WEBVTT - From the Vault: Illusory Truth Effect, Part 2 0:00:05.760 --> 0:00:07.480 Hey, you Welcome to Stuff to Blow Your Mind. My 0:00:07.560 --> 0:00:10.680 name is Robert Lamb and I'm Joe McCormick, and it's Saturday. 0:00:10.760 --> 0:00:14.560 Time for a vault episode. This episode originally published July twelve, 0:00:15.760 --> 0:00:18.439 and this is part two of our exploration of the 0:00:18.480 --> 0:00:22.600 illusory truth effect. That's right, this one, I will land 0:00:22.640 --> 0:00:25.000 the plane for you and uh and hopefully give you 0:00:25.079 --> 0:00:28.200 some tools that you might be able to employ to 0:00:28.360 --> 0:00:31.319 fight the power of the illusory truth effect. Or at 0:00:31.400 --> 0:00:34.040 least that's that's the intention. All right, let's dive right in. 0:00:37.200 --> 0:00:40.280 Welcome to Stuff to Blow Your Mind from how Stuffworks 0:00:40.320 --> 0:00:48.800 dot com. Hey, you, welcome to Stuff to Blow your Mind. 0:00:48.920 --> 0:00:51.280 My name is Robert Lamb and I'm Joe McCormick, and 0:00:51.360 --> 0:00:55.240 we're back part two of our exploration of the illusory 0:00:55.360 --> 0:00:59.760 truth effect, probably the liar's best trick. If you haven't 0:00:59.800 --> 0:01:01.960 heard in our last episodes, you'd probably go back listen 0:01:01.960 --> 0:01:04.600 to that first. But if you have int or if 0:01:04.640 --> 0:01:07.039 you have, let's just do a quick recap of what 0:01:07.080 --> 0:01:09.440 we talked about last time. We discussed all of the 0:01:09.480 --> 0:01:11.920 research on this thing that's sort of been part of 0:01:11.920 --> 0:01:15.160 folk wisdom that if you say something and if you 0:01:15.200 --> 0:01:18.360 repeat it, and repeat it and repeat it. People become 0:01:18.440 --> 0:01:21.200 over time more likely to believe that thing, and that 0:01:21.319 --> 0:01:25.440 is thoroughly validated by experimental research. Right. And we also 0:01:25.480 --> 0:01:27.520 talked a little bit about why does it even make 0:01:27.560 --> 0:01:30.080 sense that we would come to believe things that were 0:01:30.120 --> 0:01:32.920 not true about the world that we live in just 0:01:32.959 --> 0:01:36.240 because they were repeated. Yeah, and so the basis that 0:01:36.280 --> 0:01:38.640 we ultimately ended up on last time that seems to 0:01:38.640 --> 0:01:41.120 be favored by most of the psychologists to study this 0:01:41.760 --> 0:01:45.679 is based in the idea of processing fluency that for 0:01:45.760 --> 0:01:49.040 whatever reason, one researcher we talked about last time came 0:01:49.080 --> 0:01:51.560 to believe that it was because of conditioning based on 0:01:51.640 --> 0:01:54.560 real world effects. But for whatever reason, we tend to 0:01:54.600 --> 0:01:58.160 associate things that are easy to process, things with high 0:01:58.200 --> 0:02:03.320 processing fluency with truth. So something's easy to read, we 0:02:03.360 --> 0:02:06.360 think it's more true. Or if something is an idea 0:02:06.440 --> 0:02:09.560 we've seen or heard or encountered before, because that's easier 0:02:09.560 --> 0:02:12.960 to process because of familiarity, we believe that it is 0:02:13.000 --> 0:02:15.360 more likely to be true than if we're encountering it 0:02:15.400 --> 0:02:16.920 for the first time. But of course, in all of 0:02:16.919 --> 0:02:20.560 this extreme implausibility is going to be a boundary condition 0:02:20.600 --> 0:02:23.280 that's going to kick in. So this is like the 0:02:23.280 --> 0:02:28.519 Ted Cruizes, the Zodiac killer um level of of of implausibility. 0:02:28.639 --> 0:02:31.000 What's just because the ages don't match up right, well, 0:02:31.120 --> 0:02:33.320 just and it's just kind of like, alright, I'm not 0:02:33.400 --> 0:02:36.120 bleeding that that sounds ridiculous, but some people do believe that. 0:02:36.440 --> 0:02:39.359 So your boundary condition may not be where somebody else 0:02:39.560 --> 0:02:42.680 canoundary condition is well, the boundary conditions will vary from 0:02:42.720 --> 0:02:47.400 individual to individual. Um. So yeah, So the question that 0:02:47.560 --> 0:02:49.680 we should address to start off in this one is 0:02:49.680 --> 0:02:52.799 in the last episode, we discussed how this effect has 0:02:52.840 --> 0:02:56.360 been thoroughly validated in the lab. But here's a question. 