WHAT THE CONSUMER’S BRAIN TELLS THE
CONSUMER’S MIND – AND HOW WE CAN DISCOVER WHAT
Dr David Lewis-Hodgson BSc (Hons), D.Phil. FISMA, FINSTD, C.Psychol
Director of Research – Mindlab ® International Ltd
“Within the past few years…advances in psychophysiology have stimulated a number of
innovative applications of physiological testing. This approach…is likely to be the next
wave of innovation.” Bruce F. Hall On Measuring the Power of Communications; Journal of
Advertising Research June 2004
“Hey big spender, so refined, wouldn’t you like to know what’s going on in my mind?”
Shirley MacClaine’s mocking query from the 1969 musical Sweet Charity perfectly
describes the challenge long confronting marketing directors, brand managers and
How to discover precisely what is going on in the minds of consumers as they watch
television advertisements, select products off a supermarket shelf or buy holidays from
travel agents, stroll around car showrooms or, indeed, while they are engaged in any
Measuring the effectiveness of such advertising has been a highly contentious issue
for at least twenty-five years an era during which, Bruce Hall commented in a recent
article in the Journal of Advertising Research (1) testing methods have changed little if
at all over this period.
“Over the past three decades,” he comments, “copy testing has not benefited from
significant innovation…Issues that emerged over half a century ago continue to divide
So how can we discover what messages the consumer’s largest subconscious brain is
telling the consumer’s conscious mind when making purchasing decisions?
The most obvious answer might appear to be simply to ask them!
This is of course the route that traditional market research has long taken by means
of surveys and focus groups, which use questionnaires designed and analysed with ever
increasing sophistication. Unfortunately even the most carefully constructed and
painstakingly conducted such research contains significant sources of error.
(1) On Measuring the Power of Communications Journal of Advertising Research Vol 44. No 2. June 2oo4.
Pp 181 - 187
The Problem with ‘Just Asking’ in Market Research
The main difficulty with this approach is that respondents will, for a variety of
reasons, tell you something other than the truth. Indeed their comments, explanations
and reflections on past and future purchasing decisions may be so wide of the mark as
to be at best useless and at worst profoundly misleading, for two main reasons.
First, such deceptions are due less to any deliberate intention to lie than from a
desire either to provide answers they consider socially acceptable or present themselves
in the best possible light. While market researchers have developed techniques to
minimise such bias, they can do nothing to overcome the second and more serious
barrier to understanding what is going on in a consumer’s mind – the fact he or she
doesn’t really know themselves!
This is due not to any lack of intelligence or insight on their part but simply results
from the way in which the brain functions, reflecting the constant interplay between our
memory, decision making processes and emotional responses many of which will
typically occur below the level of awareness.
Research conducted by myself and other neuropsychologists has shown that at least
95% of consumer decision-making is unconscious. Studies have consistently
demonstrated that the majority of human behaviours are controlled by processes below
the level of conscious awareness. The belief that we are consciously in control of our
actions and decisions is largely illusion.
In a study of decision making involving financial gain or loss conducted by William
Gehring and Adrian Willoughby at the University of Michigan, for example, subjects
were wired up to an EEG (Electroencephalograph) and seated at a computer screen on
which was displayed two boxes, one representing a bet of 5 cents, the other of 25 cents.
Each time the subject selected a box it changed colour: red denoted they had lost the
amount displayed and green meant they had won it.
The study showed that a quarter of a second after the subject had learned of their
win or loss they showed a dip in activity an area of the brain known as the medial
frontal cortex, with a string of losses creating an ever decreasing dip. Gehring and
Willoughby argued that this was direct neurological evidence of what is known as
‘gamblers fallacy’ the decision-making error whereby after a string of losses, people
increasingly conclude that they must be ‘due’ for a win.
They demonstrated that such decision-making must be emotional in nature since it
occurred far too rapidly to be the result of conscious deliberation. It also provided an
excellent example of how QEEG (Quantified Electroencephalography) can reveal
important clues as to the unconscious nature of much consumer decision-making.
Emotions, Decisions and Memories
Emotions play a far more significant part in consumer decision-making than does
purely rational, logical decision-making. One reason for this is that because the
emotional part of our brain is far older, in evolutionary terms, than our rational brain it
is more fundamental to our decision-making. As Harvard marketing professor Gerald
Zaltman comments: “Brain scans and other physiological-function measures
demonstrate that activations among brain cells, or neurons, precede our conscious
awareness of a thought and precede activity in areas of the brain involving verbal
language. In fact, these latter neuronal areas become active only later, after a person
unconsciously chooses to represent these thoughts to herself or to others using verbal
Another area of market research in which QEEG can provide essential insights lies
in investigating the ways in which consumers attend to and make sense of commercial
messages, most of which occurs not in words, but in images and feelings.
Research into memory divides memories into two basic types: explicit and implicit.
Explicit memories are those we are able to consciously describe.
These are the type of memories we talk about, and recall when asked to remember a
particular decision. Numerous research papers have demonstrated that explicit
memories are highly malleable and often of limited validity. For example, the order,
manner and presentation of questions given to subjects in a questionnaire, survey or
focus group session will affect the answers respondents give. Research has also
demonstrated that explicit memories are unreliable: a consumer’s recall of a spending
decision will change over time.
However, implicit memories and processing are unconscious and can only be
detected indirectly through their influence on our behaviour.
Research over the last few years has demonstrated that implicit processing is a far
more accurate predictor of behaviour than explicit processing. This is especially true
when a consumer’s explicitly reported thoughts on a product are at odds with their
implicit thought processes on it.
In other words we only become consciously aware of our buying decisions once they
have been made, at which point we seek to justify them.
It is these intellectual justifications the consumer describes during traditional
market research interviews. Like an iceberg, at least nine tenths of consumer
information lies beneath the surface, unseen. Not being able to efficiently explore this
depth of customer information results in unnecessary product failures and missed
During the last decade, brain researchers discovered more about the workings of
the human mind than during all previous years. In the words of leading neuroscientist
Antonio Damasio: “More may have been learned about the brain and the mind in the
1990s – the so-called decade of the brain – than during the entire previous history of
psychology and neuroscience.”
