitable for use in developmental
studies with infant or child participants. Further details on best practice for ensuring
the safety of TMS in neuroscience research can be found in Rossi etal. (2009, 2021).
Kinematic Measures ofMovement Imitation
In this section, we consider how motion tracking can be used to measure movement
imitation. This group of techniques uses motion trackers—physical or digital—to
collect data on the location of one or more body parts in two- or three-dimensional
space across time. Physical trackers are typically infrared reectors (e.g. Kilner
etal., 2003), or electromagnetic sensors (e.g. Forbes & Hamilton, 2017), which are
attached to points on the participant’s body while the participant carries out various
movement tasks. Digital movement tracking instead uses techniques such as visual
pattern recognition, or more laboriously, video coding, to identify the location, in
images across time, of particular body parts (e.g. Niechwiej-Szwedo etal., 2018).
For the purposes of this chapter, we will consider tasks where an observer is moving
either while simultaneously observing an actor or shortly after observing an actor.
The data generated by these methods (location of body parts over time) are sub-
jected to analyses that calculate values including the location, velocity, acceleration,
and jerk (rate of change in acceleration) of each marker during a particular time
window (see Fig.2.3). Where more than one marker/location is recorded, the rela-
tive locations of multiple markers with respect to each other can also be calculated,
along with the velocity, acceleration, and jerk of these markers with respect to each
D. Khemka and C. Catmur
17
Fig. 2.3 (a) Illustration of participant performing a sinusoidal left-to-right movement in the hori-
zontal plane. (b) Velocity of participant movement (time runs left to right along the horizontal axis
for all panels). (c) Absolute acceleration of participant movement. (d) Jerk (rate of change in
acceleration) of participant movement
2 Measuring Movement Imitation
18
other. These values can then be compared to equivalent values from the actor, across
timepoints, and across experimental conditions.
To appreciate the wide range of research questions that can be addressed using
such data, we provide an overview here of some of the most inuential studies of
imitation that have used kinematic measures. For a more detailed review, we refer
the reader to Krishnan-Barman etal. (2017).
Kilner et al. (2003) used infrared motion tracking to measure the variance in
participants’ movements while they performed sinusoidal horizontal or vertical arm
movements. Participants were asked to perform these movements at a speed of
about 0.5Hz while observing either a human actor or a robot actor performing the
same type of movement in either the same or the other plane (i.e. horizontal or verti-
cal) to the participant’s own movement. The actor’s observed movements interfered
with those performed by the observer (i.e. resulted in greater variance) when the
actor performed a movement in the other plane to that performed by the observer,
even though the observed movements were incidental to those that the participant
was performing. Furthermore, this interference was found only for biological
motion and not for robotically observed movements. However, the robotically and
human-observed movements varied in a number of ways: the velocity proles were
different (biological motion decelerates towards the end of the sinusoid to ensure a
‘minimal jerk’ velocity prole, whereas robotic motion has relatively constant
velocity and thus greater change in acceleration at the end of each movement); the
observed movements were performed live; thus, it is likely that the observed human
movement was itself more variable than the observed robotic movement (see also
Chap. 10 for a discussion on the moderators of automatic imitation). Follow-up
studies demonstrated that the interference effect was indeed the result of the differ-
ent velocity proles of biological motion (Kilner etal., 2007); an intriguing series
of studies by Cook etal. (2013, 2014) suggested that participants may demonstrate
less interference from observed minimum jerk biological motion if they themselves
move with less of a minimum jerk velocity prole. This latter nding is consistent
with the idea that observed actions are represented in our own motor system as a
result of sensorimotor experience in which we observe the visual consequences of
our own actions (Heyes etal., 2005).
Kinematic techniques are not restricted to human participants. In an elegant
study, Voelkl and Huber (2007) used motion analysis from video of marmosets who
were attempting to open a small canister to retrieve a food reward. Observer marmo-
sets were previously exposed to a model opening the canister with its mouth; non-
observer marmosets did not receive this exposure. The head movements of the
observer marmosets were consistently closer, in terms of a number of kinematic
features, to the head movements of the model, compared to those of the non-observer
marmosets. This study demonstrated high delity of movement imitation in non-
human primates, casting doubt on previous claims that such high-delity imitation
is specic to humans (see also Sartori etal., 2015 for a similar demonstration in
dolphins).
Potentially ‘irrational’ imitation has also been revealed using motion tracking
techniques. Forbes and Hamilton (2017) asked participants to touch a series of
D. Khemka and C. Catmur
19
targets, copying the order in which an actor moved to their targets. The actor had to
move around obstacles between their targets, but participants had no obstacles to
avoid. Motion tracking was used to determine the peak height of participants’ move-
ments between their targets. These data demonstrated that despite the lack of obsta-
cles, participants tended to imitate the actor’s movement trajectory, even when this
was excessively high and rated by other participants as an ‘irrational’ movement by
the actor (even in the presence of an obstacle). It is not clear, however, whether this
type of result reveals that participants are deciding to imitate the actor’s irrational
movements (e.g. due to perceived social desirability effects) or whether instead par-
ticipants automatically imitate the actor’s movements as a result of a more general
process such as associative learning of the links between observed and performed
actions, or spatial compatibility.
The rich data sets generated by motion tracking do have some drawbacks. In
particular, they can lead to difculties when generating a priori hypotheses regard-
ing how an actor’s movement may impact that of an observer. As there are so many
potential parameters to compare, a potential multiple comparison problem arises,
along with the possibility that ‘shing’ in a large pool of possible parameters may
lead the researcher to come up with hypotheses after the results are known (Krishnan-
Barman etal., 2017; Kerr, 1998). Pre-registration of planned analyses and distin-
guishing these from exploratory analyses will be a helpful step in this regard.
Stimulus-Response Compatibility Measures
ofMovement Imitation
Stimulus-response compatibility tasks have been widely used to index both auto-
matic imitation and the control processes required to suppress imitative tendencies.
Such types of tasks are conducted mostly in laboratory settings (see Westfal etal.,
in preparation, for a recently validated online version of the task) and typically
involve the measurement of response times (RTs) and error rates and, in some stud-
ies, kinematics (e.g. Kilner etal., 2003), during the performance of an instructed
response while observing similar or dissimilar task-irrelevant movements (see
Cracco etal., 2018, for a recent meta-analysis). Response times in these types of
tasks can be recorded using various techniques including electromyography (e.g.
Leighton etal., 2010), light sensors (e.g. Genschow etal., 2019), response boxes
(e.g. Ainley etal., 2014), and keyboards (e.g. Sowden & Catmur, 2015). The current
section will focus on stimulus-response compatibility tasks measuring RTs and
error rates (see section “Kinematic Measures of Movement Imitation” for the mea-
surement of kinematics in imitation). Stimulus-response compatibility tasks of imi-
tation often involve action observation and execution of isolated and goalless
movements such as nger lifts, nger abduction, and foot raises (but see Catmur &
Heyes, 2019, for an example of automatic imitation of goal-directed actions). These
tasks typically involve at least two conditions (see Fig.2.4): (i.) compatible—when
2 Measuring Movement Imitation
20
Fig. 2.4 Example of stimulus-response compatibility task stimuli involving hand movements. The
2 (imitative compatibility)×2 (spatial compatibility) task design is illustrated for a trial when a
right-hand index nger lift is the instructed response. When a right-hand middle nger lift is the
instructed response, the levels of spatial and imitative compatibility are each reversed. (Adapted
from Khemka etal., 2021)
the observed action is similar to the instructed response (e.g. see index nger, raise
index nger, or see middle nger, raise middle nger); and (ii.) incompatible—
when the observed action is dissimilar to the instructed response (e.g. see index
nger, raise middle nger, or see middle nger, raise index nger). A decrease in
response times and error rates during compatible trials provides an index of the
tendency to automatically imitate observed movements. However, response times
and error rates during incompatible trials are a more complex measure because they
index not only the extent to which the observed action interferes with the instructed
response but also the processes involved in controlling or overcoming this interfer-
ence. As such, stimulus-response compatibility tasks have been used to measure
both automatic imitation and imitation-inhibition, but as we discuss below, it is
important to consider which aspects of these processes are being measured when
interpreting the results of such studies.
D. Khemka and C. Catmur
21
Stimulus-Response Compatibility Measures
ofAutomatic Imitation
The stimulus-response compatibility paradigm rst used by Brass etal. (2000) has
been widely used as a behavioural index of automatic imitation. In a forced choice
response paradigm, participants were displayed an on-screen left-hand performing
either an index or a middle nger lifting movement. Concurrently, a task-relevant
symbolic cue (a number ‘1’ or ‘2’) was presented in between the index and middle
ngers of the stimulus hand. Participants were instructed to lift either their own
index or middle nger of their right hand in response to the symbolic cue, where the
number ‘1’ instructed participants to lift the index nger and the number ‘2’
instructed them to lift the middle nger, thus making the on-screen nger lifts irrel-
evant to task performance. Contrasting conditions of compatibility between the
observed action and the instructed response during each trial create compatible and
incompatible conditions. A third trial type where the hand performed no movement
was also included as a baseline. The results indicated that compared to baseline tri-
als where the hand did not move, responses were faster and more accurate during
compatible trials and slower and more inaccurate during incompatible trials. Similar
effects have been found when using simple response tasks in which the response
participants make is blocked (e.g. make hand opening movement throughout a block
of trials) and the imperative stimulus, indicating when the participant should make
their response, is an action that is either compatible or incompatible with the
instructed response (Stürmer et al., 2000; Heyes etal., 2005). Studies utilising
stimulus- response compatibility paradigms, whether choice or simple response
tasks, have demonstrated that RTs are more sensitive to modulation compared to
error rates since these paradigms typically report very low error rates (often as low
as <5% of all trials; e.g. Brass etal., 2000; Stürmer etal., 2000).
The observation of an action likely activates the observer’s own motor represen-
tation of that action, thereby facilitating compatible and interfering with incompat-
ible trials (Brass etal., 2000; di Pellegrino etal., 1992). For example, the observation
of a nger movement activates the motor representation of that specic movement
in the observer. Participants demonstrate a tendency to copy movements regardless
of whether the observed movements facilitate or interfere with response selection,
which suggests that the imitation of movements in this task is, to a large extent,
automatic. In compatible trials where the observed action and the instructed response
lead to the activation of the same muscles, the observation of the same action as the
instructed response facilitates action execution and speeds responses, which is taken
as an index of automatic imitation of the observed action. The difference in RTs or
error rates between compatible trials and baseline trials provides a measure of this
facilitatory effect. However, although comparisons between compatible and base-
line trials are useful to provide a specic measure of the facilitation of movement
imitation, the setup of the baseline condition is a crucial consideration in stimulus-
response compatibility tasks. Initial studies using these tasks used static images of
limbs devoid of transient events as a baseline condition, as opposed to compatible
2 Measuring Movement Imitation
22
and incompatible conditions, which involve transient changes in the stimuli. Wiggett
etal. (2013) argued that it is important to control for the temporal alerting effects
that are present in trials where a specic body part moves, which was lacking in
previous versions of the task. In their study, Wiggett etal. used a baseline condition
that involved body parts transiently reducing in size, eliciting similar temporal alert-
ing effects as produced in the standard conditions, whilst allowing the measurement
of RTs unadulterated by task-irrelevant actions. Another version of the baseline trial
type, which also accounted for temporal alerting effects, was used by Sowden and
Catmur (2015) where the presentation of a resting hand image was followed by the
pixelation of the hand image. It is difcult, however, to design a ‘true’ baseline
condition, since the magnitude of the alerting effect in the baseline condition will
change as a function of the magnitude of visual change. As such, it is important to
take into account the nature of the baseline or comparison condition that is used
when one is considering stimulus-response compatibility studies of automatic
imitation.
Stimulus-response compatibility tasks have been used in the literature to demon-
strate the tendency of humans to spontaneously imitate (Brass etal., 2000, 2001;
Stürmer etal., 2000), based on the premise that the compatibility effect (measured
either as the difference between baseline and compatible trials, or between incom-
patible and compatible trials) is a reliable measure of imitation-specic processes
(e.g. Genschow etal., 2017). However, this paradigm has been met with criticism in
the past due to concerns that the compatibility effect is confounded by spatial com-
patibility (Aicken et al., 2007; Bertenthal et al., 2006; Catmur & Heyes, 2011;
Jansson etal., 2007). Classic cognitive studies have shown that participants respond
faster to visual stimuli when they are presented on the same side of space as the
response effector (e.g. the participant’s hand; Anzola etal., 1977; Simon, 1990).
Thus, in choice RT tasks involving lateralised visual stimuli (e.g. the relative posi-
tion of index and middle ngers of the on-screen hand) and lateralised responses
(e.g. the ngers of the participant’s hand), ipsilateral responses yield faster reaction
times compared to contralateral responses. In the case of the imitation-inhibition
paradigm developed by Brass etal. (2000, 2001), participants observed a left stimu-
lus hand mirroring their right response hand. As a result, trials involving index n-
ger lifts are not just imitatively compatible (e.g. see index nger lift, raise index
nger), but are also spatially compatible (e.g. see left nger lift, raise left nger).
Subsequent research demonstrated the independence of spatial compatibility and
imitative compatibility effects and conrmed that the general compatibility effect
previously used to index automatic imitation was confounded by spatial compatibil-
ity effects (Catmur & Heyes, 2011).
