21 comments

[ 3.0 ms ] story [ 36.4 ms ] thread
The Institute for Propaganda Analysis was set up almost 100 years ago to teach the American public about propaganda.

https://en.wikipedia.org/wiki/Institute_for_Propaganda_Analy...

It was abolished because the Americans wanted to use propaganda themselves.

- Aldous Huxley, Brave New World Revisited

"The Institute for Propaganda Analysis was founded in 1937, when Nazi propaganda was at its noisiest and most effective. Under its auspices analyses of non-rational propaganda were made and several texts for the instruction of high school and university students were prepared. Then came the war — a total war on all the fronts, the mental no less than the physical. With all the Allied governments engaging in "psychological warfare," an insistence upon the desirability of analyzing propaganda seemed a bit tactless."

---

Today we see "nudging" being used to manipulate. All our governments (and corporations and charities) seek to control our behaviour against our conscious wills. They won't easily give up their toolbox of tricks.

It's revealing that the article frames the discussion around "persuading" which a method how disinformaiton is spread. another kind of manipulation. Now, I don't think the article actually suggests that we should persuade people to believe what we want them to believe as it ends by recommending critical thinking, but it falls short of the mark. It fails to see how the methods and routes of propaganda are wrapped up in it.

What the Institute for Propaganda Analysis almost 100 years ago recommended was different; it was to first educate people on the methods and means that we get persuaded and manipulated and then give them tools to think around this to think critically. It respected people and viewed them as intelligent, capable, thinking rational people. Today, it appears that political institutions and many educated people (as seen in hundreds of HN comments) think "the masses" are malleable, thoughtless and brainless sub humans. If we persist in viewing our neighbour as a one of these "ignorant masses" we a) are more likely to consent to their degradation and b) more likely to be viewed that way by others and be manipulated and persuaded with the methods and ways they approve.

This is a complicated topic, but it seems especially difficult because there are cases where something has been clearly falsified, and yet people persist in believing it, and and even worse, don't seem to adjust their approach to the source of the misinformation as a result.

This leads to three interpretations, all of which are fraught for different reasons: first, that the person is the victim of a con or brainwashing operation; second, that they are strongly motivated in their assertion of beliefs, such that the belief is not actually an assertion of truth, but what they want to be true; or third, that they do in fact have some cognitive deficit.

The last is problematic for the reasons you mention, but so are the other two. The first is problematic because it suggests some vulnerability of the person, even if it's one that everyone shares. Everyone can be conned, but victims often feel shame about it, simply about being a victim. This leads to problematic situations where you can be labeled as insensitive for even suggesting that people were conned, as if you're shaming the victims rather than the perpetrator. The last is problematic maybe for obvious reasons, that you are left in a catch 22 of either accepting disingenuous statements of belief that are really hiding bigotry or other problems, or attack the underlying motives and get accused of name calling.

I'm not saying there's no way out, but it's complicated and involves a lot of mass psychology.

The fourth interpretation is that the something has not been clearly verified. The misinformation problem could lie with the accuser as well.
The 5th problem is that a broad general understanding is necessary to identify when clearly verified information, is still misinformation.

The first thing the Q handlers told them all was "to do your own research" and then led them to research exclusively their own little nonsense part of the Internet. Thru this entire process people are gas lot and manipulated in about every possible way to get the person to accept the sources for their "research discoveries" as actual sources for legitimate information.

It tricks them into thinking they are smart. It's hard to lie to someone that knows the truth - it's why smart people seem so arrogant. If they have concluded something is true, its more likely is than isn't.

Smart people that know stuff also, place the burden of proof on anyone that challenges their beliefs or understandings - that's difficult to combat with knowledge. Trumpers are like tricked into that misunderstanding of their perception of reality. All the echo chambers echoed, so it must be true, and so people then defend to the bitter endz false beliefs, accepted bc of fake information, deliberately placed where they could "independently discover it"

That's really to combat bc the information isn't real. Hard to counter a belief about something that isn't true but is the stated reason for other terrible ideas.

My point being that this same process can be true for the accusers as well, with the only difference being that the accusers have institutional support.

Is the primary criterion that a disputed fact is true whether the fact has institutional support? In order to "combat misinformation", are we going to select a bunch of oracles in institutions and have them decide what the truth is? (These are rhetorical questions, btw.)

I had to take a number of "communication" courses to get my undergraduate degree. One of the tactics covered during that course was to avoid relying on facts to convince people of an argument, as they are unlikely to be persuaded by them.
That's excellent, I'd never heard of it. I wonder how keeping the institute going would have changed the war, really.
(comment deleted)
Answer to the titular question:

  The study, which put 2,190 believers into conversation with the GPT-4 model that underpins the chatbot, reduced the self-reported strength of their beliefs by 20% after three rounds of conversation. One in four participants disavowed their beliefs entirely.

