The Cognitive Neuroscience of Social Behaviour

Alexander Easton, Nathan Emery

• Publisher: Psychology Press
• Year: 2004
• ISBN:
  9781135424848
• Edition: 1
• Topic: Psicología
• Subtopic: Historia y teoría en psicología

The potential for cognitive neuroscience to shed light on social behaviour is increasingly being acknowledged and is set to become an important new approach in the field of psychology. Standing at the vanguard of this development, The Cognitive Neuroscience of Social Behaviour provides a state-of-the-art contribution to a subject still in its infancy. Divided into three parts, the book presents an overview of research into neural substrates of social interactions, the cognitive neuroscience of social cognition and human disorders of social behaviour and cognition.

1 Introduction: What is social cognitive neuroscience (SCN)?

Nathan J. Emery and Alexander Easton

Understanding the neural mechanisms of cognitive processes, such as thought, perception, and language, or cognitive neuroscience, has been a rapidly progressing discipline for the last two decades. This expansion has been driven primarily by significant advances in the development of technology to observe the activity of the living human brain in action. Revolutionary techniques such as positron emission tomography (PET), event-related potentials (ERP), and functional magnetic resonance imaging (fMRI) have been at the heart of this revolution, and we now know much about how the human brain processes sensory information, plans and controls movement, perceives and produces speech, and experiences emotions.

Although we know infinitely more about the neural basis of human psychology than we did 20 years ago, the problems facing us in the future are immense. We highlight two main problems. First, the questions themselves are huge. For example, visual processing is a relatively minor component in the overall story of how the brain functions to give rise to complex human cognitive processes, but, in itself, vision is a horrendously complicated process involving over 50 cortical areas, and billions of neurons. Therefore, major advances must be made before we can even contemplate solving relatively “simple” processes, such as perception. Second, not all aspects of traditional psychology can be easily combined with the tools of cognitive neuroscience. This second problem applies especially to social cognitive neuroscience (SCN).

SCN is the study of the neural mechanisms of social cognition and social interactions in humans and animals, particularly nonhuman primates. It is also concerned with deficits of sociocognitive processes in humans, particularly those which have a dedicated neural basis, such as autism, schizophrenia, sociopathy, and depression. This branch of cognitive neuroscience is directed towards understanding complex aspects of social behaviour, such as mentalizing (understanding another’s mental states), empathy, attractiveness, self-awareness, moral reasoning, intentionality, and imitation. As such, it is slightly different from social neuroscience, or the study of the neurobiology of social behaviour from a comparative perspective. This branch of neuroscience is concerned with the neurobiology of motivational systems, such as aggression, sexual and parental behaviour, and play. These behaviours appear to be controlled by the interaction of neural and endocrine systems, particularly the amygdala, hypothalamus, brainstem, and basal ganglia, and are largely devoid of cognitive processing. These behaviours are displayed by virtually all vertebrates, from amphibians and reptiles to birds and mammals. In contrast, the topics covered by SCN are restricted to higher-order cognitive processes, which are 1. mainly controlled by association cortical areas, such as the prefrontal cortex, 2. not under the influence of hormones, 3. party to disruption by psychopathological disorders or discrete brain lesions, and 4. found almost predominantly (although not exclusively) in human and nonhuman primates.

Traditional social psychology is inferential or relies on pseudo-natural experimental situations, neither of which lend themselves to the tools of cognitive neuroscience, which requires tightly controlled variables. There are also other problems in applying the techniques of cognitive neuroscience to social psychology. For example, functional imaging studies (which are a common tool of cognitive neuroscientists) are difficult to apply in situations that require interaction between subjects. Such studies are very much in their infancy. Similarly, animal studies which may shed light on the neural basis of social behaviour can be problematic because human social behaviour is extremely complex, very much more so than in any of the animals that might be used in such experiments. Indeed, those animals that most closely resemble humans, the great apes, cannot be used for invasive research. Therefore, SCN has been slow to get started (see below). However, in the last few years, there has been a sudden and dramatic increase in the number of studies in the field, and technological problems are finally not only being addressed, but are also being overcome.