0:02:56.680 --> 0:02:58.920 Does it work in the real world and is it 0:02:58.960 --> 0:03:02.520 really all that powerful? Like a lot of researchers seem 0:03:02.560 --> 0:03:06.280 to assume that, surely, if you already know something about 0:03:06.280 --> 0:03:10.560 a subject, repetition of a contradictory false statement wouldn't actually 0:03:10.639 --> 0:03:15.200 undermine your real knowledge, would it. Surely they would tend 0:03:15.200 --> 0:03:18.400 to assume that this this ilusory truth effect only works 0:03:18.440 --> 0:03:21.480 for state statements that were uncertain about to begin with, 0:03:21.840 --> 0:03:25.280 and statements that seem highly plausible, like if you didn't 0:03:25.320 --> 0:03:28.560 know anything about either Ted Cruz or the Zodiac Killer 0:03:28.600 --> 0:03:30.280 really and then you would just sort of say, all right, 0:03:30.320 --> 0:03:33.880 maybe that's possible, whereas an individual who has read multiple 0:03:33.919 --> 0:03:37.000 books on the Zodiac Killer would say, no, that doesn't 0:03:37.040 --> 0:03:39.760 that doesn't match up. That is just ridiculous. Yeah, So 0:03:39.920 --> 0:03:43.880 that that's the assumption. But unfortunately some more recent research 0:03:44.000 --> 0:03:48.000 has really turned that assumption on its head. Uh So, 0:03:48.120 --> 0:03:50.800 I want to talk about an important recent study in 0:03:50.840 --> 0:03:53.480 the illusory truth effect that brings it's a bearer of 0:03:53.520 --> 0:03:57.200 bad news. The study is from the Journal of Experimental 0:03:57.200 --> 0:04:03.160 Psychology General in Fasio, Brashier, Pain and marsh and it's 0:04:03.200 --> 0:04:07.840 called knowledge does not Protect against illusory truth. So they 0:04:07.880 --> 0:04:10.400 pointed out that the illusory truth effects that we talked 0:04:10.400 --> 0:04:14.440 about last time, based on processing fluency, is widely accepted, 0:04:14.520 --> 0:04:17.680 well established, but it had been previously thought that this 0:04:17.720 --> 0:04:21.760 effect was constrained by a few things. Now, one constraint 0:04:21.960 --> 0:04:25.680 shown to actually exist in the literature is recollection of 0:04:25.720 --> 0:04:29.880 the quality of the source of the information. So previous 0:04:29.880 --> 0:04:33.400 studies have shown that if you specifically remember where a 0:04:33.440 --> 0:04:36.719 statement came from, and you consider the source of the 0:04:36.800 --> 0:04:41.240 statement a dishonest or untrustworthy source, that can produce kind 0:04:41.279 --> 0:04:44.520 of a reverse truth effect, where repetition of a statement 0:04:44.600 --> 0:04:47.440 known to come from a liar or an untrustworthy source 0:04:47.760 --> 0:04:50.080 causes us to disbelieve it. So this sounds like this 0:04:50.080 --> 0:04:53.320 should be good news, right right? Yeah? Did I ultimately 0:04:53.360 --> 0:04:56.800 the question did I hear that on the radio? Did? 0:04:57.040 --> 0:04:58.720 Or did I see it on a T shirt? Yeah? 0:04:58.800 --> 0:05:02.120 Or was this the cover of the National Enquirer? Like 0:05:02.200 --> 0:05:05.360 you remember that's where it came from, and you're you know, 0:05:05.480 --> 0:05:08.480 that's an untrustworthy source. So it actually has the reverse effect. 0:05:08.600 --> 0:05:11.640 You hear that repeated and it makes you go no, no, no, 0:05:11.720 --> 0:05:15.200 that's not true at all. But this isn't as much 0:05:15.240 --> 0:05:18.880 of a protection as we think, because honestly, how well 0:05:18.960 --> 0:05:22.039 do you remember the exact source of every bit of 0:05:22.080 --> 0:05:25.920 semantic knowledge in your head? Why no, bat Boy did 0:05:25.960 --> 0:05:28.160 not come from the New York Times, But there are 0:05:28.200 --> 0:05:30.080 lots of other things that are in your head that 0:05:30.160 --> 0:05:32.440 did come from the cover of the National Enquirer, and 0:05:32.480 --> 0:05:35.720 you don't remember that that's where it came from. I 0:05:35.760 --> 0:05:39.200 guarantee it you've stood in line at the grocery store. Well, 0:05:39.279 --> 0:05:44.919 if it's a story about any particular