Content uploaded by Adam Calderon
Author content
All content in this area was uploaded by Adam Calderon on Mar 28, 2019
Content may be subject to copyright.
The Neuroscience of Empathy, Compassion, and Self-Compassion. http://dx.doi.org/10.1016/B978-0-12-809837-0.00008-8
Copyright © 2018 Elsevier Inc. All rights reserved.
213
CHAPTER
8
Can We Change Our Mind
About Caring for Others? The
Neuroscience of Systematic
Compassion Training
Adam Calderon, Todd Ahern, Thomas Pruzinsky
Quinnipiac University, Hamden, CT, United States
INTRODUCTION AND SCOPE OF THE REVIEW
The emerging field of applied contemplative science (Desbordes &
Negi, 2013; Farb, 2014; Giorgino, 2015; Wallace, 2007, 2012) is in large mea-
sure concerned with empirically investigating how specific mental states
and traits can be made stronger, more consistent, and resilient. In this chap-
ter, we present a review of the scientific literature on the changes in brain
activation associated with systematic training in compassion, including the
seminal work of Richard Davidson and colleagues on the neurobiological
correlates of long-term systematic compassion training in highly experi-
enced contemplative practitioners. We also review the programs of research
conducted at Stanford and Emory Universities, both of which have devel-
oped and evaluated the efficacy of their manualized and secularized com-
passion training programs. Additionally, we describe the work of Tania
Singer and her colleagues at the Max Planck Institute as well as the new
intervention (PEACE) developed at Northern Arizona University.
Finally, we also provide an integrative summary of the current neuro-
scientific findings on systematic compassion training. The summary and
synthesis of these complex findings is presented in a series of figures,
including two particularly detailed brain maps, which allow for the
graphic condensation of current research findings in a clear and compel-
ling way while simultaneously maintaining a respect for the complexity
214 8. THE NEUROSCIENCE OF SYSTEMATIC COMPASSION TRAINING
of these outcomes. These brain maps graphically capture the principal
findings of our review—that research outcomes investigating compas-
sion training consistently show correlations with activity in a matrix of
brain areas that extend from the frontal/prefrontal area posteriorly to
the temporoparietal junction (TPJ) and that incorporate multiple cortical
(e.g., dorsolateral prefrontal cortex (dlPFC), inferior frontal gyrus (IFG),
medial orbital frontal cortex (mOFC), temporal gyrus, superior temporal
sulcus (STS), and TPJ), embedded (e.g., insula and amygdala [Amyg]),
subcortical (e.g., nucleus accumbens [NAcc], striatum, and ventral teg-
mental area [VTA]), and midline (e.g., medial anterior cingulate cortex
[mACC], pregenualanterior cingulate cortex [pACC], and dorsomedial
prefrontal cortex [dmPFC]) structures in a pattern that is visible in both
EEG and fMRI studies.
Questions to Consider
Evaluation of compassion training efficacy must consider the questions
below, offered here as a heuristic framework for this compelling and rap-
idly burgeoning body of research. They are very similar to the types of
questions that have very effectively guided the multi-decade research on
psychotherapy processes and outcomes. At this stage in the development
of research on compassion training, many of the following questions re-
main unanswered:
• Cansystematiccompassiontrainingreliablyresultinspecic
neurobiological changes?
• Isitreasonabletoassumethattheseneurobiologicalchangesreliably
associate with predictable changes in cognition, emotion, behavior,
and/or perception?
• Towhatdegreemighttheseneurobiologicaland/orpsychological
changes be influenced by the specific technique employed, by the
length and intensity of the practice (i.e., the “dosage” of the training),
by specific characteristics of individuals undertaking compassion
training (e.g., devoted long-term Buddhist practitioners as compared
to novices with no particular interest in Buddhism, religion, or
spirituality), and/or by the experience and skillfulness of the
compassion training teacher?
• Isitreasonabletoassumethatanyneurobiologicaland/or
psychological changes are the direct result of the putative “active
ingredients” (e.g., the cultivation of self-compassion) of the
compassion training technique? Or, might the neurobiological
correlates/sequelae be the result of more general “non-specific”
factors such as belief in the efficacy of the technique, general learning,
and/or changes in attentional focus?
INTRODUCTION AND SCOPE OF THE REVIEW 215
• Aretheredistinctneurobiologicalsequelae/correlatesofspecic
components of training (e.g., does training in empathy vs. training in
compassion have different effects)?
• Ofwhatspecicpractical/clinical/trainingvalueistheretoknowing
the neurobiological correlates/sequelae to compassion training and/
or the specific and putatively unique components of compassion
training?
Answers to these questions ideally can be addressed through a “trans-
disciplinary study of contemplative practices” (Dunne, 2016). Develop-
ment of the CBCT and CCT models of secular compassion training pro-
vide excellent vehicles for such an approach. The training manuals for
each of these programs were developed and written by geshes. A geshe
holds a doctoral degree in Tibetan Buddhism and requires approximately
20 years to complete. There is a great deal of complexity involved in un-
derstanding the very significant differences in compassion training tech-
niques and how these differences might affect the psychological and/or
neurobiological (as well as spiritual) outcomes. Each of these programs of
research also integrates varied levels of neuroscience expertise. Similarly,
the research conducted by Richard Davidson as well as Tania Singer and
collaborators has been done in conjunction with experts in contemplative
practice as well as those with neuroscience expertise. In preparing this
chapter, we have attempted to emulate this approach by bringing togeth-
er a neuroscience teacher and researcher (Ahern), a clinical psychologist
with approximately 20 years of intensive study and practice of a range of
contemplative techniques (Pruzinsky), and an advanced student in Be-
havioral Neuroscience who has conducted research in this area as well
(Calderon).
Compassion Training Techniques Reviewed
Table 8.1 provides an overview of the compassion training techniques
discussed in this chapter. To more clearly address the question “Does
compassion training reliably result in psychological and/or neurobiologi-
cal changes?” it is essential to understand distinctions among the training
programs. The table encapsulates many of the key similarities and differ-
ences among contemporary compassion training strategies.
For example, the highly experienced meditators described in the pa-
pers by Lutz et al. (2004, 2008, 2009) and Weng et al. (2013) utilized a
form of compassion training that is embedded in the Tibetan Buddhist
tradition. For many, if not most, highly experienced meditators, the re-
ligious/spiritual context is central to the compassion training process
and directly influences the motivation to devote tens of thousands of
hours to compassion meditation practice. In contrast, CBCT and CCT
TABLE 8.1 Compassion Meditation Techniques
Type of Training
Origins and Development of the
Technique Training Protocol Brief Description
Standard Amount of
Training (“Dose”) Representative Publications
Highly
Experienced
Meditators
(HEMs)
For HEMs, training is highly
individualized and part of
an ongoing teacher–student
relationship. HEMs are committed
to the tradition and also subscribe
to Tibetan Buddhist worldview
and values
HEMs trained by highly
accomplished teachers including
training in ‘wisdom-based’
practices (e.g., a deeply practiced
form of mindfulness) intended
to reduce the probability of
being personally distressed by
negative emotions associated with
compassion training (see Lutz et al.,
2008a, b)
Highly intensive
training
ranging from
10,000–50,000 h
of meditative
experience in the
context of long-
term study and/or
meditation retreat
contexts
Lutz et al. (2004, 2008 a,b,
2009 a,b); Weng et al.