We now have a suite of experimental tools for probing the unconscious of the
customer, and can explore the 90% of the consumer-information iceberg. New
Neuromarketing techniques allow us to explore the implicit side of a consumers
thinking that conventional market research techniques cannot pick up on. The only way
to get at the truth is not so much to ask individual consumers what they think and feel
as to analyse physical processes going on deep in their brain while these thought
processes are occurring.
Contrary to popular fears about neuromarketing being able to ‘control consumers
minds’, these techniques instead provide a fuller and deeper explanation of how
consumers will act. This information then has the potential to provide greater value to
consumers by understanding them deeper than the typical surface-level appraisal.
As Bruce Hall rightly comments – “the body does not lie.”
What is QEEG?
QEEG stands for Quantified Electroencephalography. This is a recording ("graph")
of electrical signals ("electro") from the brain ("encephalo") obtained by placing sensors
on the surface of the scalp to pick up electrical signals naturally generated by the brain.
The ‘Quantified’ aspect means that the raw EEG, which is impossible for anyone but a
specialists to interpret is processed through a computer to provide a more revealing and
easily understood output.
History of the Discovery of the EEG
Dr Richard Caton, a Liverpool physician and medical school lecturer, who placed
electrodes directly onto the exposed surfaces of laboratory animals, first recorded EEG
signals in 1875. To obtain his results he made use of a reflecting galvanometer, which
had been, invented some seventeen years earlier Lord Kelvin.
This amplified the tiny electrical currents generated in the brain by changing the
position of a mirror attached to galvanometer coils, so producing a far larger movement
by mean of a reflected spot of light.
In 1887, Caton told delegates to the Ninth International Medical Congress in
Washington, D.C. that when he interrupted a beam of light directed towards the
experimental animal's eye, he detected changes in the electrical activity, which occurred
in the opposite side of the brain from the eye. Most physiologists, however, remained in
ignorance of Richard Caton's work because he published his reports in medical journals
while their reading was confined to scientific journals.
In 1890, a Polish doctor named Adolph Beck of Poland repeated Caton’s studies
and found that while a sudden sensory stimulus such as a flash of light, produced
response at a single point, this was accompanied by a widespread reduction of a type of
relatively slow, regular, patterns of brain waves. However it took another sixty years
before neuroscientists identified parts of the brain involved in these responses, an area
known as the reticular activating system, which plays a key part in regulating brain
Around the same time a German, Fleischel von Marxow made similar discoveries
which he described in a report which, rather than take the risk of professional ridicule by
publishing he hid away in a bank vault. In was only after Beck published his own findings
in a physiological journal that von Marxow released them in order to claim credit for the
discovery. However his attempt failed after Professor Caton wrote to the journal informing
them of his own earlier publication. Around the same time as Caton was carrying out his
research, similar Vasili Yakovlevich Danilevsky was also conducting studies in Russia a
graduate student. Although his experiments were made in 1876 he did not publish them
until his doctoral thesis appeared the following year, two years after Caton.
Between the work of Vasili Danilevsky and Adoph Beck, another Russian, Nikolai Y.
Wedensky, used a telephone receive to eavesdrop on electrical waves in the brains of cats
and dogs. In 1912, Vladimir V. Pravdich-Neminsky published photographic recordings of
brain waves in dogs.
The honour of being first scientist to record electroencephalographs from human
brains, however, goes to Dr. Hans Berger, an Austrian psychiatrist, who for this reason is
often called ‘The Father of EEG’.
Soon after obtaining his doctorate from the University of Jena, in 1897, Berger read
about Richard Caton's studies and was inspired to conduct early, and inconclusive, studies
using animals. It was not only after World War I that he turned his attention to looking for
the EEG in the human brain. In the early 1920s, he succeeded in obtaining readings by
using patients, part of whose brains had been exposed through damage to their skulls. This
enabled him to adopt Canton’s approach of placing electrodes directly on the membrane
covering the otherwise exposed brain. Among his discoveries were regular waves at about
10 cycles per second, which he named the Alpha since they were the first to be identified in
the human EEG.
In 1929 he published the results of experiments, conducted five years earlier using
his then 15-year-old son, Klaus, as a subject. His data comprised some 73 recordings, the
first ever published of EEGs in humans. Of these he wrote: “The electroencephalogram
represents a continuous curve with continuous oscillations in which ... one can distinguish
larger first order waves with an average duration of 90 milliseconds and smaller second
order waves of an average duration of 35 milliseconds [Beta waves]. The larger deflections
measure at most 150 to 200 microvolts....we see in the electroencephalogram a
concomitant phenomenon of the continuous nerve processes which take place in the brain,
exactly as the electrocardiogram represents a concomitant phenomenon of the
contractions of the individual segments of the heart."
Within a year, Berger had made more than a thousand recordings from 76 subjects
and used these as the basis of a second paper in which he first named both Alpha and Beta
waves as well as introducing the acronym EEG to stand for ElectroEncephaloGram.
Berger reported that the amplitude of the Beta waves was smaller than Alpha,
despite the fact that he was able to demonstrate that Beta waves were related to mental
concentration and to startle reactions.
Despite these groundbreaking discoveries, Berger's work was ignored by the
majority of scientists who believed that overall brain wave recordings would just be
‘confused roars’ and later because they claimed that Alpha waves were merely "dull"
regular waves. Gradually, however, Berger’s contribution came to be more and more
widely recognised in all countries except anti-Jewish Nazi Germany. In September 1938,
Berger was forced to resign from the university and although twice considered for the
Nobel Prize, the Nazis prevented him from accepting the honour. In 1941 Berger was
driven to suicide.