aged celebrities, brave 0:05:45.040 --> 0:05:48.240 last days or sad last days, they probably came from 0:05:48.279 --> 0:05:51.080 inquire But yes, there there, there's probably there are probably 0:05:51.160 --> 0:05:53.200 some stories in there that I would not definitely be 0:05:53.240 --> 0:05:56.520 able to pin down to inquire versus other sources. Robert, 0:05:56.560 --> 0:05:59.560 I see right through your bravado. Some Inquirer stories have 0:05:59.600 --> 0:06:02.760 gotten through to you. Uh. Yeah. Other studies have backed 0:06:02.760 --> 0:06:05.160 this up. After just a period of a few weeks, 0:06:05.360 --> 0:06:07.719 what may have once been stored in the brain as 0:06:08.120 --> 0:06:12.080 false claim by an untrustworthy source could potentially, over time 0:06:12.120 --> 0:06:16.360 become just a familiar statement I remember, which of course, 0:06:16.560 --> 0:06:20.200 once it's familiar translates it into more likely to be 0:06:20.240 --> 0:06:23.280 a true fact. Uh. There was at least one study 0:06:23.279 --> 0:06:27.040 that looked into this, by beg Annas and Feignacci in 0:06:27.160 --> 0:06:32.880 Nino called Dissociation of processes and belief, source recollection, statement, familiarity, 0:06:32.920 --> 0:06:35.719 and the illusion of truth, And basically they found that 0:06:35.800 --> 0:06:38.800 when the source of a claim is not super memorable 0:06:38.839 --> 0:06:44.440 as unreliable, familiarity can be more important than truth or reliability. Okay, 0:06:44.440 --> 0:06:47.760 so it's not necessarily a like a magazine that that 0:06:47.839 --> 0:06:51.600 has a negative reputation in your mind, but it's not 0:06:51.960 --> 0:06:54.200 something that's completely reputable either. It just kind of follows 0:06:54.200 --> 0:06:56.839 in between or even if it has a negative reputation 0:06:56.960 --> 0:07:00.359 and it's just not all that memorable, you can lose 0:07:00.400 --> 0:07:03.120 track of where it came from and it will suffer 0:07:03.279 --> 0:07:06.800 from the illusory truth effect. This can happen even when 0:07:06.880 --> 0:07:10.000 you should have remembered that it came from an untrustworthy source. 0:07:10.200 --> 0:07:13.440 There are exceptions when the source is really memorable, but 0:07:13.600 --> 0:07:16.560 a lot of times it doesn't protect you. Now, the 0:07:16.640 --> 0:07:21.480 second assumption about constraints on the illusory truth effect is 0:07:21.520 --> 0:07:25.320 about knowledge. Right, We've all got knowledge already in our heads, 0:07:25.360 --> 0:07:28.160 and the idea is that pre existing knowledge will protect 0:07:28.200 --> 0:07:31.200 against the effect. And this is what came under scrutiny 0:07:31.240 --> 0:07:36.400 in this particular study by Fasio and our co co authors. So, 0:07:36.760 --> 0:07:39.880 despite being an assumption repeated again and again in the 0:07:39.920 --> 0:07:43.920 illusory truth literature, very few of the studies actually bothered 0:07:43.960 --> 0:07:47.480 to test whether knowledge protects people. I was just sort 0:07:47.520 --> 0:07:50.440 of asserted to be true as if it were obvious, 0:07:50.800 --> 0:07:53.160 And the few that did bother to tested in any 0:07:53.160 --> 0:07:56.520 way generally did so by testing how the effect presented 0:07:56.520 --> 0:08:01.120 in people who claimed subject area expertise. So, uh, these 0:08:01.160 --> 0:08:04.280 studies yielded contradictory results. But here's a couple of examples. 0:08:04.720 --> 0:08:07.640 Scroll in nineteen eighty three found that if you rate 0:08:07.680 --> 0:08:10.240 yourself as an expert on cars, Robert, would you rate 0:08:10.240 --> 0:08:13.560 yourself as an expert on cars, but some people would win. 0:08:13.760 --> 0:08:16.760 Some people around the office. Yeah. Car experts, well found 0:08:16.800 --> 0:08:21.400 suffered smaller illusory truth effects uh than non experts on 0:08:21.480 --> 0:08:24.520 car trivia. So that would suggest, Okay, knowledge gives you 0:08:24.560 --> 0:08:27.200 a little bit of an edge. You're not You're not 0:08:27.320 --> 0:08:30.800 as susceptible as amateurs. And then Parks and Tough in 0:08:30.880 --> 0:08:34.480 two thousand and six had people rate claims about known 0:08:34.840 --> 