Forecasting voter behavior from inside the brain

29/07/2014 – Neuroimaging gives researchers a look under the hood, but will the technology help campaigns?

Forecasting voter behavior from inside the brain

One of the most influential political science models of citizen participation treats voters as if they were empty buckets ready to have their heads filled by the ideas of party leaders, media elites and campaigns.

Despite all its predictive power, we know that the model is wrong. The bucket is not empty. Voters have brains, and they use them.

I don’t mean this in some metaphorical sense like “that focus group member really spoke from the heart.” Based on all that we understand from modern neuroscience, people use their brains when they make political decisions.

While we have known this for a long time, the way we typically try to get at the content of those brains is by asking people about their views directly. The problem is that people often don’t know what their own brains are up to.

About 15 years ago I had the idea to use this new tool of functional brain imaging to look at the differences between people who knew a lot about politics and those who didn’t really think about it. What we found was that people who were very interested in politics activated a set of brain regions known as the Default Mode Network.

Basically, this part of your brain is quite active when you’re isolated in a brain scanner and doing “nothing.” When you start doing something like math, conjugating verbs in a new language or imagining a rotating object, this brain network’s activity seems to decrease.

But the Default Mode Network actually kicks into high gear when you are doing something social suggesting that, for the politically active, politics is a social task. For those who rarely think about politics, the network deactivates, just as it would if they were doing some technical task like math or verb conjugation. Later on my colleagues and I tried to identify differences in how Republicans and Democrats made risky decisions.

What we found was that the gambling behavior was indistinguishable, but we spotted disparities in how particular brain regions responded to those risks. In fact, a model based on activity in just two brain regions, the amygdala and the insula, correctly classified Republicans and Democrats 82 percent of the time. These results coincide with other work showing differences in the size of the same regions that corresponds with whether young adults in Britain are liberal or conservative.

Interpreting these results requires caution. We know from studies of twins that only around half of our political ideology appears to be heritable, and that affiliation with a particular political party isn’t heritable. We’re not hardwired to be Republicans or Democrats. In fact, there’s good evidence suggesting that we’re hardwired not to be hardwired: That we’re built for mental flexibility. Rather than trying to solve the problem of deciphering what a particular pattern of brain activity means, what if we just tried to use brain-imaging data directly to predict the behavior we’re interested in?

Researchers looking at the efficacy of public health campaigns have started to do exactly that. They’ve found that particular patterns of brain activity can predict whether a person exposed to persuasive advertising about the need to use sunscreen with actually don the lotion when catching some rays a week later. The measured effect from the participants’ brain activity was distinct from the result that came simply from asking people about their view of the increased sunscreen use ads.

Later studies showed that smoking cessation could be predicted months in advance by measuring how specific brain regions responded to tailored messages.

Follow-up research supplemented standard focus group techniques by brain imaging individuals as they watched ads that were part of a 1-800-QUIT-NOW anti-smoking campaign. The focus group’s neural responses to each of the ads predicted subsequent call volume when the campaigns went live. Again, the brain predictions were distinct from and in addition to the predictive power gained by traditional methods of asking the focus group for their opinions.

Campaigns with large budgets will be able to arrange such neural focus groups using expensive equipment for national-level elections and likely get insights well beyond what the existing surveys and normal focus groups can provide.

For those on a tighter budget, measuring electrical signals from the surface of the brain (EEG), measuring visual attention (eye-tracking), measuring physiological response (heart rate variability and skin conductance), and measuring response times to questions (using CATI) can provide insights beyond traditional tools.

Fonte: campaignsandelections.com | Autore: Darren Schreiber, researches neuropolitics at the University of Exeter in the United Kingdom.

Annunci

One thought on “Forecasting voter behavior from inside the brain

Rispondi

Inserisci i tuoi dati qui sotto o clicca su un'icona per effettuare l'accesso:

Logo WordPress.com

Stai commentando usando il tuo account WordPress.com. Chiudi sessione / Modifica )

Foto Twitter

Stai commentando usando il tuo account Twitter. Chiudi sessione / Modifica )

Foto di Facebook

Stai commentando usando il tuo account Facebook. Chiudi sessione / Modifica )

Google+ photo

Stai commentando usando il tuo account Google+. Chiudi sessione / Modifica )

Connessione a %s...

%d blogger hanno fatto clic su Mi Piace per questo: