Magic trick transforms conservatives into liberals


‘Choice blindness’ can induce voters to reverse their party loyalty.

When US presidential candidate Mitt Romney said last year that he was not even going to try to reach 47% of the US electorate, and that he would focus on the 5–10% thought to be floating voters, he was articulating a commonly held opinion: that most voters are locked in to their ideological party loyalty.

But Lars Hall, a cognitive scientist at Lund University in Sweden, knew better. “His calculation, only zeroing in on 10% of voters, is a risky proposition,” he says. When Hall and his colleagues tested the rigidity of people’s political attitudes and voting intentions during Sweden’s 2010 general election, they discovered that loyalty was malleable: nearly half of all voters were open to changing their minds. The team's work is published today in PLoS ONE1.

Hall’s group polled 162 voters on the streets of Malmö and Lund during the final weeks of the election campaign, asking them which of two opposing political coalitions — conservative or social democrat/green — they intended to vote for, and how strongly they felt about their decision. The researchers also asked voters to rate where they stood on 12 political wedge issues, including tax rates and nuclear power.

The person conducting the experiment secretly filled in an identical survey with the reverse of the voter's answers, and used sleight-of-hand to exchange the answer sheets, placing the voter in the opposite political camp (see video above). The researcher invited the voter to give reasons for their manipulated opinions, then summarized their score to give a probable political affiliation and asked again who they intended to vote for.

No more than 22% of the manipulated answers were detected, and 92% of the study participants accepted the manipulated summary score as their own. This did not surprise Hall, who has previously demonstrated similar reversal effects, known as choice blindness, in people’s aesthetic preferences2 and moral attitudes3.

What is interesting about the latest study is that, on the basis of the manipulated score, 10% of the subjects switched their voting intentions, from right to left wing or vice versa. Another 19% changed from firm support of their preferred coalition to undecided. A further 18% had been undecided before the survey, indicating that as many as 47% of the electorate were open to changing their minds, in sharp contrast to the 10% of voters identified as undecided in Swedish polls at the time.

“It’s a dramatic demonstration of the potential flexibility that is there,” says Hall. “Unfortunately I don’t know how to tap into that flexibility without the magic trick. If I did I wouldn’t be talking to you. I’d be selling my secret to Hillary Clinton or [Republican New Jersey governor] Chris Christie. Or both.” (Clinton and Christie are seen by many as the front-runners for the 2016 US presidential election.)

Written By: Brian Owens
continue to source article at


  1. Personally, in the UK I think there’s nothing to choose between the parties.

    At the moment the caste system is being debated in Parliament – yes, that’s what I said, THE CASTE SYSTEM – and the House of Lords has voted it down – hurrah!

    But now it has to be debated in the House of Commons, where, I predict, like Sharia Law, it will be touch and go. Why? Because the party members won’t run the risk of losing votes!


    Short-termism is de rigueur.

  2. In reply to #1 by Stafford Gordon:

    At the moment the caste system is being debated in Parliament

    As I recall England’s always had a caste system. And only the Lordly caste make it into the House of Lords (with a few token exceptions in recent years).

  3. Without wishing to appear grumpy …

    I fail to see what is remarkable about this study. It seems to me utterly banal.

    Publishers have known for centuries, and politicians skilled in public square rhetoric probably longer, that to change someone’s mind simply requires that we short-circuit the audiences’ critical thinking. This is traditionally done in two ways:

    • An appeal to emotion

    • Group Dynamics

    An appeal to emotion can be simply shouting or shaking a fist (a big bold headline, “today’s top story”, and so on). It may be an old trick, but Fox TV presenters still use this tactic daily. The other end of the scale is circumstantial evidence – an inference is used mixed with emotive language, or a driver of emotion (today, usually a picture) in order to push the audience to a conclusion based on deduction – i.e. without reference to all the facts or alternative conclusions which may also be valid.

    Group dynamics are usually started by in-group dynamics – you’re one of us, join us! This is typically followed by a threat to ensure solidarity, because political parties tend to be two-sided, like a coin, whereas we human beings are multi-faceted. This becomes a major issue when the politics of the party are dogmatic. Parties of the extreme left and right therefore suffer frequent bitter schisms – as was so brilliantly satirised in Life of Brian. But I digress.

    At the other end of the spectrum is the use of spurious statistics – I once read that a journalist in a London newspaper, was told that the ‘management’ wanted a story that highlighted some point of view. Stumped for a way to work up a story, he had a brainwave; he simply stood up and shouted for a show of hands on the issue. Naturally, the journalists (an unrepresentative sample by number and several demographics) being largely aware of their employer’s spin requirements answered as required. The printed story began: A recent survey has revealed … Of course, once the papers started to get away with this sort of chicanery, they spotted that even the show of hands is actually a redundant step …

    All of these things are the Original Story’s sticky pad.

    I suppose the story highlights just how far we still to go to get critical thinking into the heads and habits of the average citizen. Like we didn’t know …

    But it ain’t news.


  4. You know what would be super interesting? If candidates in the US political system had to forgo their party affiliation until AFTER election. All potential candidates would have to advertise their opinions and answers to questions without the ideological backing of a giant political machine.

    In other words, get elected on your own merits and then reveal your party affiliation. Perhaps pie in the sky, but interesting to me.

  5. The voter is motivated by a desire not to look like a clumsy oaf who cannot even fill in a form accurately. This desire to look good is the pollster’s eyes is why I think polls tend to show far higher percentage of Christians than you would guess by church attendance. In a related phenomenon, soldiers going door find high support for the occupying army.

    I have often found it fun to guess the political views of pollsters, just from subtle clues when they ask the questions. I was never able to do this for Nick Nanos, the famous pollster, even in free-form conversation.

  6. This is a very depressing read, but then I suppose it’s not at all surprising. Most people don’t actually have any real opinions.

  7. Sorry, am I missing something? People were handed an answer sheet showing the precise inverse of the actual answers they’d just given, and were then asked to give reasons for each answer, and only 22% of them could remember their original answer long enough to say “That wasn’t my answer”? Seriously? I suspect this item is not giving us a complete description of the process, since that seems an extremely unlikely outcome.

  8. In reply to #7 by Jonathan Dore:

    only 22% of them could remember their original answer

    Research Article:

    At an individual level, 47% of the participants did not correct any answers, while 53% corrected between 1–4 answers. For all answers classified as corrected, the participants indicated that they had misread the question, or marked the wrong end of the scale. Only a single participant expressed any suspicion that we had manipulated her profile.

  9. In reply to #8 by Peter Grant:

    In reply to #7 by Jonathan Dore:

    only 22% of them could remember their original answer

    Research Article:

    At an individual level, 47% of the participants did not correct any answers, while 53% corrected between 1–4 answers. For all answers classified as corrected, the participants indicated that t…

    Thank you Peter, that really is extraordinary. As you say, I guess these are questions about which most respondents have never thought enough to have their own view. Gobsmacking.

Leave a Reply