0:08:39.080 versus unknown consumer brands, and the illusory truth effect was 0:08:39.120 --> 0:08:42.320 bigger for statements about brands that people were unfamiliar with. 0:08:42.760 --> 0:08:44.480 That makes sense. So like, if you didn't already know 0:08:44.520 --> 0:08:48.439 anything about this brand, you were more susceptible to illusory 0:08:48.440 --> 0:08:51.120 truth effect on statements about the brand. Yeah, that makes 0:08:51.200 --> 0:08:55.320 perfect sense. On the other hand, Arkey's, Hackett and Boem 0:08:55.360 --> 0:08:58.400 in ninety nine found the opposite, that the higher a 0:08:58.480 --> 0:09:02.480 person rated their expertise sent a subject, the more susceptible 0:09:02.520 --> 0:09:05.439 they were to the illusory truth effect in that subject area. 0:09:06.320 --> 0:09:09.240 Makes you wonder if there's like some kind of insecurity 0:09:09.320 --> 0:09:12.000 or like identity protective thing going on there. Yeah, Like 0:09:12.160 --> 0:09:13.920 I don't want I don't I don't want to be wrong. 0:09:14.000 --> 0:09:16.520 So I'm just gonna nod my head on that situation. 0:09:16.600 --> 0:09:18.480 I don't want to look bad. I've already staked my 0:09:18.559 --> 0:09:23.640 reputation on being a car expert. Also, boem In ninet. 0:09:24.240 --> 0:09:28.720 Found that psychology majors showed a larger illusory truth effect 0:09:28.840 --> 0:09:33.280 on psychology than non majors. But there's some issues with 0:09:33.320 --> 0:09:35.959 these studies. So Fasio and her co her co authors 0:09:36.000 --> 0:09:39.160 point out that these types of tests don't actually manipulate 0:09:39.240 --> 0:09:41.839 direct knowledge of whether the statements are true or false, 0:09:42.240 --> 0:09:45.760 just sort of the perception of related knowledge. So they 0:09:45.800 --> 0:09:48.520 wanted to test this directly. They created a big list 0:09:48.559 --> 0:09:51.000 of statements like we've seen in these other tests, where 0:09:51.040 --> 0:09:54.240 you'll have true statements and false statements, and they based 0:09:54.240 --> 0:09:57.079 this off existing lists of facts that have been shown 0:09:57.080 --> 0:10:01.160 in previous studies to be either generally known were generally unknown. 0:10:01.640 --> 0:10:05.240 And this created four categories of statements. You've got known truths, 0:10:05.520 --> 0:10:10.400 unknown truths, known falsehoods, and unknown falsehoods. Here's some examples. 0:10:10.559 --> 0:10:13.480 You've got a known truth quote. The cyclops is a 0:10:13.600 --> 0:10:17.360 legendary one eyed giant of Greek mythology. Robert checks out. 0:10:17.480 --> 0:10:20.160 Checks out. Okay, how about the Pacific Ocean is the 0:10:20.240 --> 0:10:23.120 largest ocean in the world. Checks out. Then you go 0:10:23.160 --> 0:10:26.920 into known falsehoods. The minotaur is the legendary one eyed 0:10:26.960 --> 0:10:30.880 giant of Greek mythology. Absolutely not the Atlantic Ocean is 0:10:30.920 --> 0:10:33.600 the largest ocean in the world, and most people are 0:10:33.600 --> 0:10:36.920 expected to know that these are not true statements. Then 0:10:36.960 --> 0:10:40.480 you've got unknown stuff. Here's an example. Unknown truth. Billy 0:10:40.559 --> 0:10:43.320 the kid's real last name? What was it? It's Bonnie. 0:10:44.720 --> 0:10:48.840 Unknown falsehood, Billy the kid's real last name is Garrett. Yeah, 0:10:48.880 --> 0:10:50.839 I would have would have been a toss up for 0:10:50.960 --> 0:10:53.840 me because I did not know Billy the kid's last name. 0:10:53.880 --> 0:10:57.200 I thought maybe it was a kid, you know, as 0:10:57.200 --> 0:10:59.480 in Kid Rock as Kid rocks first name is Billy 0:10:59.559 --> 0:11:02.880 kids like his middle name is the So there, there 0:11:02.960 --> 0:11:06.760 you go. So Experiment one, using this set of statements, 0:11:06.760 --> 0:11:09.600 forty students in the first phase. Subjects were shown a 0:11:09.679 --> 0:11:12.240 subset of statements from the list of all four types, 0:11:12.679 --> 0:11:15.360 and they were just asked to judge how interesting the 0:11:15.440 --> 0:11:18.839 statements were. You know, that sounds like a really fun task, right. 