In their study, Catmur and Heyes (2011) modied the classic stimulus-response
compatibility task to allow the measurement of each level of imitative compatibility
(compatible, incompatible) at each level of spatial compatibility (compatible,
incompatible). Participants performed index or little nger abduction movements
with their right hand in response to a discriminative stimulus (i.e. colour of a circle),
while concurrently observing a task-irrelevant right or left on-screen hand also
D. Khemka and C. Catmur
23
performing index or little nger abduction movements. This task allowed the mea-
surement of imitative compatibility since both the task-irrelevant stimuli and par-
ticipants’ responses involved congural body movements. Additionally, the
inclusion of both left- and right-hand stimuli, presented in a rst-person perspective,
also allowed the measurement of spatial compatibility since both the task-irrelevant
stimuli and participants’ responses were oriented within the same (left-right) spatial
dimension. In the case of the right-hand stimuli, the observed index nger move-
ment was on the left side of space and the observed little nger movement was on
the right side of space; in the case of the left-hand stimuli, the observed index nger
movement was on the right side of space and the observed little nger movement
was on the left side of space. Therefore, the use of both right- and left-hand stimuli
while participants responded with their right hand allowed the manipulation of the
spatial location of the stimulus independent of its nger identity (see Fig. 2.4).
Catmur and Heyes demonstrated that both types of compatibility independently
affected response times—participants were faster to respond when the moving stim-
ulus was on the same side of space as their responding nger compared to when it
was on the other side of space (spatial compatibility effect) and also when the mov-
ing stimulus comprised the same nger movement that they were making, com-
pared to the other nger movement (imitative compatibility effect), but these two
compatibility effects did not interact. They concluded that the ‘general’ compatibil-
ity effect measured in previous automatic imitation studies where spatial compati-
bility was not controlled for is a composite measure comprising both spatial and
imitative compatibilities, illustrating the importance of measuring imitative compat-
ibility independent of spatial compatibility. Further studies have demonstrated that
these two effects may have distinct neural underpinnings (e.g., Marsh etal., 2016),
making this an essential point to consider when designing or evaluating studies of
automatic imitation.
Other studies involving stimulus-response compatibility measures of imitation
have controlled for spatial compatibility by positioning the stimulus hand orthogo-
nal to the participant’s hand (Cook & Bird, 2011; Cook & Bird, 2012; Heyes etal.,
2005; Press et al., 2005). This preserves imitative compatibility but controls for
spatial compatibility since the stimulus hand is no longer positioned to mirror the
participant’s hand but instead points sideways instead of downward. However, a
disadvantage of controlling for spatial compatibility by positioning the stimulus
hand orthogonally is the reported tendency of participants to associate the ‘lower’
nger as the left nger and the ‘upper’ nger as the right nger, thereby only elimi-
nating the inuence of spatial compatibility if this orthogonal spatial compatibility
effect is also controlled for (Weeks & Proctor, 1990).
Stimulus-response compatibility measures provide a robust measurement of
automatic imitation and have been widely used to further our understanding of
action imitation and the mechanisms that support this. Crucial considerations have
been brought to light over the recent years concerning the inuence of spatial com-
patibility effects in stimulus-response compatibility tasks of imitation and the
importance of an appropriate baseline condition that has similar transient properties
2 Measuring Movement Imitation
24
as the main trials to allow the measurement of purely facilitatory processes elicited
by action observation. Further concerns regarding the relationship between stimulus-
response compatibility measures of automatic imitation, and the ‘real world’ mea-
sures of mimicry described in Section “Measuring Mimicry”, also remain to be
addressed (e.g. Ramsey, 2018; but see also Cracco & Brass, 2019).
Stimulus-Response Compatibility Measures
ofImitation-Inhibition
Stimulus-response compatibility tasks are complex in nature as they allow not just
the measurement of response facilitation elicited by action observation, but also the
processes required to control automatic imitation of the observed action. On trials
where the observed action and the instructed cue activate competing motor repre-
sentations (i.e. during incompatible trials where the observed action and the
instructed response lead to the activation of opposing motor representations), the
observer is required to inhibit the motor representation of the other person’s action.
The inhibition or regulation of the other’s motor representation is cognitively
demanding and, as such, takes time. Therefore, the response times and error rates on
incompatible trials compared to compatible or baseline trials are higher. The differ-
ence in response times or error rates between incompatible and baseline trials allows
a comparison between conditions where interference arising due to incompatible
movements must be resolved and conditions where there is no action observation,
thus indexingboth inhibition and imitation.
In a range of studies, therefore, this type of stimulus-response compatibility task
has been labelled an ‘imitation-inhibition’ task, and the response time or error rate
difference between incompatible and either baseline or, more commonly, compati-
ble trials has been considered to index the ability to inhibit imitation (Brass etal.,
2001; Santiesteban et al., 2012). A larger response time difference (i.e. a larger
imitative compatibility effect) reects a greater tendency to imitate and, by exten-
sion, reects weaker inhibition of such imitative tendencies. A smaller imitative
compatibility effect is therefore thought to reect the ability to successfully inhibit
the motor representation evoked by the other’s action and, instead, enhance one’s
own motor representation (Brass etal., 2001).
Previous studies have used the imitative compatibility effect (incompatible—
compatible trial difference) as an index of imitation-inhibition or, in other words,
the outcome of processes involved in controlling automatic imitative tendencies.
However, as previously discussed, response time differences between imitatively
incompatible and compatible trials index both response facilitation and inhibition.
This raises an important question about which dimensions of this task measure
automatic imitation and which of these measure imitation-inhibition. Furthermore,
a small imitative compatibility effect could result from at least two sources: (a) a
lack of automatic imitation of others’ actions; or (b) intact automatic imitation
D. Khemka and C. Catmur
25
alongside successful inhibition of imitation. Current studies rarely distinguish these
two possibilities. One way to address this issue would be to consider the response
time difference between baseline and compatible trials as a measure of pure facilita-
tion and the response time difference between baseline and incompatible trials as a
measure of inhibition and imitation. However, it is then crucial to incorporate appro-
priate baseline conditions that can help measure these separate processes that con-
stitute imitation and the inhibition of imitation.
Overall, the literature often lacks clarity regarding the different processes mea-
sured by stimulus-response compatibility tasks (e.g. processes involved in imitation
versus those involved in spatially compatible responding) and often fails to distin-
guish between the facilitative and inhibitory processes that contribute to the key
effects measured using these tasks. The lack of consistent terminology for key
effects (e.g. ‘automatic imitation effect’ and ‘imitation-inhibition effect’ are both
described as the response time difference between incompatible and compatible tri-
als; Heyes, 2011) has contributed to low specicity in describing the different
effects emerging from the task. Future studies should aim to use appropriate base-
line conditions together with consistent terminology that species the effect emerg-
ing from the task (e.g. ‘imitative compatibility effect’, ‘spatial compatibility effect’,
or ‘imitation-inhibition effect’).
Measuring Mimicry
The nal section of this chapter focuses on the measurement of mimicry, dened as
a type of imitation that tends to occur in more naturalistic settings, often without
awareness on the observer’s part that they are imitating the actor (Chartrand &
Bargh, 1999). Mimicry has previously been shown to play an important role in
increasing afliation and building interpersonal rapport (LaFrance & Broadbent,
1976; Chartrand & Bargh, 1999) and has thus been an important area of research
into understanding social interactions in typical and atypical development. Research
on mimicry has focused on two facets of mimicking behaviour: the mimicry of
actions and the mimicry of action timing, also known as synchrony. Synchrony is
generally considered as the tendency to perform repetitive actions (e.g. walking or
rocking a chair) at the same rate as another person. Both motor mimicry and motor
synchrony are important facets of interpersonal coordination and although synchro-
nised actions are not necessarily imitative (i.e. synchrony may not involve the pro-
duction of movements that are similar in form), it is often considered within the
umbrella of the term ‘imitation’ in the literature (see Lakin, 2013 for review). This
section will discuss the measures used in some of the most inuential studies of
automatic mimicry and synchronisation of actions.
2 Measuring Movement Imitation
26
Mimicking Actions
The empirical measurement of movement mimicry has previously been conducted
during in-person dyadic interactions and during the observation of pre-recorded
videos. For example, Chartrand and Bargh (1999, Experiment 1) recorded in-person
interactions between participants and confederates, where the extent to which par-
ticipants copied the confederate’s movements was measured. Participants performed
an unrelated ‘photograph description’ task together with two confederates, during
two separate interactions. Each confederate was instructed to perform certain move-
ments (e.g. either ‘face-touching’ or ‘leg-shaking’) during their interactions with
the participant. Video recordings of these interactions were then coded to determine
the frequency with which the participant performed the confederate’s target move-
ment, and the alternative movement, during each interaction. The frequency with
which participants performed the confederate’s movement during each interaction
was compared to that at which the participant performed the alternative movement
(i.e. that of the other confederate). An interaction effect, such that participants per-
formed more face-touching movements while observing face-touching, and more
leg-shaking movements while observing leg-shaking, demonstrates a tendency to
mimic the observed actions, as was found in the study conducted by Chartrand and
Bargh (1999). It should be noted that such mimicry tasks need to be structured in
this way, with two different target movements, in order that similarity between the
action performed by the confederate and that performed by the participant can be
established. In contrast, tasks that measure mimicry of only one action could nd
what looks like mimicry (an increase in performance of a particular action when
observing that action), but this could in principle result from a general increase in
motor activity rather than from a specic activation of the motor representation of
the target movement.
Although the naturalistic nature of this task provides high ecological validity and
enables the study of mimicry during social interactions, it also leads to low experi-
mental control, due to inevitable differences between conditions and across partici-
pants in terms of the precise movements performed by the confederates. Subsequent
studies have therefore used a video-based variation of this mimicry paradigm to
study important questions about top-down inuences on movement mimicry includ-
ing whether such mimicry can be modulated by group membership (Yabar et al.,
2006), perceived similarity (Guéguen & Martin, 2009), and afliative goals (Lakin
& Chartrand, 2003). The video-based mimicry task typically involves participants
observing video clips of a confederate performing an unrelated task such as reading
a story or describing photographs, during which the confederate performs specic
movements during regular intervals. The frequency with which participants perform
similar movements is calculated by independent video coders to provide a measure-
ment of movement mimicry. These studies using the video-based mimicry task did
not use a two-action design as described by Chartrand and Bargh (1999), which
makes it possible that the observed effects may reect a general increase in motor
activity, rather than mimicry. However, Genschow et al. (2017) utilised a
D. Khemka and C. Catmur
27
video-based mimicry task involving the observation of two videos depicting two
types of actions and demonstrated a mimicry effect, similar to the one found by
Chartrand and Bargh.
Social psychologists have been able to answer a variety of questions about how
individuals copy other people’s movements during social interactions using the
mimicry task paradigm. However, there are important methodological consider-
ations associated with the measurement of movement mimicry. The foremost meth-
odological concern is the manner in which mimicked actions are indexed. Video
recordings of interactions between participants and confederates are subjectively
rated post hoc, which can introduce measurement bias since the coders may be
aware of the task condition (i.e. which movement is being manipulated in that con-
dition) or may miss subtle movements, since unlike in automatic imitation tasks,
there is no objective measurement of when a movement is executed, or of which
movement is performed. Nevertheless, it is important to note that typically the video
recording in mimicry tasks only shows the participant’s and not the confederate’s
actions, which should remove one potential source of bias. Researchers have also
aimed to reduce biases in the task by involving two independent coders and report-
ing the inter-rater reliability of coded actions, which is generally high. More gener-
ally, studies of mimicry tend to code actions at the level of the effector rather than
the movement (e.g. coding any movement of the leg as a target ‘leg-shaking’ move-
ment), meaning that such studies may be measuring effector matching rather than
imitation of the other’s congural body movements. Future advances in automated
coding of video should allow closer investigation of the delity or precision of the
imitation that occurs during naturalistic mimicry of others’ actions.
Another concern relates to the reliability of this measure. Genschow et al.
(2017) measured the internal consistency of mimicry performance in their study
by calculating separate mimicry scores for even and odd minutes of the task.
These scores showed a negative correlation, indicating low reliability. The low
reliability of the mimicry task is difcult to address, however, due to the natural-
istic nature of the task. Unlike measures of automatic imitation, the mimicry task
does not involve a trial-by-trial measurement of mimicry, which makes it difcult
to compare the characteristics of movements made during the experimental
period. The lack of a trial- by- trial measurement of mimicry also limits the extent
to which we can determine how each individual target movement inuences the
motor system of the observer. Since the frequency of movements is calculated
across the entire task duration, rather than after each observed movement, it is
possible that participants do not execute any movements for a vast majority of the
task and may then execute multiple movements during a short span of time, which
would increase the frequency of executed movements. However, this increase may
be in response to a single observed movement, and that would not reect a reliable
mimicry effect. On a related note, the number of responses executed during mim-
icry tasks is typically very low (as low as ve movements; Genschow etal., 2017),
which raises concerns about the reliability of the effects interpreted based on a
limited range of data, and also shows the infrequent nature of mimicry during
naturalistic interactions. Furthermore, the complexity of the naturalistic
2 Measuring Movement Imitation
28
interactions involved in the mimicry task may introduce additional variability
from multiple sources. One such source is the irrelevant task that participants
perform within this paradigm. Different studies involve participants describing
various types of photographs, which introduces uncertainty regarding how the
contents of the photographs may inuence mimicry. For example, if certain stim-
uli evoke strong emotional responses, mimicry tendencies may be increased over
and above action observation (Chartrand & Lakin, 2013). However, previous stud-
ies have tried to address these concerns by choosing neutral stimuli such as pho-
tographs of natural landscapes (Yabar etal., 2006). Additionally, characteristics of
the confederate such as perceived similarity with the participant may also modu-
late the mimicry of actions. These considerations highlight the low suitability of
the mimicry task for situations requiring multiple repeated trials for each experi-
mental condition, such as when investigating the neurocognitive mechanisms
underpinning the enhancement or control of perception-action links, but it is valu-
able in understanding the role of various social factors in modulating the tendency
to mimic observed actions during more naturalistic social interactions.
