1. Introduction to social neuroscience
Hyperscanning: the simultaneous recording from two or more different brains (e.g. using fMRI or EEG)
Activity in regions of one person’s brain can reliably elicit activity in other regions of another person’s brain during
social interaction. For instance, in a trusting relationship, when one person makes a decision the other person’s brain
‘lights up’ their reward pathways, even before any reward is actually obtained.
Cognition in an individual brain is characterized by a network of flowing signals between different regions of the
brain. However, social interaction between different individuals can be characterized by the same principle: a kind of
‘mega-brain’ in which different regions in different brains can have mutual influence over each other. This is not
caused by a physical flow of activity between brains (as happens between different regions in the same brain) but by
our ability to perceive, interpret, and act on the social behaviour of others.
1.1 The emergence of social neuroscience
Social psychology: an attempt to understand and explain how the thoughts, feelings, and behaviours of individual are
influenced by the actual, imagined, or implied presence of others. (an attempt to understand and explain, using
neural mechanisms, how the thoughts, feelings, and behaviours of individuals are influenced by the actual, imagined,
or implied presence of others.
Cognitive psychology: the study of mental processes such as thinking, perceiving, speaking, acting, and planning. It
aims to decompose complex social behaviours into simpler mechanisms that are amenable for exploration using
neuroscientific methodologies.
Social neuroscience links cognitive and social psychology, as well as linking the ‘mind’ (psychology) with the brain
(biology).
Ecological validity: an approach or measure that is meaningful outside of the laboratory context
1.2 The social brain
To which extent can the social brain be considered distinct from all the other functions that the brain carries out? So
is the social brain special in any way?
One possibility is that there are particular neural substrates in the brain that are involved in social cognition but not
in other types of cognitive processing. This relates to the notions of modularity and domain specificity.
Modularity: the notion that certain cognitive processes (or regions of the brain) are restricted in the types of
information they process and the type of processing carried out.
Domain specificity: the idea that a cognitive process (or brain area) is specialized for only one particular kind of
information.
A module is the term given to a computational routine that responds to particular inputs and performs a particular
computation on them, that is, a routine that is highly specialized in terms of what is does to what. Modules are
domain specific, they process only one kind of input. Like the module responding to faces but not to bodies. In this
modular view, the social brain is special by virtue of brain mechanisms that are specifically dedicated to social
processes. Moreover, it is claimed that these mechanisms evolved to tackle specific challenges within the social
environment. Some critics see this as phrenology, but the modern approaches addressing the issue of domain-
specific social processes are subjected to experimental rigor, which didn’t happen back in the 19 th century with the
ideas of the brain localization of function.
The alternative approach is that the social brain is not specialized uniquely for social behaviour but is also involved in
non-social aspects of cognition (reasoning, visual perception). The evolution of general neural and cognitive
mechanisms that increase intellect, such as having bigger brains, may make us socially smarter too. The reverse may
also be true, namely that the evolutionary need to be socially smarter leads to general cognitive advances in other
domains. Under these accounts social cognition and non-social cognition evolved hand-in-hand (albeit with one
factor driving the other) but, crucially, they did not necessarily lead to highly specialized routes in the brain for
dealing with social problems.
There are other positions that lie in-between these two extremes.
There are certain brain regions which are activated by a wide range of social phenomena. Instead of arguing for a
narrowly defined module in this regions, it is suggested that social psychology is a natural kinds that distinguishes
itself from other aspects of cognition because it relates to concepts that are less stable and less definite than those
involved in perception and action. The social brain is special because of the nature of the information that is
processed rather than because it is social.
Another possibility is that not particular regions of the social brain are special but there are particular kinds of neural
mechanisms especially suited to social processes. The key insight, with regard to social neuroscience, is that there
may be a simple mechanism (implemented at the level of single neurons) that enables a correspondence between
self and other. Mirror neurons have been implicated in imitation, empathy, and ‘mind reading’. Although they were
,originally discovered for actions, it is possible that mirroring is a general property of many neurons and they may not
be tightly localized to one region.
Tree broad ways in which the social brain may be implemented:
1) A simple domain-specific view consisting of brain regions that are specialized for processing particular kinds
of social information (person perception) and non-social information (cognitive control).
2) Postulates networks of regions in which each region in the network has a high degree of specialization
(specific to social information).
3) Neither brain regions nor individual brain networks are functionally specialized or segregated into social and
non-social functions.
There is a variety of views concerning the broad nature of the neural mechanisms that support human social
behaviour. At one end, there is the view that there are highly specialized neural mechanisms. These may be very
limited in the type of information the process (faces, beliefs). At the other hand, there is the view that the
mechanisms that support social behaviour are used for many other functions (possibly including non-social
cognition). Whereas the highly specialized viewpoint tends to have been linked to the idea of a small number of
contributing brain regions (localizability), it is not incompatible with the idea of brain networks.