0:11:19.160 --> 0:11:21.960 Billy the kid's last name is Bonnie. How interesting was 0:11:22.040 --> 0:11:25.680 that I get more interesting than some names? Yeah? Maybe, 0:11:25.760 --> 0:11:28.240 I guess. I don't know. I didn't find that one 0:11:28.280 --> 0:11:31.280 that interesting. I don't know. I guess it sounds like 0:11:31.360 --> 0:11:34.200 Bonnie is in like pretty it sounds it sounds maybe 0:11:34.200 --> 0:11:36.360 a little odd for what based on the photos needs 0:11:36.360 --> 0:11:38.880 to be kind of like an ugly looking, you know, 0:11:39.000 --> 0:11:41.600 western outlaw. It makes me think like a Robert Burns 0:11:41.679 --> 0:11:44.880 kind of poem thing. And Bonnie Glenn or whereas Garrett 0:11:44.920 --> 0:11:46.920 sound you know, has kind of a guttural sound to it. Yeah, 0:11:47.040 --> 0:11:49.440 got right, Okay, So then they got the second phase. 0:11:49.480 --> 0:11:52.600 This happened immediately after the first phase. Students were given 0:11:52.640 --> 0:11:55.320 another subset of statements from the list, again all four 0:11:55.320 --> 0:11:58.480 types of statements, and they were warned that some statements 0:11:58.480 --> 0:12:00.839 were true and some were false, and they were also 0:12:00.920 --> 0:12:03.400 warned that they would see some repeats from the list 0:12:03.440 --> 0:12:05.920 that they had just reviewed for how interesting they were, 0:12:06.440 --> 0:12:08.240 and then they rated the claims on a scale of 0:12:08.240 --> 0:12:10.439 one to six about how true they were. There was 0:12:10.480 --> 0:12:13.400 also at the end an open ended knowledge check test 0:12:13.760 --> 0:12:16.360 with it had these open ended questions like what is 0:12:16.440 --> 0:12:20.000 the world's largest ocean? What is the one eyed monster 0:12:20.120 --> 0:12:23.240 of Greek myth uh to strengthen the experiment or's picture 0:12:23.280 --> 0:12:26.640 of the individual knowledge of each participant. So then you 0:12:26.760 --> 0:12:29.560 got the results. First of all, the original findings of 0:12:29.559 --> 0:12:33.360 the illusory truth effect were replicated. Repeated statements got higher 0:12:33.360 --> 0:12:36.199 truth ratings than new statements that the students had never 0:12:36.240 --> 0:12:41.040 seen before. But also, quite surprisingly, knowledge did not seem 0:12:41.160 --> 0:12:45.480 to prevent the illusory truth effect. Statements about both previously 0:12:45.559 --> 0:12:50.280 known and previously unknown facts were rated more true if 0:12:50.360 --> 0:12:52.920 they were repeated than if they were new. In other words, 0:12:53.120 --> 0:12:58.720 repetition increased perceived truthfulness, even for contradictions of facts that 0:12:58.800 --> 0:13:02.240 you know. So I want to quote from the author's quote. 0:13:02.760 --> 0:13:06.040 Reading a statement like a sorry is the name of 0:13:06.080 --> 0:13:11.400 the short pleaded skirt worn by Scots? Increased participants later 0:13:11.440 --> 0:13:14.400 belief that that statement was true, even if they could 0:13:14.440 --> 0:13:17.520 correctly answer the question, what is the name of the 0:13:17.559 --> 0:13:21.920 short pleaded skirt worn by Scots? Isn't that bizarre? So 0:13:22.040 --> 0:13:24.920 like you ask somebody what is the short pleaded skirt 0:13:25.000 --> 0:13:28.320 worn by Scots? And they answer kilt? But if you 0:13:28.440 --> 0:13:31.200 show them the phrase a sorry is the name of 0:13:31.200 --> 0:13:34.400 the short pleaded skirts skirt worn by Scots? And then 0:13:34.480 --> 0:13:37.400 show them the phrase again later they will they will 0:13:37.480 --> 0:13:40.880 take the repeated phrase as evidence that that statement is 0:13:40.920 --> 0:13:42.920 more true than if they saw the statement for the 0:13:42.960 --> 0:13:46.319 first time again. It it comes back to the shortcuts 0:13:46.360 --> 0:13:51.080 that our brain make. How weird is that's bizarre? I 0:13:51.080 --> 0:13:54.000 mean again, it's kind of a reminder that the human 0:13:54.000 --> 0:13:58.640 culture and human language just complicates everything. Yeah, it's crazy. 