SGU Episode 527
|This episode needs: transcription, proof-reading, formatting, links, 'Today I Learned' list, categories, segment redirects.||How to Contribute|
|SGU Episode 527|
|August 15th 2015|
|SGU 526||SGU 528|
|S: Steven Novella|
|B: Bob Novella|
|J: Jay Novella|
|E: Evan Bernstein|
|C: Cara Santa Maria|
|Quote of the Week|
|When kids look up to great scientists the way they do to great musicians and actors, civilization will jump to the next level.|
- 1 Introduction
- 2 Forgotten Superheroes of Science (3:25)
- 3 News Items
- 4 Who's That Noisy (1:01:22)
- 5 Science or Fiction (1:05:43)
- 6 Skeptical Quote of the Week (1:20:32)
- 7 Testing Cara's Sci-fi Knowledge (1:21:09)
- 8 References
- Movie discussion: Memento
You're listening to the Skeptics' Guide to the Universe, your escape to reality.
Forgotten Superheroes of Science (3:25)
- Marie Thorp: Geologist who discovered the Mid Atlantic Ridge which was key to the confirmation of Plate Tectonics
S: Bob, you're gonna tell us about this week's Forgotten Superhero of Science.
B: For this week's Forgotten Superheroes of Science, I'm covering Marie Thorp (1920 to 2006). She was a geologist who was instrumental in discovering the Mid Atlantic Ridge, which was key evidence for the discarded (at that time) theory of plate tectonics. Ever hear of her? ... Probably not!
E: I've heard of plate tectonics...
B: Yes, but that's probably all you were familiar with. Thorp faced a similar impediments in her career, like many women of her time, but she was lucky. When she and geologist Bruce Hezeen started studying and mapping the ocean floor, during the cold war. The sea floor was, at that time, considered a potential future battlefield for submarines, and money was not hard to come by at that time. So she was able to get paying jobs.
But she was not allowed to research on vessels, like her partner Hezeen was. He was doing seismic and topographic mapping. But she wasn't allowed on the research vessel because she was a woman, of course. What she did though, was that she would examine and visualize data after he had recorded it. And hiding in the data, they discovered something completely unexpected.
Scientists thought that the sea floor, at that time, was a flat, featureless expanse, with really, nothing down there. What they found was structure, and mountains, and canyons, and ridges, often rivalling anything that could be found on the land.
And the most formidable example of that was the mid-Atlantic ocean ridge. This ridge existed in all of the oceans, it was later found, an extended for fifty thousand miles, or eighty thousand kilometers. It was gargantuan!
And it's funny, that at that time, they thought this was evidence for an expanding Earth. So they didn't really know at that time what it really meant. But ultimately, it was part of the key supporting evidence for the ignored, but eventually paradigm-changing theory of plate tectonics.
So, remember Marie Thorp; mention her to your friends, perhaps when discussing ridge-pushing mantle conveyors. And of course, divergent boundaries.
S: Yeah, again, that stands as one of those sciences that completely changed our view of reality.
S: You know, going from before we understood about plate tectonics, 'till after. It still stands as one of those examples of a new scientific idea that was met with skepticism, but that eventually turned out to be true.
I also find that ironic, 'cause people give it as an example of science being broken, 'cause they didn't immediately accept plate tectonics. It's like, it's actually a perfect example of science working. Initial skepticism, followed by evidence, leading to acceptance. What else would you expect to happen?
C: Yeah, and there's almost no examples in science of people immediately accepting something. (Laughs)
B: Unless they had a vested interest.
C: Yeah, but I mean, for the most part, this is that kind of Kunian paradigm shift. Like, this is what philosophers of science write about, is that there has to be such overwhelming evidence, oftentimes from multiple lines, from different fields, and different people. And only then do people start to accept it, 'cause we are very grounded in the way that we think things are.
You know, when you're thinking that there's a protoplasm – or, that's not the word I'm looking for – an ectoplasm, that exists – and this has been the prevailing scientific view for years – in order to kind of turn that on its head, it takes a lot of evidence.
B: And prior plausibility is important. That's one of the main reasons it was discarded, was because people considered it completely implausible.
S: It raises the bar,
S: But even when you raise the bar,
S: if you get enough evidence, then
S: people will still accept it.