My own interest in EEG and its possible role in market research started while I was
working in the department of Experimental Psychology at the University of Sussex during
the mid 1980’s. This arose from a desire to find a suitable stimulus when studying the
response of both highly stressed and normal subjects. Television commercials struck me as
an ideal medium, since not only had they been carefully crafted to be as memorable and
persuasive as possible, but their short duration – around 30 seconds – ensured that not
too much data was generated for analysis by the relatively crude software then available to
me, most of which I was obliged to write myself in the absence of commercially available
Dr Lewis at work with a subject during his early studies of EEG.
For many years my use of commercials continued as an entirely academic pursuit
and it is only much more recently that it has started to attract interest from commercial
organisations and advertising agencies.
So what exactly is EEG? How does it arise, how can we measure it and what does it
Our Electric Brains
At around 1,500 grams the human brain represents just 2% of our body weight yet
consumers 60% of the glucose produced and requires 20% of the total energy. Much of
this is used to produce electricity. The brain cell mainly responsible for storing and
transmitting information is the neuron. Communications between neurons involves the
flow of various ions, such as sodium, potassium, chlorine, or calcium, into and out of each
neuron. Since ions have an electric charge, communication brings about changes in the
electrical potential inside relative to the outside of the cell. This activity can be detected by
means of sensors attached to the scalp and connected to an EEG.
Because mental and neural activities are one and the same, every thought that we
have, every emotion we feel and each action we take is associated with different patterns of
activity in vast networks of interconnected neurons that make up our brain.
This means that every cognition, even those occurring below our normal level of
consciousness, has a corresponding electrical signature that is, at least in theory, capable
That the task of acquiring, interpreting and using EEG signals is far from
straightforward is hardly surprising given the staggering complexity of the human brain
with its hundreds of billions of neurons each of which forms synapses (chemical junctions)
with and average of 10,000 other neurons and leads to more than 7 million kilometres of
‘wiring’ all within an organ around the size of a coco-nut!
As a consequence, even the most trivial mental activity, such as wondering where
you have left your car keys, involves large, highly distributed networks of millions of
neurons all working together. This electrical activity can be detected as "brain waves"
whose frequency is measured in Hertz (Hz) - or cycles per second) - and power
(amplitude) are continually varying depending on mental state and the intellectual
challenges being faced.
Since their discovery by Hans Berger brain waves have been classified into a
number of bands according to their frequency. While there is still no universal agreement
on exactly which frequencies belong in which category, most research papers explicitly
define the frequency ranges associated with each label, and rarely deviate by more than 1
Hz from the labels described below.
Delta (0.5 – 4 Hz) reflects very slow, high amplitude oscillations of neural populations
and may be seen in many brain regions. It is particularly associated with sleep.
Theta (4-8 Hz) also appears widely distributed across the brain. It is associated with very
relaxed or deep meditative states and is prevalent during some sleep stages.
Alpha (8-13 Hz) is most widespread over the occipital (vision) and parietal (sensory)
regions of the brain, and is typically associated with relaxed wakefulness and appears in
lighter sleep stages.
Mu (8-13 Hz) is the name applied to alpha waves when present in the frontal regions of
the brain. These are most evident over the posterior frontal lobe, the region that contains
motor processing areas. The mu rhythm is most pronounced when subjects are motionless
and becomes desynchronized when subjects either see someone else moving or
contemplate moving themselves again consistent with the view that rhythmic
synchronized activity reflects an idling rhythm.
Beta (13-40 Hz) usually comprises irregular, very small amplitude waves, which are most
prevalent over frontal regions when the subject is alert and engaged in a task requiring
Gamma (40 – 100 Hz) high frequency waves associated with memory consolidation and
Neuromarketing researchers employ their knowledge of the presence or absence of
these frequency, together with their strength (amplitude) and location to answer such
practical questions as how viewers are responding, second by second, to a television
commercial, what is going through their minds while window shopping in a shopping mall,
how they are reacting to the designs of a new car, artwork or packaging designs.
These answers are sought within a very substantial body of research literature in
which the EEG record is recognized as a by product of the neural activity responsible for a
specific thought, emotion or action – especially those which arise below the level of
Such mental activity is, by definition unable to be put into words and, as a result,
unavailable to traditional qualitative research no matter how experienced the researcher
or how co-operative the respondent.
How EEG Readings are obtained before recordings can be obtained the subject must first
be prepared (“prepped”) so as to obtain clean signals. This involved lightly – and entirely
painlessly - abrading the area under each electrode to remove the top layer of skin cells,
which are dead and therefore poor conductors of electricity. This reduces the noise
generated by changes in electrodermal activity, called the galvanic skin response or GSR.
The next step is to introduce a conductive gel between the skin and the electrode after
which it is usually necessary to ‘fine-tune’ the contact by adding more gel and gently
manipulating the electrode while monitoring electrical resistance until this drops to an
acceptable threshold (typically 5,000 Ohms).
EEG signals are often recorded from an electrode cap containing an array of electrodes.
These may contain as many as 256 recording sites, though typical caps use 16, 32, 64, or
128 sites. High-density arrays can yield more information, especially about the spatial
distribution of the electrical activity.
In our studies we typically use 20 electrodes although the total can vary depending
on the requirements of a particular study. Clearly the more electrodes the more precise are
the measurements than can be made from specific regions of the brain. When only the
balance of output between left and right hemispheres is required, for example, four
electrodes (plus a reference and ground electrode) may suffice. This would be useful, for
example, if one wanted to find out whether the features and benefits of a particular
product or message were being assessed mainly factually (high left brain activity in the
majority of right handed people) or evoking a largely imaginative response (high right
Subject wearing an elctro-cap during a study of driver behaviour for the Ministry
In addition to scalp electrodes, experimenters often record from one or two
electrodes near the eyes to detect the electrical activity associated with eye movements
(called the electrooculogram or EOG), as well as an electrode to serve as a ground.
Data containing excess artifact from eye movements or other activities such as
fidgeting or swallowing is sometimes thrown out as “artifact,” though it is also possible to
remove to remove such noise from the data using modern software.