0:13:58.760 --> 0:14:01.560 Uh So again, there has found that the repetition effect 0:14:01.600 --> 0:14:03.920 also emerged for truth. So it wasn't just false statements, 0:14:03.920 --> 0:14:06.120 it was true statements to whether it's true or false, 0:14:06.160 --> 0:14:08.960 if you repeat it, people believe it more. So the 0:14:09.000 --> 0:14:11.840 takeaway from this first experiment is whether a statement is 0:14:11.840 --> 0:14:15.480 true or false, and whether you already no better or not. 0:14:15.679 --> 0:14:18.920 If somebody repeats the statement to you, on average, you're 0:14:19.000 --> 0:14:22.040 more likely to believe it. And then the second part 0:14:22.040 --> 0:14:24.360 of their study was kind of interesting. So they're discussing 0:14:24.400 --> 0:14:27.400 their own finding and they say, quote, the data suggests 0:14:27.440 --> 0:14:32.680 a counterintuitive relationship between fluency. Remember that's the fluency processing 0:14:32.680 --> 0:14:36.440 fluency how easy it is to process information between fluency 0:14:36.520 --> 0:14:40.680 and knowledge. Prior work assumes that people only rely on 0:14:40.800 --> 0:14:45.600 fluency if knowledge retrieval is unsuccessful i e. If participants 0:14:45.680 --> 0:14:48.440 lack relevant knowledge or fail to search memory at all. 0:14:48.840 --> 0:14:52.800 Experiment one demonstrated that the reverse may be true. Perhaps 0:14:52.840 --> 0:14:57.440 people retrieve their knowledge only if fluency is absent. So 0:14:57.560 --> 0:14:59.720 to test this out, they did a second experiment, and 0:14:59.720 --> 0:15:02.640 they repeated a modified version of the experiment to test 0:15:02.680 --> 0:15:05.600 it uh. They believe that the results indicate that people 0:15:05.680 --> 0:15:09.960 sometimes use a fluency conditional model, which means they would 0:15:10.040 --> 0:15:13.200 rely on fluency even if knowledge is available to them. 0:15:13.240 --> 0:15:16.680 Do you start with fluency and influency fails, you fall 0:15:16.720 --> 0:15:20.720 back on what you actually know. We shouldn't over interpret it, 0:15:20.720 --> 0:15:24.480 but in a limited way. There may be processes in 0:15:24.520 --> 0:15:26.760 the brain that say, I'm going to go for what 0:15:26.960 --> 0:15:30.320 feels easy before I even check my memory to see 0:15:30.360 --> 0:15:32.600 what I know. What kind of lines up with there 0:15:33.040 --> 0:15:36.800 the mind's tendency to want to offload memory to people 0:15:36.800 --> 0:15:39.880 and gadgets, like I do I have to remember that 0:15:39.920 --> 0:15:41.640 anymore if the machine is going to do it or 0:15:41.680 --> 0:15:43.880 my spouse is going to do it, And the brain 0:15:43.960 --> 0:15:47.640 says no, I think, well, that's completely prune that section. 0:15:47.840 --> 0:15:50.920 Here's a question, how often have you used a calculator 0:15:51.040 --> 0:15:55.360 to do math that you could yourself easily do? Um? 0:15:55.440 --> 0:15:57.640 You know what I mean, like, not not problems that 0:15:57.680 --> 0:16:00.160 would be really hard, but something that if you us 0:16:00.200 --> 0:16:03.280 took ten seconds, you could probably solve in your head. Yeah. 0:16:03.280 --> 0:16:05.560 I do that in Dungeons and Dragons sometimes when we 0:16:05.600 --> 0:16:08.560 get into hit points and whatnot. You know, I could 0:16:08.600 --> 0:16:10.880 certainly easy. I could either do it in my mind 0:16:11.120 --> 0:16:13.200 or just do it, you know, and pen and pencil 0:16:13.200 --> 0:16:15.560 real quick. But I'll go ahead and type it into 0:16:15.720 --> 0:16:18.920 my calculator just to yeah, I get it done. I've 0:16:18.920 --> 0:16:20.880 done the same thing too. It's weird. It's a little 0:16:20.960 --> 0:16:25.440 disturbing why or search engines, you know, just just throwing 0:16:25.480 --> 0:16:29.400 in the mathematical equation something really simple, um so, such 0:16:29.400 --> 0:16:32.320 as just determining how old a particular actor is or 0:16:32.360 --> 0:16:34.520 how old they would have been during a certain movie. 