Neuromarketing: a whole lot of fluff?

The camera pans across a dimly lit swamp. It picks up a bullfrog letting out a deep, loud “BUD”. Another frog joins in with a shrill “Weisssss”; their friend finishes off with a baritone “Er”. At first they call out haphazardly before synchronously calling to each other, “Bud” – “Weis” – “Er”. The camera zooms out, revealing a neon sign with the insignia, Budweiser.

This was a famous TV commercial from the beer manufacturer Budweiser that debuted during the Super Bowl in 1995. The memory of this advert remains with me today, and always puts a smile on my face, I’m not sure why, I’m not sure whether it necessarily makes me more likely to buy a Bud either, but it does stay with me.

Marketing has traditionally been thought of as an art. A creative business fronted by creative types who work hard to develop amusing, emotional and memorable campaigns which convince us we want/need to buy their product. Any metrics assigned to this process have classically been via standard market researcher questionnaires.

The problem with this, so the argument goes, is that people lie. They either tell you what they think you want to hear, genuinely can’t remember or just cannot imagine themselves in a real-life scenario which would allow them to give an accurate answer.

Becoming more and more prominent in this industry is a segment of advertising that claims to eliminate these problems by basing the research on science. Welcome to the increasingly lucrative world of neuromarketing.

Neuromarketing uses neuroscience techniques to try to understand why we buy what we buy, what is that certain je ne sais quoi that turns a product into a must-have?

One of the basic techniques is the use of eye-tracking software. Sensors on the edges of your eyes can track where you are looking at any time. Portable versions have been developed that allow companies to track your eyes as you look around a supermarket or watch a commercial. Companies can tell from this whether you’re looking at what they want you to look at.

In the video below you can see a 2011 advert from the car manufacturer Volkswagen where a child is dressed as Darth Vader and tries to use ‘the force’ to move things around the house. This is overlaid with research carried out at Sands Research Inc. in Texas, United States. In the top left of the video you can see the results of eye-tracking showing what subjects are likely to be looking at any one time.

From this we can see that viewers were looking at faces more than anything else. Also present in this analysis are brain recordings using electroencephalography (EEG). EEG electrodes can be placed all over the scalp and used to record electrical activity from various brain regions. Sands Research’s analysis ranked this advert from Volkswagen as the most engaging in their analysis of all adverts from the 2011 Super Bowl.

EEG Neuromarketing: a whole lot of fluff?Dr. Sands, Chairman and Chief Scientific Officer of Sands Research, said, “As you will see in the Volkswagen ad, the positive and negative emotional response flows with the commercial and ends on an extremely positive point. By creating an engaging and emotional storyline with strong positive response, viewers were extensively engaged and strongly recalled the spot and more importantly, specifically recalled the brand associated with the commercial.”

This is where the field of neuromarketing gets hazier. Very few people would dispute the relevance of eye-tracking to make sure that viewers are focusing on what they should be focusing on. If the scene is too busy and there are too many distractions then the message will be lost. But, there is much scepticism around the idea that EEG recordings can tell us when people are more engaged.

For one thing, EEG recordings have poor spatial resolution. EEG electrodes are attached to the scalp, this means that electrical changes deep within the brain struggle to reach these electrodes and the signals that do reach them smear out to the point where you can’t really isolate the exact origin of this activity. Secondly, there is significant debate in the neuroscience community about what ‘activity’ in a certain region even means… For examples, see herehere and here – a more scientific explanation of some of the issues behind imaging experiments can be found here.

The main reason why scientists are sceptical of this type of analysis is that a number of the methods have not been published in peer-reviewed journals. There is some interesting published work (here for example) and some companies do publish some details of their methods, but scepticism is always necessary, even for published works.

Those in the scientific community who discuss these issues daily disagree about the best ways to analyse this type of data and what interpretations can be made. The idea that regions perform specific jobs and that measuring these areas can give us a score of complex human behaviour, such as how engaged or emotional we are, is therefore debatable.

Even so, the corporate world seems to be lapping these techniques up. Many campaigns are built on these data. Volvo had a large campaign at the end of 2013 claiming that their “car design [was] proven to be on a par with the most basic of human emotions”. Brain imaging is being used to understand what makes us enjoy a blockbuster film. It has also been used to see what effect celebrities have in a marketing campaign’s success.