And a very interesting result from another study.

  an analysis published in August by researchers at the University of Wisconsin-Madison suggests the debunking messages posted by medical experts on TikTok are *more effective if overlaid with high-tempo music.*
> an analysis published in August by researchers at the University of Wisconsin-Madison suggests the debunking messages posted by medical experts on TikTok are more effective if overlaid with high-tempo music.

Everybody know the best place to preach your ideas to your friends is on the dance floor of a jungle rave

I suppose at the end, when you have a discussion with God about why you did the things you did and try to explain yourself, a few rounds of conversation will have had you surrender your prior beliefs.

In a sense, there’s no argument against the LLM, or God. Take that how you want.

> The academics believe the music helps swamp the brain’s ability to present counter-arguments, making the message more persuasive to the listener. Deploying strong narratives to accompany a particular message, including characters and rich description, is another way of overwhelming the brain’s ability to battle back against spurious claims, prior research has shown.

You know what's even better than this? Meeting people on the same level instead of seeing them as animals that need to get herded around. But this isn't new, we've been saying this for a while now. Condescension is politically toxic. This whole tone and method of delivery is a proven failure. It's probably easier for these people to lose on election day than meet voters where they are, because ego matters more here than winning

> You know what's even better than this? Meeting people on the same level instead of seeing them as animals that need to get herded around. But this isn't new, we've been saying this for a while now.

As a moderate liberal: The better take is to remind liberals that they’re also “animals [that are] herded around”.

> Condescension is politically toxic.

I agree. There was a lack of self-awareness on the liberal side, but the conservatives are now at risk of the same mistake.

The WSJ was and still is rife with this.

I agree, but I think we have to accept that there isn't really a reliable way to convince people of something. People (and I include myself) struggle to discard beliefs borne of intuitive, irrational processes. And personally, when I realize my attempts at logic are failing, it's easy to give up and condescend, that or stop trying. People don't want treatises, no matter how dressed up they are. They want tasty morsels. Forthrightness and composure, "taking the high road", all the time is draining, if not mostly ineffective.
(comment deleted)
This is already happening and people are becoming more discerning with their news sources. Look at Joe Rogan vs CNN viewer numbers.
The may/must distinction in belief is a plausible cause for the power of propaganda.

Human beings want to believe things they like and want to be skeptical of things they don't like. This was so obscenely obvious to me as I was going up with religious extremist parents after studying some epistemology in college. After living with it on my mind, I think it takes a really, really stoic person to avoid this.

When we see a fact we like presented, we often just think that makes sense, and believe it. When we see some study backing it up, it gives us permission to believe it. For things we like, we use the "may believe" threshold for belief.

When we see a fact we don't like presented, we often think it's bullshit. When we see some study backing it up, we immediately dig into the study to find potential methodology problems. Until these issues are dealt with, we can refuse to believe it. With thinks we don't like we use the "must believe" threshold for belief.

Want to believe drinking wine makes us healthier: we accept that we may believe it, until decades later after numerous skeptics finally demonstrate the studies were flawed.

Want to doubt a controversial political: we refuse to believe it unless we must, when it is corroborated by outlets we trust, who may downplay or not dedicate resources to verifying the story.

Until and unless we hold the same level of skepticism for our facts, and unless we have verification services that don't play favorites in the same way, we will suffer greatly from this may/must belief dichotomy. It's why bad faith news sources are so pernicious. I don't think a chatbot has any hope of fixing this for us.

What is the term "things we like" representing here? It's about background theories, isn't it? Inclinations to agree with a theory, or doubt a theory, because of a mess of peripheral theories such as this is our way of life or they're in it for the money.

In short, bias. And to be free of bias is rationality. However:

* This is not possible.

* Or it is possible, but the process is endless.

* Rationality is made out of theories, and we can always learn better ones.

* Bias is also made out of theories, so it's just inferior rationality.

* Some of those theories can be very bad, so bad that they sabotage the process of getting better ones. Then the person has entrenched anti-rational theories, a sticky situation.

But generally speaking, bias is expected. To be completely free of it is unimaginable. Even the most rational of us would look extremely biased from the perspective of the distant future. To have no bias is so unrealistic that it's undesirable, akin to solving all problems that there will ever be. The process we're engaged with, our problem situation, includes our biases. Changing our minds reluctantly is how we function, and I kind of like it.

This does make propaganda effective, it's true: people pick up a battle cry. If only these people were reasonable, like me! But I think they mostly are.

>What is the term "things we like" representing here?

Generally, things that comport with the (by definition, arbitrary) axioms we've chosen to ground our world view.

truth is like a miracle, in that it is very hard to demonstrate, and might even take generations of effort, to finaly prove and describe something that is still hard to understand ,demonstrate and believe for anyone else. but a lie,like sin, is right there, the very worst lie can have power instantly, and unlike miracles, the most abhorant sin takes no real effort at all