(2013)
Cognitively-
Based
Compassion
Training
(CBCT)
Developed at Emory University
by Geshe Negi Lobsang and
colleagues based on the Tibetan
Buddhist Lojong (Mind Training)
tradition
CBCT is taught in a sequence of
6 weeks addressing specific
components which are described in
Pace et al. (2009)
6 weeks of training,
twice weekly for
1 h with at-home
meditation practice
Mascaro et al. (2012, 2013,
2016); Pace et al. (2009,
2010, 2013); Reddy et al.
(2013)
Compassion
Cultivation
Training (CCT)
Developed at Stanford University by
Geshe Thupten Jinpa based on the
Tibetan Buddhist Lojong (Mind
Training) tradition
CCT is taught in a sequence of
8 weeks addressing specific
components described in Jazaieri et
al. (2013)
8 weeks of training,
once weekly 2 h
class with daily
compassion-
focused meditation
practice
Jazaieri et al. (2013, 2014,
2015); Chapin et al. (2014)
Project for
Empathy and
Compassion
Education
(PEACE)
Developed by Lisa Doskocil at
Northern Arizona University in
order to engender compassion for
others and for oneself, particularly
for use in the public school system
PEACE is presented over 6 weeks
in which participants are taught
compassion, empathy, and self-
compassion through lectures and
guided practices
Participants meet
once a week for 2 h
for 6 weeks with
daily homework
This is a new compassion
training program
implemented and tested
over a 5-year period with
outcome data collected
but not yet published
COMPASSION TRAINING PROGRAMS 217
are completely secularized, despite having been directly derived from
the same Lojong (Mind Training) tradition of Tibetan Buddhism, and
these protocols have been developed for individuals with no back-
ground and/or interest in Tibetan Buddhist epistemology, cosmology,
ontology, or psychology.
COMPASSION TRAINING PROGRAMS
Cognitively-Based Compassion Training (CBCT)
CBCT was created by Geshe Lobsang Tenzin Negi and colleagues
at Emory University. The CBCT definition of compassion includes five
separate and distinct components: (1) cognitive (recognizing suffering in
oneself or another); (2) affective (a sense of concern or affection for the oth-
er); (3) aspirational or motivational (the wish to relieve the suffering of the
other); (4) attentional (the degree of immersion and focus); and (5) behav-
ioral (the compassionate response; an action that stems from compassion)
(Dodson-Lavelle, Ozawa-de Silva, Negi, & Raison, 2015). As noted above,
while CBCT is based on the Lojong (Mind Training) tradition of Tibetan
Buddhism, it is entirely secular in its presentation.
The initial studies of CBCT focused on whether or not CBCT would
improve psychosocial functioning among adolescents in foster care (Reddy
et al., 2013). These researchers found that CBCT, despite the lack of
significant between-groups differences, was helpful to some of the ado-
lescents in this pilot study. For instance, CBCT practice sessions were cor-
related with reduced C-reactive protein (an immune marker associated
with stress) (Pace et al., 2013). In addition, CBCT has also been found to
decrease loneliness and depression when compared to a control condition
(Mascaro, Kelley, Darcher, Negi, Worthman, Miller, & Raison, 2016).
Pace et al. (2009, 2010) observed significant correlations between
amount of meditation practice and immune and behavioral responses to
psychosocial stress for adults who engaged in CBCT but not for normal
controls. The study showed that compassion meditation is an effective
method to learn how to cope with social stress and the individuals who
practiced this specific type of meditation had lower cortisol levels.
Compassion Cultivation Training (CCT)
Geshe Thupten Jinpa, in collaboration with researchers at Stanford
University and the Center for Compassion and Altruism Research and
Education (CCARE), has created a compassion intervention program
called CCT. Similar to CBCT, CCT is a manualized and secularized ap-
proach to compassion training that derives from the Tibetan Buddhist
218 8. THE NEUROSCIENCE OF SYSTEMATIC COMPASSION TRAINING
Lojong (Mind Training) tradition. CCT follows a structured protocol that
consists of a 2-h introductory orientation, 9 (once-weekly) 2-h classes,
and daily compassion-focused meditation practice. Each class includes:
(1) pedagogical instruction with active group discussion, (2) a guided
group meditation, (3) interactive practical exercises related to the spe-
cific step of the week, and (4) exercises designed to prime feelings of
openheartedness or connection to others, either through reading poetry
or through reflecting on inspiring stories (Jazaieri et al., 2013).
The first published study evaluating the efficacy of CCT in a ran-
domized-control trial found that compassion training led to increases
in compassion felt for others, compassion offered from others, and
compassion for the self, and that the amount of meditation practice
was related to increased compassion for others (i.e., a dose–response
effect) (Jazaieri et al., 2013). Another paper by Jazaieri et al. (2014), us-
ing the same sample as the 2013 study, reported increases in mindful-
ness and well-being in addition to improvements in mood (e.g., worry)
and emotional suppression. Experienced meditators had significantly
better outcomes than inexperienced meditators on measures of worry
and emotional suppression (Jazaieri et al., 2014). Furthermore, a third
clinical study also found that individuals had a reduction in pain se-
verity and anger and increased pain acceptance after participating in
CCT (Chapin et al., 2014).
Project for Empathy and Compassion Education (PEACE)
In 2012, a new compassion training program called “Project for Empa-
thy and Compassion Education (PEACE)” was developed by Lisa Dos-
kocil at Northern Arizona University. The goal of PEACE is to create
educational opportunities for communities to learn about compassion
and empathy and to intentionally foster and strengthen compassionate
interactions in our daily lives, particularly in the public schools. PEACE
is a 6-week course on compassion during which participants meet for
training once-a-week for 2 h where they attend lectures and discussions
on compassion, empathy, and positive psychology. Toward the end of
each session, the instructor leads guided compassion-related practices
(30–60 min) which vary from week to week. These activities include
yoga, compassionate role-playing, Metta meditation, mindfulness medi-
tation, and compassionate communicating. Outside of class, participants
complete a community volunteer assignment and weekly homework
tasks, including readings from the scholarly literature on compassion
and empathy as well as applying meditation and behavioral practices
to one’s life (e.g., Tonglen meditation, Vipassana meditation, explor-
ing Non-Violent Communication, doing one positive thing for oneself
each day, and engaging in one mindful daily routine). Psychosocial
THE NEUROSCIENCE OF SYSTEMATIC COMPASSION TRAINING 219
and neuroscience investigations of the effects of PEACE are underway
and are anticipated to be published over the coming year.
Highly Experienced Meditators
Highly experienced meditators represent a unique research cohort for
neuroscience investigations as they offer quasi-experimental longitudinal
investigations of the effects of long-term meditation practice, sometimes
relative to no-treatment or brief-practice controls. From a contemplative
science perspective, the kinds of meditation that these highly experienced
meditators engage in require: (1) highly focused attention; (2) sustained
over long periods of time; (3) the steady deepening of concentration over
time; (4) mindful development of metacognitive awareness; and (5) an
acute awareness of the suffering of all living beings. These meditation
techniques also often involve quite elaborate visualization practices, fur-
ther adding to the complexity of the cognitive processing which is occur-
ring. Convenience samples of such long-term meditators have been made
available through the benevolence and scientific interest of His Holiness,
The Dalai Lama, as well as through unique community samples. Results
of those studies are reviewed below.