It is hard to know how reliable this research really is as some of it has not been scientifically reported or scrutinised. There is a heavy amount of bias attached to these claims and if not properly reported, ‘neuromania’ can ensue. For now, be sceptical about what claims companies make about what your brain is telling you that you want. Even if a ‘neuromarketed’ magazine cover can increase sales.

Post by: Olly Freeman @ojfreeman

  • This post was altered on 30 March 2014. The original implied that all of this work was “based on methods which have not been published in peer-reviewed literature”. This is incorrect and reference has now been made to some peer-reviewed literature.

button print gry20 Neuromarketing: a whole lot of fluff?

About thebrainbank

The brain bank comprises a group of Manchester based scientists eager to enthuse and entertain with their scientific banter. To learn more about who we are see the our 'about' page. You can also find us on twitter @brainbankmanc or email us brainbankmanc@gmail.com.
This entry was posted in Oliver Freeman. Bookmark the permalink.

17 Responses to Neuromarketing: a whole lot of fluff?

  1. Michelle says:

    Great critique.

    As a neuroscientist working in market research I tend to cringe when I hear the word “neuromarketing”. I do believe neuroscientific methodologies can be used to study consumer reactions. But I do think that it has to be done properly and cautiously. And the results cannot be stretched.
    The problem I see is that far too often, people want to use these more advanced neuro techniques in settings where it hasn’t been proven out and then want the results to “read the mind”.
    With a properly designed study, one can definitely get insights. But using the wrong tool is like using a sledgehammer to put a nail in the wall. The nail goes in, but the results are a bunch of mush.

    • thebrainbank says:

      Hi Michelle,

      Thanks very much for your response. I completely agree with you that neuromarketing can have a place but caution must be taken. I really enjoyed your final phrase!

  2. Hi Dr. Freeman,
    As an academic also working commercially, I constantly find myself between the proverbial rock and a hard place. On the one hand, I am infuriated by the simplistic way in which neuroscience is being overpromised, underdelivered and oversimplified for business – and I see this daily in media (journalists are horribly simplistic species).

    However, as one that tries to bridge the gap between neuroscience and business, I find it equally annoying when these approaches are presented equally simplistic and being dumbed down. Please allow me to put a solid distance between our own approach and others such as Lindstrom and Pradeep, who in no way understand the subject matter nor have any interest in doing so.

    As you are mentioning Sands Research Inc, I feel compelled to reply. My own company, Neurons Inc, is in the process of merging with SRI. Please note the following:
    1) As opposed to your claim, we do publish our findings in peer-review journals, and we rely on peer-reviewed published works – we operate under an open book principle, not black box. This is, we acknowledge, not typical for the industry, but this is one of our competitive advantages. But we do think that it IS indeed possible to be a geeky scientists AND at the same time use these methods to provide true insights for the public and private sectors.

    2) You claim “there is much scepticism around the idea that EEG recordings can tell us when people are more engaged”. This is absolutely a false statement! THere is now more than 20 years of literature using EEG, fMRI and other modalities to demonstrate a relationship between brain responses and motivation/engagement. Your claim is completely out of sync with this literature, and I have collected a compendium of literature related to this and other features: https://www.dropbox.com/s/ejls7ubretkr4aa/NEUROMARKETING%20COMPENDIUM%202013.pdf

    3) Despite the argument that what “activity” really means in the literature (e.g. whether both the BOLD fMRI signal and the EEG signal seems more related to dendritic potentials rather than action potentials) is interesting, but less relevant when considering whether a response is functionally relevant. If we can repeatedly show – using a so-called functional localizer – that a particular response is highly predictive of, say, cognitive load, then it is perfectly legitimate to use this response as a metric of cognitive load. Of course, basic and advanced experimental design knowledge applies here, as it should everywhere.

    4) In the same vein, although many (most) companies seem to hold an overly simplistic neo-phrenological model about the brain and the mind, many of us are still able to carve out meaningful data and studies that can help companies make better choices. Just in the same way that I employ the same methods and insights to improve the condition of pathological gamblers and compulsive buyers.