B: Even evolution could conceivably be overturned, if you had that much evidence. It's just that at this point, it's fairly inconceivable that that could even possibly happen.
Skin Cells into Brain Cells (7:20)
Registering Studies (16:01)
Search Engines Influence Elections (25:49)
S: Jay, you're gonna tell us about how internet search engines like Google are influencing our elections.
J: It's not just elections, it's everything. But we'll be using the election in this news item as an example. But this is really interesting, guys – a little scary as well.
So, there is a researcher named Robert Epstein, and he's a psychologist at the American Institute for Behavioral Research in Vista, California. And he concluded that the higher up in results a politician ranks in a page, the more likely you're gonna vote for them. And here's, the three experiments.
The first one, he started with a hundred and two volunteers. And these were all from San Diego, California. And these people were picked to represent the general voting population in the US, taking into consideration factors like age, race, political affiliation, other traits. The researchers tested their ability to influence who the test group would have voted for in the 2010 election for Prime Minister of Australia.
Now, why'd they pick Australia? That's because they wanted to make sure that the test subjects didn't know much about the people involved. So they built a fake search engine, and they called it “Bob's Biceps.”
(Bob chuckles, Cara snickers)
J: Nope, they called it Kadoodle. I thought a lot of people would be interested to, Bob's Biceps. And apparently, not. When used,
B: Oh yeah?
J:the fake search engine showed a list of thirty web sites. (Laughs) Come on, Bob. So they had thirty web sites for the two top candidates: Fifteen for Tony Abbot, and fifteen for Julia Guilliard. Now, most of the people in the experiment, they didn't know much about the candidates, like I said. And the test would be their introduction to these candidates, and it would be the first time that they saw legitimate, or just data on those candidates. Now, of course, like Bob's Biceps, the whole thing was rigged, nothing.
J: It displayed results,
B: So implausible!
J: It displayed results that were based in favor of one of the two candidates, depending on who the subject was. So, different people saw different results. Now, and this is what I – the thing that was obvious to most people is, what, the researchers also, obviously predicted this, that the subjects spent the vast majority of their time reading the websites on the people that showed up.
So here's your results list, and you're gonna get predominantly Person A over Person B. Well, you're gonna read those websites, those are the things that were put in front of them. Now, what was not anticipated by the researchers was how powerful the ranking of the websites influenced the undecided voters, their favorite candidate, that they ended up seeing ranked at the top.
So, what I'm saying is that it had a significant influence on who they reported that they would vote for, simply by, not even that the websites were skewed to put down the other candidates, or stuff like that. It was just whose name is put in front of them, which I thought was very interesting.
And what was even scarier was that people noticed that the search engine was quote-unquote, was manipulative, meaning that it was selecting what they see, but people thought that that was a good thing. They thought that the search engine's bias was the result of the search engine doing its job, which is to make good decisions on what information is the most valid.
C: Uh oh.
E: Oh, we've seen this before in subjects we've talked about. Homeopathy, for example. If somebody really doesn't know what's going on with homeopathy, and they plug in homeopathy for web search, and their first entry happens to be a pro-homeopathy website, they're gonna be influenced by that, even if they continue to read and research. But that'll have influence.
J: Forty-eight percent of the people under this circumstance were swayed, compared to the control group, meaning that they saw a result set, and they were forty-eight percent likely to vote for that person, which is huge, in this circumstance. That's a huge sway.C: You know what's interesting, we actually got an email from a listener – let me see if I can find it here … yeah. It's from David, in British Columbia, who left this message. “Hey guys and gal.” Love that, thank you.
“Love your show. Just a comment on a Google search for a science topic. When I type quote 'fossils in sedimentary rocks' into Google, the first article that shows up, even before Wikipedia, and the US Geological Society, is the Institute for Creation Research website.”
J: Absolutely. Yeah, Cara, I email people that I know at Google when I see this, and when I'm made aware of these things, and they are aware that this type of stuff is happening, and they report it. So you could actually email Google to let them know if you feel like something is out of sorts. That being said, there's millions of things that
J: people search on, that are manipulated by a lot of factors. And I'm not saying it's bad, because it works for the quote-unquote “good guys” too, you know, right? We're not gonna complain when you see the correct stuff show up, which you think is correct. So, the algorithms that they use in order to determine what shows up are massively complicated, hyper-complex systems that do a lot of data crunching and data analysis and decision making.