THE NEUROSCIENCE OF SYSTEMATIC COMPASSION
TRAINING
Highly Experienced Meditators Exhibit Distinct Patterns
of Brain Activity
In a 128-channel EEG investigation, Lutz et al. (2004) found that high-
ly trained Buddhist monks (with 10,000–50,000 h of meditation practice)
were able to induce synchronized, high frequency and relatively high-am-
plitude gamma band oscillations (25–42 Hz) during meditation. These dif-
ferences were most pronounced in the frontal and parietal/temporal lobe
regions of the brain (Fig. 8.1). Gamma activity has been shown to be asso-
ciated with hyperfocused attention and “neural binding”, that is, the rapid
integration of information across multiple sensory domains (Opitz, 2010).
Historically, gamma wave band oscillations have often been overlooked
due to their similarity to high-frequency muscle activity. However, their
presentation in the Lutz et al., study following careful editing and control
of muscle artifacts in these seasoned meditators, suggests hyperattention
during the integration of complex, multisensory information, perhaps
similar to an elevated information processing state.
While Lutz and colleagues’ EEG studies revealed distinct gamma activ-
ity in medial and lateral frontoparietal electrodes for expert meditators
220 8. THE NEUROSCIENCE OF SYSTEMATIC COMPASSION TRAINING
(Lutz, Greischar, Rawlings, Ricard, & Davidson, 2004), they did not look
for activation of more specific brain areas. In 2010, Engstrom and Soder-
felt used fMRI to investigate the neural correlates of compassion medita-
tion in a single experienced meditator and found activity in left medial
prefrontal cortex (mPFC) extending to the anterior cingulate gyrus (ACC)
(Engström & Söderfeldt, 2010).
Additional studies have added to the constellation of brain areas re-
lated to expert meditation and meditation training. For example, one in-
vestigation used fMRI to examine how experienced meditators react to
emotional stimuli while meditating (Lutz, Brefczynski-Lewis, Johnstone,
& Davidson, 2008a). The stimuli were emotionally valenced sounds of
human vocalizations, such as a woman screaming or a baby crying. The
results showed that experienced meditators exhibited higher Amyg, right
TPJ, and right posterior STS (pSTS) activation compared to novices. Fur-
thermore, experts had greater activation in the insula in response to nega-
tive sounds than to positive or neutral sounds in comparison to novice
meditators. The implication of these outcomes was that, when compared
to novice meditators, experts experience more empathy, and empathic
arousal, when observing or hearing suffering. The same research group
also examined which fMRI-measured brain activity changes were also
FIGURE 8.1 Adapted from Lutz et al. (2004): (A) depicts the change in EEG brain wave
activity of a meditator who has had over 10,000 h of compassion meditation training as he
moves from a rest state to a meditative state. The fast, higher-amplitude brain waves are
Gamma oscillations. (B) Illustrates that highly experienced meditation practitioners have
greater Gamma wave activity at baseline (IB), but they also significantly increase their Gam-
ma activity as they move into a meditative state, whereas novice (Control) meditators do not.
(C) With the nose pointing toward the top of the page, this figure shows the approximate
locations of the EEG electrodes and areas that showed high levels of Gamma activity in ex-
perienced practitioners. (D) This is a graphic depiction of the lateral perspective of the right
hemisphere with Red and Orange areas representing the high Gamma wave activity. Note
that frontal and temporal-parietal areas show the greatest increase in Gamma wave activity.
THE NEUROSCIENCE OF SYSTEMATIC COMPASSION TRAINING 221
associated with heart rate (Lutz et al., 2009 a,b). Under compassion medi-
tation conditions, expert meditators exhibited increased heart rates com-
pared to novice meditators and showed increased activity in the Insula
and ACC. These outcomes are somewhat paradoxical to the notion that
expert meditators would show lower heart rates, decreased activation of
midline structures, and less empathic distress when exposed to suffering.
However, the authors suggest that, unlike novice meditators, experts have
attained through training unique regulatory capabilities (e.g., cultivating
equanimity) to allow a deeper experiencing of compassion without undue
personal or empathic distress.
In summary, examining the responses of highly experienced meditators
reveals several important neuroscientific findings. Prefrontal cortex (PFC)
and TPJ cortical regions appear to show enhanced EEG gamma bind-
ing activity and increased fMRI blood oxygen-level dependent (BOLD)
responses in expert meditators compared to novice meditators when
both are engaged in compassion meditation, particularly in response to
distressing stimuli (Lutz, Greischar, Perlman, & Davidson, 2009a). fMRI
studies have also revealed that a number of midline (e.g., ACC), embed-
ded (e.g., Insula), and subcortical structures (e.g., Amyg) are also activated
differentially by experts and novices, and that some of this activity (Insula
and ACC) is closely coupled to changes in peripheral state (e.g., heart rate)
in response to compassion-inducing stimuli. See Fig. 8.2.
Brief Compassion Training and Neurobiology
Given that the neuroscientific study of meditation is in its infancy,
relatively few studies have assessed how experimental manipulation of
compassion meditation can alter brain activity. Further, current studies
rarely compare different types of training. Taken together, however, the
extant literature does reveal a putative network of brain areas and electro-
physiological changes affected by compassion training. The use of EEG
and fMRI to study expert meditators described above has revealed impor-
tant changes in the brain wave activity of PFC and TPJ regions as well as
increased fMRI BOLD signals in pSTS, ACC, Insula, and Amyg when en-
gaged in compassion meditation in response to distressing stimuli when
compared to novices. Many of these same areas have shown altered activ-
ity in a number of subsequent studies of much less experienced medita-
tors.
For example, two studies (Desbordes et al., 2012; Mascaro et al., 2013)
found that compassion training in novices can apparently induce chang-
es in some of these previously identified brain areas. In the Desbordes
et al. (2012) study, participants randomly assigned to Mindful Attention
Training (MAT) after 8 weeks of training exhibited significantly decreased
Amyg BOLD activity in response to emotional images, even though
222 8. THE NEUROSCIENCE OF SYSTEMATIC COMPASSION TRAINING
participants were in a non-meditative rest state during the MRI proce-
dure. This decrease was not observed in either the control group or the
group engaged in CBCT training. Interestingly, these researchers also saw
a trend toward increased Amyg activation in the CBCT group, suggesting
that this CBCT tended to increase Amyg activity in response to distressing
images. There was also a trend toward a relationship between these Amyg
increases and general affective state (i.e., depression score). However, the
sample size was small and not gender matched. Therefore, these apparent
CBCT-related changes must be interpreted very cautiously.
Similarly, the second study also used fMRI but in this study they as-
sessed how CBCT might alter both empathic accuracy and BOLD signals.