    We’re in an age where it is becoming technologically, economically and scientifically feasible to move neuroscience tools out of the lab. Nobody are claiming that it is an easy job. But it is part of the future to get away from a noisy, claustrophobic MR scanner and highly artificial test environments. After all, are we not trying to understand everyday behaviours, for good or bad?

    • thebrainbank says:

      Hi Dr. Ramsøy,

      Thanks very much for your response. It is great to get a debate going about these kinds of topics on our blog. I agree with a number of your points and think you present a good argument but do not agree entirely.

      I agree that media reports of neuroscience are often overblown. That is a large reason as to why we write this blog, to try to promote a scientists viewpoint on things alongside the mainstream media. However, I do believe that the vast majority of neuromarketing companies go about things in a similar way. Marketing, by definition, is an industry designed to sell. When combined with science, this can quickly get into the realms of overegging the techniques and/or results.

      I do not know the industry well enough to comment on the practices of individual companies compared to others but I am pleased if your company bases your research on peer-reviewed methods. I think this is rare in the industry and should be upheld with openness encouraged.

      I think neuroscience techniques can be used to understand consumer behaviour but I am inclined to side with Michelle above. Giving a measure of ‘engagement’ or ‘motivation’ is a dangerous and haphazard terminology.

      My argument against EEG recordings telling us about inner most thoughts of the brain is two-fold. Firstly, we simply do not know enough about the neuroscience of decision making or human desire to be simplifying it into an electrical activity signature in the brain. Secondly, even if there were one region of the brain responsible for ‘engagement’ then statistical analyses of brain imaging/scanning techniques are immensely complicated – for casual readers please see http://blogs.discovermagazine.com/neuroskeptic/2013/05/03/brain-voodoo-goes-electric/#.UzSI8PmKWSo for an example of the kind of debates I am referring to.

      Thank you again Dr. Ramsøy for your response. I appreciate you openly discussing these issues. I’d like to reiterate that I think neuromarketing can have a place and a valid purpose, but it needs to be handled very cautiously indeed.

      • Dear Dr. Freeman,

        You are absolutely right that vast majority of companies have a completely black box approach and rarely if ever refer to or use actual peer-reviewed research (or contribute to this research).

        But I strongly disagree with your contention that these methods and our knowledge is not sufficiently sophisticated to allow us to talk about arousal (“engagement”) or motivation, even “cognitive load”. As you know, the literature on working memory and emotional arousal today is so vast that we can reproduce their responses reliably and predictably. EEG as a method can be used reliably to diagnose a vast array of neurological disorders.

        Let’s take your points here:
        “we simply do not know enough about the neuroscience of decision making or human desire to be simplifying it into an electrical activity signature in the brain”

        This is completely wrong, and as an academic you should know better than providing such claims that do not fit with the actual scientific literature. Today, we see a converging literature demonstrating that certain features of the brain are highly related to choices. For example, the role of the basal ganglia, although not completely understood, is so well understood that we know that it has a direct effect on choice. Works by, for example, Kent Berridge on wanting-liking, is worth mentioning here. Other features such as the prefrontal asymmetry and its role in approach-avoidance behaviour has more than two decades of academic research. Work demonstrating the relationship between dorsolateral prefrontal activation (both using fMRI, EEG or even leasion studies) and working memory (performance) is one of the well established processes in cognitive neuroscience.

        The point is that even though we do not have a perfect understanding of the exquisite details of the brain bases of emotions, memory, choice etc, then we can still crate models that are highly indicative and predictive of behaviour. If you can reliably show that the frontal theta engagement increases as a person is solving an increasingly difficult task (e.g., an n-back task), then hey, you very likely have an index of working memory load. If you find that the prefrontal asymmetry is highly predictive of actual approach or avoidance behaviour, then you can infer that the prefrontal asymmetry is indeed (in part) an index of approach-avoidance.

        A key point is what counts as evidence. Of course, in a perfect world, our understanding of the lingua franca of the brain would be 100%, and by definition we could apply reverse inference all the time. The problem with reverse inference is, after all, not a logical fallacy but is problematic since we cannot fulfil the premise that we have a 1:1 understanding of brain and function.

        So as academics, we still allow ourselves to engage in a weak kind of reverse inference – we say that certain brain responses and areas are likely to be more involved in certain functions than others. In commercial applications of neuroscience, we try to do the very same thing: we try to link responses to actual behaviour, to see if we can understand the underpinnings of real-world choices.

