Very interesting video, and I really like the practical demonstration of how the same "dataset" can mean something completely different depending on what you expect from it.
Its a nice way to show that even though you might consider yourself "completely unbiased" and taking the scientific approach, you might get the wrong result anyways.
FYI Simon Singh has lost the last couple years of his life being sued by the UK's chiropractic association for telling the truth - ie, it's a load of codswallop.
Please, if you are not satisfied with the present
state of things, sign it. If you are unaware,
perhaps have a quick read, and see what you think?
They have a report here,
but do have a read around the net too. Simon has
even come up with a pyramid scheme (for a good
cause) in order to get 100 000 petitions. I have
included his email below (verbatim). Please
give this some consideration, thanks.
==============================
Dear Friends,
I’ve had an idea – an unusual idea, but I think it might just
work.
As you know, England’s chilling libel laws need to be reformed. One
way to help achieve this is for 100,000 people to sign the petition
for libel reform before the political parties write their manifestos
for the election. We have 17,000 signatures, but we really need
100,000, and we need your help to get there.
My idea
My idea is simple: if everyone who has already signed up persuades
just one more person each week to sign the petition then we will reach
our goal within a month!
One person per week is all we need, but please spread the word as
much as you can. In fact, if you persuade 10 people to sign up then
email me (simon <<at>> simonsingh.net) and I promise to thank you by printing your name in my next book
… which I will start writing as soon as I have put my own libel case
behind me. I cannot say when this will be, but it is a very real
promise. My only caveat is that I will limit this to the first
thousand people who recruit ten supporters.
When persuading your friends remember to tell them:
(a) English libel laws have been condemned by the UN Human Rights
Committee.
(b) These laws gag scientists, bloggers and journalists who want to
discuss matters of genuine public interest (and public health!).
(c) Our laws give rise to libel tourism, whereby the rich and the
powerful (Saudi billionaires, Russian oligarchs and overseas
corporations) come to London to sue writers because English libel laws
are so hostile to responsible journalism. (In fact, it is exactly
because English libel laws have this global impact that we welcome
signatories to the petition from around the world.)
(d) Vested interests can use their resources to bully and intimidate
those who seek to question them. The cost of a libel trial in England
is 100 times more expensive than the European average and typically
runs to over £1 million.
(e) Three separate ongoing libel cases involve myself and two medical
researchers raising concerns about three medical treatments. We face
losing £1 million each. In future, why would anyone else raise similar
concerns? If these health matters are not reported, then the public is
put at risk.
My experience has been sobering. I’ve had to spend £100,000 to
defend my writing and have put my life on hold for almost two years.
However, the prospect of reforming our libel laws keeps me cheerful.
Thanks so much for your support. We’ve only got one shot at this
– so I hope you can persuade 1 (or maybe 10) friends, family and
colleagues to sign.
Massive thanks,
Simon
The Libel Reform Campaign is a coalition of English PEN, Index on
Censorship and Sense About Science.
So far, 188 MPs have signed our Parliamentary Early Day Motion
calling for libel reform and the Justice Secretary Jack Straw has
formed a working party that the Libel Reform Coalition is represented
on.
There's a great book by Joseph O'Connor called "Introducing NLP: Psychological Skills for Understanding and Influencing People "
Which deals a great bit with "framing", as well as other concepts that really open up the understanding of human communication and perception. It was initially intended for psychologists/psychatrists/etc.. (I believe, based on the numerous references to using this in "sessions"), but it is really appropriate for just about every person who wants to understand human communication a little better.
This is a great subject. Framing is essentially manipulating our bayesian priors. I am interested in the variation of this thought line that we can consciously vary our prior. I can do this with, for instance, a Necker cube. I can deliberately switch my anticipated model and have the rest of my experience line up with this switch. Although curiously, it has a rather slow switching time.
It's not just the fact that he's compelling us to hear something, he's also pointing out what words we're going to hear on the screen. If he just promised we'd hear something and then not show them as it's playing, a lot of people would hear much less meaningful stuff still.
He specifically talked about the fact that when the brain has a "strong enough" bias, that's when it ends up telling you that its false error correction data is indeed fact.
Amazing how the mind locks on to the given pattern. Maybe that's related to the psychological fact that people tend to overestimate the control they have over the physical world. You could also call it optimism.
But, could the demonstrator have given any other lines and still caused the audience to observe them in the song? Those lines were deliberately designed from careful study of the song.
Yes, but I think the interesting part of this (and similar) demonstrations is that it shows illusions that cross more than one sense. It's not just tricking your eyes; it's using your eyes to trick your hearing.
It would be neat to find reverse examples -- where sound changes visual perception. (Perhaps the typical dramatic false-punch is one; the added sound of contact makes an otherwise implausibly short or off-center strike seem much more realistic.)
Its a nice way to show that even though you might consider yourself "completely unbiased" and taking the scientific approach, you might get the wrong result anyways.