Another naturalistic type of mimicry is contagious yawning, which refers to the
onset of a yawn triggered by seeing, hearing, or even thinking about another person
yawning (Platek et al., 2003, 2005; Provine, 1986; 1989). Although contagious
yawning can be triggered even in the absence of perceptual inputs, the execution of
a yawn in response to observing another person’s yawn is often considered as a form
of imitation. The way contagious yawning is measured in experimental settings is
analogous to the way movement mimicry is measured. In one study, participants
observed seven-second video recordings of a model in either a yawning, laughing,
or neutral condition (Platek etal., 2003, Experiment 1). The inclusion of a laughing
condition acts as a control to ensure that the yawning behaviour is not triggered by
the observation of mouth-opening behaviour, whereas the inclusion of a neutral
condition provides a baseline condition where no elicited yawns should be expected.
The experimenter coded participants’ behaviour (yawn, laugh, other, or no behav-
iour) and found that the incidence rate of evoked yawning (i.e. number of partici-
pants who yawned at least once) after watching yawning videos was 41.5%, while
that after non-yawning videos was 9%. Furthermore, 60% of individuals who
yawned at least once engaged in yawning more than once. The inclusion of a control
condition is essential to determine the specicity of yawning behaviour, and not all
studies include such control conditions (e.g. Helt etal., 2010). Previously discussed
limitations of the measurement of movement mimicry such as low number of trials,
lack of a trial-by-trial structure, and subjective measurement are also relevant to the
measurement of contagious yawning. Nonetheless, previous research on contagious
yawning has revealed several associations between contagious yawning and social
cognitive performance. Platek etal. (2003) found that individuals who engaged in
contagious yawning were quicker in identifying their own faces and better at infer-
ring mental states, compared to those who did not engage in contagious yawning. It
has also been hypothesised that contagious yawning may reect a basic capacity for
empathic processes such as emotion contagion (Palagi etal., 2020; Platek etal.,
D. Khemka and C. Catmur
29
2003, 2005). However, the evidence supporting a link between contagious yawning
and empathy is inconclusive (Massen & Gallup, 2017; Gallup, 2021).
Mimicking Action Timing (Synchrony)
Synchrony has been dened as when individuals coordinate their movements (which
may or may not be similar in form) to coincide with those of others in terms of tim-
ing or rhythm (Lakin, 2013). Although research on the synchronisation of actions is
not as prevalent as research on the mimicry of actions, the measurement of synchro-
nised actions has evolved from initial measures that relied on subjective ratings
during the observation of interactions (e.g. Zivotofsky & Hausdorff, 2007) to more
sophisticated methods such as motion capture systems, which can index the tempo-
ral organisation of movements and the dynamic characteristics of movement pat-
terns. The current section focuses on the latter group of methods for measuring
movement synchrony. There are two modes of movement coordination that are rel-
evant to research on synchrony: in-phase and anti-phase. In-phase coordination
refers to movements that are similar in both timing and form (e.g. two people walk-
ing together at the same rate of movement while performing the same right-left foot
movement). Anti-phase coordination, on the other hand, refers to movements that
are similar in timing but not in form (e.g. two people walking together at the same
movement rate but one person performing a right-left foot movement while the
other person performing a left-right foot movement or vice versa). Both in-phase
coordination and anti-phase coordination reect stable forms of synchrony.
The measurement of synchrony of actions typically involves two individuals per-
forming a repetitive movement (e.g. walking on a treadmill, stepping on a standard
exercise step, or tapping their ngers) while wearing a motion tracking system (see
section “Kinematic Measures of Movement Imitation” for an overview of the differ-
ent types of motion tracking systems). The key measure of interest in these tasks is
the relative phase relationship between the movements of both individuals calcu-
lated separately to compare oscillatory end effectors (e.g. left-left and right-right leg
movements). Relative phases are standardised to a range of 0–180°, reecting in-
phase and anti-phase coordination, respectively. For each participant, estimates of
the time spent on each relative phase region (ranging in xed intervals between 0°
and 180°) are calculated when the participant and the other individual perform the
key movement together. Synchrony between movements is indicated by a concen-
tration of relative phase angles in the phase regions close to 0° and/or 180°. Another
measure that enables the quantitative measurement of automatic synchronisation of
actions is the power spectrum overlap between movements, which measures the
percentage of movement frequencies common to interacting pairs.
The synchronisation of actions has previously been investigated with different
types of movements. For example, in Miles etal. (2010), participants completed a
task involving stepping on a standard exercise step where the movements of the
2 Measuring Movement Imitation
30
participant and the confederate were measured using electromagnetic motion sen-
sors attached to each leg above the knee. The exercise steps were aligned adjacent
to each other with the confederate’s exercise step positioned in front of the partici-
pant to provide participants a clear view of the confederate. Both individuals wore
headphones where white noise was played for the participant and metronome tones
were played for the confederate, which was used to time the steps. Movements were
recorded during both a baseline stage (when only the participant was moving; used
to account for chance coordination) and a test stage when both the participant and
confederate were moving. Participants’ movements were concentrated in the 0–20°
range more during the test than during the baseline stage, suggesting that synchrony
between the movements of the participant and the confederate during the test stage
is attributable to the individuals coordinating their steps with each other. In a differ-
ent study, participants walked on a treadmill with infrared reectors attached to
record movements of the left and right lower legs (van Ulzen et al., 2008).
Participants with varying default stride frequencies walked side by side on the tread-
mill uninstructed while xating their gaze on a white square, which provided a view
of the other participant’s leg in the peripheral view. This study demonstrated that
individuals tend to synchronise their movements while walking together despite no
instructions.
An important consideration while discussing the measurement of synchrony is
the extent to which the interindividual coupling of movement timing and rhythm
can be considered as imitative in nature. According to the denition of imitation
provided at the start of this chapter, the production of a matching movement, at least
in naturalistic settings, rst requires the observation of another person’s movement,
which precludes executed movements to be synchronous with observed movements.
However, there is evidence to suggest that imitative processes, in some cases, may
be initiated even before the observation of the target movement (see also Chap. 8;
this volume). This evidence comes from research on the anticipatory nature of imi-
tation, which suggests that the matching of observed motor representations to one’s
own corresponding motor representation may be important to not just mirror
observed movements but also anticipate the potential course of the other’s action
(Pster etal., 2013). For example, Genschow and Brass (2015) demonstrated that
observing an event in someone (e.g. nose wrinkling or hair falling on one’s face)
triggers the anticipated action in the observer (e.g. nose scratching or hair stroking).
In this study, although participants did not observe the model perform the target
movement (i.e. nose scratching or hair stroking), they were more likely to engage in
nose scratching after observing nose wrinkling and hair stroking after observing
hair falling. A subsequent TMS study revealed preliminary evidence showing that
observing nose wrinkling leads to increased motor-evoked potentials in the biceps
muscle, which suggests that anticipation of movements may elicit activity in the
motor system in a similar manner as when perceived movements elicit neurophysi-
ological activity in the corresponding muscle group (Genschow etal., 2018). Given
that imitative processes may be initiated in anticipation of another person’s move-
ments, it is possible that imitation may be synchronous provided that the observer
D. Khemka and C. Catmur
31
can anticipate the subsequent action. However, there is currently a gap in our under-
standing of the potential overlap between movement imitation and synchrony, from
both behavioural and neuroscientic perspectives. Studies on automatic imitation
and mimicry have predominantly focused on the copying of movements elicited by
perceived movements and those investigating anticipatory imitation have not mea-
sured whether individuals tend to synchronise the target movements with those of
the model. At the neural level, it is currently unclear whether automatic imitation
and synchronisation of movements share overlapping mechanisms, which could
provide insight into the relationship between imitation and synchrony. Nevertheless,
within the existing parameters of what constitutes imitation and synchrony, it is
unlikely that the execution of spontaneous movements elicited by observed move-
ments can be synchronous. However, investigating whether individuals tend to syn-
chronise the anticipated movements of the model would be interesting to uncover a
potential overlap between imitation and synchrony.
Finally, although studies of synchrony tend to measure phase relationships
between similar actions (e.g. similar leg movements in the case of walking speed),
it may be that the similarity between the individuals’ actions is less important than
the synchrony (and thus that similar results would be obtained for the synchronisa-
tion of non-matching actions such as leg movements with arm movements; see
Catmur & Heyes, 2013 for a related nding). As such, more research is needed to
uncover the extent to which synchrony and mimicry share overlapping neurocogni-
tive mechanisms.
Summary
This chapter has reviewed a range of different methods for measuring movement
imitation. Researchers will wish to consider which methodology is most appropri-
ate for their research question, depending on the requirements of their particular
study (see Table2.1 for a summary). Such considerations may relate to practicalities
associated with data collection: whereas some methods (e.g. TMS) cannot be used
outside of the research laboratory, other measures have been or are being adapted
for use in online testing platforms (e.g. stimulus-response compatibility measures)
and in more naturalistic settings (e.g. kinematic measures with subsequent video
analysis). Other considerations relate to the type of stimuli for which one wishes to
measure imitative responses: certain measures lend themselves to study designs
involving more naturalistic interaction, whereas others require multiple trials for
each experimental condition and, as such, trade-off ecological validity against
experimental control. Finally, when designing studies of imitation or when reading
and assessing such studies, it is essential to consider appropriate control conditions
and the use of muscle-specic designs which allow one to draw conclusions regard-
ing the specicity of the responses that are being recorded.
2 Measuring Movement Imitation
32
Table 2.1 Considerations regarding the use of the different imitation measures
Measure Muscle−/action-specic? Strengths Weaknesses When to use it Other considerations
Electromyographic
(EMG) measures of
muscle activity
Yes, as long as at least two
muscles are recorded from
Response can be
averaged over time,
allowing measurement
during observation of
ongoing movements;
Can be used during
observation of either live
movements or recorded
movements (e.g. video)
Can only be used in
certain muscles (e.g.
face and neck) where
activity can be detected
even at rest;
Difcult to use outside
of laboratory settings;
Signal-to-noise ratio
can be quite low
To measure activity in
face and neck muscles
during observation of
ongoing movements
involving those
muscles
Important to control for
individual differences in
EMG signal magnitude
TMS-evoked
measures of motor
responses
Yes, as long as at least two
muscles are recorded from;
evoking simultaneous
responses in two muscles is
only possible if the motor
cortical representations are
very close (e.g. two hand
muscles)
Pulses can be delivered
at different times after
the onset of observed
actions to determine
timecourse of motor
responses;
Can be used to measure
responses to both single
movements and ongoing
actions, either live or
recorded
Can only be used in
laboratory settings
which may lower
ecological validity;
Multiple trials of the
same observed action
are required for
averaging
To measure motor
activity in a wide
range of muscles
during observation of
actions
Early responses (< c.
200ms after observed
action) are non-specic
and likely reect a
general alerting
mechanism;
MEP amplitudes reduce
when TMS pulses are
predictable;
Contra-indications to the
use of TMS in certain
populations
Kinematic measures Can be, but studies are not
always designed to look at
action specicity
Suitable for both
laboratory-based and
more naturalistic
studies;
Generates rich datasets
with large numbers of
potential parameters
Video coding/motion
analysis can be
time-consuming;
Rich datasets may lead
to multiple comparison
problems
To measure actions
during naturalistic
social interaction, in
the laboratory and (via
video coding)
elsewhere
Pre-registration of
planned analyses may
help aid replicability of
ndings
D. Khemka and C. Catmur
33
Measure Muscle−/action-specic? Strengths Weaknesses When to use it Other considerations
Stimulus-response
compatibility (SRC)
measures
Yes—In most cases, the
task involves observing two
types of movements and
responding with
corresponding actions/
muscles
High experimental
control;
Trial-by-trial design
enabling the
measurement of action
execution elicited by
specic movements
Generally, can only be
used to measure
responses to single
movements, leading to
low ecological validity;
Multiple trials per
condition required for
averaging;
Difcult to disentangle
processes involved in
imitation and
imitation-inhibition
In controlled
laboratory or online
settings to investigate
(neuro)-cognitive
mechanisms
underpinning
automatic imitation
and
imitation-inhibition;
Experimental studies
to understand
conditions under
which perception-
action links may be
modulated
Appropriate baseline
conditions are important
to distinguish between
facilitatory and inhibitory
processes
Movement mimicry
measures
Not usually—Mimicry tasks
encode the copying of
non-specic movements
using similar body parts
which may not be action−/
muscle-specic
High ecological validity;
Allows measurement of
imitation during
in-person dyadic
interactions that are
more similar to real-life
interactions compared to
SRC tasks
Low experimental
control;
Imprecise
measurement of
mimicked actions;
Poor (psychometric)
reliability
To study conditions
under which
perception-action links
are modulated using
in-person dyadic
interactions
Video versions of
mimicry tasks provide
increased experimental
control but do not resolve
the issues related to the
imprecise measurement
of mimicked responses
Action synchrony
measures
Not usually—synchrony
tasks rarely ask participants
to perform more than one
type of movement (e.g.
walking)
Measured during
in-person dyadic
interactions;
Motion trackers enable
the measurement of the
precise strength of
synchronisation
Not necessarily
imitative, limiting the
extent to which
ndings involving such
measures can be
generalised to imitative
behaviour
To measure movement
coordination between
individuals, in the
laboratory and (via
video coding)
elsewhere
Unclear to what extent
synchrony involves the
same neurocognitive
mechanisms as those
involved in imitation
2 Measuring Movement Imitation
34
References
Aicken, M.D., Wilson, A.D., Williams, J.H. G., & Mon-Williams, M. (2007). Methodological
issues in measures of imitative reaction times. Brain and Cognition, 63, 304–308. https://doi.
org/10.1016/j.bandc.2006.09.005
Ainley, V., Brass, M., & Tsakiris, M. (2014). Heartfelt imitation: High interoceptive awareness is
linked to greater automatic imitation. Neuropsychologia, 60, 21–28. https://doi.org/10.1016/j.
neuropsychologia.2014.05.010
Anzola, G.P., Bertoloni, G., Buchtel, H. A., & Rizzolatti, G. (1977). Spatial compatibility and
anatomical factors in simple and choice reaction time. Neuropsychologia, 15, 295–302. https://
doi.org/10.1016/0028- 3932(77)90038- 0
Bertenthal, B.I., Longo, M. R., & Kosobud, A. (2006). Imitative response tendencies following
observation of intransitive actions. Journal of Experimental Psychology: Human Perception
and Performance, 32, 210–225. https://doi.org/10.1037/0096- 1523.32.2.210
Brass, M., Bekkering, H., Wohlschläger, A., & Prinz, W. (2000). Compatibility between observed
and executed nger movements: Comparing symbolic, spatial, and imitative cues. Brain and
Cognition, 44, 124–143. https://doi.org/10.1006/brcg.2000.1225
Brass, M., Zysset, S., & von Cramon, D.Y. (2001). The inhibition of imitative response tendencies.