        You also write “even if there were one region of the brain responsible for ‘engagement’ then statistical analyses of brain imaging/scanning techniques are immensely complicated”

        Well yes and no. Nobody today in their full mind will claim that there is one region responsible for one function. I suggest you skip the neo-phrenological straw man. It’s all about probability, and whether you can predict actual real-life behaviour. I do not subscribe to a view where this “truth” will be found in the lab, and I have spent my share of hours running fMRI and EEG experiments to believe that the SPM preprocessing and GLM analyses are REALLY going to show the truth about the brain (fortunately, now my graduate students are doing that for me). That would be equally self-delusional as believing that we know the buy button of the brain.

        I do acknowledge that the vast majority of the commercial application of neuroscience is completely and utterly BS. But I firmly believe that it is still a necessary step – and technology now allows us – to bring neuroscience out of the lab and into the everyday situations we so desperately try to predict. Whenever we encounter a new solution to or criticism of preprocessing and analysis methods in fMRI and EEG, we try to accommodate to this. But just as these methods do typically not invalidate a whole field of research, it does not invalidate the commercial applications, either.

        What I see is a lot of pushback from people who simply cannot believe that neuroscience is capable of explaining or predicting behaviour, let alone outside the lab! I think nothing can be farther from the truth. Again and again, history has shown us that real progress is NOT happening inside the ivory tower lab environment, but out in the real world.

        • Michelle says:

          I wouldn’t say that neuroscience denies that neuroscience can predict behavior.

          Having done my PhD in a behavioral neuroscience lab, this just doesn’t equate.

          I will agree with you that academia shouldn’t be afraid of the applied use of neuroscience. It does further the field in ways that can be difficult if not impossible for a governmentally funded academic lab to accomplish.

          But I think that caution still must be taken on the applied side. While things are getting better, in our industry it is still the “Wild West” in that there are “neuro-cowboys”. And without a real structure on how to police it, bad science is being done.

          This doesn’t absolve academia though. They’ve been just as lax recently (http://blogs.scientificamerican.com/talking-back/2014/03/27/statistical-flaw-punctuates-brain-research-in-elite-journals/).

          But the real concern is that the cowboys are winning the media battle. People like headlines saying “scientists” can “read the mind” (http://gizmodo.com/scientists-can-reconstruct-faces-by-reading-your-mind-1553577287). Fox News wrote about this story titling it “We know what you’re thinking: Scientists find a way to read minds”. But when you look at the study you see that the participants were rigorously trained while being scanned on chosen faces. This study isn’t really about reading the mind. It’s about training a brain pattern and then recognizing that brain pattern later. And while that’s great and will of course be a big step in understanding how we recognize faces… it is not reading the mind and predicting or seeing what the person has seen. You can call this bad journalism, but it’s more than that, it’s over-reaching with your findings. And we are all doing it to our own detriment.

          I think it’s great that applied & industry neuroscientists can take the risks that academic neuroscientists can’t. But when I meet a business person that’s just entered into the world of “neuromarketing” and is using very high tech very complicated neuroscience methodologies and they justify their “neuromarketing expertise” by saying they’ve read some “neuromarketing” books (and yes, this has happened), I’m both offended as an academic neuroscientist and scared as an applied neuroscientist. I think that is just far too risky for any science to agree to or tolerate.

          It was recently stated in the Neuromarketing World Forum meeting a few weeks ago that we shouldn’t tolerate bad science. I think that was an awesome and important point that had to be made. But that means being able to take the criticism as well as dish it.

          • @Michelle: “I wouldn’t say that neuroscience denies that neuroscience can predict behavior”

            *** I’m not entirely sure how to respond to this. Neuroscience sure does not deny this…not my claim at all. Rather, my point is that neuroscience should be able to explain and predict everyday behaviours, and that this cannot be found by lab studies alone. Sorry if my point was vague.

            Also: “But the real concern is that the cowboys are winning the media battle”

            *** And my response is AMEN to that! This is part of my own battle. I firmly believe that good basic AND applied science can be done, but that cowboys and snake-oilers are stealing the show, which is not making our work any easier.

