Empathy—the was selectively recruited when subjects read

Empathy—the ability to understand and relate to the perspectives of others—is an essential
human act. As it requires conceptualizing and identifying with the mental states of other people while
maintaining the distinction between self and other, empathy necessitates complex social cognition. There
is a general consensus in the literature that empathy is comprised of several subordinate processes,
including affective empathy (or experience sharing) and cognitive empathy (or mentalizing) (Overwalle
& Baetens, 2009; Zaki & Ochsner, 2012). Affective empathy involves resonating with another person’s
emotional experience, and is linked to activation in a circuit that includes the insula, amygdala, and
anterior cingulate cortex (ACC) (Decety & Jackson, 2006). Cognitive empathy, on the other hand, refers
to intentionally taking the perspective of the other person, and has been shown to engage midline
structures such as the superior temporal sulcus (STS) and the temporoparietal junction (TPJ) (Frith &
Frith, 2006; Spunt & Lieberman, 2012). The TPJ in particular has been emphasized as a key region for
empathetic cognition. Many studies have found evidence implicating it in comprehending another’s
viewpoint while maintaining the distinction between self and other. However, the TPJ has also been
demonstrated to be active during more basic cognitive tasks, prompting the question of whether there is
anything particularly “social” or “empathy-driven” about this brain region. This paper will address the
debate over whether the TPJ is inherently social by reviewing the literature on its role in mentalizing,
discussing its function in cognition more broadly, and examining the evidence in the context of a network

The TPJ is consistently recruited during tasks that involve contemplating the mental state of
another person. In one study, the TPJ was selectively recruited when subjects read stories about a
protagonist’s thoughts and beliefs (Saxe & Powell, 2006). In another experiment using positron emission
tomography (PET), researchers found a hemodynamic increase in the TPJ when participants were

Best services for writing your paper according to Trustpilot

Premium Partner
From $18.00 per page
4,8 / 5
Writers Experience
Recommended Service
From $13.90 per page
4,6 / 5
Writers Experience
From $20.00 per page
4,5 / 5
Writers Experience
* All Partners were chosen among 50+ writing services by our Customer Satisfaction Team

Soderberg 1

instructed to take a third person perspective compared to taking their own first person view (Ruby &
Decety, 2004). As well as mediating these explicitly empathetic acts, the TPJ has been shown to be
engaged during the identification of the goals of others’ actions. In a meta-analysis of functional imaging
studies on social cognition, tasks that required the determination of another actor’s goals and perspective
consistently engaged the TPJ (Overwalle, 2009). Along with this imaging evidence, a lesion study
provides a stronger causal link between TPJ activity and perspective-taking. Three patients with damage
to the left TPJ did no better than chance on a task that required the understanding of another person’s
false belief; importantly, they performed similarly to controls on tasks matched for cognitive demand that
did not involve false beliefs (Samson, Apperly, Chiavarino, & Humphreys, 2004). This finding implies
that the left TPJ is necessary for generating representations of someone else’s belief, and that this effect is
not attributable to the cognitive load such a task requires. These results and others provide evidence that
the TPJ is essential for comprehending the mental states of others.

While it has been established that the TPJ is an important component of the mentalizing system,
there remain differences in the interpretation of its specific function. Broadly, cognitive empathy is
understood to entail the act of grasping the mindset of another while maintaining the distinction between
oneself and that other (Overwalle & Baetens, 2009). One view emphasizes the TPJ’s role in preserving an
individual’s agency as a perceiver (Decety & Jackson, 2006). Empathy would be disorienting rather than
valuable if the boundary between these two perspectives was eroded. As this could potentially lead to
emotional contagion and maladaptive behavior, it is likely that during the course of evolution, processes
that kept a strong partition between self and other would be selected for. Another interpretation focuses
on the temporal domain of empathic thinking. Van Overwalle proposes that the TPJ is responsible for
“transient mental inferences” about the other person (i.e., their present thoughts and goals), as opposed to
more permanent traits and characteristics (Overwalle, 2009). It seems that the TPJ is a site of empathetic
processing in real time, and as such, has the capacity to dynamically react to changing social input.

Although the TPJ is clearly implicated in perceiving the mental states of others, neglecting its
involvement in other realms of cognition would be detrimentally selective. In the domain of attention, the

Soderberg 2

TPJ has been classified as an element of the ventral frontoparietal network, which is engaged during the
reorienting of attention to a salient stimulus (Corbetta, Patel, & Shulman, 2008). This ventral network acts
as a “circuit breaker” for the top-down dorsal attentional system, signaling the presence of behaviorally
relevant stimuli and directing attention towards them. It is consistently recruited in “oddball” tasks, in
which subjects are instructed to respond to abnormal stimuli in a stream of repetitive inputs (Marois,
Leung, & Gore, 2000). Although much of the work on attentional control has been conducted in the
visual domain, there is evidence to suggest that the TPJ functions in its reorienting capacity independent
of sensory modality (Corbetta & Shulman, 2002). In addition to its role in the ventral frontoparietal
attention network, the TPJ has also been implicated in the perception of one’s own body relative to the
external environment. Electrically stimulating the TPJ can lead to the feeling of being extracorporeal,
known as an out-of-body experience (Blanke & Arzy, 2005). An experiment analyzing evoked potentials
from electroencephalography readings mapped potentials to the TPJ when subjects were instructed to
imagine themselves in the position of a human figure presented on a screen (Arzy, Thut, Mohr, Michel, &
Blanke, 2006). Taken together, these findings suggest that the TPJ is responsible for maintaining the
distinction between a corporeal self and the external world.