NeuroImage, 14, 1416–1423. https://doi.org/10.1016/j.neuropsychologia.2004.06.018
Catmur, C., & Heyes, C. (2011). Time course analyses conrm independence of imitative and spa-
tial compatibility. Journal of Experimental Psychology: Human Perception and Performance,
37, 409–421. https://doi.org/10.1037/a0019325
Catmur, C., & Heyes, C. (2013). Is it what you do, or when you do it? The roles of contingency
and similarity in pro-social effects of imitation. Cognitive Science, 37, 1541–1552. https://doi.
org/10.1111/cogs.12071
Catmur, C., & Heyes, C. (2019). Mirroring ‘meaningful’ actions: Sensorimotor learning modu-
lates imitation of goal-directed actions. Quarterly Journal of Experimental Psychology, 72,
322–334. https://doi.org/10.1080/17470218.2017.1344257
Catmur, C., Walsh, V., & Heyes, C. (2007). Sensorimotor learning congures the human mirror
system. Current Biology, 17, 1527–1531. https://doi.org/10.1016/j.cub.2007.08.006
Cattaneo, L., Fabbri-Destro, M., Boria, S., Pieraccini, C., Monti, A., Cossu, G., & Rizzolatti,
G. (2007). Impairment of actions chains in autism and its possible role in intention understand-
ing. Proceedings of the National Academy of Sciences of the United States of America, 104,
17825–17830. https://doi.org/10.1073/pnas.0706273104
Cavallo, A., Heyes, C., Becchio, C., Bird, G., & Catmur, C. (2014). Timecourse of mirror and
counter-mirror effects measured with transcranial magnetic stimulation. Social Cognitive and
Affective Neuroscience, 9, 1082–1088. https://doi.org/10.1093/scan/nst085
Chartrand, T.L., & Bargh, J.A. (1999). The chameleon effect: The perception–behavior link and
social interaction. Journal of Personality and Social Psychology, 76, 893–910. https://doi.
org/10.1037//0022-3514.76.6.893
Chartrand, T. L., & Bargh, J. A. (2002). Nonconscious motivations: Their activation, opera-
tion, and consequences. In A.Tesser, D.A. Stapel, & J.V. Wood (Eds.), Self and motiva-
tion: Emerging psychological perspectives (pp.13–41). American Psychological Association.
https://doi.org/10.1037/10448- 001
Chartrand, T. L., & Lakin, J. L. (2013). The antecedents and consequences of human behav-
ioral mimicry. Annual Review of Psychology, 64, 285–308. https://doi.org/10.1146/
annurev- psych- 113011- 143754
Chen, R., Cros, D., Curra, A., Di Lazzaro, V., Lefaucheur, J.P., Magistris, M.R., Mills, K., Rösler,
K. M., Triggs, W. J., Ugawa, Y., & Ziemann, U. (2008). The clinical diagnostic utility of
transcranial magnetic stimulation: Report of an IFCN committee. Clinical Neurophysiology:
Ofcial Journal of the International Federation of Clinical Neurophysiology, 119, 504–532.
https://doi.org/10.1016/j.clinph.2007.10.014
D. Khemka and C. Catmur
35
Cook, J., & Bird, G. (2011). Social attitudes differentially modulate imitation in adoles-
cents and adults. Experimental Brain Research, 211, 601–612. https://doi.org/10.1007/
s00221- 011- 2584- 4
Cook, J.L., & Bird, G. (2012). Atypical social modulation of imitation in autism Spectrum condi-
tions. Journal of Autism and Developmental Disorders, 42, 1045–1051. https://doi.org/10.1007/
s10803- 011- 1341- 7
Cook, J.L., Blakemore, S.J., & Press, C. (2013). Atypical basic movement kinematics in autism
spectrum conditions. Brain, 136, 2816–2824. https://doi.org/10.1093/brain/awt208
Cook, J., Swapp, D., Pan, X., Bianchi-Berthouze, N., & Blakemore, S.J. (2014). Atypical interfer-
ence effect of action observation in autism spectrum conditions. Psychological Medicine, 44,
731–740. https://doi.org/10.1017%2FS0033291713001335
Cracco, E., & Brass, M. (2019). Reaction time indices of automatic imitation measure imitative
response tendencies. Consciousness and Cognition, 68, 115–118. https://doi.org/10.1016/j.
concog.2019.01.001
Cracco, E., Bardi, L., Desmet, C., Genschow, O., Rigoni, D., De Coster, L., Radkova, I., Deschrijver,
E., & Brass, M. (2018). Automatic imitation: A meta-analysis. Psychological Bulletin, 144,
453–500. https://doi.org/10.1037/bul0000143
di Pellegrino, G., Fadiga, L., Fogassi, L., Gallese, V., & Rizzolatti, G. (1992). Understanding motor
events: A neurophysiological study. Experimental Brain Research, 91, 176–180. https://doi.
org/10.1007/BF00230027
Dimberg, U. (1982). Facial reactions to facial expressions. Psychophysiology, 19, 643–647. https://
doi.org/10.1111/j.1469- 8986.1982.tb02516.x
Fadiga, L., Fogassi, L., Pavesi, G., & Rizzolatti, G. (1995). Motor facilitation during action obser-
vation: A magnetic stimulation study. Journal of Neurophysiology, 73, 2608–2611. https://doi.
org/10.1152/jn.1995.73.6.2608
Forbes, P., & Hamilton, A. (2017). Moving higher and higher: Imitators’ movements are sensitive
to observed trajectories regardless of action rationality. Experimental Brain Research, 235,
2741–2753. https://doi.org/10.1007/s00221- 017- 5006- 4
Gallup, A. C. (2021). On the link between emotional contagion and contagious yawn-
ing. Neuroscience & Biobehavioral Reviews, 121, 18–19. https://doi.org/10.1016/j.
neubiorev.2020.11.023
Genschow, O., & Brass, M. (2015). The predictive chameleon: Evidence for anticipated social
action. Journal of Experimental Psychology: Human Perception and Performance, 41,
265–268. https://doi.org/10.1037/xhp0000035
Genschow, O., van Den Bossche, S., Cracco, E., Bardi, L., Rigoni, D., etal. (2017). Mimicry
and automatic imitation are not correlated. PLoS One, 12, e0183784. https://doi.org/10.1371/
journal.pone.0183784
Genschow, O., Bardi, L., & Brass, M. (2018). Anticipating actions and corticospinal excitabil-
ity: A preregistered motor TMS experiment. Cortex, 106, 81–92. https://doi.org/10.1016/j.
cortex.2018.04.014
Genschow, O., Schuler, J., Cracco, E., Brass, M., & Wänke, M. (2019). The effect of money prim-
ing on self-focus in the imitation-inhibition task. Experimental Psychology, 66, 423–436.
https://doi.org/10.1027/1618- 3169/a000466
Guéguen, N., & Martin, A. (2009). Incidental similarity facilitates behavioral mimicry. Social
Psychology, 40, 88–92. https://doi.org/10.1027/1864- 9335.40.2.88
Halaki, M., & Ginn, K. (2012). Normalization of EMG signals: To normalize or not to normal-
ize and what to normalize to? In G.R. Naik (Ed.), Computational intelligence in electromy-
ography analysis– A perspective on current applications and future challenges. https://doi.
org/10.5772/49957
Helt, M. S., Eigsti, I. M., Snyder, P. J., & Fein, D. A. (2010). Contagious yawning in
autistic and typical development. Child Development, 81, 1620–1631. https://doi.
org/10.1111/j.1467- 8624.2010.01495.x
Heyes, C. (2011). Automatic imitation. Psychological Bulletin, 137, 463–483. https://doi.
org/10.1037/a0022288
2 Measuring Movement Imitation
36
Heyes, C. (2021). Imitation. Current Biology, 31, R228–R232. https://doi.org/10.1016/j.
cub.2020.11.071
Heyes, C., Bird, G., Johnson, H., & Haggard, P. (2005). Experience modulates automatic imitation.
Cognitive Brain Research, 22, 233–240. https://doi.org/10.1016/j.cogbrainres.2004.09.009
Hofree, G., Ruvolo, P., Bartlett, M. S., & Winkielman, P. (2014). Bridging the mechanical and
the human mind: Spontaneous mimicry of a physically present android. PLoS One, 9, e99934.
https://doi.org/10.1371/journal.pone.0099934
Jansson, E., Wilson, A. D., Williams, J. H. G., & Mon-Williams, M. (2007). Methodological
problems undermine tests of the ideo-motor conjecture. Experimental Brain Research, 182,
549–558. https://doi.org/10.1007/s00221- 007- 1013- 1
Kaiser, J., Crespo-Llado, M.M., Turati, C., & Geangu, E. (2017). The development of spontaneous
facial responses to others’ emotions in infancy: An EMG study. Scientic Reports, 7, 17500.
https://doi.org/10.1038/s41598- 017- 17556- y
Kerr, N.L. (1998). HARKing: Hypothesizing after the results are known. Personality and Social
Psychology Review: An Ofcial Journal of the Society for Personality and Social Psychology,
Inc, 2, 196–217. https://doi.org/10.1207/s15327957pspr0203_4
Keupp, S., Behne, T., & Rakoczy, H. (2018). The rationality of (over)imitation. Perspectives on
Psychological Science, 13, 678–687. https://doi.org/10.1177/1745691618794921
Khemka, D., Ahmadilari, N., Bird, G., & Catmur, C. (2021). Imitation in one’s own presence:
No specic effect of self-focus on imitation. Acta Psychologica, 212, 103194. https://doi.
org/10.1016/j.actpsy.2020.103194
Kilner, J. M., Paulignan, Y., & Blakemore, S. J. (2003). An interference effect of observed
biological movement on action. Current Biology, 13, 522–525. https://doi.org/10.1016/
s0960-9822(03)00165-9
Kilner, J., Hamilton, A.F. D.C., & Blakemore, S.J. (2007). Interference effect of observed human
movement on action is due to velocity prole of biological motion. Social Neuroscience, 2,
158–166. https://doi.org/10.1080/17470910701428190
Krishnan-Barman, S., Forbes, P., & Hamilton, A. (2017). How can the study of action kinemat-
ics inform our understanding of human social interaction? Neuropsychologia, 105, 101–110.
https://doi.org/10.1016/j.neuropsychologia.2017.01.018
Künecke, J., Mokros, A., Olderbak, S., & Wilhelm, O. (2018). Facial responsiveness of psycho-
paths to the emotional expressions of others. PLoS One, 13, e0190714. https://doi.org/10.1371/
journal.pone.0190714
LaFrance, M., & Broadbent, M. (1976). Group rapport: Posture sharing as a nonver-
bal indicator. Group & Organization Studies, 1, 328–333. https://psycnet.apa.org/
doi/10.1177/105960117600100307
Lakin, J. L. (2013). Behavioral mimicry and interpersonal synchrony. In J. A. Hall &
M.L. Knapp (Eds.), Nonverbal communication (pp.539–575). De Gruyter Mouton. https://
doi.org/10.1515/9783110238150.539
Lakin, J. L., & Chartrand, T. L. (2003). Using nonconscious behavioral mimicry to cre-
ate afliation and rapport. Psychological Science, 14, 334–339. https://doi.org/10.1111/
1467-9280.14481
Leighton, J., Bird, G., Orsini, C., & Heyes, C. (2010). Social attitudes modulate automatic imi-
tation. Journal of Experimental Social Psychology, 46, 905–910. https://psycnet.apa.org/
doi/10.1016/j.jesp.2010.07.001
Marsh, L.E., Bird, G., & Catmur, C. (2016). The imitation game: Effects of social cues on ‘imi-
tation’ are domain-general in nature. NeuroImage, 139, 368–375. https://doi.org/10.1016/j.
neuroimage.2016.06.050
Marsh, L. E., Ropar, D., & Hamilton, A. (2019). Are you watching me? The role of audience
and object novelty in overimitation. Journal of Experimental Child Psychology, 180, 123–130.
https://doi.org/10.1016/j.jecp.2018.12.010
Maslovat, D., Chua, R., & Hodges, N.J. (2013). When unintended movements “leak” out: A star-
tling acoustic stimulus can elicit a prepared response during motor imagery and action observa-
tion. Neuropsychologia, 51, 838–844. https://doi.org/10.1016/j.neuropsychologia.2013.01.016
D. Khemka and C. Catmur
37
Massen, J. J., & Gallup, A. C. (2017). Why contagious yawning does not (yet) equate to
empathy. Neuroscience & Biobehavioral Reviews, 80, 573–585. https://doi.org/10.1016/j.
neubiorev.2017.07.006
McIntosh, D. N., Reichmann-Decker, A., Winkielman, P., & Wilbarger, J. L. (2006). When
the social mirror breaks: Decits in automatic, but not voluntary, mimicry of emo-
tional facial expressions in autism. Developmental Science, 9, 295–302. https://doi.
org/10.1111/j.1467- 7687.2006.00492.x
Miles, L.K., Grifths, J.L., Richardson, M.J., & Macrae, C.N. (2010). Too late to coordinate:
Contextual inuences on behavioral synchrony. European Journal of Social Psychology, 40,
52–60. https://doi.org/10.1002/ejsp.721
Naish, K.R., Houston-Price, C., Bremner, A.J., & Holmes, N.P. (2014). Effects of action observa-
tion on corticospinal excitability: Muscle specicity, direction, and timing of the mirror response.