            Finally: “It was recently stated in the Neuromarketing World Forum meeting a few weeks ago that we shouldn’t tolerate bad science. I think that was an awesome and important point that had to be made. But that means being able to take the criticism as well as dish it.”

            *** I believe this refers to my own talk: “The Path to Validation: How to Continuously Bridge Science and Business of Neuromarketing”, and you can see the PDF here: https://www.dropbox.com/s/9szp4fyczy6evla/NMSBA-validity-2014-NEW.pdf

  3. Pingback: Mitch’s Rules, Perfect Testimonials, Dream Control, & More – Roger’s Picks | Neuromarketing

  4. Pingback: Who’s the bullshitter? Applied neuroscience and their critics | Neurons Inc

  5. thebrainbank says:

    Hi Dr. Ramsøy,

    I think we are close to agreeing on a fair few points. You make some good arguments but scattered with a couple I think are overblown.

    Your claim that “EEG as a method can be used reliably to diagnose a vast array of neurological disorders.” for example is not entirely true. When it is true (e.g. in the case of sleep disorders and epilepsy studies) it does not really validate your stance since these diagnostics are often more global and tend not to provide functional information concerning specific regions.

    I have altered the article slightly to include some reference to published works as my original implied that none of the work was peer-reviewed. I think the main point of my scepticism is also backed by your last comment which says “I do acknowledge that the vast majority of the commercial application of neuroscience is completely and utterly BS”.

    I certainly agree that neuroscience can be used ‘in the real world’ (as you put it) and that people should continue to push forward with increasing its applications. But continued scrutiny is paramount and commercial motifs should never be allowed to cloud good science.

    Michelle has a healthier view on the industry in my opinion. Like me she believes there is potential there but scepticism is necessary at every turn.

    It has been great talking with you both, and thanks for your interest in our blog.

  6. Hi thebrainbank

    I don’t think we do differ much at all. My skepticism is at least as profound whenever I hear words like “neuro-this” and “neuro-that,” and news items about the brain are IMO by default entirely wrong, simplistic, overblown and misleading.

    As a person with my feet planted solidly in both applied and basic research camps, I do believe that there is a qualitative difference between lab-based studies and “real world” studies. I intentionally set this up, as I have spent many years believing that our deepest problems about the mind and brain can be solved in the lab. I don’t hold that belief any longer.

    I understand that this is a very different perspective than the typical goal of your criticism. However, I find that the criticism you raise – as many other neuro-skeptics – becomes too eager to dumb down *all* aspects of applied neuroscience, and does not serve the field at all. Basically, your criticism, if taken literally, can easily be interpreted as going against the commercial application of neuroscience (well, at least beyond pharmacology, a highly commercial industry, too). This is my problem, and where I disagree.

  7. Michelle says:

    Thanks for the link to your talk Thomas!
    I did enjoy that talk. And I remember really appreciating you bringing up validity and my two favorite studies to bring up (the dead fish fMRI and the doll face).

    I was actually referring to Caroline’s talk where she had that one slide up, as a message to clients that simply said “don’t tolerate bad science”.

    I’m going to continue to be skeptical of all the methodologies, be they yours or my own or whom ever’s, because I think that’s the only way the field can move forward, for all of us. And that’s what I took home from your presentation.

    We need to not be afraid of people questioning our validity. We need to embrace that challenge and prove it out… over and over and over again.

    • Hi Michelle

      I’m sure we’re on the same page. Unfortunately I did miss the talk by Caroline /Wynett I assume), and her violin playing skills. Suffice to say that she did found and represent NeuroFocus, one of the biggest black box neuromarketing companies on this planet.

      Again: “We need to not be afraid of people questioning our validity. We need to embrace that challenge and prove it out… over and over and over again”
      I completely agree!

      Thanks all for a great discussion and a great blog!

  8. Steve Genco says:

    Thanks for the spirited discussion. Since my book, Neuromarketing for Dummies, apparently has been used as a prop to support the notion of neuromarketing as “fluff”, I just wanted to point out to anyone who’s interested that they can go to the book’s website (here: http://bit.ly/1dObt9G) and download a 70-page “References and Notes” document that contains citations for the hundreds of peer-reviewed studies that underlie every statement in the book. (Unfortunately, they’re not in the book because the “dummies” format doesn’t allow footnotes.) The idea that neuromarketing is not based on real science is an old and tired cliche that needs to be put to rest.

    In any field where there is an asymmetry of information between buyers and sellers, snake oil salesmen will appear (think of used car salesmen vs. buyers before the Internet leveled the information playing field). As better understanding of what neuromarketing really is (and what it’s good for) continues to spread, the neuro-cowboys will continue to vanish from the scene. Most of the more egregious offenders have already left the field, moving on to new pursuits where information asymmetry and magical thinking provide a more fertile ground for their style of commerce.

    Good scientists like Thomas and Michelle (assuming Michelle is the same Michelle I think she is :) are turning neuromarketing into a respectable field based on solid, transparent science, not secret black boxes. Potential critics need to get current.

    To that end, I hope Dr. Freeman and the brainbankers will take the time to actually read my book, rather than just use it as a prop. I believe they will be pleasantly surprised.

    • Michelle says:

      Yes, Steve, I think I am who you think I am… or is it I think therefor I am?
      And thank you for the good scientist compliment!

      I totally agree. With presentations like yours and Thomas’ really aiming to educate consumers of applied neuroscience, the field is evolving. That’s a great thing.

      But without our critics, we would get lazy. I appreciate and support all that critics like @neuroskeptic, @drisis, @rebeccaskloot, @scicurious, this brain bank and like do to better science and keep us all on our toes.

    • thebrainbank says:

      Hi Dr. Genco,

      Thanks for your reply. I don’t make any reference to whether your book supports or denies the notion that neuromarketing may be ‘fluff’ and so I think your worry is misplaced. Your book is intentionally shown so that those who are interested can go and read it to learn more. I hope that people do.

      Following this discussion with the three of you, I hold two predominant views. The first is that all science, not matter what, should be continually challenged and scrutinised. This is how good science is upheld and how we progress and eliminate the ‘cowboys’ (as you call them). Michelle has a healthy appreciation of this and welcomes the challenge to preserve legitimacy in this field. Thomas and yourself seem to dislike the challenge.

      The second is that still, no matter how many peer-reviewed papers there might be, many, many commercial companies do not say what their methods are and what exactly they are measuring. It is this that I am challenging more than anything, not the published work.

      I really like Michelle’s comment – “We need to not be afraid of people questioning our validity. We need to embrace that challenge and prove it out… over and over and over again.”

      • Steve Genco says:

        I have no problem with challenges or scrutiny. In fact, I was just challenging your challenges, which I thought were rather weak arguments that needed to be challenged! I think it’s clear that we all agree that there is a lot of improvement to be done in our little field. I simply believe you have painted with too broad a brush and repeated some criticisms that were incorrect (like the point about no peer-reviewed research), and some that were misplaced (like the criticisms of fMRI, which are directed at neuroscience methods in general, not neuromarketing, where fMRI is very much a minority method).

        All the old “cowboy” arguments and “black box” arguments still have some currency, but the point I wanted to make was that the cowboys are disappearing and the black boxes are opening up (Thomas is indeed a trailblazer in this regard). One of the biggest differences between academic and commercial research is that the latter is subject to the winnowing effect of demand. And what we have seen in recent years is a significant self-correction in the market as buyers stop responding to the exaggerated promises of the cowboys, start asking smarter questions, and migrate their research dollars to more responsible and scientifically transparent vendors. As I mentioned above, I think this trend is an inevitable result of leveling the information asymmetries that used to dominate the field. In a commercial context, it will ultimately be buyers who change the behavior of vendors, not critics like you or me. And it is my belief that they will force it to change in the direction of greater business value and accountability, not in the direction of more “fluff”.

        I’ll accept your explanation that you posted a picture of my book for informational purposes only. But perhaps you can acknowledge that in the midst of an article entitled “Neuromarketing: A Whole Lot of Fluff?,” the picture of the book, with no mention or explanation in the body of the article, could be misinterpreted as an illustration of the “fluff” referred to in the title. Since one of the main purposes of Neuromarketing for Dummies (despite its cheeky title) was to provide an authoritative refutation of that misperception, I hope you can understand why I felt the need to comment.

Leave a Reply

Your email address will not be published. Required fields are marked *

* Copy This Password *

* Type Or Paste Password Here *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>