In this introductory chapter, we will attempt to introduce the field of SCN, first from a historical perspective, focusing particularly on the recent development of the field and its specialization within cognitive neuroscience. Then we discuss what we predict might be the major research themes of the future, and finally we address Adolphs’ (2003) 10 questions for SCN, putting it into context.

A brief history of SCN

The field of SCN has had a long gestation with two parents—cognitive neuroscience and social psychology. Attempts to understand the neural basis of cognition have a history longer than modern cognitive neuroscience, with their foundation in studies of brain-lesioned patients, such as soldiers sustaining gunshot wounds to the head during battle. However, the development of cognitive neuroscience as an experimental science was dependent on the invention of a new methodology—functional neuroimaging. Developed in the 1980s, these techniques, particularly PET and the subtraction method, provided researchers with the capability of localizing function to specific brain areas. These early studies were crude, largely replicating simple cognitive experiments in the scanner; however, the technology has become more complex, using sophisticated experimental designs to remove noise from the system. The formation of cognitive neuroscience was therefore driven primarily by improvements in technology rather than producing an entirely new theoretical approach.

Within cognitive neuroscience, studies in nonhuman primates were more sophisticated and elegant than those performed in humans. Of particular relevance to SCN was the finding, in the early 1980s, that neurons in the anterior temporal cortex of rhesus monkeys are selectively responsive to biologically important stimuli, such as faces (Bruce, Desimone, & Gross, 1981; Perrett, Rolls, & Caan, 1982). Later studies revealed that different neurons in this region are responsive to facial expressions, facial identity, gaze direction, facial movements, walking, and even intentional actions (for reviews, see Emery & Perrett, 2000; Jellema & Perrett, this volume).

Perhaps the most significant paper in the establishment of SCN as a tangible field of study was “The Social Brain: A Project Integrating Primate Behaviour and Neurophysiology in a New Domain” by Leslie Brothers (1990). This paper introduced the idea that it was possible to investigate the neurobiology of social interaction by integrating studies of neurophysiology, behaviour, and psychopathological disorders which specifically affect social behaviour, such as autism. What was particularly significant about this paper was the focus on a specific neural circuit essential to processing social information, which Brothers called the “social brain”. This circuit included the anterior temporal cortex and temporal pole, the nuclei of the amygdala and the orbitofrontal cortex. As we will discover throughout this book, these areas and others connected to them play a central role in social behaviour.

A fundamental process which functions during social interaction is the ability to read another individual’s mental states (“theory of mind” [ToM]). This “mind-reading” ability develops in human infants around 3–4 years of age, and is affected strikingly in various psychopathological disorders, such as autism, Asperger’s syndrome, and schizophrenia. The first functional neuroimaging studies to address this issue of social problem solving were focused specifically on the neural basis of mental attribution (Baron-Cohen, Ring, Moriarty, Schmitz, Costa, & Ell, 1994; Fletcher, et al., 1995). These two studies found that different parts of the prefrontal cortex were activated during a story-comprehension ToM task. We now know that ToM is a suite of abilities (Emery, this volume) that are likely to be based on the integration of different neural systems (Heberlein & Adolphs, this volume). From the seeds sown by these two neuroimaging studies, similarly complex aspects of social interaction are now being investigated by this technique, such as selfknowledge, moral behaviour, empathy, stereotyping, gender or racial biases, attractiveness, and humour appreciation.