The arrangement of electrodes is known as a ‘montage’ and the type selected varies
depending on the requirements of the task.
In a ‘monopolar montage’, each electrode is paired with a common reference. This
common reference may be a single electrode placed at a “neutral” site (i.e. the mastoid or
earlobe) whose activity is assumed not to be affected by brain activity; equally it may be an
average of two or more electrodes.
In what is called a ‘bipolar montage, electrodes are grouped in non-overlapping
pairs, and potentials are recorded between each pair. This means that a bipolar montage
intended to record 16 waveforms requires 32 electrodes.
Because neuromarketing studies are, typically, conducted in noisy environments,
where electrical devices can create considerable interference, bipolar differential
recording, in which the activity at one site is measured relative to another nearby scalp site
are often used. Since both scalp sites tend to pick up about the same noise, most noise is
cancelled out, and the remaining signal reflects the difference in activity between two scalp
Processing EEG Signals
EEG signals, which are typically of the order of 100 microvolts must first be
amplified by a factor of between 5,000 to 10,000 times. Next they are filtered to remove
frequencies not of interest in a particular study. This is a crucial step in noise reduction,
since certain types of artifact occur at known frequencies and cognitive activity very rarely
occurs outside of the 3-40 Hz range. Four types of filters are typically employed:
1] Low Pass filters allow low frequency information to pass, but removes higher
frequencies. A typical low pass filter setting is 100 Hz.
2] High Pass filters allow high frequencies and remove lower ones. Typical high pass
setting is 0.01 Hz.
3] Band Pass filters, which combine low and high pass filters, allows only a band of
frequencies to pass.
4] Notch filters remove all frequencies in a certain range. The most common notch filter is
either 50Hz (UK) or 60 Hz (USA) to remove the significant noise created by electrical
currents [50Hz UK/60 Hz US].
In the final stage signals are converted from analog to digital, using an ADB
[Acronym for Analog to Digital Board], which enables them to be processed by a digital
computer. Advances in microelectronics have resulted in the development of small and
inexpensive chips, which are capable of digitizing EEG data, and these have now replaced
the ADB in most modern EEG equipment.
Though A/D conversion equipment can easily sample on the order of GHz, EEG
data is rarely sampled above 512 Hz because the brain simply does not operate fast enough
to justify higher frequencies. Sampling too fast provides unnecessary data that could be
interpolated if necessary from adjacent data points. Sampling too slowly creates the risk of
missing certain frequencies.
QEEG Analytical Techniques
In QEEG, multi-channel recordings are analyzed using a mathematical procedure
known as Fast Fourier Transform (FFT) to quantify the power at each frequency of the
EEG averaged across the entire sample, known as the power spectrum. The test-retest
replicability of power spectra computed in this way has been shown to be highly
reproducible. As mentioned above, the power spectrum of commercial interest is generally
considered to extend from about 1Hz to 30Hz (cycles per second).
Results from each electrode can be represented as absolute power in each band,
relative power in each band (percentage of total power in each channel), coherence (a
measure of synchronization between activity in two channels) or symmetry (the ratio of
power in each band between a symmetrical pair of electrodes).
Research has shown that the power spectra of large samples of healthy, individuals
across a wide age range undergo systematic changes occur between the ages of 17 and 64
in the average power in the delta, theta, alpha, and beta frequency bands.
More recently normative data have been extended to cover the age range from 1 to
95 years for each electrode in the standardized International 10-20 System and have been
broadened to include measures of absolute power, relative power, mean frequency,
coherence, and symmetry, as well as covariance matrices that quantify normal brain
relationships, multivariate composite measures which are unique to QEEG.
The independence of the normative QEEG descriptors from cultural and ethnic
factors enables objective assessment of brain integrity in persons of any age, origin, or
background. This independence and specificity, as well as high replicability, has been
established in studies from Barbados, China, Cuba, Germany, Holland, Japan, Korea,
Mexico, Netherlands, Sweden, the United States, and Venezuela.
In summary a voluminous literature attests to the robustness of QEEG as a
powerful and reliable neuromarketing methodology.
Modern analytical procedures offer commercial organisations an unsurpassed
capacity to measure and to analyse consumer brain patterns under a wide variety of
environmentally realistic scenarios.
The term ‘Neuromarketing’ was only coined - by Professor Ale Smidts -in 2002 and
it was not until 2004 that the first ever Neuromarketing conference was held at Baylor
College of Medicine in Houston. Today many leading market research professionals are
describing Neuromarketing as representing the greatest advance in their industry for more
than quarter of a century. As I have already explained, my own interest in the subject
started during the late 1980’s with experiments conducted at the University of Sussex.
At this time I also began by developing equipment which was used to translate
brain activity (EEG) into a control signal which would allow people, especially the severely
disabled, to control various electrical devices by means of ‘thought power’ alone. Today
this technique, known as Brain Computer Interfacing is a major research topic which
offers considerable hope for those with serious physical disabilities.
Later I examined ways of enabling stressed and anxious individuals to relax, to
become more creative and production by learning how to control brain activity by
means of strobe lights and neurofeedback. The latter is a process whereby what is
happening inside the brain is fed-back to that brain enabling the individual to lean ways
of increasing or decreasing specific frequencies.
During the past ten years, developments in both electronics and software mean the
torrent of information can be kept to manageable proportions without loss of vital
information while the data is presented in an easily interpreted manner making it easy
for non-specialists to understand how different individuals are responding to their
commercial messages and instantly to identify both the strengths and weaknesses of
Furthermore because the equipment is easily portable subjects can be recorded in
real life situations, such as when shopping, wandering around a car show room,
watching a cinema advertisement or a television commercial in the comfort of their own
When combined with other specialist equipment it is also possible to correlate
precisely what that individual was seeing and hearing at the precise moment brain
activity was being recorded.
