Chapter 1
Introduction
The nature of the discipline
Our vision is to bring together three previously distinct disciplines [education, psychology and
neuroscience] – to focus on a specific common problem: how to promote better learning.
Three disciplines: Education, Psychology, Neuroscience
Education seeks answer to two main questions:
1. What are the sources of individual differences in learning?
2. What are the optimal contexts for the learner?
Phase 1. Education and psychology
Psychology first pointed to two main sources of individual differences in learning. First, there
were differences in intrinsic cognitive capacity, for example as measured by IQ tests. Second,
psychological as well as sociological studies revealed experiential sources of individual
variation, for example differences in home environment. Finally, with regard to optimal
learning contexts, psychology has provided methodologies for investigating and comparing
teaching methods.
Thorndike (1922) took ideas from associationist theories of psychology, and
emphasized drilling simple number bonds. In the 1930s, Brownell, in several important
papers, applied psychological ideas about meaningful practice to how math should be taught.
In the 1950s and 1960s, Piaget’s “constructivist” theories about the nature of cognitive
development were very influential. Constructivism emphasizes the child’s construction of
new schemas (accommodation) when new stimuli cannot be understood using existing
schemas.
Phase 2. Psychology and neuroscience
This phase is characterized by the collaboration between neuroscience and the cognitive,
affective, and developmental branches of psychology, to create cognitive neuroscience.
The critical impetus for the most relevant aspect of neuroscience for education,
cognitive neuroscience, came with availability of in vivo imaging of neural processes as they
happened. Neuroimaging has revealed important aspects of domain-general cognitive
processes, such as performance on IQ tests. Advances have also been made in curriculum-
relevant cognitive capacities.
, It has now become feasible to carry out large-scale studies of the development of the
brain, and to understand better the genetic and environmental factors that affect it.
Phase 3. Emergence of educational neuroscience
Phase 3 is where we are now: we are seeking to use neuroscience to inform educational
practice as a way to improve learning. John Bruer famously argued that this was “a bridge too
far.” Bruer based this position on critiques of three aspects of very basic neuroscience usually
derived from studies of non-human species: the time course of synaptogenesis and synaptic
pruning, critical periods for learning, and the role of enriched early environments.
In fact, new methodologies have enabled scientists to plot the developmental
trajectories with much more precision than previously. Of course, three disciplines are
involved, each with their own methodologies, that cannot easily be unified. Therefore, the
critical move is what Laurillard has termed “methodological interoperability”. That is,
although the methods of the three disciplines are different, it is possible, and indeed
necessary, for each discipline to test the findings of the others.
Issues and problems in developing educational neuroscience
From a scientific perspective, the rationale is clear cut, even if the collaboration between –
and ultimately integration of – disciplines that it requires presents a range of theoretical and
methodological challenges.
The picture becomes more complex, however, when we turn to educational
perspectives on the purpose of this enterprise. Education is itself a hugely complex activity
with social, economic, political, and individual goals – and a corresponding variety of views
on how successful outcomes should be defined.
If the translational goals of educational neuroscience parallel those of public health
science, then the implication would seem to be that we should (a) begin by targeting a key
area of educational need where good theory is able to make an obvious difference, (b) build
outward from this initial example via core teams of individuals representing the different
contributing strands of activity (i.e., the equivalent to epidemiologists, biostatisticians, local
and national government officials, and health service professionals), whose activity is focused
on mutually identified areas of need or risk and methods of counteracting these, (c) promote
public knowledge of effective practices (without necessarily worrying too much about grasp
of why these are effective), and (d) let governments take control ultimately, whilst continuing
to feed them good, relevant evidence.
This is a complex and difficult balancing act. To start with the science itself, the public
health model suggests that researchers have a critical role to play in providing reliable and
systematic evidence that can steer effective action. Equally important is the need to progress
as a community in a number of different senses. As noted already, though, to be effective we
need to recognize that researchers can only be one part of a wider community of engagement
and exchange that helps set the research agenda, and maintains a focus on the implications for
practice, including delivery. Finally, there will also need to be engagement with policy
makers and policy shapers, in order to help ensure that educational neuroscience has socially
perceived value, and that team members are therefore in some sense sanctioned to contribute
to the development and deployment of novel forms of provision.