Earlier in this section, we made the claim that perhaps the most significant paper in the history of SCN was that of Brothers (1990). To investigate the claim that SCN has developed primarily in the years since this paper, we searched the PubMed on-line citation reference database (MEDLINE; www.ncbi.nlm.nih.gov/PubMed/) from 1990 to 2003, using the keywords “social” AND “brain”. This produced a total of 5645 documents. As many of these papers were either not related to SCN or reported studies in social neurobiology, a more refined search using the keywords “social” AND “cognitive” AND “brain” was performed, producing 1013 documents. The titles and abstracts of the papers in the second search were examined for the following criteria. The papers were either theoretical, review, or experimental. The experimental papers used traditional methods of cognitive neuroscience (neuroimaging, lesion analysis, transcranial magnetic stimulation, and electrophysiology) and reported research performed on humans and nonhuman primates. The theoretical focus of the papers included the neurobiology of social interaction, ToM, imitation, empathy, facial attractiveness, selfawareness, moral cognition, cooperation, deception, humour, and gaze processing. Papers on the processing of emotion, including facial expressions, were not included if they did not focus on the role of social communication. Papers on language and face perception (with respect to visual processing) were not included. Finally, papers on disorders of social cognition were included only if they addressed both the social aspects of the disorder and the neural basis. Additional searches using other related criteria (“social” AND “amygdala”; “social” AND “prefrontal cortex”; “empathy” AND “brain”; “imitation” AND “brain”, etc.) were also performed to determine whether any papers had failed to be selected with the search terms of the earlier searches.

We found a total of 312 documents that fulfilled these criteria. These documents were sorted by date, and the totals for each year from 1990 to 2003 are represented in Figure 1.1. As can be seen in Figure 1.1, throughout the 1990s, research in SCN was slowly increasing; however, there was an explosion of research at the beginning of the twenty-first century.

There are many possible explanations for this explosion of interest in the neural basis of social interaction. We make three suggestions. First, mental health has become a research priority for funding councils and government agencies. This increase in funding has grown steadily across the 1990s, the socalled Decade of the Brain. Second, cognitive neuroscience has become mainstream, using sophisticated methods for observing the brain in action, and techniques for extracting the huge amounts of information produced. Third, cognitive neuroscientists and social psychologists have started talking to one another, and they are now collaborating. This final point is crucial. For SCN to work, it is vital that there is collaboration between cognitive scientists (such as linguists, computer scientists, philosophers, and cognitive psychologists), neuroscientists, ethologists, and social psychologists. The requisite skills for the successful completion of projects in this area can be achieved only by a strong interdisciplinary approach.

Figure 1.1 Number of journal articles directly related to social cognitive neuroscience (SCN) published every year between 1990 and 2003. Results were produced by an on-line PubMed (MEDLINE) and PsychInfo search, initially using the search criteria “social” AND “brain”, and then “social” AND “cognitive” AND “brain”. The titles and abstracts were studied to determine whether the papers were relevant to SCN. Other more refined searches were then made, using more specific search criteria, such as “empathy”, “theory of mind”, “self-awareness”, “imitation”, etc. There has been a steady increase in the number of papers published, with a dramatic increase after 2000.

What are the methods of SCN?

Although SCN was founded on work performed in animals, particularly nonhuman primates, is there still a place for this type of research? As will be seen in Chapters 2–5, animal studies remain vitally important to the neurobiology of social interaction. We see three main reasons why animal research is still essential. First, functional neuroimaging technology, although greatly advanced in the last 10 years, is still limited in how it can be applied to situations in which two or more protagonists are physically interacting. At present, our investigations are based on how individuals (constrained within a scanner or attached to electrodes) respond to static or moving video images of faces, hands, and bodies, and on questionnaires and self-report. Neuropsychological studies in human patients are equally constrained by the presentation of “simple” social stimuli, rather than involvement in real-life social scenarios. There are fewer methodological constraints with nonhuman animals. Animals can be observed during social interactions, or socially relevant cognitive tests can be presented after or during invasive manipulation. Techniques such as selective neurotoxic and reversible lesions, multiunit electrophysiology, microdialysis, psychopharmacology, and genetic knockouts have been used successfully for a number of years, providing a level of analysis currently impossible in studies of human beings. Some of these techniques, such as experimental lesions and genetic knockouts, can be used only in animals.