C: And here's – just to interject really quickly – there's a tip that a friend of mine actually pointed out the last time she was on my podcast. If you use Chrome, you can actually go to the browser, and you can click File – instead of New Window, click New Incognito Window.
C: And instead of opening up a window that is tailored to you, an incognito window basically takes away all of the part of the algorithm that tailors it to you. Now it's still gonna use the algorithms based on everybody else's internet searches, but you won't see as many targeted things. Like, if you're somebody who's always looking up unicorns, it's not gonna pull you to all these unicorn websites when you type in “horn.” You might get other things first. That was such a terrible example, but it's a cool trick to use.
C: I definitely recommend it. If you want to go incognito, open up an incognito window, and you'll see things that aren't as perfectly tailored to you.
J: And you could also go to a non country specific version of Google as well. So there are ways to nullify those effects, to put 'em down. Incognito is also great because it automatically does not track anything that you're doing, locally, on your machine. So it's a good thing to use at work or whatever, if you feel like you need to do that. But, all right, so let me continue. There's a lot more to talk about, guys.
J: They conducted a second, larger experiment, with twenty-one hundred subjects, and they found these people online through an Amazon site called Mechanical Turk. And again, they chose the test subjects to represent the spectrum of voters in the US, and the really cool thing here, is that they were able to factor in the results, more specific information about the test subjects, which I thought was great, 'cause I was thinking about that when I was reading about the first test.
So this gave insight into the demographics. So prepare yourselves, young Jedi, for what I'm about to tell you! Who was the most susceptible to the search engine's manipulation?
S: You mean what demographic?
C: Uhh … undecided voters?
J: What demographic?
B: Age? Kids?
J: Just keep in mind, the SGU is agnostic to the results of this study. I'm just gonna read out the results here. “Divorcees, Republicans, and subjects who reported low familiarity with the candidates. Subjects who were better informed, married, or reported an annual household income between forty and fifty thousand were harder to sway. Moderate Republicans were the most susceptible of any group. The manipulated search results increase the number of undecided voters who said they would choose their favorite candidate by eighty percent!
J: Oh!! Yes, that's huge. Now, they did a third test, in India, with a real election that was happening. And they were able to sway voters by twelve percent, but that lower number was due to their familiarity with the candidates. But keep this in mind guys: Awareness of the manipulation enhanced the effect. And also, most elections are won by margins that are smaller than one percent!
J: A large number of elections are won by margins that small. So these results – twelve percent – you think that's a lot less than previous numbers that I stated. They are, but so what? They're still huge! They're incredibly profound. This means that search engine results can and do have a strong effect on voter selection, and everything else, everything else that you do in a browser. So you kinda have to have your bullshit detector up at all times. Not kinda. I recommend you do have your bullshit detector massively up when you're perusing the internet.
S: Yeah, it's really interesting that knowing you're being manipulated makes it more powerful. 'Cause that suggests that there's a social influence there, like a bandwagon effect. Say, “Oh, this is really popular. It's ranking very high. That means it must be a good choice. And I don't know what's really going on. Maybe I'll just go along with that.”
C: Yeah, it kind of allows you
C: to be a little bit more intellectually lazy, which is appealing to a lot of people who are just busy! I mean, this is the reason, I think, that these kinds of effects have such staying power, is it's very hard for people who work full time, who have kids, who have jobs, to also be worried about elections, and be worried about trying to always be skeptical, and it's a lot! It's overwhelming for people. And so when they say, “That's kind of what Google's telling me,” or that “That's what the majority seems to think,” they're probably right. I might bank on that.
C: It's scary!
J: Yeah, I agree with that. I mean, the less decisions we need to make, you just kind of let certain things go, you know. And as skeptics, we don't let a lot know.
S: Are you saying that they're sheeple?
S: Is that what you're saying?
J: Well, guys, keep this in mind, right? So, I'm gonna immediately come out with my bias. I'm a fan of Google. Now, I will change my opinion of Google on a dime if I feel like that company does anything sketchy, or something that would compromise the free flow of data, and if I felt like they were manipulating things, which is what they came out to say that they will do no evil, that was the motto of Google. So I still believe in this company.