Although these low-level cognitive processes seem far from the complicated cognition of
empathy, their computational mechanisms may be crucial for the complexity of social cognition.
Throughout the course of evolution, as understanding the intentions and emotions of conspecifics became
increasingly adaptive, it is possible that the TPJ’s reorienting and body-perception functions were co-
opted to serve social cognition. Its capacity to draw attention to behaviorally relevant information seems
particularly suited to empathy, as recognizing and interpreting the perspectives of others in social
interactions is extremely pertinent to one’s behavior in those interactions. Its role in separating the self
from the environment could be extended to maintain an individual’s sense of self while they are
empathizing with another person. Several theorists have taken this perspective, positing that attention
signals in the TPJ are central to the act of switching between an internal perspective and another’s
viewpoint (Corbetta et al., 2008). In this view, the TPJ is defined as carrying out a “domain-general

Soderberg 3

computational mechanism:” one that is essential for the higher-order social process of mentalizing
(Decety & Lamm, 2007).

This compelling interpretation provokes the question of whether the temporoparietal junction is
especially attuned to social perspective taking, or if it is simply performing its low-level cognitive
function in the social domain. This dichotomy can be difficult to tease apart, as findings that might seem
to prove the uniquely social character of the TPJ can be interpreted to support the opposite theory. For
example, an fMRI experiment found a greater BOLD signal when subjects read stories about characters’
false beliefs than when they read about false physical representations, while this difference was not seen
for expected versus unexpected stories (Young, Dodell-Feder, & Saxe, 2010). The authors conclude that
TPJ response is selective for content about mental state, over and above its function for reorienting
towards salient stimuli. However, if the TPJ’s functional domain is seen as responding to a continuum of
behavioral relevance, and perspective-taking is recognized as requiring a greater cognitive load than
object perception, the finding that it is more activated during the mental state condition could be
explained by the behavioral relevance and complexity of understanding that mental state. In addition, this
study did not examine TPJ activation during a low-level attention paradigm, so it cannot rule out the
possibility that the TPJ’s functional domain encompasses both attention shifting and perspective-taking.
Recognizing that omissions of this kind were an issue in the literature, another experiment compared a
similar belief task with an attention reorienting task within one group of participants, and found that the
same region was modulated by both tasks (Mitchell, 2008). This result indicates the need for a conception
of TPJ function that integrates across attentional, social, and agential domains, rather than claiming
selectivity for any one task.

Although not directly concerned with taking the perspective of another person, a study was
conducted that proposed a uniquely social function for the TPJ by examining social versus non-social
decision-making in a simplified poker game. Participants received a card, saw an image of their opponent
(either a human face or a computer), and decided whether to bet or fold. When comparing the BOLD
signal using a calculation of unique combinatorial performance (UCP), which measures the degree of a

Soderberg 4

brain region’s predictive performance, the authors found a strong independent contribution of the TPJ for
human opponent versus computer opponent trials. When a “social bias” metric was computed using UCP
for regions across the entire brain, the TPJ emerged above all other regions in preferential processing of
information concerning a social agent. On the basis of this strikingly different social bias, the authors
conclude that the TPJ has a unique sensitivity to the “perceived behavioral relevance of other agents”
(Carter, Bowling, Reeck, & Huettel, 2012). Applying the continuum model discussed above, one could
similarly attribute this increased social responding solely to the behavioral relevance of the social
stimulus, thus maintaining that the TPJ’s processing of social information is quantitatively, but not
qualitatively, different from its attentional function. However, the fact that out of the entire brain, the
temporoparietal region alone stood out on the social bias metric is hard to ignore. Presumably, the low-
level computational mechanisms of some other brain regions are at play in social decisions, so why do
they not exhibit a significant level of social bias? Perhaps there is something about the particular
integrative power of the TPJ that makes it distinctly suited to conceptualizing another person’s point of

This dilemma exemplifies a larger controversy in the field: whether any region (or network of
regions) in the brain is inherently “social.” Although there is an intuitive tendency to search for specific
neural locations for our psychological concepts, it seems increasingly likely that there is no one-to-one
match of function to region. It is probable that the quest for regions that are completely selective for
social input will turn up no results. So while it might be tempting to look for the “seat of empathy,” it is
likely that empathetic processing is accomplished through the combined efforts of many different regions
(including the TPJ), each performing more basic functions which are then integrated to generate a unified
percept. This aligns with the growing emphasis in the field on prioritizing network interactions over
individual structures (Kennedy & Adolphs, 2012). Although it is undoubtedly useful to elucidate the
specific contributions of each brain region, attention should be paid to how they operate in functional