While this may sound complicated, all it means is that the information can be
analysed over a required period of time and the final output readily understood by non-
When analysing a television commercial, for example, time frames can be selected
corresponding to the length of individual scenes. This enables the impact generated by
each scene, together with its relationship to the rest of the commercial to be evaluated
Technology of Neuromarketing
There are two main techniques of brain analysis currently being used – fMRI and
QEEG. In my view the former, although it has received a great deal of publicity in the
general, technical and trade press is not a viable option for market research. Why this is
so I will explain in a moment after describing what the two technologies involve.
What is fMRI?
The acronym stands for Functional Magnetic Resonance Imaging and in order to
understand how it works, it is first necessary to understand something of the way
conventional magnetic resonance imaging (MRI). MRI scanning uses a very strong
magnet and radio waves to produce astonishingly detailed images of the brain or other
These images which reveal brain anatomy at high resolution (a typical MRI scan
will divide the brain into some 124 wave thin slices) are used in virtually all modern
hospitals to diagnose a wide variety of brain disorders, including brain tumours,
multiple sclerosis, and stroke.
In order to obtain these images the patient must lie motionless on the examination
table and slide their head inside a giant magnetic ring. This causes some of the atoms
(or, more precisely, particles inside the atoms, called protons) inside the patient's head
to align with the magnetic field. A pulse of radio waves is then directed at the patient's
head and some of it is absorbed by the protons, knocking them out of alignment. The
protons, however, gradually realign themselves, emitting radio waves as they do.
The radio waves are captured by a radio receiver and are sent to a computer, which
constructs the brain image. Different parts of the brain respond to the radio waves
differently, and emit slightly different radio signals depending, among other things, on
the local water and fat content. The computer that receives all of the signals, therefore,
is able to distinguish one brain structure from another. The patient can sense neither
the magnet nor the radio waves and only knows the machine is working at all due to the
sounds made while scanning.
MRI and fMRI
Functional MRI (fMRI), developed in the early 1990s, is a variation of magnetic
resonance imaging. It uses a conventional MRI scanner, but takes advantage of two
additional phenomena. The first is that blood contains iron, which is the oxygen-
carrying part of haemoglobin inside red blood cells, and that these atoms cause small
distortions in the magnetic field around them. (1)
(1) Not all iron atoms do this. In the case of blood it is only iron which has not been bound to oxygen
(deoxyhaemoglobin) which produces this effect.
The second key phenomenon underlying fMRI is the fact that whenever any part of
the brain becomes active, the small blood vessels in that specific region dilate, causing
more blood to rush in order to provide extra oxygen and fuel (glucose) for the active
brain cells. As a result a large amount of freshly oxygenated blood pours into any active
regions of the brain reducing the amount of oxygen-free (deoxy-) haemoglobin and
causing a small change in the magnetic field and, consequently, in the MRI signal, in
the active region. In the early 1990s, it was shown that an MRI scanner can be used to
detect this small change in the signal and, in doing so, identify those areas of the brain
have been activated.
If, for example, a patient lying in a scanner is suddenly shown a flash of light, the
visual cortex in his brain will become activated, blood flow to that region will rapidly
increase, and the MRI signal will change.
The result is usually displayed as a patchy area of colour, representing the brain
area activated, superimposed upon a conventional, high-resolution, grey-scale image of
the subject's brain.
Although undoubtedly seductive the colourful brain images produced are the
result of high-level computer processing and cannot be interpreted “without a detailed
understanding of the analytical methods by which they are generated.” (2)
In fMRI, the same scanner is optimized to detect small changes in blood flow in the
brain in response to scientifically designed stimuli. In principle, fMRI can be used to
observe the activation of brain structures in response to almost any kind of brief
stimulation, ranging from sounds, to visual images, to gentle touching of the skin.
At Baylor College of Medicine in Houston, for example, neuroscientists have shown
that the brain registers a preference for Coke of Pepsi similar to that chosen by the
subjects in blind taste tests.
Similarly, studies conducted by Richard Silberstein, a neuroscientist with the Brain
Sciences Institute at the Swinburne University of Technology in Melbourne, Australia
has demonstrated that successful advertisements generate both high levels of emotional
engagement and long-term memory encoding: “People who are more likely to purchase
a product show significantly higher memory encoding than those who are less likely,” he
has been reported as stating.
The Problems of using fMRI for Neuromarketing
While fMRI provides a detailed record of brain activity at any particular time the
procedure is fraught with problems when it comes to using it for the purposes of
commercial neuromarketing research. fMRI machines are cumbersome pieces of
equipment, which have to be used within fixed, specialist, locations. This means that
subjects must travel to the machine for testing and only one person can be tested at a
time. While commercial messages can be sent to the volunteers either by means of a
mirror which direct images into their field of view or using goggles fitted with miniature
television screens or.
Furthermore the subject cannot react to the images with any sudden movement,
which would distort the scan. Both methods are, of course, highly artificial and may well
introduce a significant bias into any results. As the Nature Neuroscience commented the
results are: "Invariably produced under controlled laboratory conditions and it is a major
leap to extrapolate to a genetically and culturally diverse population of people in an almost
infinite variety of real world situations."
(2) Nature Neuroscience 7, 683 (2004)
Although the process is considered absolutely harmless this noisy clanging makes
many people nervous while any who are claustrophobic may start to panic. Indeed
worldwide some 7 per cent of subjects are unsuitable for fMRI scanning for this very
reason. Even when they are made only mildly apprehensive by the procedure, such anxiety
will of course exert a profound influence on their response to commercial messages.
Two further major drawbacks are cost and time. MRI scanners are extremely
expensive to buy (around $1 million) and must be operated by highly trained technicians.
Currently running a single subject through an fMRI scanner costs around $1,000 with the
bill for large-scale studies frequently exceeding $250,000. Although the time taking to
record the image is reasonably rapid – up to about 10 minutes for a very detailed picture –
it can then require up to 20 hours to process the images through a computer.
The length of time taken to record the image, although reasonably short, still means
that the best a market researcher can hope for is a snapshot, taken over a period of up to
ten minutes, of how the brain is responding. It would not be possible, for instance, to
analyse how the brain was responding moment by moment to the rapid sequence of
changing images and sounds that make up a typical television commercial.