0:16:35.080 --> 0:16:37.200 I feel like I do that all the time, Like 0:16:37.320 --> 0:16:41.400 you're saying you do that even though you could easily 0:16:41.560 --> 0:16:44.920 know the answer if you checked your own memory. M M. 0:16:45.560 --> 0:16:47.480 I feel like I do that less with search engine, 0:16:47.560 --> 0:16:51.000 Like I definitely do the calculator thing. Yeah, not so 0:16:51.080 --> 0:16:53.760 much that I would remember, say how old Robert de 0:16:53.800 --> 0:16:57.400 Niro was during Godfather too, but I would just but 0:16:57.440 --> 0:16:59.560 I was it would suddenly wonder how old he was, 0:16:59.640 --> 0:17:04.119 and so do the simple mathematical scenario of you know, subtracting, 0:17:04.280 --> 0:17:07.040 subtracting one year from the other. Let's plant a lie 0:17:07.040 --> 0:17:09.920 in everybody's mind right now, Robert de Niro was four 0:17:10.000 --> 0:17:12.760 hundred and twenty three years old when he did Godfather too. 0:17:12.840 --> 0:17:15.359 And now you'll remember that that's implausible, that that's the 0:17:15.359 --> 0:17:18.280 implausibility barrier in action. Oh yeah, maybe I should do 0:17:18.320 --> 0:17:21.200 something else. Yeah, we'll come back to that. But anyway, 0:17:21.400 --> 0:17:25.080 So the conclusion of this experiment by Fasio and co 0:17:25.200 --> 0:17:30.040 authors is that quote participants demonstrated knowledge neglect, or the 0:17:30.119 --> 0:17:33.320 failure to rely on stored knowledge in the face of 0:17:33.440 --> 0:17:37.600 fluent processing experiences, so they'd rather go for what was 0:17:37.640 --> 0:17:41.000 easy to process than what was the correct answer based 0:17:41.040 --> 0:17:43.440 on their own knowledge. At the same time, it's really 0:17:43.480 --> 0:17:45.840 important to note that this doesn't happen every time, it 0:17:45.880 --> 0:17:48.880 doesn't happen with every person, it doesn't happen with every question, 0:17:49.000 --> 0:17:52.680 and it doesn't necessarily happen with huge effects, so the 0:17:53.000 --> 0:17:57.159 effect is relatively small. This was actually pointed out pretty 0:17:57.160 --> 0:18:00.440 well in a BBC article in sten by Tom's Afford. 0:18:01.000 --> 0:18:04.160 He pointed out that while repeated exposure to statements increase 0:18:04.200 --> 0:18:07.720 their believability. The biggest influence on whether a statement was 0:18:07.840 --> 0:18:10.639 rated true or not was whether it was actually true. 0:18:11.200 --> 0:18:15.199 So the the illusory truth effect is valid, and it 0:18:15.280 --> 0:18:18.840 does change the averages of the answers, but it's not 0:18:18.920 --> 0:18:22.040 like the only thing that matters, and it doesn't overpower 0:18:22.200 --> 0:18:25.440 our real knowledge about the truth. It's just weird that 0:18:25.520 --> 0:18:28.600 it does have some effect in the face of actual 0:18:28.720 --> 0:18:31.360 knowledge we have when actual knowledge should mean it has 0:18:31.400 --> 0:18:34.679 no effect. Does that make sense? Yeah? Again, I just 0:18:34.720 --> 0:18:37.560 come back to the you know, to to to the 0:18:37.600 --> 0:18:40.879 fact that the mind is going to offload whatever information 0:18:40.920 --> 0:18:43.760 it can or whatever processing it can. Yeah, those lazy 0:18:43.800 --> 0:18:45.640 brains of ours. Okay, well we should take a quick 0:18:45.640 --> 0:18:47.360 break and then when we come back, we will discuss 0:18:47.440 --> 0:18:51.199 more recent research on the illusory truth effect and some 0:18:51.280 --> 0:18:54.800 related concepts and what it means for our lives. Than 0:18:55.560 --> 0:18:59.119 thank alright, we're back. So we've discussed the subject of 0:18:59.119 --> 0:19:01.840 false memories, but for the many ways in which false 0:19:01.880 --> 0:19:06.520 memories can form um Psychologist Daniel Shackter identified seven in 0:19:06.600 --> 0:19:10.720 fact his his work The Seventh Sins of Memory, transient 0:19:10.760 --> 0:19:16.840 sam's absent mindedness, blocking, misattribution, bias, persistence, Uh, and I 0:19:16.920 --> 0:19:19.280 like to think of it this way. Memory is is 0:19:19.320 --> 0:19:22.320 not something that is carved in stone, but rather uh, 0:19:22.400 --> 0:19:24.840 something that is sculptured from clay, and the clay of 0:19:24.920 --> 0:19:28.520 memory remains valuable every time we retrieved from the drawer 0:19:28.520 --> 0:19:32.720 and handle it. As psychologist Pascal Boyer, who referenced in 0:19:32.840 --> 0:19:36.399 our last episode pointed out, um examples of this range 0:19:36.480 --> 0:19:41.640 from wordless recall intrusions and