Neuropsychologia, 64, 331–348. https://doi.org/10.1016/j.neuropsychologia.2014.09.034
Niechwiej-Szwedo, E., Gonzalez, D., Nouredanesh, M., & Tung, J. (2018). Evaluation of the leap
motion controller during the performance of visually-guided upper limb movements. PLoS
One, 13, e0193639. https://doi.org/10.1371/journal.pone.0193639
Palagi, E., Celeghin, A., Tamietto, M., Winkielman, P., & Norscia, I. (2020). The neuroethology of
spontaneous mimicry and emotional contagion in human and non-human animals. Neuroscience
and Biobehavioral Reviews, 111, 149–165. https://doi.org/10.1016/j.neubiorev.2020.01.020
Pster, R., Dignath, D., Hommel, B., & Kunde, W. (2013). It takes two to imitate: Anticipation
and imitation in social interaction. Psychological Science, 24, 2117–2121. https://doi.
org/10.1177/0956797613489139
Pitcher, D., Parkin, B., & Walsh, V. (2021). Transcranial magnetic stimulation and the under-
standing of behavior. Annual Review of Psychology, 72, 97–121. https://doi.org/10.1146/
annurev- psych- 081120- 013144
Platek, S.M., Critton, S.R., Myers, T.E., & Gallup, G.G., Jr. (2003). Contagious yawning: The
role of self-awareness and mental state attribution. Cognitive Brain Research, 17, 223–227.
https://doi.org/10.1016/S0926- 6410(03)00109- 5
Platek, S. M., Mohamed, F. B., & Gallup, G. G., Jr. (2005). Contagious yawn-
ing and the brain. Cognitive Brain Research, 23, 448–452. https://doi.org/10.1016/
j.cogbrainres.2004.11.011
Press, C., Bird, G., Flach, R., & Heyes, C. (2005). Robotic movement elicits auto-
matic imitation. Cognitive Brain Research, 25, 632–640. https://doi.org/10.1016/
j.cogbrainres.2005.08.020
Provine, R.R. (1986). Yawning as a stereotyped action pattern and releasing stimulus. Ethology,
72, 109–122. https://doi.org/10.1111/j.1439- 0310.1986.tb00611.x
Provine, R.R. (1989). Faces as releasers of contagious yawning: An approach to face detection
using normal human subjects. Bulletin of the Psychonomic Society, 27, 211–214. https://doi.
org/10.3758/BF03334587
Ramsey, R. (2018). What are reaction time indices of automatic imitation measuring? Consciousness
and Cognition, 65, 240–254. https://doi.org/10.1016/j.concog.2018.08.006
Romani, M., Cesari, P., Urgesi, C., Facchini, S., & Aglioti, S.M. (2005). Motor facilitation of the
human cortico-spinal system during observation of bio-mechanically impossible movements.
NeuroImage, 26, 755–763. https://doi.org/10.1016/j.neuroimage.2005.02.027
Rossi, S., Hallett, M., Rossini, P. M., Pascual-Leone, A., & Safety of TMS Consensus Group.
(2009). Safety, ethical considerations, and application guidelines for the use of transcranial
magnetic stimulation in clinical practice and research. Clinical Neurophysiology: Ofcial
Journal of the International Federation of Clinical Neurophysiology, 120, 2008–2039. https://
doi.org/10.1016/j.clinph.2009.08.016
Rossi, S., Antal, A., Bestmann, S., Bikson, M., Brewer, C., Brockmöller, J., Carpenter, L.L.,
Cincotta, M., Chen, R., Daskalakis, J.D., Di Lazzaro, V., Fox, M.D., George, M.S., Gilbert,
D., Kimiskidis, V.K., Koch, G., Ilmoniemi, R.J., Lefaucheur, J.P., Leocani, L., & Lisanby,
S.H. (2021). Safety and recommendations for TMS use in healthy subjects and patient popu-
lations, with updates on training, ethical and regulatory issues: Expert guidelines. Clinical
2 Measuring Movement Imitation
38
Neurophysiology: Ofcial Journal of the International Federation of Clinical Neurophysiology,
132, 269–306. https://doi.org/10.1016/j.clinph.2020.10.003
Ruggiero, M., & Catmur, C. (2018). Mirror neurons and intention understanding: Dissociating
the contribution of object type and intention to mirror responses using electromyography.
Psychophysiology, 55, e13061. https://doi.org/10.1111/psyp.13061
Santiesteban, I., White, S., Cook, J., Gilbert, S.J., Heyes, C., & Bird, G. (2012). Training social cog-
nition: From imitation to theory of mind. Cognition, 122, 228–235. https://doi.org/10.1016/j.
cognition.2011.11.004
Sartori, L., Bulgheroni, M., Tizzi, R., & Castiello, U. (2015). A kinematic study on (un)inten-
tional imitation in bottlenose dolphins. Frontiers in Human Neuroscience, 9, 446. https://doi.
org/10.3389/fnhum.2015.00446
Scarpazza, C., Làdavas, E., & Cattaneo, L. (2018). Invisible side of emotions: Somato-motor
responses to affective facial displays in alexithymia. Experimental Brain Research, 236,
195–206. https://doi.org/10.1007/s00221- 017- 5118- x
Simon, J.R. (1990). The effects of an irrelevant directional cue on human information processing.
In R.W. Proctor & T.G. Reeve (Eds.), Stimulus-response compatibility: An integrated perspec-
tive (pp.31–86). North-Holland.
Sowden, S., & Catmur, C. (2015). The role of the right temporoparietal junction in the control of
imitation. Cerebral Cortex, 25, 1107–1113. https://doi.org/10.1093/cercor/bht306
Strafella, A. P., & Paus, T. (2000). Modulation of cortical excitability during action observa-
tion: A transcranial magnetic stimulation study. Neuroreport, 11, 2289–2292. https://doi.
org/10.1097/00001756- 200007140- 00044
Stürmer, B., Aschersleben, G., & Prinz, W. (2000). Correspondence effects with manual gestures
and postures: A study of imitation. Journal of Experimental Psychology: Human Perception
and Performance, 26, 1746–1759. https://doi.org/10.1037/0096-1523.26.6.1746
Turati, C., Natale, E., Bolognini, N., Senna, I., Picozzi, M., Longhi, E., & Cassia, V.M. (2013).
The early development of human mirror mechanisms: Evidence from electromyographic
recordings at 3 and 6 months. Developmental Science, 16, 793–800. https://doi.org/10.1111/
desc.12066
van Ulzen, N. R., Lamoth, C. J., Daffertshofer, A., Semin, G. R., & Beek, P. J. (2008).
Characteristics of instructed and uninstructed interpersonal coordination while walking side-
by- side. Neuroscience Letters, 432, 88–93. https://doi.org/10.1016/j.neulet.2007.11.070
Villiger, M., Chandrasekharan, S., & Welsh, T.N. (2011). Activity of human motor system during
action observation is modulated by object presence. Experimental Brain Research, 209, 85–93.
https://doi.org/10.1007/s00221- 010- 2522- x
Voelkl, B., & Huber, L. (2007). Imitation as faithful copying of a novel technique in marmoset
monkeys. PLoS One, 2, e611. https://doi.org/10.1371/journal.pone.0000611
Walsh, V., & Cowey, A. (1998). Magnetic stimulation studies of visual cognition. Trends in
Cognitive Sciences, 2, 103–110. https://doi.org/10.1016/s1364- 6613(98)01134- 6
Weeks, D.J., & Proctor, R.W. (1990). Salient-features coding in the translation between orthogo-
nal stimulus and response dimensions. Journal of Experimental Psychology: General, 119,
355–366. https://doi.org/10.1037/0096- 3445.119.4.355
Westfal, M., Cracco, E., Crusius, J., & Genschow, O. (in preparation). Implementation of an online
imitation inhibition task. https://osf.io/q7fju/
Whiten, A., Horner, V., Litcheld, C. A., & Marshall-Pescini, S. (2004). How do apes ape?
Learning & Behavior, 32, 36–52. https://doi.org/10.3758/bf03196005
Wiggett, A.J., Downing, P.E., & Tipper, S.P. (2013). Facilitation and interference in spatial and
body reference frames. Experimental Brain Research, 225, 119–131. https://doi.org/10.1007/
s00221-012-3353-8
Yabar, Y., Johnston, L., Miles, L., & Peace, V. (2006). Implicit behavioral mimicry:
Investigating the impact of group membership. Journal of Nonverbal Behavior, 30,
97–113. https://psycnet.apa.org/doi/10.1007/s10919-006-0010-6
Zivotofsky, A. Z., & Hausdorff, J. M. (2007). The sensory feedback mechanisms enabling
couples to walk synchronously: An initial investigation. Journal of Neuroengineering and
Rehabilitation, 4, 1–5. https://doi.org/10.1186/1743-0003-4-28
D. Khemka and C. Catmur
39
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), 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 license and
indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative
Commons license, unless indicated otherwise in a credit line to the material. If material is not
included in the chapter's Creative Commons license 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.
2 Measuring Movement Imitation
41© The Author(s) 2025
O. Genschow, E. Cracco (eds.), Automatic Imitation,
https://doi.org/10.1007/978-3-031-62634-0_3
Chapter 3
Emotional Mimicry
UrsulaHess andAgnetaFischer
The present chapter will focus on emotional mimicry, that is, the imitation of non-
verbal behaviors that signal emotions. Emotional mimicry has been a focus of
research and theorizing about empathy and mutual understanding since Lipps
(1907) proposed a role for imitation for the understanding of others. Lipps sug-
gested that people automatically adopt the behavior of others and that this imitation
leads—via a feedback process—to shared mental states, thereby facilitating the rec-
ognition of these mental states in others. For Lipps, the mediating process was intro-
spection. This general notion was taken up by Freud (1921) and entered the
psychotherapy literature, where it was referred to as the role of empathy in the
therapeutic process (e.g., Haase & Tepper, 1972). Given the focus on empathy and
understanding, as well as the prevalent research interests of the time, emotional
mimicry was referred to mainly in the clinical context.
In this vein, the importance of mimicry for therapy (for a detailed review, see
Chap. 15; this volume) was also underlined by Carl Rogers and others (Rogers,
1957; Scheen, 1964). They focused, however, more on the notion that the adoption
of congruent nonverbal behaviors leads to increased rapport, because it signals
rather than causes understanding à la Lipps. As such, these early views already laid
the groundwork for two of the theoretical approaches to mimicry that are still rele-
vant today: mimicry as a means to recognize the emotions of others and mimicry as
a social signal of understanding and afliation.
Much of the research discussed in this chapter will focus on the mimicry of emo-
tions via facial expressions (Hess & Fischer, 2013), but we expect similar processes
U. Hess (*)
Faculty of Life Sciences, Department of Psychology, Humboldt-University, Berlin, Germany
e-mail: Ursula.Hess@hu-berlin.de
A. Fischer
Faculty of Social and Behavioral Sciences, Department of Psychology, University of
Amsterdam, Amsterdam, Netherlands
e-mail: a.h.scher@uva.nl
42
for other nonverbal channels, such as vocal expressions (Neumann & Strack, 2000)
or emotional postures (Magnée etal., 2007). It is important to note that many behav-
iors, such as posture changes (Feese etal., 2012) and gestures (e.g., face touching;
Chartrand & Bargh, 1999), can be mimicked, but not all carry emotional meaning.
This is why we distinguish between behavioral mimicry and emotional mimicry.
Although the antecedents and consequences of behavioral and emotional mimicry
tend to overlap to some degree (Hess & Fischer, 2017), these two forms of mimicry
also show considerable differences, which will be our focus in the present chapter.
In humans, emotional mimicry is a ubiquitous phenomenon that can be readily
observed in everyday life. Interestingly, though, emotional mimicry is not restricted
to humans. Primates also mimic facial expressions. For example, play-face mimicry
has been observed in orangutans (Davila Ross etal., 2008) and in more egalitarian,
but not in more “despotic,” macaque species (Scopa & Palagi, 2016), suggesting a
social context moderator. Mimicry of play signals has also been found in dogs
(Palagi etal., 2015) and meerkats (Palagi etal., 2019). This common occurrence of
mimicry in different species suggests that mimicry is a basic element of communi-
cation in species that overtly show emotions.
Because mimicry is dened as imitation or matching, the main criterion for the
presence of mimicry implies that two (or more) individuals show the same behavior
at (roughly) the same time. In emotional mimicry, there is usually a 300- to 500-ms
delay between the visibility of the stimulus expression and the mimicry of that
expression (Dimberg etal., 2002) although slightly longer lag times can also be
observed. Yet, there are in fact many reasons why two people may show the same
behavior, which we would not dene as mimicry (Elfenbein, 2014). One example is
when two people observe the same emotion-eliciting event and react in the same
manner, without seeing each other’s emotional expression (parallel emotional
induction). In what follows, we will therefore rst differentiate mimicry from other
phenomena that may also result in matched behaviors. We will then briey outline
different theories of mimicry. Finally, we will discuss more recent theoretical
approaches that focus on top-down effects on mimicry.
Theories ofMimicry
What Is Emotional Mimicry?
As noted above, both emotional and non-emotional behaviors are mimicked. The
important difference between mimicry of emotional and non-emotional expres-
sions, such as foot-tapping or face touching, is that the latter are not intrinsically
meaningful. They tell us little about the person or their intentions, unless they are
interpreted as an emotional signal, for example, as a sign of anxiousness.