But keep this in mind, their search engines use a couple of things they call relevance and credibility, and these are ideas that they factor into their algorithm in order to show you what they think you want to see. And I don't want to say “What they want you to see.” I'm hoping, and I really do believe this, it's what they think you want to see.
Now, that being said, let's just make a public service announcement here. We're heavily influenced by search engines, and by results of any kind on the internet, even on Youtube, will show you the most popular videos, like Cara was saying, and we'll watch them. Now, one of the researchers, Epstein, said, “Without any intervention by anyone working at Google, it means that Google's algorithm has been determining the outcome of close elections around the world.” That's an incredible statement
J: to make. I'd like you guys to consider that. Epstein also believes that it'd be very hard to tell if Google was deliberately manipulating the algorithms, and affecting the results we see. We just don't really have an insight into if they're just turning it off. I will remind you, I think that Google is a great company. And I don't think, personally, I don't think that they're doing this. It's just, it's more of a thought experiment, like, we should be aware of how powerful that data stream into our minds is.Another person involved with the project, Diacapolus (I'm sorry if I'm pronouncing your name wrong), he said,
“It's easy to point the finger at the algorithm, because it's supposedly this inert thing. But there are a lot of people behind the algorithm. I think that it does pose a threat to the legitimacy of the democracy that we have. We desperately need to have a public conversation about the role of these systems in the democratic process.”
Now, I'm gonna take this to a slightly absurd level, but …
B: Please do!
J: follow me.
J: You know, some day, we're gonna have more and more brain to internet connectivity. You know, we're gonna have virtual reality, and heads up displays, and augmented reality. And imagine you're wearing a pair of glasses, kind of like Google glasses. You're wearing a pair of glasses
E: That actually work.
J: in the next fifteen years,
J: and it's popping up relevant information about the area that you're in. “Oh! Your favorite restaurant is right around the corner. And this guy walking up to you is Jack, and you met him at a convention two years ago.” And that data is just gonna be flowing into your head constantly, like augmenting your reality, and telling you things, and giving you tidbits of information, and whatnot.
Now that's kind of scary if you think that there is some type of malfeasance behind any of that, or any kind of, whatever, cajoling, or skewing of the numbers, or however you want to put it. So I don't think it's a bad thing for us to talk about it. And Google might not survive, and another company might pop up, and who are they? You know what I mean? It is something that we should be talking about.
S: Yeah, when there's massive data controlling what we do, what we think, what we know, with algorithms, that has an actual effect. For example, there are stock buying and selling algorithms.
E: Oh yeah.
S: And that affects the stock market, you know? And people have to take into account what the algorithms are gonna do. The degree to which computers are influencing our lives is growing tremendously, and this is just one aspect of it that we're just starting to see.
(Commercial at 39:29)
Picture of Earth and Moon (41:13)
Scotland Bans GMOs (48:17)
Who's That Noisy (1:01:22)
- Answer to last week: Cicadias
(Commercial at 1:04:29)
Science or Fiction (1:05:43)
Item #1: A recent study shows that when faced with a female opponent, candidates, either male or female, with a higher pitched voice fared better. Item #2: The most recent Gallup poll shows, for the first time, that atheists are not the least likely religious category that people would vote for, being beat out by Muslims. – Item #3: Studies show that only 54 out of 600 people (9%) change their mind during the course of an election campaign.
Skeptical Quote of the Week (1:20:32)
'When kids look up to great scientists the way they do to great musicians and actors, civilization will jump to the next level.' Brian Greene
Testing Cara's Sci-fi Knowledge (1:21:09)
Transitions into a membership drive
S: The Skeptics' Guide to the Universe is produced by SGU Productions, dedicated to promoting science and critical thinking. For more information on this and other episodes, please visit our website at theskepticsguide.org, where you will find the show notes as well as links to our blogs, videos, online forum, and other content. You can send us feedback or questions to firstname.lastname@example.org. Also, please consider supporting the SGU by visiting the store page on our website, where you will find merchandise, premium content, and subscription information. Our listeners are what make SGU possible.