1.3 Is neuroscience an appropriate level of explanation for studying social behaviour
The most general criticism is that the brain is not the most appropriate level of explanation for understanding social
processes. Surely social processes need to be studies and understood at the social level – that is, the level of
interactions between people, groups of people, and societies.
Reductionism: one type of explanation will become replaced with another, more basic, type of explanation over time.
In a reductionist framework the language of social psychology will be replaced by the concepts of neuroscience.
However, most researchers in social neuroscience are attempting to create bridges between different levels of
explanation rather than replace one kind of explanation with another.
Another common way in which neuroscience data are used to bridge levels of explanation is the reverse inference
approach.
Reverse inference: an attempt to infer the nature of cognitive processes from neuroscience (notably neuroimaging)
data.
The reliability of this reverse inference depends on what is known about the functions of given regions. If these
regions turned out to have very different functions then the inference would be flawed. Also the function of regions
is not resolutely fixed but depends on the context in which they are employed. Another more general methodological
point is the importance of not being over-reliant on neuroimaging data, but to look at other sources of evidence such
as TMS in which behaviour itself is normally measured (and hence does not suffer from the problems of reverse
inference in the same way).
There is one scenario in which brain-based data could have no significant impact on our understanding of social
processes, and that is the blank slate scenario.
Blank slate – the idea that the brain learns environmental contingencies without imposing any biases, constraints, or
pre-existing knowledge of that learning.
According to the blank slate, the brain is not completely redundant (it still implements social behaviour) but the
nature of social interactions themselves is entirely attributable to culture, society, and the environment. The
structure of our social environment is created entirely within the environment itself, reflecting arbitrary but
perpetuated historical precedents. Thus, culture, society, and the nature of social interactions invent and shape
themselves. A more realistic scenario is that the brain, and its underlying processes, creates constraints on social
processes. Social processes are all in the brain, but some of them are created by environmental constraints and
historical accidents (and learned by the brain) whereas others may be caused by the inherent organization, biases,
and limitations of the brain itself.
1.3.1 Aggression as an example of interacting levels of explanation
While income inequality is itself cultural, and not biological, the fact that aggression is linked to resource control and
perceived injustice is likely to be independent of culture. Cultural differences may acts as an ‘accelerator’ or ‘brake’
on biological tendencies.
1.3.2 A biological basis for culture
What causes culture?
A set of mechanisms that enables people to transfer skills, beliefs, and knowledge from each other and retain
these as a relatively stable pattern across individuals. (Cognitive mechanistic explanation)
, Neural mechanisms that respond to the repeated patterns of behaviour in others, whom we affiliate
positively with, and increase the likelihood that our own neural mechanisms will generate those behaviours.
(Neuroscientific explanation)
It would be an entirely circular argument to say that culture creates itself. Culture cannot create itself in the absence
of appropriate biological entities. As to the question of what creates variability in culture, the answers could be quite
different. It may reflect the different environments that people live in and arbitrary historical precedents. However,
the number of cultural variants may not be limitless. There could be some cultural forms that will never be created,
or, if they are, will rapidly die out because they are too difficult to require – that is, biology may go as far as to specify
which cultural variants are likely, possible, or virtually impossible. An impossible culture could be a system of slavery
associated with high levels of empathy and human cognitions towards the slave group. The impossibility is created by
the nature of brain-based mechanisms, even though it manifests itself in terms of the nature of social processes.
1.3.3 Gene-culture co-evolution
Cultural neuroscience: an interdisciplinary field bridging cultural psychology, neurosciences, and neurogenetics.
Cultural neuroscience explains how neurobiological processes give rise to cultural values, practices, and beliefs as
well as how culture shapes neurobiological processes. It explicitly assumes that not only will cultural differences
influence the brain but also that the brain will impact on culture itself.
Gene-culture co-evolution: culture can influence gene frequencies in a population, and genes have an impact on
cultural evolution via psychological predispositions.
Certain genotypes may predispose people to create particular features in their environment (thus influencing cultural
selection) and, at the same time, aspects of a given culture may tend to favour individuals of a given genotype (thus
influencing genetic selection). The outcome of this iterative process is that there is a good fit between a particular
genotype and a particular cultural practice.
2. The methods of social neuroscience
Social neuroscience hasn’t yet developed a distinct methodology of its own, so its methods are borrowed from other
disciplines.
Temporal resolution: the accuracy with which one can measure when an event is occurring. (Not with brain damage,
because it is permanent and thus has no temporal resolution).
Spatial resolution: the accuracy with which one can measure where an event is occurring.
Invasiveness: whether or not the equipment is located internally or externally.
