|
The use of functional neuroimaging techniques, such as positron
emission tomography (PET), to study neuropsychiatric diseases has
a long history. These techniques were applied to diseases such as
schizophrenia and depression in the early days of their development.
More recently, with the advent of functional magnetic resonance
imaging (fMRI), the use of imaging to study psychiatric diseases,
as well as normal physiology, has increased dramatically, due in
part to the widespread availability of MRI to the neuroscience and
medical communities. The purpose of this review is to summarize
recent experiments using either PET or fMRI to explore cognitive
changes that are associated with psychiatric diseases. In the past
few years, we have paid considerable attention to the study of emotion
and social cognition, which closely relate and most likely function
in a highly integrated fashion. Social cognition has been defined
as the ability to interpret and predict others’ behaviour in
terms of their beliefs and intentions and to interact in complex
social environments and relationships (1). Evidence is accumulating
that emotion and social cognition depend on some of the same brain
regions, thus having the ability to influence each other in many
ways. Of particular interest is that several psychiatric disorders
are accompanied by deficits in these functions, including autism,
schizophrenia, depression, and posttraumatic stress disorder (PTSD).
In light of the recent interest in these disorders and the emotional
and social impairments in each, this review focuses on the use of
imaging to examine how emotion and social cognition, and the brain
regions mediating these functions, are altered in these 4 disorders.
The first section introduces the brain regions that are involved
in emotion and social cognition and that comprise a broadly defined
social cognition network. Each of these regions plays a role in
a relevant cognitive process, and we propose that together they
form the network that allows us to use these processes, such as
emotional processing and working memory, to function appropriately
in social situations, and to maintain interpersonal relationships.
In subsequent sections, we discuss each of the 4 disorders and the
task paradigms used to study them. Finally, we conclude with a discussion
of the gaps that remain in our understanding and some of the problems
that are encountered in doing imaging research on patient populations.
The Emotional and Social
Brain Network
In our view, higher brain function is not the result of activity
in any single brain region, but rather, it is the outcome of integrated
activity in groups or networks of brain regions that function interdependently
(2). Social cognition, using a broad definition of the term, involves
regions that mediate face perception, emotional processing (including
both perception of emotional information in the environment and
regulation of mood); theory of mind (TOM) (understanding others
beliefs and motives); self reference; and working memory. Together,
the functioning of these regions would support the complex behaviours
necessary for social interactions. There are 10 regions associated
with these various functions that are critical for social cognition
(Figure
1).
|
|
Given the focus of this review, the functional neuroimaging evidence
for the involvement of these areas in the relevant cognitive processes
is emphasized, but note that a large amount of information from
human lesion and animal work is also available on the role of these
regions in cognition and on their anatomical interconnections.
Several brain regions involved in perceiving faces are included
in the network; it seems clear that multiple aspects of face perception
are important for understanding the intentions of conspecifics.
Activity in the fusiform gyrus (Region 1, Figure 1) is found consistently
in face discrimination or face identification tasks (3), and this
area responds to faces, relative to other body parts and other objects,
with a fairly high degree of specificity (4). Regions in the superior
temporal sulcus (STS) (Region 7), are sensitive to gaze direction
of another person’s eyes (5,6) and to motion of other body
parts such as the hands (7). The amygdala (Region 3), more than
any other region, is considered critical for emotional processing
(8). This area is consistently active when subjects view faces depicting
emotional expressions, particularly the left amygdala (9,10). Its
role in emotion extends beyond faces, however, and encompasses modulation
of memory (11,12) and fear conditioning (13,14).
The remaining regions in the proposed network are all in frontal
portions of the brain, either in the prefrontal cortex (PFC) or
the anterior cingulate. There are 3 regions of the anterior cingulate
that relate to functions with potential importance for social cognition.
The first of these is the subgenual cingulate (Region 5), which
is critical for autonomic responses (15,16) and involved in reward
mechanisms (17). This region, along with a more rostral but still
ventral area, is often referred to as the emotional sector of the
anterior cingulate (18). The rostral cingulate (Region 6) is active
during tasks requiring processing of emotional stimuli (19, 20).
The dorsal portion of the anterior cingulate, the so-called “cognitive
cingulate” (Region 8), is involved in error monitoring and
selecting among competing responses (18,21,22). Thus, the anterior
cingulate gyrus appears to play multiple roles in social cognition.
Finally, there are 3 prefrontal regions that we have included in
the social network. Orbitofrontal cortex (Region 2) is important
for decision making in the context of emotional situations (23–25),
and lesions in this region also result in TOM deficits (26). Ventrolateral
prefrontal cortex (VLPFC) (Region 4) mediates responding-to-reward
contingencies, along with the subgenual cingulate, and alters its
activity as reward parameters change (17, 24). A region of dorsomedial
prefrontal cortex (Region 10) is active when subjects engage in
tasks that emphasize self-reference (27) or an internal vs an external
focus (28). A similar region of cortex is activated during TOM tasks
(29–31). The remaining frontal region is the dorsolateral prefrontal
cortex (DLPFC) (Region 9), which is involved in executive functions
and working memory (32,33–35). At first, this region may seem
less related to social cognition than others, but it has been shown
that TOM development depends on working memory development in children
(36).
|