Emotional signals are dened as behaviors that are perceived by an observer as
signaling an emotional state. Although there is an ongoing debate about the degree
to which emotional expressions are actually related to internal emotional states
U. Hess and A. Fischer
43
(e.g., Crivelli & Fridlund, 2019; Hassin etal., 2013; Hess, 2017), we argue that
there is ample evidence based on the use of these expressions in the arts, lms, and
literature as well as scientic evidence that people infer emotions from such expres-
sions and that they react in function of this understanding (see Niedenthal & Brauer,
2012). A given expression does not have to be perceived as emotional. For example,
a frown can signal bothconcentration and irritation. When a nonverbal expression
is appraised as an emotional signal, it carries information about the expresser’s
intentions toward the perceiver (or some other person), such as to back away,
approach, attack, or ignore (see Scarantino etal., 2022, on the attributed meanings
of emotional expressions). As we will outline below, only when this emotional sig-
nal is perceived as afliative, will emotional mimicry be a likely consequence.
Furthermore, these signals have to be considered in context. For example, in
general, happy expressions are considered afliative and people who show happi-
ness are judged positively; however, showing happiness at a funeral is not an afli-
ative signal and leads to a negative judgment and the reduction or absence of
mimicry (Kastendieck etal., 2020). By contrast, some emotional expressions such
as disgust are generally not considered to signal afliation (Knutson, 1996) and
these are not typically mimicked (see Hess & Fischer, 2013). Mauersberger etal.
(2015) found that only a small group of participants in their study mimicked disgust
and that this effect was moderated by individual differences such that participants
higher in neuroticism were more likely to mimic disgust. Similarly to disgust, anger
can be antagonistic (Knutson, 1996) and result in reduced mimicry, but this effect
depends on the context—for example, when anger seems to be directed at a com-
mon foe it may support closeness rather than reduce it and hence be mimicked
(Bourgeois & Hess, 2008).
Non-emotional signals do not have the property to foster afliation, and even
though emotionally meaningless behaviors such as face touching are mimicked,
some of the interpersonal sequalae of emotional mimicry we discuss here do
not apply.
Related Phenomena That Are Not Emotional Mimicry
A number of phenomena have been conceptualized as either causally linked to emo-
tional mimicry or as forms of mimicry (with the associated overlapping terminol-
ogy). We argue, however, that these phenomena should not be conated with
emotional mimicry.
Emotional Contagion
A phenomenon that is often confused with mimicry is emotional contagion, and in
this vein, mimicry has sometimes been referred to as motor contagion (e.g., Becchio
et al., 2007; Blakemore & Frith, 2005). Hateld, Cacioppo, and Rapson (1993)
3 Emotional Mimicry
44
dene emotional contagion as the “catching” of someone else’s emotional state, and
they consider mimicry a causal antecedent to contagion. Yet, emotional contagion
refers to a feeling state, whereas emotional mimicry refers to a (nonverbal) behav-
ior. Hence, conceptually, the two are independent. In fact, whereas both mimicry
and emotional contagion have been found in the same studies, they do not necessar-
ily co-occur (e.g., Hess & Blairy, 2001; Lundqvist & Dimberg, 1995).
Synchrony
Another concept, which has especially been used in group contexts, is interpersonal
synchrony. This is typically dened as the matching of behaviors and the coordina-
tion of movement between individuals in a temporally organized fashion during
interpersonal communication (Bernieri et al., 1988; Miles et al., 2010;
Vacharkulksemsuk & Fredrickson, 2012; Valdesolo & DeSteno, 2011). However,
whereas in mimicry there is an initiator of the behavior—the mimicked person—
followed by a time-locked response by the mimicker, synchrony can also refer to
behaviors that occur simultaneously and does not depend on the time-locked match-
ing of specic behaviors.
Automatic Imitation
Heyes (2011, p.463) denes automatic imitation as “a type of stimulus-response
compatibility effect in which the topographical features of task-irrelevant action
stimuli facilitate similar, and interfere with dissimilar responses.” A typical para-
digm involves participants making a hand movement in response to a cue while at
the same time observing another hand making the same or a different movement
(Cracco et al., 2018). Notably, the mechanisms underlying automatic imitation
(which focuses on the automatic effects of observing a movement on an intentional
movement effectuated by the observer) and mimicry (which is an automatic reac-
tion to an observed movement) are not the same. Specically, mimicry is a direct
automatic reaction to the movement, whereas automatic imitation is the modulation
of an intentional movement. In addition, as mentioned above, the signal in auto-
matic imitation does not carry emotional meaning. That said, there is nonetheless
some overlap between these phenomena, as some moderators seem to operate simi-
larly. For example, both phenomena can be found when the observed behavior is
effectuated by an avatar (Pan & Hamilton, 2015; Weyers etal., 2006) and both are
facilitated by social priming (Leighton etal., 2010; van Baaren etal., 2003). Mutual
gaze can facilitate both emotional mimicry (Mauersberger etal., 2022a; Rychlowska
etal., 2012) and automatic imitation (Wang etal., 2010), but this effect has not been
consistently found (Carr etal., 2021). By contrast, whereas emotional mimicry is
facilitated for in-groups (Bourgeois & Hess, 2008; van der Schalk etal., 2011), the
same effect does not seem to be present for automatic imitation (Genschow etal.,
2022a, b). As such, the degree of overlap between these phenomena remains
uncertain.
U. Hess and A. Fischer
45
Reactive Emotions
Finally, two people may show the same emotional expression in a time-locked man-
ner, because one person reacts emotionally to the expression of the other. Thus,
when person A shows an angry expression and person B feels insulted and reacts
with anger as well, this is not an example of emotional mimicry, even though the
expressions and timing might be very similar.
In sum, although the phenomena discussed so far all refer to matched behaviors,
they differ in whether they constitute emotional signals, occur in reaction to one
another, or result from an automatic tendency to synchronize. These differences are
important, because they may imply different underlying processes and may there-
fore also occur in different contexts and have different boundary conditions.
Different Accounts ofMimicry
Over the years, a number of different accounts of the role and function, as well as
the underlying processes related to mimicry, have been proposed. It is important to
emphasize that these accounts are generally not contradictory. Rather, we argue that
they focus on different aspects of mimicry.
Mimicry asEmbodiment
The early account by Lipps (1907) proposed a model according to which individu-
als tend to imitate the emotional displays of their interaction partners, which induces
a corresponding state that in turn informs, via introspection, the interaction partner
about the other’s emotional state. Modern-day accounts of embodied emotion rec-
ognition via mimicry (Niedenthal etal., 2017) focus on the action of mirror neurons
rather than introspection. These accounts do not necessarily stipulate overt mimicry
as a necessary component, but allow for a mediation via efferent copies (Goldman
& Sripada, 2005). The basic notion is that when people make social judgments they
simulate relevant aspects of the stimulus in a form of embodied cognition (Niedenthal
etal., 2005). That is, when judging emotional expressions, such as a smile, people
simulate this expression in sensorimotor cortex. If this simulation results in a motor
output, this output would then be (facial) mimicry (for more detail, see Wood
etal., 2016).
The notion of a simulation process that underpins social perception, in particular
with regard to emotions, has been more recently supported by research on EEG mu
responses. Specically, the mu frequency band of the EEG, measured over senso-
rimotor cortex, is suppressed not only when a person performs a motor act but also
when the person observes motor acts performed by someone else (Oberman etal.,
2007a; Pineda, 2005). Based on this nding, the mu response has been linked to
3 Emotional Mimicry
46
mirror neuron activity. A more recent study found a distinct mu suppression response
during the observation of positively and negatively valenced emotional faces (Moore
etal., 2012). These ndings suggest a role for mirror neurons for the interpretation
of social stimuli. However, there is some controversy as to whether mu suppression
is indeed a reliable indicator of mirror neuron activity (Hobson & Bishop, 2017). In
addition, mu suppression does not imply an actual motor output. Hence, the ques-
tion of how mimicry is linked to these simulation processes and whether blocking
mimicry can in fact hinder simulations remains open.
Mimicry asaMatched Motor Response
The standard view on behavioral mimicry is compatible with the mirror neuron
account above (which, however, does not require an overt mimicry response). From
this account, mimicry is an automatic, matched motor response, based on a
perception- behavior link (Chartrand & Bargh, 1999; Preston & de Waal, 2002).
Hess and Fischer (2013) refer to this idea as the Matched Motor Hypothesis, which
assumes that merely perceiving a specic nonverbal display automatically entrains
the same expression in the perceiver.
Various mechanisms have been proposed to underlie this link between percep-
tion and behavior, which include, in addition to mirror neurons, shared schemas
(Barresi & Moore, 1996), shared representations (Prinz, 1997), or spreading activa-
tion (see Chartrand & Dalton, 2009). In either case, the perceptual activity is pre-
sumed to spread to behavioral representations, which in turn increases the probability
of imitating that same behavior, without conscious awareness, control, or intent
(Chartrand & Bargh, 1999). Emotional mimicry would then just be one instantiation
of such motor behavior. Following the original Matched Motor Hypothesis, the
movements in the face are thus spontaneously copied, independent of the intentions
of the observer or expresser (see Chartrand & Bargh, 1999). More recent theorizing
allows for some level of top-down social perception processes as a moderator (e.g.,
Chartrand & Lakin, 2013).
Mimicry asaSocial Regulator
The Mimicry as Social Regulator view (Fischer & Hess, 2017; Hess, 2021; Hess &
Fischer, 2013, 2014, 2017, 2022) is different from the Matched Motor Hypothesis
in that it is based on the observation that the motivation to develop social bonds to
fulll our universal need to belong is one of the most powerful drivers of human
behavior (Baumeister & Leary, 1995). Emotional mimicry is the unconscious pro-
cess that serves this need by supporting our aim to establish social and emotional
connections and to fulll our basic need for shared understanding (Fischer & Hess,
2017). The core assumption of this view is that emotional mimicry has the function
U. Hess and A. Fischer
47
to foster afliative interactions and is dependent on the goal to afliate and to com-
municate with others that we understand them (see also Rogers, 1957; Bavelas
etal., 1986).
This view implies that the mimicry of emotional signals requires a (rapid and
usually automatic) appraisal of an emotional expression in the social context in
which it occurs before it will be imitated. Is this an angry frown or concentration?
Is this happy or malicious laughter? Whether mimicry follows or not will depend on
this appraisal. This view is fundamentally different from the embodiment perspec-
tive (see Wood etal., 2016), which assumes that mimicry contributes to emotion
decoding (see Wood et al., 2016), because the Mimicry as Social Regulator view
sees mimicry as based on emotion understanding.
According to this view, emotional mimicry is not merely based on the perception
of a facial display, but on the interpretation of the motives underlying this display in
a specic context, and thus on understanding the emotion and its meaning in context
(Hess & Fischer, 2013, 2014). Rather than merely seeing a movement of the corner
of the lips, people may understand this movement to be playful amusement, or
schadenfreude (the pleasure in the misfortune of others) or even sadistic pleasure,
depending on the context, and whereas they mimic the perceived amusement, they
do not mimic the identical expression when the movement is interpreted as sadistic
pleasure (Mauersberger etal., 2022b). In other words, emotional mimicry requires
the interpretation of signals as emotions, conveying emotional intentions in a spe-
cic context (Hess & Fischer, 2022). This is in line with one of the main functions
of mimicry, namely smoothing social interactions and establishing or maintaining
social bonds.
The Functions ofEmotional Mimicry
Four different functions of emotional mimicry have been discussed in the literature,
which are associated with the different theoretical accounts described above.
Overall, these functions are not necessarily mutually exclusive. Rather, different
theories focus more on the one or the other function.
Facilitating Emotion Understanding
The evidence on whether mimicry facilitates emotion understanding as proposed by
embodiment theories of mimicry (see Wood etal., 2016) is complex. A number of
well-controlled studies in which participants saw a series of standardized facial
expressions found no relationship between mimicry and emotion recognition accu-
racy (Blairy etal., 1999; Bogart & Matsumoto, 2010; Hess & Blairy, 2001). There
is some evidence that mimicry can speed up the emotion recognition process
3 Emotional Mimicry
48
(Niedenthal etal., 2001; Stel & van Knippenberg, 2008), but the reverse effect has
also been found (Hawk etal., 2011).
The most consistent evidence on the facilitating role of mimicry on emotion
recognition regards studies that demand subtle judgments regarding smiles, either
because the smiles are weak (Oberman etal., 2007b) or because more difcult judg-
ments are required, such as genuineness (Ipser & Cook, 2015; Maringer etal., 2011;
Rychlowska etal., 2014). However, other studies found conicting results (Hess
etal., 1998; Stel etal., 2009). Most of these studies aimed to block mimicry by a
variety of means and then compared accuracy in blocked versus unblocked trials.
Interestingly, however, some of the methods used to block mimicry (such as holding
a pen with puckered lips) do not actually block mimicry efciently (Hess & Blaison,
2016; Hess etal., 2018), but block subvocalization. In this context, it is interesting
that Ipser and Cook (2015) found that smile decoding accuracy was reduced when
participants produced a vowel—a very efcient way to block subvocalization—but
not necessarily one that would impede smiling. In short, the evidence favors no
general effect of mimicry on emotion recognition, but points to the possibility that
mimicry might be helpful for smile-related judgments in difcult decoding tasks.
Yet, there is evidence for the notion that mimicry may nonetheless contribute to
a feeling of emotion understanding. For example, Yabar and Hess (2007) found that
an interaction partner who shows congruent sad affect during an interaction is per-
ceived as more understanding—even when the person is an out-group member.
More recently, Mauersberger etal. (2015) found that the tendency to mimic sadness
(an afliative emotion) in a laboratory task predicted the positivity of daily interac-
tions in a following diary task over 7days. Conversely, the tendency to mimic dis-
gust (which was much rarer) predicted negative interactions. These data suggest that
indeed, one positive function of some forms of mimicry may be to create an atmo-
sphere of mutual understanding, which then may well result in actual better under-
standing as suggested by Rogers (1957).