experiments, to therapy induced imaginings 0:19:41.640 --> 0:19:45.440 of past lives and or ritual abuse, which we've we've 0:19:45.480 --> 0:19:47.400 discussed on the episode on the on the show before 0:19:47.400 --> 0:19:50.520 in past episodes. Uh. So, memory retrieval is a very 0:19:50.560 --> 0:19:55.800 delicate stage. There's actually a line from the television series 0:19:55.840 --> 0:19:58.600 The Expanse and I think captures this perfectly well. The 0:19:58.920 --> 0:20:03.119 character Miller played Thomas jane Um. He sums up that 0:20:03.119 --> 0:20:05.600 they have the character sum up this rather perfectly says, 0:20:05.760 --> 0:20:08.480 you know, every time you remember something, your mind changes 0:20:08.520 --> 0:20:11.080 it a little, until your best and worst memories are 0:20:11.080 --> 0:20:14.679 your biggest illusions. So in the two thousand and eleven paper, 0:20:15.000 --> 0:20:19.160 remembering makes evidence, compelling retrieval from memory can give rise 0:20:19.200 --> 0:20:22.840 to the illusion of truth. From Jason d Azubko and 0:20:22.960 --> 0:20:27.240 Jonathan Fogel, saying, The authors conclude that quote memory retrieval 0:20:27.280 --> 0:20:31.000 is a powerful method for increasing the perceived validity of 0:20:31.040 --> 0:20:34.560 statements and subsequent illusion of truth, and that the illusion 0:20:34.600 --> 0:20:37.520 of truth is a robust effect that can be observed 0:20:37.600 --> 0:20:41.919 even without directly pulling the factual statements in question. WHOA, 0:20:42.080 --> 0:20:44.120 So this is sort of the same effect, but not 0:20:44.440 --> 0:20:48.080 statements coming in from the outside. Right. So they conducted 0:20:48.080 --> 0:20:51.639 a two fifty seven person study, all individuals from the 0:20:51.720 --> 0:20:55.040 University of Waterloo. So we're, you know, relatively small study, 0:20:55.040 --> 0:20:57.560 and they and they admit that they quote may have 0:20:57.720 --> 0:21:00.679 made it particularly difficult to observe any different is between 0:21:00.720 --> 0:21:04.360 our control condition and our experimental conditions. So as always, 0:21:04.880 --> 0:21:08.760 you know, more studies are required. But uh, here's how 0:21:08.760 --> 0:21:11.600 it shakes out. Quote. If this account is correct, the 0:21:11.640 --> 0:21:15.800 current work demonstrates that information retrieved from memory cannot only 0:21:15.840 --> 0:21:18.720 be viewed as relatively more important than more difficult to 0:21:18.720 --> 0:21:22.240 retrieve information, but can also be viewed as more important 0:21:22.240 --> 0:21:26.560 than information that is explicitly provided. In particular, information that 0:21:26.680 --> 0:21:30.959 is retrieved from memory may actually be more fluently processed 0:21:30.960 --> 0:21:34.800 in general than information that is directly perceived. So the 0:21:34.840 --> 0:21:39.400 idea here is that repetition entailed in memory retrieval need 0:21:39.480 --> 0:21:41.720 not be from an external source. It can be internal. 0:21:41.760 --> 0:21:43.840 In the form of memory retrieval, it is it is 0:21:43.920 --> 0:21:48.159 quote naturally more familiar and fluent than information that is perceived. Wow, 0:21:48.320 --> 0:21:53.280 that that is profound. Actually, like the idea that you 0:21:53.520 --> 0:21:56.960 that your memory is the haze of your memories is 0:21:57.960 --> 0:22:01.239 greater evidence sometimes to your own mind. Then what's in 0:22:01.320 --> 0:22:03.800 front of your eyes right now? Yeah, and it and 0:22:03.840 --> 0:22:06.800 it means that like for the for the lie or 0:22:06.840 --> 0:22:11.680 the the untruth to to resonate, Uh, it only needs 0:22:11.720 --> 0:22:15.040 to be memorable, like something that you'll continually retrieve. Oh yeah, 0:22:15.040 --> 0:22:17.880 and that forms that serves as a form of repetition. Oh. 0:22:17.880 --> 0:22:20.200 And this is so true of so many of these 0:22:20.240 --> 0:22:23.520 lies they get repeated so often in public conversations. Is 0:22:23.560 --> 0:22:28.080 that they're the really memorable, weird outlandish ones that stick around. 0:22:28.800 --> 0:22:31.399 I think about in the last episode, we talked about 0:22:31.400 --> 0:22:35.080