2.1 Measuring behaviour and cognition: psychological methods
Almost all experiments in social neuroscience measure behaviour in some way, given that it’s social behaviour that
they are trying to explain. In functional imaging experiments, the participant is given a set of instructions on how to
respond even if the main dependent measure is brain activity rather than behaviour per se. in social neuroscience, it
is also common to correlate neuropsychological responses when performing a task with individual differences on a
psychological measures such as empathy or personality.
Three different ways of measuring behaviour and cognition:
Performance-based measures – where the dependent measures are typically response times or error rates
Observation-based measures – where the dependent measure is often a frequency count of how often
something occurs
Firs-person-based measures- where the dependent measure may be scores on a questionnaire
2.1.1 Performance-based measures: response times and accuracy rates
Mental chronometry: the study of the time-course of information processing in the human nervous system
The basic idea is that changes in the nature or efficiency of information processing will manifest themselves in the
time it takes to complete a task. Aside from response times, the other main performance measure is accuracy. This
can be measured in terms of error rates, percentage correct, or percentile performance in which individual scores are
recalculated relative to the population mean. Accuracy is crucially related to whether certain knowledge is
present/absent rather than to processing efficiency (which is more related to response time). However, accuracy and
efficiency are related in certain circumstances. For example, if people are forced to respond faster they will tend to
be less accurate speed-accuracy trade
Performance measures:
+ They reflect actual behaviour; they are simple to analyse and interpret
, - They are hard to link directly to neural substrates (unless combined with other measures). There is not always a
clear relationship between laboratory tasks and real-world behaviour.
2.1.2 Observational measures
Observational measures tend to code ‘what’ is being done or ‘how often’ something is done through one person
observing the behaviour of others. There are certain situations in which observational measures are used in place of
the more common performance-based measures:
- Infancy research, infants cannot be trained or instructed to perform a task.
- Non-human species, for the same reasons as stated above. Although training is possible, there is still a need
to know how (untrained) animals behave in the wild.
- When the experimenter does not want the participant to know the true nature of a task, e.g. imitation.
Two specific observational methods in the infant literature are preferential looking and habituation.
Preferential looking: in infant research, a number of stimuli (normally two) are presented and the amount of time
that the infant spends looking at each of them is scored).
A deviation from change implies that the infant is able to discriminate between the stimuli and has a preference.
Habituation: in infant research, the same stimulus (or same kind of stimulus) is presented repeatedly and the infant’s
attention towards the stimulus (measured in terms of looking time) diminishes.
The critical phase of a habituation experiment occurs when a new stimulus is presented. If the infant’s attention
increases, this implies that he/she can recognize a difference, whereas if it’s not it implies he/she treats it as the
same.
Results using preferential looking and habituation reveal that infants have a preference for social stimuli over non-
social stimuli. Another method is coding the imitative behaviour (tongue protrusions, lip rounding).
There are several methodological problems with this method, primarily because the scoring system is open to human
error.
1) Inter-rater (or inter-observer) reliability: the extent to which two independent observers generate the same
answers.
This is typically dealt with by recording the experiment and having two people independently scoring a randomly
selected subset of the behaviours.
2) Whether the observer knows the hypothesis and might be biased to report what they expect to see
Blind scoring: the observer is unaware of the status of the event that is being scored.
Observational methods:
+ They can be used when it is impossible/inappropriate to give instructions to a participant; they can be used in
naturalistic settings
- There are difficulties associated with scoring and observer biases (although methods as eye tracking can limit this)
How to measure the unconscious
Masking: the presentation of junk visual material after a stimulus (to eliminate persistence of a visual image)
You can ask participants to guess what they saw, or use SCR or eye blink startle responses.
2.1.3 Survey measures: questionnaires and interviews
Questionnaires – questions and a set of responses that are fixed in advance
Interviews – questions and a range of responses that are open-ended
First-person methods – the participant is expressing his/her own thoughts that cannot be objectively labelled right or
wrong.
Reliability: the extent to which the same measure would yield the same results if repeated.
Ask the participants to repeat the same questionnaire at another time point
Acquiescence bias: a tendency to respond affirmatively in surveys, irrespective of the content of the question
Formulate the question differently (I like … / I don’t like …)
Factor analysis: a statistical method for reducing a data set (e.g. in questionnaires, 20 questions may be
grouped into a smaller number of factors)
External validity: the extent to which a measure relates to something useful in ‘real life’
Survey methods ask people how they think they might behave, so they have a lower external validity than
observational methods. However, they have some advantages. Many researchers are also interested in what
people think and feel, besides how they behave. They are easier to carry out and the external validity may be
improved by administering the surveys anonymously and confidentially, so participants will be more inclined
to give an honest answer.