Mimicry Promotes Human Afliation
Both motor mimicry (Chartrand & Lakin, 2013) and emotional mimicry (Hess &
Fischer, 2013, 2014) have been shown to not only depend on afliation but also
foster afliation. Hess and Fischer (2013, 2014) reviewed evidence that people
mimic others’ emotions more in contexts where participants have positive rather
than negative attitudes toward each other (Likowski etal., 2008), or when they are
similar rather than dissimilar (Olszanowski etal., 2022), or when they belong to the
same group rather than a different group (Bourgeois & Hess, 2008; van der Schalk
et al., 2011), or want to cooperate rather than compete with each other (Weyers
etal., 2009). This is not only the case for emotional mimicry; Lakin and Chartrand
(2013) also reported more behavioral mimicry when participants have a goal to
afliate. Thus, both behavioral mimicry and emotional mimicry are sensitive to the
nature of the relationship with the mimickee. Whether, or at least the extent to
U. Hess and A. Fischer
49
which, people mimic emotional expressions depends on the perceived intentions of
the expresser and on the observer’s goals and values. These intentions can be
inferred from the direction and type of the emotional signal, the relationship between
observer and target, and the emotional state or disposition of the observer. Moreover,
the relationship is not uni-directional, because emotional mimicry also serves to
increase perceived similarity and liking (Hess etal., 1999; Stel etal., 2008; van der
Schalk etal., 2011; Yabar & Hess, 2007).
Mimicry Enhances Social Standing
The STORM (social top-down response modulation) model (Wang & Hamilton,
2012) takes up the notions expressed above, in that it emphasizes the social function
of mimicry and its dependence on social context. However, STORM sets a different
emphasis for the function of mimicry. Here, mimicry is a Machiavellian strategy for
enhancing one’s social standing or a strategic intervention to change the social
world for self-advancement. Wang and Hamilton base their model on the observa-
tion that people increase mimicry toward those who are important for their social
welfare. Some of the evidence for this notion has also been adduced by the aflia-
tion theories mentioned above, such that people preferentially mimic others who are
nice (Likowski etal., 2008) or those who are in-group members (Bourgeois & Hess,
2008; van der Schalk etal., 2011). They also note that people increase mimicry
when they feel that their social relationship is endangered such as when they fail to
afliate with other individuals (Lakin & Chartrand, 2003) or when they are ostra-
cized by their group members (Brandenburg etal., 2022; Lakin etal., 2008).
However, much of the evidence for the model does not stem from research on
mimicry (i.e., the imitation of nonverbal behaviors), but rather is based on a variant
of the standard paradigm used in automatic imitation research (see above). In this
variant, participants rst learn social information about a hand, which then shows a
nger movement that is either congruent or incongruent with one that the partici-
pant is required to perform. The degree of interference with the participant’s move-
ment is then a sign of imitation. Given the differences in mechanisms between
automatic imitation and emotional mimicry, these ndings offer at best circumstan-
tial evidence. A later study on emotional mimicry by contrast (Carr etal., 2014) is
more in line with Wang and Hamilton’s argument in that they found emotion-spe-
cic effects of both observer and target power, congruent with the notion that social
hierarchy inuences mimicry in meaningful ways.
In essence, however, the main message of the model is that mimicry processes
(and these include in this case automatic imitation) serve to regulate the social dis-
tance to socially attractive versus unattractive targets. As such, despite many differ-
ences in conceptualization, the model is surprisingly compatible with the Mimicry
as Social Regulator model.
3 Emotional Mimicry
50
Mimicry Supports Implicit (Social) Learning
Another important potential function of mimicry regards (social) learning. This
aspect is emphasized by Kavanagh and Winkielman (2016). In fact, one of the rst
reviews on mimicry by Hess etal. (1999) noted an older developmental literature
that conceptualized mimicry as a “primitive motor code,” which might be a pri-
mary cognitive medium for learning about other people during early development.
That is, children imitate the behavior of adults and thereby learn the effects of that
behavior on others. Similarly, Kavanagh and Winkielman (2016) consider mim-
icry as a tool for implicit social learning, because it leads to the acquisition of
culturally appropriate bodily and emotional behaviors (see also Fischer, 2019).
They emphasize that this learning process and the resulting knowledge are
implicit. Thus, it cannot easily be rejected, criticized, revised, or employed by the
learner in a deliberative or deceptive manner. The function of mimicry as a mech-
anism for social learning also explains why people generally preferentially mimic
in-group members who by denition are more trusted to have the proper knowl-
edge. As such, they conclude that mimicry can be considered an honest signal of
group afliation.
According to Kavanagh and Winkielman (2016), spontaneous mimicry can be
costly when there is no focus on the in-group, because it would imply the learning
of maladaptive behaviors. Given that the in-group is the group with the same val-
ues and priorities, it is likely that humans also share their feelings with this group
and in-group members’ feelings are thus considered more informative than that of
out- group members. A child’s fear of strangers emphasizes this point. Mimicry of
in- group members therefore is benecial to the mimickee and to the mimicker,
because it supports not only mutual bonds but also the learning of culturally
appropriate behaviors by the mimicker and the observation by the mimickee that
new and appropriate behaviors are being learned. They further point out that mim-
icry that is too precise may become blatantly obvious to the mimickee and thus
appear strategic and that the actual subtle and approximate expressions that are
typically shown are more likely to serve as an honest signal. From this view, the
fostering of afliation is more of a side effect to the learning of appropriate group
signals.
In sum, different theories of mimicry converge by highlighting two functions—
mutual understanding and social afliation. They differ in the emphasis given to
each and in the exact processes that are presumed. For example, whereas embodi-
ment theories consider mimicry a means for understanding via emotion recognition,
the Mimicry as Social Regulator view presumes that emotion understanding pre-
cedes mimicry, but because mimicry signals understanding it invites a more open
emotion communication.
U. Hess and A. Fischer
51
Top-Down Inuences
According to all theories described above, facial mimicry is an automatic process
(Dimberg & Thunberg, 1998) that is difcult to suppress (Dimberg etal., 2002) and
does not necessarily require explicit awareness of the stimulus (Dimberg etal.,
2000). Theories that assume a matched motor response, based on a perception-
behavior link (Chartrand & Bargh, 1999; Preston & de Waal, 2002), originally pos-
ited that merely perceiving a specic nonverbal display automatically entrains the
same display in the perceiver. Nonetheless, there is mounting evidence for the inu-
ence of social context on both emotional mimicry and behavioral mimicry (for a
review, see Chartrand & Lakin, 2013; Fischer & Hess, 2017; Hess & Fischer, 2013).
In line with this evidence, the Mimicry as Social Regulator model considers
mimicry a social act that is inuenced by the social context of the interaction and the
social goals of the mimicker (Fischer & Hess, 2017; Hess, 2021; Hess & Fischer,
2013, 2014, 2017, 2022). It posits that mimicry is automatic but goal dependent.
The goals that are served by mimicry are to communicate with others to foster afli-
ation and to regulate interpersonal closeness. Because of this, we do not mimic our
enemies, people we do not like, or competitors. From this view, it is not the expres-
sion per se but the social interpretation of the expression in its context that drives
mimicry. This strongly implies that emotional mimicry is shaped by top-down pro-
cesses as well.
Specically, there is increasing evidence that the meaning of a given expression
in a given context impacts on mimicry. As noted above, smiles are generally consid-
ered to be afliative and therefore smile mimicry often is preserved in contexts
where other types of mimicry would be reduced or absent, for example, when the
other is an out-group member or a disliked other (e.g., Bourgeois & Hess, 2008;
Hess etal., 2017; Seibt etal., 2013; van der Schalk etal., 2011). However, there are
many types of smiles (Niedenthal etal., 2010) that are not afliative in nature (Hess
etal., 2002). Thus, people may smile as an expression of schadenfreude—the plea-
sure in the misfortune of others—or as an expression of sadistic pleasure in anoth-
er’s pain (Mauersberger et al., 2022b). These smiles are malicious rather than
afliative, and the Mimicry as Social Regulator model predicts that such smiles
would be mimicked to a lesser degree or not at all.
This notion was studied by Kastendieck etal. (2020), who showed participants
videos of individuals who were embedded in an iconic social context associated
with clear social norms regarding the prescribed emotional expressions: weddings
and funerals. As expected, participants who smiled at weddings were mimicked, but
those who smiled at funerals were mimicked less or not at all. The level of mimicry
was mediated by perceived closeness, which in turn was mediated by the perceived
appropriateness of the expression. To the degree that participants considered the
expression inappropriate to the context, they felt more distant toward the expresser
3 Emotional Mimicry
52
and mimicked their smiles to a lesser extent. Similarly, Mauersberger etal. (2022b)
found that participants mimicked individuals who laughed at funny scenes more
than those who laughed at schadenfreude scenes or disgusting scenes. Again, the
level of mimicry was mediated by perceived closeness and appropriateness of the
expression.
For the mimicry of sadness, the results seem less consistent. For example,
Kastendieck etal. (2020) and Kastendieck etal. (2022b) found that mimicry of sad
expressions does not depend on perceived closeness. This could be explained by the
strong appeal to show empathy that is signaled by sad expressions (Scarantino etal.,
2022). Still, another study has found an effect of context, showing that sad expres-
sions are not mimicked when shown by disliked others (Likowski etal., 2008).
Also, some studies found that perceived contextual appropriateness may play a
role for sadness expressions as well. For example, Fischer and Hess (2018) found
that a sad face not showing tears was mimicked, but not when showing tears. This
suggests that sad expressions that are too intense might not be mimicked (probably
because they were deemed inappropriate as well). These ndings also suggest an
effect of contextual appropriateness, if we consider that the social signal conveyed
by a sad expression is an appeal to empathize and to help (Scarantino etal., 2022).
Tears, however, are often perceived as difcult to control or overwhelming. As such,
a person who cries at a funeral may be forgiven for the “fault pas” and still liked, yet
the sadness may reduce perceived closeness as people shy away from the social cost
of helping that closeness may require of them.
In sum, there is evidence for top-down effects on emotional mimicry that depend
on the perceived motives of the expresser and the resulting meaning that is attrib-
uted to the expression in a particular context. In fact, some of the studies referred to
above, in which mimicry is reduced or absent, can be similarly interpreted even
though these authors did not measure the mediating variables. For example, in a
study by Lanzetta and Englis (1989) participants mimicked another person’s smile
only when they expected a collaboration with this person, but not when they
expected competition. One explanation for the lack of mimicry in the competition
condition could be that the smile of a competing other was interpreted as malicious-
ness or schadenfreude rather than as a signal of afliation.
Future Perspectives
The conceptualization of mimicry has changed over time, from a view that under-
stood mimicry of all behaviors—including emotional ones—as a process whereby
“one’s behavior passively and unintentionally changes to match that of others in
one’s current social environment” (Chartrand & Bargh, 1999, p.893) to a process
that heavily depends on the social context and the social motives of the interaction
partners. Both emotional mimicry and behavioral mimicry act as “social glue”
(Lakin etal., 2003), but for the most part only in contexts that are afliative (Hess
& Fischer, 2022; Lakin & Chartrand, 2013). Fortunately, for most interactions it
seems that the default stance is one where afliation is assumed. Only when there
U. Hess and A. Fischer
53
are clear signals of potential non-afliation, such as the dislike or hostile intentions
of an interaction partner (Likowski etal., 2008) or emotional deviance (Kastendieck
etal., 2020), does mimicry fail.
Notably, as noted in the introduction, the effects of afliation and social context
on mimicry are most clearly evidenced for emotional mimicry, in comparison with
behavioral mimicry, and even though some effects observed for mimicry—such as
eye-gaze effects (Wang & Hamilton, 2014) or task relevance of the mimicked
behavior (Hemed etal., 2022)—apply also to automatic imitation, others such as the
effect of in-group status do not (Genschow etal., 2022a, b). Such differences may
be expected, given the different degrees of social engagement afforded by emotional
expressions versus hand movements and the differences in social signal value
between the two. This further supports the notion that these are different processes
even though they do share some common ground.
What is an important lesson from the research and theorizing on emotional mim-
icry is that we should acknowledge that perceivers are not passive. People are not
emotion readout machines who look at a face and attach a suitable label indepen-
dent of the context and of whom the face belongs to. They also are not automatons
who move the muscle they see someone else move regardless of circumstance.
Rather, they engage in active sensemaking that takes into account the context and
the presumed goals of the interaction partner (Hess & Hareli, 2019). Thus, people
do not simply look at a face and label an expression as a smile and move the corners
of their mouth in imitation, but they judge the expression in light of the context and
what they know about the expresser. Thus, the same smile may be considered pleas-
ant or malicious and evoke divergent reactions.
However, this does not mean that facial expressions do not have any intrinsic
meaning. Rather, both context and expression contribute to the social judgment by
the perceiver (Hess & Hareli, 2018). In fact, emotional expressions can actually
provide information about context as well. In one study, participants were able to
deduce the rules of a made-up ball game based on the facial responses of the “spec-
tators” (Hareli etal., 2019). Instead of trying to decide whether facial expressions or
context dominate the judgments of perceivers, it is more realistic to propose that
observers engage in active sensemaking based on the available information. If infor-
mation about the expression is available, perceivers may use this to draw conclu-
sions about the situation, and when information about the situation is available, it
can be used to predict the likely expression. When both are available, the informa-
tion is integrated in a way that makes sense to the perceiver (Hess & Hareli, 2019).