Quantified Electroencephalography (QEEG)
Of all the imaging modalities currently being employed in the fledgling field of
neuromarketing date provided by EEG and QEEG studies is the most practical using
relatively simple, inexpensive, compact equipment capable of quantitatively assessing
brain activity with a sensitivity and temporal resolution superior to those of any other
As previously explained, this involves measuring and recording electrical activity in
the brain while the individual is looking at and/or listening to and/or experiencing the
desired commercial experience.
Electrical activity of each brain region is homeostatically regulated, resulting in
predictable frequency composition of the background EEG. Furthermore research has
also established that the EEG power spectrum is independent of ethnic background.
Artefact-free EEG evaluated relative to such norms displays few deviant values in healthy,
normally functioning individuals. Evidence from brain imaging methods such a
Electroencephalography (QEEG) and event-related potentials (ERP) analysis, topographic
QEEG and statistical probability mapping has unequivocally established that aspects of
consumer cognition and emotional responses to commercial messages, occurring below
the level of conscious awareness, can be successfully monitored in real time and analysed
with sufficient depth and accuracy to provide an invaluable window on their inner decision
Using the most sophisticated available equipment and algorithms the process is
capable of virtually “reading the mind” of consumers at the most basic level of brain
function by observing the type of mental activity occurring and the regions which are most
active as the individual is experiencing the marketing or advertising message.
As mentioned above, this can be done either in a fixed location, such as a work or
home environment, or while the consumer is actually making his or her purchasing
decisions in a supermarket, shopping mall or showroom.
Exactly what is going on in their minds can be recorded in real time on a second by
second basis on each side of the brain and within specific regions of the brain. The process
can reveal, for example, whether a message is being thought about logically and
analytically or at a more imaginative and emotional level.
A television commercial for a car, for example, might be viewed at either a logical
level, where the viewer is thinking about those technical aspects of the vehicle described in
the advertisement (left brain activity in a majority of right handed individuals) or to evoke
a sense of excitement, inviting the viewer to fantasise being behind the wheel and vividly
imagining what it would be like to drive that vehicle.
Among the many essential items of marketing information which QEEG and
associated physiological measures provide are:
1. The precise colour, shape and design that works best for a new product.
2. Which one of a number of preliminary edits or “animatic” versions of a
television commercial is most likely to ensure brand recognition and/or
generate positive emotions towards the product.
3. The extent to which key marketing messages are likely to be consolidated
into long-term memory.
4. The type of displays most likely to generate the desire to buy.
5. The sort of music that will prove is most persuasive when accompanying a
commercial or being played in a store.
6. Whether or not the viewer’s attention during a commercial message is
maintained at the point of branding
7. What is happening inside a potential customer’s brain as he, or she, studies
different aspects of a large and sophisticated product such as a car, a
computer or hi-fi?
8. Whether a radio commercial or a poster advertisement will best capture the
attention of a target market?
9. Which groups of potential customers will prove more open to a visual or an
10. Which colour, shape and design works best for a new product?
11. Which of a number of initial edits on a television commercial is most likely to
ensure brand recognition and/or generate positive emotions towards the
12. Which type of window displays is most likely to arouse interest and generate
a desire to buy?
13. What type of music is most persuasive when accompanying a commercial or
being played in store.
14. Whether or not the viewer’s attention during a television commercial is
maintained at the point of branding.
15. What is happening inside a potential customer’s brain as he, or she, studies
different aspects of a large a complex product such as a car?
The Technology Employed
In our studies we make wide use of highly portable EEGs, which can easily be
carried around by respondents. About the size of a paper back book (see
illustration above) this stores up to 24 hours of data on a built in hard drive and is
capable of measuring other physiological parameters such as heart rate, rate of
breathing and muscle tension.
A young female shopper tours a UK shopping mall equipment with electrodes and
carrying an EEG in her shopping bag.
During shopping trips and on others when it is necessary to correlate brain activity
with some specific aspect of such an experience the volunteers also wear a pair of
seemingly ordinary sunglasses, which contain a pinhole camera in the bridge. This enables
digital video recordings, in full colour and with sound that provides a volunteer’s eye view
of the unfolding event. Time coding enables specific brain patterns to be precisely related
to what the subject was looking at – and hearing – at any point in the trip.
Output from these brain monitors (EEGs) can be displayed in a wide variety of
ways, for example as a graphic of the brain which uses different colours to reveal the
patterns of electrical activity in different regions, as illustrated below.
Topographic brain maps produced during a study of responses to different facial
expressions conducted for Hewlett Packard. In the first a female respondent is looking at a
picture of a (male) actor displaying an expression of happiness and in the second one of
disgust. The frequency shown here comprises the Alpha wave band and the various colours
show the amplitude (power) of the signal with red and yellow colours indicating higher
amplitudes than green and blue. The first map shows a far higher level of relaxation –
especially in the right frontal lobes – than does the second.
About Dr David Lewis-Hodgson
Dr David Lewis at work in his laboratory. (Photo Steven Matthews)
Dr David Lewis has been working in the field of neuroscience for the past thirty
years and is a Chartered Member of the British Psychological Society. He has a First
Class Honours degree in Psychology from the University of Westminster and a
doctorate from the Department of Experimental Psychology at the University of Sussex
where he later lectured in clinical psychology and psychopathology. He is a Fellow of
the Royal Society of Medicine, the International Stress Management Association and
the Institute of Directors and a member of the American Association for the
Advancement of Science, the International Association of Neuronal Regulation & the
International Society for Bioelectromagnetism. A frequent lecturer at the international
business conferences he is the author of more than a dozen international best sellers on
aspects of brain function and consumerism in The Soul of the New Consumer:
Authenticity – What We Buy & Why in the New Economy.
Addison, T. (2005) More science: more sense or nonsense? Ad-Map, May, Issue 461.
