The goal of market research is to understand consumer needs, preferences, and behaviors. Traditional techniques rely on information collected from surveys, focus groups, and behavioral experiments. These methods work well, but they scratch the surface of how we respond to stimuli and typically lead to biased results. For example, simply asking a consumer for product feedback may deliver misleading responses, as they may unknowingly provide biased or made-up answers.
As more accurate methodologies have emerged, many market researchers now rely on neuroscience by looking at brain activity to better understand how consumers respond to ads and products. That is not to say that neuromarketing reads minds; rather, neuromarketing investigates how our bodies and brains respond to stimuli unbiasedly.
To understand how advertisements influence purchasing decisions, market researchers ask consumers a series of questions to evaluate the effectiveness of an ad: what caught your attention initially? What kept you engaged? What made you feel a certain way? And ultimately, what made you want to purchase the product? With traditional methods, researchers find it difficult to answer these questions, whereas neuromarketing methods focus specifically on these questions.
Don't let the name neuromarketing make you believe it only studies the brain… Neuromarketing research also uses biological and physiological methods to assess the body's response to stimuli. The popular techniques include electroencephalography (EEG), functional magnetic resonance imaging (fMRI), heart rate tracking, respiration, skin conductance, and eye tracking. Each method provides different insights. Think of them as tools in a toolbox. If you need to drive a nail, you grab a hammer; if you need to cut wood, you grab a saw.
Currently, no single neuromarketing method can answer every research question, but each is useful in different ways. Additionally, while neuromarketing supplements and improves traditional methods, it is not a replacement. The following sections provide a brief overview of each technique.
EEG is used to measure brain activity and as the oldest neuroscience tool, has the most published research backing the methodology. Hans Berger first developed EEG in 1929 to measure the electrical signals generated by human brain cells. EEG is a non-invasive system that only requires electrodes placed on the participant’s scalp. When neurons on the brain’s surface activate, they generate an electrical signal that escapes the skull. The EEG electrodes record that signal as frequencies.
Although there are different approaches to analyzing EEG data, they all examine frequency bands and how they relate to cognitive processes such as attention, emotion, memory, and reward. For example, the alpha frequency (8-12Hz) relates to visual attention. However, pinpointing one specific frequency band to one cognitive function is difficult because certain networks that use gamma (30-80Hz), for example, might involve reward processing. At the same time, neural networks involved in attention also use gamma. To add to the complexity, the power of these oscillations might increase or decrease when someone attends to an object, depending on which electrodes are analyzed.
A method that bypasses this complexity is correlating brain activity across several participants. This method, called cross-brain-correlation (CBC), measures how a group of people think alike. Using CBC does not matter if a frequency band like alpha increases or decreases. When people engage in the same way, their brains respond similarly. ThinkAlike Laboratories® uses this patented method to measure consumer engagement to predict real-world outcomes. Barnett and Cerf collected EEG data from moviegoers while they watched the trailers. The researchers were able to predict opening weekend ticket sales using CBC. Did this EEG analysis "read their minds?" Well, they could expect that the more people were engaged, the higher the CBC and the greater the ticket sales.
EEG is the most cost-effective methodology that measures brain activity, which is why it is widely used in market research today (a research-grade EEG system costs around $50k and is inexpensive to maintain). The system is relatively easy to use, although the analysis and study design can be tricky. The best use for EEG is to answer the "when" question because EEG measures the precise moment the brain is active to the millisecond. However, EEG is subject to noise from movement and other electronics and has difficulty knowing which part of the brain is active. The electrical signal read by EEG comes from the brain's surface and bounces around in the skull, so it is difficult to precisely pinpoint where it came from. One would use fMRI to answer the "where" stimuli are activated in the brain.
Functional Magnetic Resonance Imaging (fMRI)
Magnetic resonance imaging (MRI) is used to measure the structure of a body part, for example, to see if a ligament is torn in the knee, to assess brain damage, etc. Whereas functional MRI (fMRI) uses the same equipment, just a different approach to measure function or brain activity.
To understand fMRI, we must first talk about MRI, which uses powerful magnets to align hydrogen ions (this is important because we have a lot of hydrogen in our bodies). Then another magnetic field flips these ions. The rate at which they flip depends on the medium. That is, hydrogen ions will flip differently in muscle than in bone. The MRI will pick up the vibrations of the ions flipping (resonance) to create an image of the body part.
When the neurons in the brain activate; they require oxygen, which is pulled from hemoglobin in the blood. This deoxygenated hemoglobin (which is iron) distorts the magnetic signal. fMRI measures this distortion. Therefore, fMRI data is a correlate of neuron activity, whereas EEG measures actual neuron activity. This blood oxygen level-dependent response is slow -- it could take a few seconds to obtain the measure after the brain is activated. Therefore, fMRI has a poor temporal resolution (when the brain is active), but it can measure "activity" anywhere in the brain within a few millimeters (great spatial resolution).