This notion of an active perceiver—what for visual stimuli is referred to as social
vision (Adams etal., 2010)—is central to emotion communication. This can be seen
in parallel to the understanding of 4E cognition (see, e.g., Newen etal., 2018). That
is, emotion perception is a process that is embodied, embedded, enacted, and
extended. Of these 4 Es, emotion research has addressed most explicitly embodi-
ment (Niedenthal etal., 2017). Calls to understand emotions from the context in
which they occur address the notions of the importance of extrabodily processes
that underlie the notions of embeddedness and extendedness. With regard to mim-
icry, research such as by Kastendieck and colleagues (Kastendieck etal., 2020,
2022a; Mauersberger etal., 2022b) that aims to study mimicry with stimuli that are
embedded in a specic (and meaningful) context is the rst step in that direction.
3 Emotional Mimicry
54
However, it is just as important to keep in mind that emotions are enacted as
well—we decode the emotions of others not just for the sake of applying a label, but
in order to successfully interact (Hess & Kafetsios, 2022). Future research should
investigate the presumed motives of the expresser and the goals of the interaction.
Finally, mimicry research to date has, except for research on out-group mimicry,
paid little attention to the question of who interacts with whom. This is in part a
heritage of the notion that mimicry is a simple reex-like automatism. Yet, when we
consider the importance of context and the social knowledge that we have about our
interaction partners, the situation becomes more complex. Only more recently have
researchers started to be more explicitly concerned with the intersection of social
group identities. Specically, many social processes play out differently for mem-
bers of different groups and people tend to be members of more than one group.
This implies that the combination of groups that people belong to may result in very
specic effects. This in turn limits generalization across groups. For example, emo-
tion stereotypes suggest that Black men are aggressive, but this does not apply in the
same way to black women (for a review, see Hedgecoth etal., 2023). On the percep-
tual level, the wrinkles and folds of old age reduce perceived attractiveness dispro-
portionally for women compared to men. Attractiveness in turn correlates positively
with liking and perceived closeness (Sutherland & Young, 2023). As such, it will be
important for future research to consider not only the social identities of mimicker
and mimickee but also their intersection.
Summary
The present chapter focuses on emotional mimicry, that is, the mimicry of nonverbal
behaviors that signal emotions. Emotional mimicry differs from behavioral mimicry
and automatic imitation in that the actual signal—the emotional expression—carries
meaning that is relevant for the relationship between expresser and mimicker. This is
important because emotional mimicry depends crucially on perceived closeness or
afliation. We reviewed different functions of emotional mimicry that have been pro-
posed in the literature, such as facilitating afliation, emotion recognition when sig-
nals are ambiguous, social standing, and more broadly social-cultural learning. In
addition, we summarized research on top-down effects on emotional mimicry show-
ing how social judgments and the interpretation of emotional signals in a given context
inuence perceived closeness and afliation and in turn support emotional mimicry.
References
Adams, R.B., Jr., Ambady, N., Nakayama, K., & Shimojo, S. (2010). The science of social vision.
Oxford University Press.
Barresi, J., & Moore, C. (1996). Intentional relations and social understanding. Behavioral and
Brain Sciences, 19, 107–122. https://doi.org/10.1017/S0140525X00041790
U. Hess and A. Fischer
55
Baumeister, R.F., & Leary, M.R. (1995). The need to belong: Desire for interpersonal attach-
ments as a fundamental human motivation. Psychological Bulletin, 117, 497–529. https://doi.
org/10.1037/0033- 2909.117.3.497
Bavelas, J.B., Black, A., Lemery, C. R., & Mullett, J. (1986). “I show how you feel”: Motor
mimicry as a communicative act. Journal of Personality and Social Psychology, 50, 322–329.
https://doi.org/10.1037/0022- 3514.50.2.322
Becchio, C., Pierno, A., Mari, M., Lusher, D., & Castiello, U. (2007). Motor contagion from gaze:
The case of autism. Brain, 130, 2401–2411. https://doi.org/10.1093/brain/awm171
Bernieri, F.J., Reznick, J.S., & Rosenthal, R. (1988). Synchrony pseudosynchrony and dissyn-
chrony: Measuring the entrainment process in mother-infant interactions. Journal of Personality
and Social Psychology, 54, 243–253. https://doi.org/10.1037/0022- 3514.54.2.243
Blairy, S., Herrera, P., & Hess, U. (1999). Mimicry and the judgment of emotional facial expres-
sions. Journal of Nonverbal Behavior, 23, 5–41. https://doi.org/10.1023/A:1021370825283
Blakemore, S.-J., & Frith, C. (2005). The role of motor contagion in the prediction of action.
Neuropsychologia, 43, 260–267. https://doi.org/10.1016/j.neuropsychologia.2004.11.012
Bogart, K.R., & Matsumoto, D. (2010). Facial mimicry is not necessary to recognize emotion:
Facial expression recognition by people with Moebius syndrome. Social Neuroscience, 5,
241–251. https://doi.org/10.1080/17470910903395692
Bourgeois, P., & Hess, U. (2008). The impact of social context on mimicry. Biological Psychology,
77, 343–352. https://doi.org/10.1016/j.biopsycho.2007.11.008
Brandenburg, J.C., Albohn, D.N., Bernstein, M.J., Soto, J.A., Hess, U., & Adams, R.B. (2022).
Facing social exclusion: A facial EMG examination of the reafliative function of smiling.
Cognition and Emotion, 36, 741–749. https://doi.org/10.1080/02699931.2022.2041404
Carr, E.W., Winkielman, P., & Oveis, C. (2014). Transforming the mirror: Power fundamentally
changes facial responding to emotional expressions, 143, 997–1003. https://doi.org/10.1037/
a0034972
Carr, E.W., Bird, G., Catmur, C., & Winkielman, P. (2021). Dissociable effects of averted “gaze”
on the priming of bodily representations and motor actions. Acta Psychologica, 212, 103225.
https://doi.org/10.1016/j.actpsy.2020.103225
Chartrand, T.L., & Bargh, J.A. (1999). The chameleon effect: The perception-behavior link and
social interaction. Journal of Personality and Social Psychology, 76, 893–910. https://doi.
org/10.1037/0022- 3514.76.6.893
Chartrand, T.L., & Dalton, A.N. (2009). Mimicry: Its ubiquity, importance, and functionality.
In Oxford handbook of human action (pp. 458–483). Oxford University Press. https://doi.
org/10.1093/oso/9780195309980.003.0022
Chartrand, T. L., & Lakin, J. L. (2013). The antecedents and consequences of human behav-
ioral mimicry. Annual Review of Psychology, 64, 285–308. https://doi.org/10.1146/
annurev- psych- 113011- 143754
Cracco, E., Bardi, L., Desmet, C., Genschow, O., Rigoni, D., De Coster, L., Radkova, I., Deschrijver,
E., & Brass, M. (2018). Automatic imitation: A meta-analysis. Psychological Bulletin, 144,
453–500. https://doi.org/10.1037/bul0000143
Crivelli, C., & Fridlund, A. J. (2019). Inside-out: From basic emotions theory to the behav-
ioral ecology view. Journal of Nonverbal Behavior, 43, 161–194. https://doi.org/10.1007/
s10919- 019- 00294- 2
Davila Ross, M., Menzler, S., & Zimmermann, E. (2008). Rapid facial mimicry in orangutan play.
Biology Letters, 4, 27–30. https://doi.org/10.1098/rsbl.2007.0535
Dimberg, U., & Thunberg, M. (1998). Rapid facial reactions to emotional facial expressions.
Scandinavian Journal of Psychology, 39, 39–45. https://doi.org/10.1111/1467-9450.00054
Dimberg, U., Thunberg, M., & Elmehed, K. (2000). Unconscious facial reactions to emotional
facial expressions. Psychological Science, 11, 86–89. https://doi.org/10.1111/1467-9280.00221
Dimberg, U., Thunberg, M., & Grunedal, S. (2002). Facial reactions to emotional stimuli:
Automatically controlled emotional responses. Cognition and Emotion, 16, 449–472. https://
doi.org/10.1080/02699930143000356
3 Emotional Mimicry
56
Elfenbein, H. A. (2014). The many faces of emotional contagion: An affective process the-
ory of affective linkage. Organizational Psychology Review, 4, 326–362. https://doi.
org/10.1177/2041386614542889
Feese, S., Arnrich, B., Tröster, G., Meyer, B., & Jonas, K. (2012). Quantifying behavioral mimicry
by automatic detection of nonverbal cues from body motion. In 2012 International Conference
on Privacy, Security, Risk and Trust and 2012 International Conference on Social Computing.
Fischer, A. (2019). Learning from others’ emotions. In D.Dukes & F.Clement (Eds.), Foundations
of affective social learning: Conceptualizing the social transmission of value (pp.165–184).
Cambridge University Press. https://doi.org/10.1017/9781108661362.008
Fischer, A., & Hess, U. (2017). Mimicking emotions. Current Opinion in Psychology, 17, 151–155.
Fischer, A., & Hess, U. (2018, April 4–5). Facial mimicry of crying. Paper presented at the CERE
2018, Glasgow, UK.
Freud, S. (1921). Group psychology and the analysis of the ego. In The standard edition of the
complete works of Sigmund Freud (Vol. 18, pp.65–143). The Hogart Press.
Genschow, O., Pauels, E., Krugmann, K., & Winter, A. (2022a). Group membership does not mod-
ulate goal-versus movement-based imitation. Quarterly Journal of Experimental Psychology,
76, 827–837. https://doi.org/10.1177/17470218221132842
Genschow, O., Westfal, M., Cracco, E., & Crusius, J. (2022b). Group membership does not
modulate automatic imitation. Psychological Research, 86, 780–791. https://doi.org/10.1007/
s00426-021-01526-1
Goldman, A., & Sripada, C.S. (2005). Simulationist models of face-based emotion recognition.
Cognition, 94, 193–213. https://doi.org/10.1016/j.cognition.2004.01.005
Haase, R.F., & Tepper, D.T. (1972). Nonverbal components of empathic communication. Journal
of Counseling Psychology, 19, 417–424. https://doi.org/10.1037/h0033188
Hareli, S., Elkabetz, S., & Hess, U. (2019). Drawing inferences from emotion expressions: The
role of situative informativeness and context. Emotion, 19, 200–208. https://doi.org/10.1037/
emo0000368
Hassin, R.R., Aviezer, H., & Bentin, S. (2013). Inherently ambiguous: Facial expressions of emo-
tions, in context. Emotion Review, 5, 60–65. https://doi.org/10.1177/1754073912451331
Hateld, E., Cacioppo, J. T., & Rapson, R. L. (1993). Emotional Contagion. Current Directions in
Psychological Science, 2, 96–100. https://doi.org/10.1111/1467-8721.ep10770953
Hawk, S. T., Fischer, A. H., & Van Kleef, G. A. (2011). Taking your place or matching your
face: Two paths to empathic embarrassment. Emotion, 11, 502–513. https://doi.org/10.1037/
a0022762
Hedgecoth, N., Strand, N., & Adams, R.B., Jr. (2023). The intersection of race, gender/sex, and
age in emotion perception from faces and bodies. In U.Hess, R.B. Adams Jr., & R.B. Kleck
(Eds.), Emotion communication by the aging face and body (pp. 106–139). Cambridge
University Press. https://doi.org/10.1017/9781009209656.006
Hemed, E., Mark-Tavger, I., Hertz, U., Bakbani-Elkayam, S., & Eitam, B. (2022). Automatically
controlled: Task irrelevance fully cancels otherwise automatic imitation. Journal of
Experimental Psychology: General, 151, 996–1017. https://doi.org/10.1037/xge0001117
Hess, U. (2017). Emotion categorization. In C.Lefebvre & H. Cohen (Eds.), Handbook of cat-
egorization in cognitive science (2nd ed., pp. 107–126). Elsevier. https://doi.org/10.1016/
B978-0-08-101107-2.00005-1
Hess, U. (2021). Who to whom and why: The social nature of emotional mimicry. Psychophysiology,
58, e13675. https://doi.org/10.1111/psyp.13675
Hess, U., & Blairy, S. (2001). Facial mimicry and emotional contagion to dynamic emotional
facial expressions and their inuence on decoding accuracy. International Journal of
Psychophysiology, 40, 129–141. https://doi.org/10.1016/S0167- 8760(00)00161- 6
Hess, U., & Blaison, C. (2016). Blocking mimicry—Or not? In 56th Annual Meeting of the Society
for Psychophysiological Research, Minneapolis, Minnesota, September 21–25.
Hess, U., & Fischer, A. (2013). Emotional mimicry as social regulation. Personality and Social
Psychology Review, 17, 142–157. https://doi.org/10.1177/1088868312472607
Hess, U., & Fischer, A. (2014). Emotional mimicry: Why and when we mimic emotions. Social
and Personality Psychology Compass, 8, 45–57. https://doi.org/10.1111/spc3.12083
U. Hess and A. Fischer
57
Hess, U., & Fischer, A. (2017). The role of emotional mimicry in intergroup relations. In
H. Giles & J. Harwood (Eds.), Oxford encyclopedia of intergroup communication. Oxford
University Press.
Hess, U., & Fischer, A. (2022). Emotional mimicry as social regulator: Theoretical considerations.
Cognition and Emotion, 36, 785–793. https://doi.org/10.1080/02699931.2022.2103522
Hess, U., & Hareli, S. (2018). On the malleability of the meaning of contexts: The inuence of
another person’s emotion expressions on situation perception. Cognition and Emotion, 32,
185–191. https://doi.org/10.1080/02699931.2016.1269725
Hess, U., & Hareli, S. (2019). The emotion-based inferences in context (EBIC)
model. In The social nature of emotion expression (pp. 1–5). Springer. https://doi.
org/10.1007/978-3-030-32968-6_1
Hess, U., & Kafetsios, K. (2022). Infusing context into emotion perception impacts emo-
tion decoding accuracy. Experimental Psychology, 68, 285–294. https://doi.org/
10.1027/1618- 3169/a000531
Hess, U., Philippot, P., & Blairy, S. (1998). Facial reactions to emotional facial expres-