Ahn H, Prichep LS, John ER, et al., Developmental equations reflect brain
dysfunction. Science 1980; 210:1259–1262
Alper K., Quantitative EEG and evoked potentials in adult psychiatry, in
Advances in Biological Psychiatry, vol 1, (Panksepp J., Greenwich, CT. Eds) JAI Press,
1995, pp 65–112
Alvarez A, Pascual R, Valdes, US.,EEG developmental equations
confirmed for Cuban schoolchildren. Electroencephalogr Clin Neurophysiol
American Electroencephalographic Society: Statement on clinical use of
quantitative EEG. J Clinical Neurophysiology 1987; 4:75
American Academy of Neurology: Assessment: EEG brain mapping. Neurology
American Psychiatric Association Task Force on Quantitative EEG
Techniques: Quantitative electroencephalography: a report on the present
state of computerized EEG techniques. Am. J Psychiatry 1991; 148:961–964
Anderer P, Saletu B, Klöppel B, et al. Discrimination between patients and normals
based on topographic EEG slow wave activity: comparison between Z statistics,
discriminant analysis and artificial neural network classifiers. Electroencephalogr Clin
Neurophysiol 1994; 91:108–117
Babiloni F, Babiloni C, Carducci F, et al., High resolution EEG: a new model dependent
spatial deblurring method using a realistically shaped MR-constructed subject's head
model. Electroencephalogr Clin Neurophysiol 1997; 102:69–80
Baillet S, Garmero L., A Bayesian approach to introducing anatomo-functional priors
in the EEG/MEG inverse problem. IEEE Tran Biomed Eng 1997; 44:374–385
Berger H.A., On the electroencephalogram of man. Archiven des Psychiatrie und
nerven Krankenheis 1937; 106:577–584
Brautigam, S, et al (2001) Magnetoencephalographic signals identify stages in real life
decision processes, Neural Plasticity, 8: 241-53.
Bricolo A, Turazzi S, Faccioli F, et al., Clinical application of compressed spectral array
in long-term EEG monitoring of comatose patients. Electroencephalogr Clin
Neurophysiol 1978; 45:211–225
Buckner H, Knoll G, Fuchs M, et al., Inverse localization of electric dipole current
sources in finite element models of the human head. Electroencephalogr Clin
Neurophysiol 1997; 102:267–278
Coles MGH, Gratton G, Donchin E (1988) Detecting early communication: using
measures of movement-related potentials to illuminate human information processing.
Biol. Psychol. 26:69-89
Diaz de Leon AE, Harmony T, Marosi E, et al: Effect of different factors on EEG
spectral parameters. Int. J Neuroscience 1988; 43:123–131
Duffy FH, Albert MS, McAnulty GB., The pattern of age-related differences
in electrophysiological activity of healthy subjects. Neurobiology of Aging
Duffy FH, Jones K, McAnulty GB, et al., Spectral coherence in normal adults:
unrestricted principal components analysis. Clinical Electroencephalography 1995;
Fernandez-Bouzas A, Casanova R, Harmony T, et al., EEG frequency domain
distributed inverse solutions in brain lesions (abstract). Electroencephalography Clin.
Neurophysiology 1997; 103:195
Fisch BJ, Pedley TA., The role of quantitative topographic mapping on
"neurometrics" in the diagnosis of psychiatric and neurological disorders: the cons.
Electroencephalogr Clin Neurophysiol 1989; 73:5–9
Gasser T, Bacher P, Mochs J., Transformation towards the normal
distribution of broad-band spectral parameters of the EEG. Electroencephalogr Clin
Neurophysiol 1982; 53:119–124
Gevins A, Le J, Martin NK, et al., High resolution EEG:124 channel recording, spatial
deblurring and MRI integration methods. Electroencephalogr Clin Neurophysiol 1994;
Gibbs FA, Gibbs EL., Atlas of Electroencephalography Reading, MA, Addison-Wesley,
Gorodnitsky IF, George JS, Ras BD., Neuromagnetic source imaging with FOCUSS: a
recursive minimum norm algorithm. Electroencephalogr Clin Neurophysiol 1995;
Grave de Peralta Menendez R, Hauk O, Audino SG, et al., Linear inverse solutions with
optimal resolution kernels applied to electromagnetic tomography. Human Brain
Mapping 1997; 5:454–467
Greger R, Windhorst U. Berlin, Springer-Verlag, 1996, pp 509–531
Harmony T, Alvarez A, Pascual R, et al., EEG maturation of children with
different economic and psychosocial characteristics. Int. J Neuroscience
Hall, B.F. A New Model for Measuring Advertising Effectiveness, Journal of
Advertising Research, Vol 42. No. 2., 2004, pp 23-32
Hall, B. F. On Measuring the Power of Communications Journal of Advertising
Research, Vol 44. No. 2., 2004, pp 181-187
Hooshmand H, Beckner E, Radfor F., Technical and clinical aspects of topographic
brain mapping. Clin Electroencephalogr 1989; 20:235–247
Hughes JR: The EEG in psychiatry: an outline with summarized points and
references. Clin Electroencephalogr 1995; 26:92–101
Ingvar DH, Sjolund B, Ardo A., Correlation between dominant EEG frequency,
cerebral oxygen uptake and blood flow. Electroencephalogr Clin Neurophysiol 1976;
Jerret SA, Corsak J., Clinical utility of topographic EEG brain mapping. Clin
Electroencephalogr 1988; 19:134–143
John ER, Ahn H, Prichep LS, et al., Developmental equations for the
electroencephalogram. Science 1980; 210:1255–1258
John ER, Prichep LS, Fridman J, et al., Neurometrics: computer assisted
differential diagnosis of brain dysfunctions. Science 1988; 293:162–169
John ER, Prichep LS., Principles of neurometrics and neurometric analysis
of EEG and evoked potentials, in EEG: Basic Principles, Clinical
Applications, and Related Fields, (Niedermeyer E, Lopes da
Silva F. eds) Baltimore, Williams and Wilkins, 1993, pp 989–1003
Kalat, J. W (1998) Biological psychology. Pacific Grove, CA, Brooks/Cole.