, Chapter 2
Neuroimaging methods
In this chapter, we review the three major neuroimaging techniques currently used with
children, namely electroencephalography (EEG) and associated event-related potentials
(ERPs), near-infrared spectroscopy (NIRS), and magnetic resonance imaging (MRI).
Electroencephalography and event-related potentials
EEG/ERPs are relatively inexpensive to record and do not require an overt response from the
participant. It is currently the most practical neuroimaging method for studying
developmental changes in brain activity in individuals such as infants and young children who
are not easily tested using other brain imaging techniques.
Principles of EEG recording and averaging ERPs
By placing small metal sensors (electrodes) at different locations over the scalp, scientists can
record and analyse the electrical activity of the brain. Electroencephalography records
changes in brain activity over time by measuring the difference in voltage between two
electrode sites sampled at regular time intervals. Neural activity oscillates at various
frequencies linked to different states of alertness, leading to the popular name for EEG of
“brain waves.”
In contrast to EEG, which measures ongoing brain activity, ERPs are averages of
epochs of EEG at each electrode site, time-locked in response to specific stimuli, such as
pictures or sounds. Typically, and as seen in Figure 2.2, ERPs are illustrated as changes in
voltage (measured in microvolts along the y-axis) over time (in milliseconds along the x-
axis).
Placement of electrodes Electrode application time varies markedly depending on the number
of electrodes and the type of system. The ideal number of electrodes to use depends on the
research questions being asked. Studies using source localization programs will require
a large number of electrodes. In contrast, studies examining modulation of an ERP component
with a wide distribution may elect to record from fewer electrodes.
Most current systems use an elastic cap (see Figure 2.1) or net designed to fit securely
on the head. The amount of time spent placing electrodes on the participant can have a major
impact on the data recorded. Entertaining younger participants can be quite helpful.
, EEG recording As noted above, EEG is measured as the difference in voltage between two
electrodes. To obtain these recordings, most systems select one of the electrode sites as a
reference channel, i.e., an electrode channel that is subtracted from the other active electrode
sites during data acquisition. A data point is recorded at each site every 0.5–2ms.
To eliminate electrical noise not related to brain activity, high- and low-pass filters are
also frequently applied during recording. Somewhat counterintuitively, filters are described
according to the frequencies they allow to pass for recording. Therefore, high-pass filters are
used to eliminate contamination of the signal from low-frequency artefact such as movement
and skin potentials. In contrast, low-pass filters are used to eliminate high-frequency noise
caused by external sources such as electrical equipment. So-called “bandpass” filters set both
high- and low-pass filter settings. It is important to note that filtering during data acquisition
permanently eliminates certain frequencies from the recordings.
Event-related potential signal averaging In contrast to EEG, ERPs reflect changes in brain
activity across a specific time period (epoch) of activity elicited by a specific set of stimuli
such as pictures of familiar versus unfamiliar faces. ERPs are created by averaging together
epochs of EEG that are time-locked to the onset of a particular category of stimuli (e.g.,
familiar faces).
The epoch length includes a prestimulus interval for a baseline, and a specific amount
of time following the stimulus. In general, ERP components that are large in amplitude and
widely distributed require fewer trials per condition than smaller and earlier sensory
potentials.
In general, ERP components that are large in amplitude and widely distributed require
fewer trials per condition than smaller and earlier sensory potentials. ERPs can be markedly
distorted by large artefacts and particularly by artefacts that are time-locked to the stimuli –
e.g., if the participant blinks each time s/he hears a sound.
Blinks are readily detected with an ocular electrode, as they are opposite in polarity
over and under the eye. There are two common approaches for artefact removal, although
which method to use can be controversial. Traditionally, trials contaminated with artefact are
eliminated from the averages through artefact rejection. Additionally, setting individual
thresholds is more time consuming but a better practice.
Making sense of ERP components
Establishing the functional significance of ERP findings can be mystifying to the newcomer
and expert alike. This section will describe how ERP researchers display, measure, and
interpret brain waves. luctuations in voltage that have been linked with sensory and
psychological processes across multiple experiments are called ERP components.
Measurements of ERP components typically reflect the latency (the time in
milliseconds at which the maximum positive or negative activity occurred), peak or mean
amplitude (the maximal or mean positive or negative going activity in microvolts within a
given time window), and distribution of the activity across the scalp (the location of electrode
sites at which the activity is observed).