FIGURE 8.2 A graphic depiction/brain map illustrating the approximate location of
increased EEG gamma activity in expert practitioners of compassion meditation (red and
orange shading with no lines; compare to Fig. 1D), as well as specific brain areas that show
increases in fMRI BOLD signals in expert meditators when compared to baseline or novice
trainees. Note the overlap in peak Gamma activity and several cortical (—) areas. Embedded
(— · · —), subcortical (− − −), and midline cortical (······) structures also showed increases in
fMRI BOLD activity in experts, and the dlPFC and NAcc showed greater coupling (↔). Some
changes in brain activity occurred bilaterally, but the most prominent changes occurred in
right hemisphere, which are illustrated above. Each area is accompanied by articles that in-
vestigated changes in brain activity: First author and year. Abbreviations: ACC, anterior cin-
gulate cortex; AI, anterior insula; mACC, medial anterior cingulate cortex; Amyg, amygdala;
dACC, dorsal ACC; dlPFC, dorsolateral prefrontal cortex; dmPFC, dorsomedial PFC; IFG,
inferior frontal cortex; IPL, inferior parietal lobule; INS, Insula; mOFC, medial orbitofrontal
cortex; NAcc, nucleus accumbens; OFC, orbitofrontal cortex; pACC, pregenual ACC; pSTS,
posterior superior temporal sulcus; pTPJ, posterior temporal parietal junction; SII, postcen-
tral somatosensory gyrus (Brodman 40); SN, substantia nigra; TPJ, temporal parietal junc-
tion; VFO, ventral frontal operculum; VTA , ventral tegmental area.
THE NEUROSCIENCE OF SYSTEMATIC COMPASSION TRAINING 223
This study too had a small sample size with gender imbalances (N = 21;
CBCT = 7 females, 6 males; Control group = 2 females, 6 males), still the
results indicated that CBCT could enhance empathic accuracy as well as al-
ter brain function. In particular, CBCT training significantly increased IFG,
STS, dmPFC, and anterior paracingulate cortex activity, and these changes
correlated with empathic accuracy (Mascaro, Riling, Negi, & Raison, 2013).
There are also data suggesting that the dlPFC, right inferior parietal
cortex (IPC), and the NAcc are affected by compassion-related meditation
training. Since compassion training encourages meditators to take the per-
spective of the sufferer, Weng et al. (2013) hypothesized that compassion
training would result in increased altruistic behavior along with chang-
es in brain activity. Using a money-distribution game, participants wit-
nessed one player providing an unfair distribution of money to another
player. The participant then had a chance to correct the apparent wrong
by contributing a sum that had to be matched by the first player. The more
money the participant gave, the higher the altruism score. The group that
received compassion training was then compared with a control group
that had undergone reappraisal training. Not only did the compassion
group redistribute more money (greater altruistic behavior) than the re-
appraisal group but also the compassion-trained participants exhibited
greater fMRI BOLD activity in the right IPC and right dlPFC. Moreover,
the degree of signal change was significantly correlated with the level of
altruistic redistribution.
Two other findings from this study are important to note. First, the
authors reported greater “coupling” between the right dlPFC and NAcc.
That is, as the activity of one region increased there was a correspond-
ing increase in the other. The degree of coupling was also associated with
altruistic behavior. In short, Weng et al. (2013) observed that compassion
training appeared to influence an interconnected network of brain areas
simultaneously rather than independently.
Second, the reappraisal training altered the BOLD signals and dlPFC-
NAcc coupling in the opposite way. Instead of a positive correlation be-
tween brain activity and redistribution behavior, there was a negative
correlation. This finding suggests that, while the same brain areas might
be affected by cognitive/meditative engagement, the type of training sig-
nificantly influenced the direction of change.
Compassion training seems to induce greater levels of positive affect
and approach behavior. Klimecki et al. (2013) investigated the affective
and neurobiological consequences of short-term compassion training.
Based on the nature of the training and prior research, they predicted that
they would see increased activation in the anterior insula and mACC in
comparison to baseline and that by empathizing more with individuals
in distressing situations, pain network activity would be enhanced, espe-
cially in response to high distress stimuli. Given the focus on compassion,
224 8. THE NEUROSCIENCE OF SYSTEMATIC COMPASSION TRAINING
they also predicted that brain areas associated with approach, affiliation,
and positive affect would show greater activation (Klimecki, Leiberg,
Lamm, & Singer, 2013). These researchers found that short-term compas-
sion training induced increased activity in the mOFC, VTA/SN (ventral
tegmental/substantia nigra area, both major sources of dopamine), puta-
men, and pallidum. Interestingly, the most pronounced changes occurred
in the right hemisphere, consistent with the above-cited research, where
the increases for the novice trainees paralleled those of expert meditators
engaging in compassion (Klimecki et al., 2013).
The goals of empathy and compassion training are similar in some ways
but distinct in others. Despite their similarity as social emotions, they tend
to activate different, and in some important ways, opposing affective and
cognitive process. Specifically, empathy training increases empathic con-
cern, perspective taking, and a sharing in the emotions of another (Singer
& Lamm, 2009). However, when witnessing another’s suffering, partici-
pants can experience empathic (personal) distress, marked by negative af-
fect, avoidance behavior, and burnout, particularly if there is insufficient
differentiation between the self and the other (Klimecki et al., 2014). Com-
passion training, on the other hand, involves a sensitivity to the suffering
of another but not necessarily a sharing of their emotional experience. In-
dividuals who undergo compassion training may exhibit greater empathy
and negative affect, but compassion is also accompanied by positive affect
in the anticipation of the alleviation of suffering, as well as approach and
helping behaviors (Klimecki, Leiberg, Ricard, & Singer, 2014). This dif-
ference between empathy and compassion suggested to Klimecki, Singer,
and colleagues that perhaps these different types of training could lead to
different patterns of brain activity.
By directly comparing empathy training with compassion training,
Klimecki et al. (2014) aimed to tease apart which patterns of brain activ-
ity were associated with which type of training. Such distinct patterns of
neurological activation might help explain the affective and behavioral
outcomes associated with each type of training. On the neural level, short-
term empathy training increased fMRI BOLD signals in the Insula (spe-
cifically, AI) and Cingulate Cortex (specifically, medial ACC) compared to
compassion training and a control group (i.e., participants who received
memory training). They also found increased activity in the temporal gy-
rus, dlPFC, operculum, and parts of basal ganglia (posterior putamen and
head of the caudate). Klimecki et al. (2014) noted that this constellation of
structures comprises the pain matrix, which becomes activated when ex-
periencing pain or witnessing other people in pain. The increased activity
of the dlPFC and middle temporal gyrus may be related to distress man-
agement, as both regions are important for emotion and pain regulation.
Compassion training, however, resulted in increased fMRI BOLD activity
in the mOFC, pregenual ACC (pACC), and striatum, including the ven-
THE NEUROSCIENCE OF SYSTEMATIC COMPASSION TRAINING 225
tral striatum-NAcc. As Klimecki et al. noted, these areas have consistently
been linked to positive affect, reward, and pleasure, as might be expected
to occur from compassion training (Klimecki et al., 2013, 2014). Given
that, when compared directly to empathy training, compassion training
resulted in less activation of the mACC and Insula and more activation
of the mOFC, pACC, and NAcc, the results suggest that different types
of meditation training have distinct effects on how the brain responds to
distressing stimuli.
Overlapping but Distinct Systems
The survey of how empathy and compassion training relate to brain
activity above suggests a number of important conclusions. First, as dis-
cussed at the beginning of this chapter, studying experts who have over
10,000 h of compassion training is a valuable approach for identifying
brain systems that might show different activity after training (e.g., Lutz
et al., 2004). Given the inherently correlational nature of those studies and
the inability to test a specific kind of compassion training as an interven-
tion limits the conclusions we can draw. Nonetheless, the painstaking ef-
forts to conduct these studies with Tibetan monks are indeed laudatory
and clearly have advanced the field.
