presents


The Political Brain

How the Final Presidential Debate
Played on the Subconscious Minds of Voters

0min1min2min3min4min5min3. Hover to Explore Audience Subconscious ResponseSupreme CourtNotable QuotesCandidate AttacksKey Moments & ThemesVideo Transcript2. Click to Select Demographic GroupAbove 30Under 30AmericanNon-AmericanFemalesMalesSupreme CourtImmigrationEconomyFitness to be presidentForeign HotspotsNational DebtNotable Quotes15min30min45min60min75min90min"Bad hombres""Nobody has more respect,for women than I do"Rigged election"Such a nasty woman"1. Drag to Select Debate Moment

Why We Made this Tool

Polls now show Hillary Clinton with a commanding lead, but voters should approach these numbers with caution. It's been a campaign soaked in emotion and irrationality – misogyny, scandal, racism, xenophobia, arson, misinformation, and violence. Under these conditions, it’s hazardous to fully trust the stated response methodology of political polling to get a read on voter intention. What we say we’ll do, and what we actually do, are often very different.

So, to peel back this conscious layer and reveal the subconscious workings of voters and citizens, Brainsights deployed its audience brain measurement platform to measure the subconscious brain activity of 60 citizens as they watched the final Presidential debate live.

Using electroencephalography (EEG) – devices that use small electrodes to measure brain waves – Brainsights is able to pick up on the tiny signals in the brain that fire off every millisecond without our knowing, and which drive our decision-making. Specifically, Brainsights’ technology measures audience Attention, emotional Connection and Encoding to memory – quite literally, what we attune to, what we identify or bond with, and what we find value enough to store away in our memory. Together, these measures combine for a read on the engagement or persuasiveness of specific stimuli.

The technology platform syncs this brain data at the millisecond level to the visual stimuli each viewer is processing – in this instance, the third and final Presidential debate. The Brainsights team coded every second of the debate, tagging each moment with which candidate was on screen, who was speaking, and about what topics. Where there was a ‘zinger’ or a candidate attack, we coded those, too.

All of this was linked to the profile data of participants. We’ve broken this down into six groups you can compare – Males and Females, for which there were equal numbers; under and over 30 years old (another 50/50 split) and Americans and non-Americans, where there were 20 of the former, and 40 of the latter (the event happened in Toronto). But while these numbers may seem small, for neuroscience-based studies, they’re actually huge – by comparison, most PhD theses using EEG have 15-20 participant samples.

How to read this visualization

The time series charts have two default lines. The horizontal line is the average score for everyone across the entire debate. The single line is a combined score – moment-to-moment - of audience Attention, Connection and Encoding, which ebbs and flows in response to the stimuli on screen. Where this rises above the average, it means the audience is engaged or persuaded by what they’re seeing and hearing. Where you see it dip below, it’s the opposite – voters don’t care much about what the candidate is saying. You can select specific demographic groups to map against these average scores*.

Select specific moments to dive into more detail further below, including how the audience responds to specific quotes from the candidates, or as they discuss different topics. Colour shading in the chart refers to who’s speaking – Blue is Hillary; Red is Trump; Yellow is the moderator Chris Wallace. And where you see White, that’s where the candidates and the moderator are speaking over one another and it was difficult for us to determine who was saying what**.

Finally, if you have any ideas as to how we can make this a better tool, tell us. Let us know if we’ve missed anything or any moments of interest. We’re eager to make this as useful a tool as possible for you, so you can better understand how the subconscious mind responds to political communications – what’s persuasive, what’s engaging – and thus make a more informed decision come Election Day.

*Note: Positive/Negative valence is not intrinsic to this data. In other words, because a response is above or below mean doesn't mean that it's liked/disliked. However, the interaction between Attention, Connection and Encoding metrics - which can be found in the details - can help to tease this out: Where Connection rises and Attention drops, people are withdrawing from the stimuli on screen, and considering the implications of what's been said/shown. In the past, this has indicated confusion, disdain, and other negative emotions. For an example of this phenomena, look at the details for Trump's "Nobody has more respect for women" claim for females. Connection rises more than Attention, and stays higher than Attention even as overall engagement drops.

**You'll see we pulled the official transcript from the closed captioning of the debate. We wanted to give you full context of what was happening. However, as with closed captioning, the syncing of the text to what was said can be off by 1 or 2 seconds. Also - here's a link to the debate file we screened.

Data & Analysis by Brainsights | Visualization & Design by Shirley Wu