Donald Trump’s victory shocked political pundits, pollsters, and “experts” of all kinds, who almost unanimously predicted that Trump would get clobbered by Hillary Clinton. Perhaps we shouldn’t be so surprised by their failure; none of them were basing their opinions on the real thing doing the voting and making the decisions—the human brain. If they had been looking at all the science, which is just a cool term for “that which we can measure,” rather than just poll numbers, they may have gotten it right.
According to neuroscientist and VP of Research at SPARK Neuro, Ryan McGarry, PhD, Trump’s win could have been predicted based on the results of a series of political neuroscience studies the company conducted last year. Spark Neuro is a startup from Bethesda, Maryland that uses the tools of neuroscience to measure audience engagement, with clients that range from major brands and TV networks to the U.S. military. Rather than relying only on self-report measures like focus groups and surveys, which are subjective in nature, they go straight to the source by measuring nervous system activity.
In 2016, SPARK Neuro spent six months probing voters’ neural and physiological responses to political ads and debate footage by measuring brain activity (EEG), skin conductance (GSR), facial expressions, and eye movements. The company claims that their proprietary algorithms crank through hundreds of data points per second, converting all this messy biological data into easy-to-understand scores of attentional and emotional engagement. The result is two simple numbers, one for attention and the other for emotion, which can range from 0 to 10 (or -10)
SPARK believes that these two numbers are the most important measure of effectiveness for any message. Engaging attention is critical since, as they say, “Without attention, it’s as if no one is there to hear it at all.” On the attention scale, a score of 1 means the subject is tuned out, a 5 indicates that they are moderately engaged, and a 10 means they are hyper-focused. For the emotion scale, a 0 denotes feeling neutral, positive numbers correlate with positive emotions (10 is sheer joy), and negative numbers up to -10 indicate emotions like anger, fear, disgust, or sadness. McGarry stresses that eliciting negative emotion isn’t always a bad thing for politicians—sometimes the goal is to make someone feel angry or fearful.
In addition to SPARK’s neuroscience studies that included potential voters from across the political spectrum watching 40 minutes of political video content, voter brain data was also gathered from two battleground states—Florida and Pennsylvania—just two weeks prior to voting day.
McGarry said, “In the first study, from the EEG emotion data, we were able to tell who participants would vote for, and in what order. In the second study we could see that our emotion data often showed that a participant liked something Trump said, even though they claimed not to.” When McGarry confronted participants about this discrepancy, making them aware of what their emotion and attention data was showing, they would often admit that they in fact did agree with some of Trump’s points. Attention was also a key indicator of who people were really listening to—and the data showed that most just weren’t paying attention to Hillary Clinton.
SPARK Neuro’s neuroscience results show that what people say about how they feel and how they intend to vote doesn’t always match up with what their brain and body physiology is showing. This explains why the polls were so wrong. The company’s CEO, Spencer Gerrol, says that it’s more complicated than simply saying participants were lying about the candidate they intended to vote for. To help us understand exactly what happened, Gerrol has identified three distinct types of voters that were not accurately accounted for by polling methods.
The Dishonest Decided
These hidden Trump voters were aware that they weren’t giving the full story to the experimenters, but they weren’t trying to be deceptive or malicious. The deceit was a result of “response bias,” or more specifically, “social desirability bias.” Gerrol says that some participants assumed the experimenter was liberal, and adjusted their answer to resemble something they thought the researcher would want to hear. This is a way to avoid the discomfort of an awkward social situation, and is something that happens frequently with controversial topics, like politics. “No one wants to be judged,” Gerrol says.
One woman told the experimenter, “I know some of the things Trump says are out there. I don’t necessarily agree with the stuff about Mexicans.” However, after she was showed graphs representing her strong emotional response to anti-immigration comments, she fessed up. “I am pretty concerned about the jobs we are losing. Regular people are just getting squeezed and these people are coming over illegally.” Later she admitted that she most definitely planned to vote for Trump all along.
The Subconscious Decided
Many of the participants in the swing states put themselves firmly in the undecided camp. McGarry thinks these people weren’t lying—they just legitimately hadn’t figured out whom they wanted to vote for yet, at least consciously. Astonishingly, the brain data showed that some of their minds were more made up than they thought. Their emotions were guiding them towards Trump on a visceral level, although they hadn’t fully admitted that to themselves yet. These people weren’t being dishonest, they just didn’t think much about how their emotions could influence or override their logical thinking.
The truth of the matter is that every decision you will ever make is influenced by emotions. Polls alone only take verbal responses at face value. But measuring emotions can reveal the subconscious feelings that can drive behaviors and decisions.
The Decidedly Apathetic
These are the people who intend to vote but ultimately aren’t compelled enough to get out and do it. For example, one participant was a twenty something male whose verbal reports and brain data showed that Bernie Sanders and his messages strongly resonated with him. Whenever Sanders spoke about income inequality, his attention levels jumped to about a 7, while they never went above 4 when Clinton spoke. When asked about whether he’d vote for Clinton if Sanders didn’t make it past the primary, he said, “I prefer Sanders, but I will vote democrat no matter what.”
While he intended to vote Clinton, he ultimately lacked the passion needed to motivate him to actually get out and vote. Suspecting that this may have happened based on their brain data, SPARK Neuro contacted the man after the election was over. Sure enough, he never made it to the voting booth. “I’ll be honest. I just decided my vote didn’t matter anyway,” the man said.
The results of the SPARK Neuro studies clearly demonstrate the importance of objective measures. Politics is extremely messy and so is human behavior. Data taken directly from the brain and body regarding states of emotion and attention can tell us things about the decision-maker that polling and questionnaires simply cannot.
If think tanks and media outlets across the nation had been gathering data using the tools of psychology and neuroscience, the American people may have had a better understanding of how likely a Trump victory was in advance. However, this warning might not have made a difference, as many of the brain processes tapped into by Donald Trump were largely automatic, much like a reflex. Whether or not that is true, it is obvious that pollsters and pundits should be looking at the brain too, rather than relying so much on the self-report methods used in focus groups and surveys. By doing so, we will all have a much clearer understanding of where society stands on political candidates and important issues.
Bobby Azarian is a freelance science writer with a PhD in neuroscience. His research has been published in journals such as Cognition & Emotion and Human Brain Mapping, and he has written for The Atlantic, The New York Times, BBC, Scientific American, Psychology Today, Slate, The Huffington Post, and others. He is the creator of the blog Science Is Sexy. Follow him @BobbyAzarian.