Second, to effectively compare different types of meditation treat-
ments, it is essential that they be compared within the same experimen-
tal framework—same training duration, similar training protocols, same
measurement techniques, etc. Without such an approach, researchers will
continue to see what has been described above, overlapping but not di-
rectly comparable measures of brain activation. In the absence of direct
comparisons, it will be difficult to identify whether the differences are due
to the type of training or simply to an artifact of experimental design.
Understanding these caveats leads us to the third conclusion: many
brain areas seem to be affected by training, but identification of specific
effects can be daunting. Fig. 8.3 provides an image that is based on the cur-
rent cutting edge of empathy and compassion training research discussed
in this chapter. By no means is it exhaustive, but we believe that it illus-
trates the pattern of empirical findings related to research on the neurosci-
ence of compassion training. This current mapping suggests a sequence of
brain areas that extends from the frontal/prefrontal area posteriorly to the
temporo-parietal junction and incorporates multiple cortical (e.g., dlPFC,
IFG, mOFC, temporal gyrus, STS, and TPJ), embedded (e.g., Insula and
Amyg), subcortical (e.g., NAcc, Striatum, and VTA), and midline (e.g.,
mACC, pACC, and dmPFC) structures in a pattern that is represented by
both EEG and fMRI studies.
The complexity of this system can be reduced to some degree by group-
ing brain areas based on the networks, affective states, and behaviors to
226 8. THE NEUROSCIENCE OF SYSTEMATIC COMPASSION TRAINING
which they typically contribute. For example, the Insula and mACC are
part of the network that underlies pain perception and management,
while mOFC, pACC, VTA, and NAcc are part of the network that reg-
ulates reward, pleasure, and positive affect. Likewise, the dmPFC and
TPJ have been associated with mentalizing or thinking about the men-
tal states of others, also called Theory of Mind (ToM: see Chapter 3 for
a more comprehensive presentation of ToM) (Mascaro, Rilling, Negi, &
Raison, 2013; Weng et al., 2013).
Such anatomical and functional linkages are essential for making sense
of how different types of training might differentially influence brain
FIGURE 8.3 This graphic depiction adds more detail than presented in Figs. 1D and 2,
and provides a more comprehensive map of all of the empathy and compassion findings
detailed in this chapter. The graphic clearly demonstrates that there is substantial overlap
between studies of the relationship between empathy/compassion meditation and changes
in brain activity. There are cortical (—), embedded (— · · —), subcortical (− − −), and mid-
line cortical (······) structures. Most regions show increases in brain activity (↑), some showed
decreases (↓), and two areas showed greater coupling/connectivity (↔), depending
on whether the study was assessing empathy (Emp) or compassion (Comp) training. Areas
illustrated in Blue/Red hashing show increased activation (Red) in response to one training
regime and less activation (Blue) under a different training protocol. Abbreviations: ACC,
anterior cingulate cortex; AI, anterior insula; mACC, medial anterior cingulate cortex; Amyg,
amygdala; dACC, dorsal ACC; dlPFC, dorsolateral prefrontal cortex; dmPFC, dorsomedial
PFC; IFG, inferior frontal cortex; IPL, inferior parietal lobule; INS, Insula; mOFC, medial
orbitofrontal cortex; NAcc, nucleus accumbens; OFC, orbitofrontal cortex; pACC, pregenual
ACC; pSTS, posterior superior temporal sulcus; pTPJ, posterior temporal parietal junction;
SII, postcentral somatosensory gyrus (Brodman 40); Put, putamen; SN, substantia nigra; Str,
striatum; TPJ, temporal parietal junction; VFO, ventral frontal operculum; vStr, ventral stria-
tum; VTA , ventral tegmental area.
DIRECTIONS FOR FUTURE RESEARCH 227
activity and in turn alter affect, cognition, and behavior. The linkages also
provide a basis for future hypotheses or for making predictions about
how the training “treatments” that are being developed might enhance
one type of brain pattern or another. Klimecki (2015) and Mascaro, Darcher,
Negi, and Raison (2015) have both reviewed the distinct patterns of
brain activity associated with empathy versus compassion training with
Klimecki proposing that the type of training in which participants engage
shapes the functioning of two antagonistic systems. Empathy training tips
the balance toward one system, while compassion training tips the bal-
ance in favor of the antagonistic system. It is proposed that these changes
underlie the measured differences in emotional state and behavioral re-
sponses. Fig. 8.4A and B illustrate these findings.
DIRECTIONS FOR FUTURE RESEARCH
As impressive as many of the studies we have reviewed are, they are
also a good reminder that we are only beginning to probe the questions
outlined at the start of the chapter. To outline our view of what to explore
next, it is worth looking at a few of the limitations of the current literature.
These will provide a foundation for the future studies we propose in the
second section.
Understanding the Current Limitations
Techniques like EEG and fMRI are remarkable investigative tools.
Moreover, given the overlap between so many studies, it is reasonable
to currently conclude that meditation, empathy, compassion, and other
cognitive and affective changes really are associated with changes in brain
activity. But we must also be appropriately cautious. For example, fMRI
requires sophisticated technology, corrections for artifacts (e.g., motion,
individual differences in brain morphology, slice acquisition asynchrony,
etc.), and the correct choice of statistical methods, any of which has the po-
tential to introduce errors or unknown biases into the data. For example,
a recent report by Eklund and colleagues indicated that approximately
one in 10 published fMRI studies may be flawed due to processing er-
rors, particularly spatial autocorrelation functions (Eklund, Nichols, &
Knutsson, 2016). Thus, we recommend being cautious about taking the re-
sults of any single study as irrefutable evidence for a particular outcome.
Additionally, fMRI is an indirect measure of brain activity, recording
BOLD blood flow responses associated with an experience. While reflecting
metabolic demands of neurons and, by inference, increased activation, fMRI
does not directly measure neuronal activity. The same concern can be lev-
ied at single-photon emission computed tomography and positron-emission
228 8. THE NEUROSCIENCE OF SYSTEMATIC COMPASSION TRAINING
tomography studies. Interestingly, although not yet generating high-resolu-
tion neuroimages of neural tissue, EEG is actually a more direct and tempo-
rally precise measurement of the activation of large neural networks. With
the development of improved cortical and subcortical current source density
(CSD) algorithms directly associated with surface EEG potentials, particu-
larly with denser arrays, EEG has the potential to offer more direct measures
of electrocortical activity (compared to relatively indirect hemodynamic mea-
sures) as well as reliable and valid localization methods.
There is also the concept of neuroplasticity to consider. Such physi-
cal brain changes in response to experience comprise some of the most
exciting discoveries of contemporary neuroscience. These changes occur
throughout life, have been demonstrated at molecular, neuronal, circuit,
system, and whole-brain levels, and seem to be the basis of most, if not
all, learning and behavioral change (Sweatt, 2016). Though not discussed
here, in the compassion literature, as in much neuroscience research,
there are many references to neural plasticity. It is important to note,
however, that not a single study in this review directly measured neu-
ral alterations; such changes are inferred from blood flow or electrocor-
tical modifications. This inference is well-considered in that changes in
level of activity following training are likely due to changes in strength
of neural connections (e.g., Hebb’s Rule, Long-Term Potentiation, Long-
Term Depression, etc.; see Sweatt, 2016 for a review). Notwithstanding
the challenges in directly measuring in human studies specific alterations
in neural networks associated with interventions, it is hoped that techno-
logical advances on the horizon will someday allow such measures in a
nonintrusive manner.