Krugman, H.E. (1971) Brain wave measures of media involvement. Journal of Advertising
Research, 11: 3-10.
Kutas, M., & Dale, A. (1997). Electrical and magnetic readings of mental functions. In
M. D. Rugg (Ed.), Cognitive Neuroscience (pp. 197-242). Hove, East Sussex:
LaBarbera, P.A. & Tucciarone, J.D., GSR Reconsidered: A Behaviour Approach to
Evaluating and Improving the Sales Potency of Advertising Journal of Advertising
Research, Vol 35. No. 5. 1995, pp 35-53
Lehmann D, Michel CM., Intracerebral source localization for FFT power maps.
Electroencephalogr Clin Neurophysiol 1990; 76:271–276
Lewis, D. & Bridger, D. (2005) Market researchers making increasing use of brain
imaging, Advances in Clinical Neuroscience & Rehabilitation (in Press)
Llinás RR., The intrinsic properties of mammalian neurons: insights into
central nervous system function. Science 1988; 242:1654–1664
Lopes da Silva FH., The generation of electric and magnetic signals of
the brain by local networks, in Comparative Human Physiology, (Small JG., Ed)
Lütkenhöner B, Grave de Peralta Menendez R., Theoretical model for case of
concurrent or diffusely distributed sources in EEG/MEG. Electroencephalogr Clin
Neurophysiol 1997; 102:326–334
McCormick DA: Neurotransmitter actions in the thalamus and cerebral
cortex and their role in neuromodulation of thalamocortical activity.
Prog Neurobiol 1992; 39:337–388
Matousék M, Volavka J, Roubícek J, et al: EEG frequency analysis related
to age in normal adults. Electroencephalogr Clin Neurophysiol 1967;
Matousék M, Petersén I: Norms for the EEG, in Automation of Clinical
Electroencephalography, edited by Kellaway P, Petersén I. New York,
Raven, 1973, pp 75–102
Nagata K, Tagwa K, Hiroi S, et al: Electroencephalographic correlates of blood flow and
oxygen metabolism provided by positron emission tomography in patients with
cerebral infarction. Electroencephalogr Clin Neurophysiol 1989; 72:16–30
Niedermeyer E: EEG and clinical neurophysiology, in Electroencephalography: Basic
Principles, Clinical Applications and Related Fields, edited by Niedermeyer E, Lopes da
Silva F. Baltimore, Urban and Schwarzenberg, 1987, pp 97–117
Niedermeyer E, Lopes da Silva F. Baltimore, Williams and Wilkins, 1993,
Nuwer M: Assessment of digital EEG, quantitative EEG and EEG brain
mapping: report of the American Academy of Neurology and the American
Clinical Neurophysiology Society. Neurology 1997; 49:277–292
Oken BS, Chiappa KH: Short-term variability in EEG frequency analysis.
Electroencephalogr Clin Neurophysiol 1988; 69:191–198
Pascual-Marqui RD, Michel CM, Lehmann D: Low resolution electromagnetic
tomography: a new method for localizing electrical activity in the brain. Int J
Psychophysiol 1994; 18:49–65
Pascual-Marqui RD: Low resolution brain electromagnetic tomography
(LORETA) (abstract). Electroencephalogr Clin Neurophysiol 1997; 103:36
Pineda, J. A. (2004) Understanding Actions: Mu Rhythms and Mirror Neurons.
Lecture Notes. Cognitive Neuroscience Laboratory.
Prichep LS, John ER, Ferris SH, et al: Quantitative EEG correlates of cognitive
deterioration in the elderly. Neurobiol Aging 1994; 15:85–90
Prichep LS, John ER: Neurometrics: Clinical applications, in Clinical
Applications of Computer Analysis of EEG and Other Neurophysiological
Variables, edited by Lopes da Silva FH, Storm van Leeuwen W, Remond A;
vol 2 of Handbook of Electroencephalograpy and Clinical Neurophysiology.
Amsterdam, Elsevier, 1986, pp 153–170
Rothschild,M. et al. (1986) EEG activity and the processing of television commercials,
Rothschild,M. and Hyun, Y.J. (1990) Predicting memory for components of TV
commercials from EEG, Journal of consumer research, 16(4): 472-478.
Scherg M, Picton TW: Separation and identification of event-related potential
components by brain electric source analysis. Electroencephalogr Clin Neurophysiol
1991; 87(suppl. 42):24–37
Scherg M: Functional imaging and localization of electromagnetic brain activity. Brain
Topography, 1992; 5:103–111
Steriade M, Gloor P, Llinás RR, et al: Basic mechanisms of cerebral
rhythmic activities. Electroencephalogr Clin Neurophysiol 1990; 76:481–508
Scherg M, & Von Crainon, D: Two bilateral sources of the late AEP as identified by a
spatio-temporal dipole model. Electroencephalogr Clin Neurophysiol 1985; 62:32–44
Smith, M.E. and Gevins, A. (2004) Attention and brain activity while watching
television: components of viewer engagement, Media Psychology, Vol 6, 285-305.
Valdes-Sosa P, Bosch J, Grave R, et al: Frequency domain models of the
EEG. Brain Topogr 1992; 5:309–319
Van Veen BD, Joseph J, Hecox K: Localization of intracerebral sources of electrical
activity via linearly constrained minimum variance spatial filtering. IEEE Workshop on
Statistical Signal and Array Processing, 1992, pp 526–529
van Drongelen W, Yutchman M, Van Veen BD, et al: A spatial filtering technique to
detect the localize multiple sources in the brain. Brain Topogr 1996; 939–49
Veldhuizen RJ, Jonkman EJ, Poortvliet DCJ: Sex differences in age
regression parameters of healthy adults-normative data. Electroencephalogr Clin
Neurophysiol 1993; 86:377–384
Wells, M. In Search of the Buy Button, Forbes September 1 2003.