The benefit of better spatial resolution is the ability to measure brain structures under the surface (like the amygdala) and the functional connectivity between structures. Using fMRI to measure brain activity offers unbiased information into what psychological processes are active during certain behaviors or decision-making.
One interesting study used fMRI to examine which brain areas activate different buying motivations. Researchers found that experiential motivations (emotional benefit to interacting with the product, e.g., the product is fun) activated the reward network in the brain, but symbolic motivation (social benefits, e.g., status, self-image, etc.) did not relate to reward processing. Therefore, one would find it more rewarding to purchase a product they see themselves enjoying as opposed to the idea that it would increase status or improve self-image.
Although fMRI offers fantastic information about the structure and function of the brain, it is uncomfortable and expensive. If you have ever had an MRI, you know how uncomfortable it is to remain completely still in a cigar-shaped tube that is very noisy. This reduces real-world experiences and may affect how people process information, leading to inaccurate results. The cost of running an fMRI experiment costs around $2,000 per hour. A reasonable amount of time for one scan is about an hour, and researchers typically need about 20-30 participants -- $40k - $60k per study, not including the researcher's time, participant compensation, supplies, etc.
Functional MRI is an effective measure but not practical for applied neuromarketing. That is, fMRI is best used for theoretical understandings, like which neural networks are involved in processing general consumer experiences. Using this method to assess the effectiveness of a single ad may not be worth the cost and risks to the participants (although rare, the risks include claustrophobia, harm to hearing, and heating of the body). For this type of applied study (single ad analysis), researchers use much less costly and less risky methods to measure brain activity, such as EEG and physiological responses.
Many neuromarketing researchers measure physiological responses such as heart rate, skin conductance, and respirations. When a researcher uses all these measures, it is called a polygraph because it has many measures (the literal meaning of "polygraph"). Although a polygraph is terrible at catching people in a lie, it is a reliable method to detect a response from the autonomic nervous system. For example, if someone gets excited from something they like or fear, their heart rate and respirations increase, and their hands get a bit sweatier as measured by skin conductance.
The reason for the increase in heart rate and breathing is easy to guess as more oxygen is sent to the muscles for fight or flight. But why do your hands get sweaty? The answer is that our hands get sweaty when we are in fight or flight mode -- so we can have a better grip when climbing or holding on to something.
One study examined how words per minute in commercials affect heart rate, skin conductance, perception, and memory. The researchers found that commercials with 180 words per minute caused the highest arousal measured by skin conductance and the lowest heart rate. Additionally, recall and perceived effectiveness were highest at 180 words per minute. The perception and memory results are straightforward, but they do not tell why people responded differently. The physiological results suggest that the 180 words per minute offered the perfect balance by reducing cognitive load while maintaining enough task difficulty. When the task is too difficult (200 words per minute) or too easy (160 words per minute), people become disengaged from the task.
Measuring physiological responses offers supplementary information to behavioral results. These methods are relatively inexpensive and easy to use. However, one must be careful in interpreting the results. Take the lie detector as an example. It is easy to assume that an increase in heart rate and skin conductance could mean someone is lying, but the causes of these responses are more complicated. You can ask someone their name, they answer correctly, and they have little physiological response because you assume they are telling the truth. Then you ask them if they murdered someone, and they say no, but their skin conductance, heart rate, and respirations significantly increase; therefore, they are lying. Or, perhaps, they just had a physiological reaction to an emotional question about murder.
The previous methods measure more internal or covert responses. Whereas eye-tracking measures precisely where the subject is looking. One benefit to eye-tracking is that the researcher can use the technique for anything visual: commercials, single-page ads, billboards, websites, and in-store shopping. The previous methods are not as versatile as eye-tracking as the tasks require consistent timing across participants, whereas with eye-tracking, the participant can control how long they look at something.
A recent review discussed the general findings and potential pitfalls from years of eye-tracking research. They found that with stimulus-driven attention, people tend to fixate on large labels and product images, products on the middle shelf, and a greater number of product facings (more rows of cereal on the shelf). Additionally, a person's emotional state, knowledge, motivation, or goal influences eye gaze. For example, a person will fixate on nutritional information when motivated to eat healthily. Although the authors discussed many pitfalls with eye-tracking, the biggest one is "eye-mind assumptions," which is that researchers cannot deduce what the mind does based on the eye's behavior. For example, if you stare at an object for a long time but ignore it while thinking about something else or when you cognitively attend to something using your peripheral vision.
Eye tracking's affordability and versatility attract many consumer researchers, and rightfully so, but you must use caution when interpreting the results.
Each method supplements behavioral or self-reported data by offering unbiased results. Most people do not purposely report inaccurate information, but they simply make a forced response within a research setting. Through the use of neuroscience methods, we correctly measure the exact moment people engage in the video (EEG), which neural network is activated (fMRI), when the video arouses the person (physiological measures), and where people fixate their eyes (eye tracking). But even though we measure how the brain and body respond to products or ads, we do not measure minds. None of these methods tell us what a person thinks or feels.