It is also important to point out that even for studies in which there
was an experimental manipulation (e.g., empathy training vs. compas-
sion training vs. memory training; e.g., Klimecki et al., 2014), the rela-
tionship between brain measures (e.g., mOFC activity) and the cogni-
tive and affective states are only correlational. None of the studies have
shown directly that artificially inducing increased activity in the mOFC,
pACC, and ventral striatum, while simultaneously decreasing the activ-
ity of the left and right insula and anterior medial cingulate cortex will
induce a state of compassion that is accompanied by positive affect and
decreased negative affect. A correlational relationship has been shown
which suggests that this effect might be possible, but the causal relation-
ship between distinct profiles of brain activity and distinct cognitive and
affective states has not been directly demonstrated, as concerns empathy
and compassion research. Advancements in transcranial magnetic stim-
ulation (TMS) and deep brain stimulation (de Weijer et al., 2014; Kahan
et al., 2014) may soon allow us to perform the types of experiments we
propose below to make causal inferences, but as yet, they have not been
conducted.
DIRECTIONS FOR FUTURE RESEARCH 229
Finally, it is important to note that all of the studies referenced above
have limitations with regard to external validity (e.g., the generalizability
of the findings). As Davidson’s group notes in multiple papers: “novices
and experts differ in many respects other than simply the extent of medi-
tative training (such as culture of origin and first language),” as well as be-
lief systems, teachers and their expertise, an assumed prerequisite training
of highly experienced meditators (HEMS), etc. (Lutz, Brefczynski-Lewis,
Johnstone, & Davidson, 2008a). In addition to culture of origin and first
language differences between novices and HEMS, there may also be self-
selecting biases when recruiting study participants. Some may self-select
for meditative training. Cautions with regard to such volunteer biases are
important for consideration in future research (Demir, Haynes, Orthel-
Clark, & Ozen, 2016).
Equally important is the fact that several early experimental training
studies utilized groups that were not gender matched (i.e., different ra-
tios of males and females in the training vs. control groups [Desbordes
et al., 2012; Mascaro et al., 2013]) or studied only female participants
(Klimecki et al., 2013, 2014). In practice, recruiting sufficiently large
groups of both genders is a challenge and focusing on only one (e.g.,
females who are more empathetic) enhances homogeneity and provides
greater control. However, many of the expert studies, which were based
on predominantly male experts, also make generalizing across genders
more difficult. Future research needs to address these imbalances in
order to clarify our understanding of brain function as it pertains to em-
pathy and compassion (Sacher, Neumann, Okon-Singer, Gotowiec, &
Villringer, 2013).
Future Studies
As research advances on compassion training, the field could benefit
by focusing on three directions, by conducting studies that are more: (1)
descriptive; (2) experimental; as well as (3) deeply collaborative.
Descriptive: Being more descriptive here suggests that research-
ers compare and contrast different types of training (e.g., CBCT, CCT,
PEACE). Specific questions that can be addressed include: Is one type
of compassion training more effective than another? Is there a sequence
of training components (e.g., the specific components of CBCT, CCT,
PEACE) that is more efficacious (e.g., the CBCT training program em-
phasizes cultivation of equanimity as foundational to compassion train-
ing, whereas CCT places somewhat less emphasis on this particular
component)? Is there a way to match the specific type of training to the
specific individual characteristics (e.g., some individuals might benefit
from practicing self-compassion earlier in the training, whereas others
may not)?
230 8. THE NEUROSCIENCE OF SYSTEMATIC COMPASSION TRAINING
An increase in the overall specificity in describing the neurobiological
effects of compassion training would also be helpful. For example, utiliz-
ing more fine-grained analyses of EEG data (e.g., using CSD) as well as
comparing, contrasting, and combining EEG assessment with fMRI data
could help describe cortical and subcortical brain activation differences
between training regimens and during segmented time periods to estab-
lish if there are specific changes from baseline measures. Additionally,
it would be informative to increase the amount and type of neurophe-
nomenological forms of assessment conducted. If we could make only
one suggestion for future research on compassion training, it would be to
further develop and to refine a range of phenomenological/qualitative/
first-person methodologies (Englander & Folkesson, 2014; Lutz, Slagter,
Dunne, & Davidson, 2008b; Lutz & Thompson, 2003; Petitmengin, 2006)
in order to link subjective and objective EEG (van Lutterveld et al., 2016)
and fMRI data (Garrison et al., 2013). Such an approach could potentially
provide unique information regarding how to enhance the efficacy of the
training. For example, when some individuals engage in cultivating grati-
tude and/or self-compassion in the context of compassion training, if not
applied correctly, participants can be harshly judgmental of themselves
which can be quite a negative subjective experience that may not be
captured by current psychometric measures. Additionally, being able to
“map” this type of experience onto the neurobiological assessment may
help to clarify the neuroimaging conducted thus far.
Experimental: A compelling experimental question is whether there is
a “critical period” of development during which one can potentially maxi-
mize the efficacy of compassion training. For example, in terms of adult
developmental outcomes, would it be more beneficial to provide develop-
mentally appropriate forms of compassion training for young adolescents
when compared with older adolescents? Since adolescent brains are more
“plastic,” it is reasonable to posit that compassion training may have more
positive psychosocial and neurobiological effects if started earlier in life.
The PEACE program described earlier was developed expressly for this
purpose and we await publication of neuroscience outcomes to clarify the
differential developmental effects suggested above.
A potentially illuminating experimental approach to enhancing com-
passion could include using targeted noninvasive stimulation tech-
niques such as TMS (e.g., de Weijer et al., 2014) and Transcranial Direct
Current Stimulation (e.g., Zheng, Alsop, & Schlaug, 2011) to selectively
activate or deactivate specific brain areas in the compassion/empathy
networks. Such strategies could evaluate whether these interventions di-
rectly cause a change in the experience of compassion/empathy feelings,
cognitions, or behaviors and perhaps, even more intriguing, could de-
termine if such interventions bring about neurobiological changes that
have putatively arisen from compassion training. Ideally, since fMRI
DIRECTIONS FOR FUTURE RESEARCH 231
data show the importance of multiple brain areas (networks) changing
activity together rather than individually, it would be exciting to extend
such a study further to directly manipulate multiple areas simultane-
ously. In short, could one electromagnetically induce the effects noted in
Fig. 8.4A versus B without the need for systematic compassion training?
Deeply Collaborative: To make valid, reliable, and practical progress to-
ward the goal of maximizing the efficacy of compassion training, we believe
that researchers involved in this work must have expertise in three separate
and distinct disciplines: neuroscience, contemplative practice, and clinical
psychology. It is rare that any single individual will have true expertise in
two of these three areas and exceptionally rare for any single individual to
be expert in all three areas. Therefore, it is imperative when using neuro-
science data to evaluate the efficacy of compassion training that an expert
perspective from these three disciplines be included in order to maximize
our understanding and cultivation of the benefits of such training and to
prevent the occurrence of any potential psychological distress that might
occur as a result of intensive compassion training (i.e., primum non nocere).
FIGURE 8.4 Distinct brain activity patterns post-empathy training (A) and post-compassion
training (B) (Klimecki et al., 2013, 2014). There are cortical (—), embedded (— · · —), subcortical
(− − −), and midline cortical (······) structures. In response to each type of training, areas colored Red
represent increases in activation when the participant is confronted with distressing video clips,
whereas areas colored Blue represent less or decreased activation in response to the same stimuli.