Second, a comparative approach to SCN may inform us about the evolutionary history of social interaction, and how certain neurocognitive processes may function similarly to or differently from those in humans. For example, there is increasing convergent evidence that western scrub-jays may possess an avian analogue to human ToM (Emery & Clayton, 2001). This is particularly striking because the common ancestor of scrub-jays and humans lived over 300 million years ago, and the brains of birds and mammals are completely different in structure. The medial prefrontal cortex appears to be the primary structure involved in the attribution of mental states in humans (Heberlein & Adolphs, this volume); however, birds do not have a prefrontal cortex (Emery & Clayton, 2003). Understanding the neural basis of “ToM” in corvids by using neurophysiology, lesions, and network models may inform us about whether understanding another’s mental states is dependent on similar neural circuitry in terms of efficiency of coding, or whether completely different mechanisms are employed in distantly related species.

Third, the development of behavioural tests in nonhuman animals may be used in human populations which are impaired in their use of language, such as preverbal infants, patients with aphasia, the mentally handicapped, or autistics.

The physical and analytical techniques currently employed in human SCN studies are increasing in complexity. fMRI, in particular, may be the most revolutionary method available to help us understand the human brain. However, with respect to SCN, there are still many problems and limitations in applying these techniques to understand social behaviour. First, social behaviour involves the interaction of two individuals. Although presentation of images on a monitor provides us with information about how we perceive and categorize social stimuli, into different facial expressions, races, or genders, the interactive component essential to social cognition is missing. Currently, a number of laboratories have developed methods by which an individual interacts with either another person or a computer in a cooperative game (McCabe, Houser, Ryan, Smith, & Trouard, 2001; Rilling, Gutman, Zeh, Pagnoni, Berns, & Kilts, 2002), or two individuals in two separate MRI scanners interact in various deceptive games while their brains are scanned simultaneously (Montague et al., 2002). In both cases, the participants cannot see one another; however, with the implementation of web cams and the Internet, this does not seem too difficult to implement. Outside the realms of science fiction and portable MRI scanners, this new technology appears to have incredible potential, if the equipment is made available through collaborative effort.

These newly developed technologies should bring about unique opportunities to investigate previously difficult or impossible problems. As we described earlier, it has been impossible to investigate the neurobiology of human social interaction with the technology currently available, so that we rely instead on studies of social perception, particularly responses to facial stimuli. In the future, we see extensive use of video images (live and prerecorded) to present social stimuli, possibly with the use of “actors”. This advance would directly utilize the methods of social psychological research. In this instance, images with sound could be presented as scenarios in the “real world” (for example, from popular soap operas), or live images of an actor in an adjacent room, or facial images of another individual also in a scanner by hyperscanning (Montague et al., 2002).

The future of SCN will be driven by advances not only in technology but also in sociocognitive theory. Most studies in SCN, to date, have focused on: 1. social perception, resulting from electrophysiological studies of faceresponsive neurons in rhesus monkeys; 2. ToM, resulting from studies on autism; 3. social emotions. What is starting to occur, as will be highlighted in Chapters 6 and 7, is explicit collaboration between cognitive neuroscientists and social psychologists to investigate particular problems in social cognition, such as social exclusion (Eisenberger, Lieberman, & Williams, 2003), selfconscious emotions (Beer, Heerey, Keltner, Scabini, & Knight, 2003), social evaluation (Cunningham, Johnson, Gatenby, Gore, & Banaji, 2003), selfknowledge (Kelley, Macrae, Wyland, Caglar, Inati, & ...

Table of contents

• Cover Page
• Title Page
• Copyright Page
• Acknowledgements
• Contributors
• 1 Introduction: What is social cognitive neuroscience (SCN)?
• Part I Neural substrates of social interactions
• 2 The neurobiology of social-emotional cognition in nonhuman primates
• 3 Behavioural flexibility, social learning, and the frontal cortex
• 4 Neural basis for the perception of goal-directed actions
• Part II Cognitive neuroscience of social cognition
• 5 The evolution of social cognition
• 6 Functional anatomy of human social cognition
• 7 The self and social perception: Three kinds of questions in social cognitive neuroscience
• Part III Human disorders of social behaviour and cognition
• 8 Autism and the origins of social neuroscience
• 9 The neurobiology of social cognition and its relationship to unipolar depression
• 10 The neurobiology of antisocial behaviour and psychopathy