Klimecki et al. (2014) have noted that empathy training (A) increases activation of a pain relevant
network, while compassion training (B) increases the activity of pleasure and reward areas associ-
ated with approach behavior. Abbreviations: ACC, anterior cingulate cortex; AI, anterior insula;
mACC, medial anterior cingulate cortex; Amyg, amygdala; dACC, dorsal ACC; dlPFC, dorsolateral
prefrontal cortex; dmPFC, dorsomedial PFC; IFG, inferior frontal gyrus; IPL, inferior parietal lob-
ule; INS, insula; mOFC, medial orbitofrontal cortex; NAcc, nucleus accumbens; OFC, orbitofrontal
cortex; pACC, pregenual ACC; pSTS, posterior superior temporal sulcus; pTPJ, posterior temporal
parietal junction; SII, postcentral somatosensory gyrus (Brodman 40); SN, substantia nigra; TPJ,
temporal parietal junction; VFO, ventral frontal operculum; VTA , ventral tegmental area.
232 8. THE NEUROSCIENCE OF SYSTEMATIC COMPASSION TRAINING
References
Chapin, H. L., Darnall, B. D., Seppala, E. M., Doty, J. R., Hah, J. M., & Mackey, S. C. (2014).
Pilot study of a compassion meditation intervention in chronic pain. Journal of Compas-
sionate Health Care, 1, 4. doi: 10.1186/s40639-014-0004-x.
Demir, M., Haynes, A., Orthel-Clark, H., & Ozen, A. (2016). Volunteer bias in re-
search on friendship among emerging adults. Emerging Adulthood, 5, 53–68. doi:
10.1177/2167696816641542.
Desbordes, G., & Negi, L. T. (2013). A new era for mind studies: training investigators in both
scientific and contemplative methods of inquiry. Frontiers in Human Neuroscience, 7, 741.
doi: 10.3389/fnhum.2013.00741.
Desbordes, G., Negi, L. T., Pace, T. W., Wallace, B., Raison, C. L., & Schwartz, E. L. (2012). Ef-
fects of mindful-attention and compassion meditation training on amygdala response to
emotional stimuli in an ordinary, non-meditative state. Frontiers in Human Neuroscience, 6,
292. doi: 10.3389/fnhum.2012.0029.
Desbordes, G., Gard, T., Hoge, E. A., Hölzel, B. K., Kerr, C., Lazar, S. W., et al. (2014). Mov-
ing beyond mindfulness: Defining equanimity as an outcome measure in meditation and
contemplative research. Mindfulness, 6(2), 356–372. doi: 10.1007/s12671-013-0269-8.
de Weijer, A. D., Sommer, I. E. C., Bakker, E. J., Bloemendaal, M., Bakker, C. J. G., Klomp,
D. W. J., et al. (2014). A setup for administering TMS to medial and lateral cortical ar-
eas during whole-brain FMRI recording. J Clin Neurophysiol, 31, 474–487. doi: 10.1097/
WNP.0000000000000075.
Dodson-Lavelle, B., Ozawa-de Silva, B., Negi, G. T., & Raison, C. L. (2015). Cognitively based
compassion training for adolescents. In V. M. Follette, J. Briere, D. Rozelle, J. W. Hopper,
D. I. Rome, V. M. Follette, & D. I. Rome (Eds.), Mindfulness-oriented interventions for trau-
ma: Integrating contemplative practices (pp. 343–358). New York, NY: US: Guilford Press.
Dunne, J. (2016). The transdisciplinary study of contemplative practices. Challenges and Op-
portunities November, Invited Address: Mind and Life Institute, San Diego, CA.
Eklund, A., Nichols, T. E., & Knutsson, H. (2016). Cluster failure: Why fMRI inferences for
spatial extent have inflated false-positive rates. Proceedings of the National Academy of sci-
ences, 113(28), 7900–7905. doi: 10.1073/pnas.1602413113.
Engen, H. G., & Singer, T. (2015). Compassion-based emotion regulation up-regulates ex-
perienced positive affect and associated neural networks. Social Cognitive and Affective
Neuroscience, 10(9), 1291–1301. doi: 10.1093/scan/nsv008.
Englander, M., & Folkesson, A. (2014). Evaluating the phenomenological approach to empathy
training. Journal of Humanistic Psychology, 54(3), 294–313. doi: 10.1177/0022167813493351.
Engström, M., & Söderfeldt, B. (2010). Brain activation during compassion meditation: A case
study. J. Altern Complement Med, 16(5), 597–599. doi: 10.1089/acm.2009.0309.
Farb, N. S. (2014). Can contemplative science bring meditation to (Western) life? In S.
Schmidt, H. Walach, S. Schmidt, & H. Walach (Eds.), Meditation—Neuroscientific approach-
es and philosophical implications (pp. 243–259). Cham, Switzerland: Springer International
Publishing doi: 10.1007/978-3-319-01634-4_14.
Garrison, K., Santoyo, J., Davis, J., Thornhill, T., Kerr, C., & Brewer, J. (2013). Effortless
awareness: Using real time neurofeedback to investigate correlates of posterior cingu-
late cortex activity in meditators’ self-report. Front Hum Neurosci, 7, 440. doi: 10.3389/
fnhum.2013.00440.
Giorgino, V. B. (2015). Contemplative methods meet social sciences: Back to human ex-
perience as it is. Journal for the Theory of Social Behaviour, 45(4), 461–483. doi: 10.1111/
jtsb.12078.
Jazaieri, H., Jinpa, G., McGonigal, K., Rosenberg, E. L., Finkelstein, J., Simon-Thomas, E.,
et al. (2013). Enhancing compassion: A randomized controlled trial of a compassion cul-
tivation training program. Journal of Happiness Studies, 14(4), 1113–1126. doi: 10.1007/
s10902-012-9373-z.
DIRECTIONS FOR FUTURE RESEARCH 233
Jazaieri, H., McGonigal, K., Jinpa, T., Doty, J. R., Gross, J. J., & Goldin, P. R. (2014). A random-
ized controlled trial of compassion cultivation training: Effects on mindfulness, affect,
and emotion regulation. Motivation and Emotion, 38(1), 23–35. doi: 10.1007/s11031-013-
9368-z.
Jazaieri, H., Lee, I. A., McGonigal, K., Jinpa, T., Doty, J. R., Gross, J. J., et al. (2015). A wander-
ing mind is a less caring mind: Daily experience sampling during compassion meditation
training. The Journal of Positive Psychology, 11(1), 37–50. doi: 10.1080/17439760.2015.1025418.
Kahan, J., Urner, M., Moran, R., Flandin, G., Marreiros, A., Mancini, L., et al. (2014). Resting
state functional MRI in Parkinson’s disease: The impact of deep brain stimulation on ‘ef-
fective’ connectivity. Brain, 137, 1130–1144. doi: 10.1093/brain/awu027.
Klimecki, O. M. (2015). The plasticity of social emotions. Social Neuroscience, 10(5), 466–473.
doi: 10.1080/17470919.2015.1087427.
Klimecki, O. M., Leiberg, S., Lamm, C., & Singer, T. (2013). Functional neural plasticity and
associated changes in positive affect after compassion training. Cerebral Cortex, 23, 1552–
1561. doi: 10.1093/cercor/bhs142.
Klimecki, O. M., Leiberg, S., Ricard, M., & Singer, T. (2014). Differential pattern of functional
brain plasticity after compassion and empathy training. Social Cognitive and Affective Neu-
roscience, 9, 873–879. doi: 10.1093/scan/nst060.
Lamm, C., Decety, J., & Singer, T. (2011). Meta-analytic evidence for common and distinct
neural networks associated with directly experienced pain and empathy for pain. Neuro-
Image, 54(3), 2492–2502. doi: 10.1016/j.neuroimage.2010.10.014.
Lutz, A., & Thompson, E. (2003). Neurophenomenology: Integrating subjective experience
and brain dynamics in the neuroscience of consciousness. J Conscious Stud, 9(10), 31–52.
Lutz, A., Greischar, L. L., Rawlings, N. B., Ricard, M., & Davidson, R. J. (2004). Long-
term meditators self-induced high-amplitude gamma synchrony during mental prac-
tice. Proceedings of the National Academy of Sciences, 101, 16369–16373. doi: 10.1073/
pnas.0407401101.
Lutz, A., Brefczynski-Lewis, J., Johnstone, T., & Davidson, R. (2008a). Regulation of the neu-
ral circuitry of emotion by compassion meditation: Effects of meditative expertise. PLoS
One, 3, e1897. doi: 10.1371/journal.pone.00018972004.
Lutz, A., Slagter, H. A., Dunne, J. D., & Davidson, R. J. (2008b). Attention regulation and
monitoring in meditation. Trends in Cognitive Science, 12(4), 163–169. doi: 10.1016/j.
tics.2008.01.005.
Lutz, A., Greischar, L. L., Perlman, D. M., & Davidson, R. J. (2009a). BOLD signal in insula
is differentially related to cardiac function during compassion meditation in experts vs.
novices. NeuroImage, 47(3), 1038–1046. doi: 10.1016/j.neuroimage.2009.04.081.
Lutz, A., Slagter, H. A., Rawlings, N. B., Francis, A. D., Greischar, L. L., & Davidson, R. J.
(2009b). Mental training enhances attentional stability: Neural and behavioral evidence.
The Journal of Neuroscience, 29(42), 13418–13427. doi: 10.1523/JNEUROSCI.1614-09.2009.
Mascaro, J. S., Rilling, J. K., Negi, L. T., & Raison, C. L. (2012). Compassion meditation en-
hances empathic accuracy and related neural activity. Social Cognitive and Affective Neuro-
science, 8(1), 48–55. doi: 10.1093/scan/nss095.
Mascaro, J. S., Rilling, J. K., Negi, L. T., & Raison, C. L. (2013). Pre-existing brain function
predicts subsequent practice of mindfulness and compassion meditation. NeuroImage, 69,
6935–6942. doi: 10.1016/j.neuroimage.2012.12.021.
Mascaro, J. S., Darcher, A., Negi, L. T., & Raison, C. (2015). The neural mediators of
kindness-based meditation: A theoretical model. Front Psychol, 6, 109. doi: 10.3389/
fpsyg.2015.00109.
Mascaro, J. S., Kelley, S., Darcher, A., Negi, L. T., Worthman, C., Miller, C., et al. (2016). Medi-
tation buffers medical student compassion from the deleterious effects of depression. The
Journal of Positive Psychology, 13(2), 133–142. doi: 10.1080/17439760.2016.1233348.
Opitz, B. (2010). Neural binding mechanisms in learning and memory. Neuroscience and Biobe-
havioral Reviews, 34, 1036–1046. doi: 10.1016/j.neubiorev.2009.11.001.
234 8. THE NEUROSCIENCE OF SYSTEMATIC COMPASSION TRAINING
Pace, T. W., Negi, L. T., Adame, D. D., Cole, S. P., Sivilli, T. I., Brown, T. D., et al. (2009).
Effect of compassion meditation on neuroendocrine, innate immune and behavioral
responses to psychosocial stress. Psychoneuroendocrinology, 34(1), 87–98. doi: 10.1016/j.
psyneuen.2008.08.011.
Pace, T. W., Negi, L. T., Sivilli, T. I., Issa, M. J., Cole, S. P., Adame, D. D., et al. (2010). In-
nate immune, neuroendocrine and behavioral responses to psychosocial stress do not
predict subsequent compassion meditation practice time. Psychoneuroendocrinology, 35(2),
310–315. doi: 10.1016/j.psyneuen.2009.06.008.
Pace, T. W., Negi, L. T., Dodson-Lavelle, B., Ozawa-de Silva, B., Reddy, S. D., Cole, S. P., et al.
(2013). Engagement with cognitively-based compassion training is associated with re-
duced salivary C-reactive protein from before to after training in foster care program ado-
lescents. Psychoneuroendocrinology, 38(2), 294–299. doi: 10.1016/j.psyneuen.2012.05.019.
Petitmengin, C. (2006). Describing one’s subjective experience in the second person: An in-
terview method for the science of consciousness. Phenomenology and the Cognitive Sciences,
5, 229–269 10.1007s11097-006-9022-2.
Reddy, S. D., Negi, L. T., Dodson-Lavelle, B., Ozawa-de Silva, B., Pace, T. W., Cole, S. P., et al.
(2013). Cognitive-based compassion training: A promising prevention strategy for at-risk
adolescents. Journal of Child and Family Studies, 22(2), 219–230. doi: 10.1007/s10826-012-
9571-7.
Sacher, J., Neumann, J., Okon-Singer, H., Gotowiec, S., & Villringer, A. (2013). Sexual dimor-
phism in the human brain: Evidence from neuroimaging. Magnetic Resonance Imaging,
31(3), 366–375. doi: 10.1016/j.rmi.2012.06.007.
Singer, T., & Lamm, C. (2009). The social neuroscience of empathy. Annals of the New York
Academy of Sciences, 1156, 81–96. doi: 10.1111/j.1749-6632.2009.04418.x.
Sweatt, J. D. (2016). Neural plasticity and behavior: Sixty years of conceptual advances. Jour-
nal of Neurochemistry, 139(2), 179–199. doi: 10.1111/jnc.13580.
van Lutterveld, R., Houlihan, S. D., Pal, P., Sacchet, M. D., McFarlane-Blake, C., & Brewer,
J. D. (2016). Source-space EEG neurofeedback links subjective experience with brain ac-
tivity during effortless awareness meditation. NeuroImage, 151, 117–127. doi: 10.1016/j.
neuroimage.2016.02.047.
Wallace, B. A. (2007). Contemplative science: Where Buddhism and neuroscience converge. New
York, NY, USA: Columbia University Press.
Wallace, B. A. (2012). Meditations of a Buddhist skeptic: A manifesto for the mind sciences and
contemplative practice. New York, NY, USA: Columbia University Press.
Weng, H. Y., Fox, A. S., Shackman, A. J., Stodola, D. E., Caldwell, J. Z. K., Olson, M., et al.
(2013). Compassion training alters altruism and neural responses to suffering. Psychologi-
cal Science, 24, 1171–1180. doi: 10.1177/0956797612469537.
Zheng, X., Alsop, D. C., & Schlaug, G. (2011). Effects of transcranial direct current stimulation
(tDCS) on human regional cerebral blood flow. NeuroImage, 58, 26–33. doi: 10.1016/j.
neuroimage.2011.06.018.