Cognitive Biases and “Doing Your Own Research”, Part 3 – Memory Errors and Biasesby Christopher_NW // April 30, 2013
Finally the third installment of my cognitive bias articles is finished. I must apologize for the delay I have been so busy learning cloning techniques for my new research and dealing with many students, that when I have had free time it has been taken up napping (don’t tell my professor haha)! So in keeping with the theme of what cognitive biases you really need to be aware of when you are doing your own research, particularly citizen-scientists using the internet, I thought I would go over the predominant memory errors and biases that affect all of us. A memory bias is a cognitive process that alters the way, the time, or how we recall memories. As opposed to logical fallacies, memory biases are subtler to pick up on, and as the Wikipedia list in the references shows, there are a lot of them. However, I intend to show you using real examples (as I have done previously), the five memory errors I believe to be both easier to spot, and common in occurrence.
The Bizarreness Effect:
Once you have read and understood what the bizarreness effect is, I am sure you (like myself) will go “ahh, I do that”. The bizarreness effect is the tendency to recollect strange and uncommon information as opposed to more common or ordinary material. Some of the earliest published research on this phenomenon was carried out during the 1980’s by McDaniel and Einstein (no, not that Einstein!). This is just like when you read odd facts about something (i.e. the human body), such as that out bodies are made up of mostly empty space and would fit into cube less than 1/500th of a centimeter per side. I bet most of you will remember this, and go tell your friends, that’s normal. The reason for doing this can be postulated to be purely an evolutionary adaptation, even a massive advantage. Think about it, if we didn’t learn to pick up on strange and bizarre images early in our evolution, we may not have realized that a tiger or another predatory animal was nearby via the identification of their tracks over our own near a campsite (for example).
The problem occurs when it interferes with our logical reasoning due to placing higher stock in the strange things we read or see. One example is when we read of a new patents being granted for what sounds like ridiculous things (humorous examples provided here and here). Some of you might remember the recent furor over Monsanto being granted a patent on “regular” tomato and other vegetable seeds, and you may even have received a petition to sign (I am not hyperlinking here since I don’t want to give crank websites more traffic). Initially reading any of the news that you can google sounds shocking and bizarre, and this is what most people take away, that Monsanto is an evil corporation trying to gain a monopoly over our regular fruit and veg. Now I am not going to defend nor be a proponent of Monsanto, discuss whether or not their alleged monopoly is unethical business practice, and I am not going to discuss patent law, as I am not appropriately qualified to do so. However most people forget that what Monsanto is actually patenting are seed varieties they have crossbred for a specific purpose; to generate a new variety of crop with desirable traits. They are not going around to random gardens claiming to have invented those vegies. As Professor Ross Barnard, Director of the Biotechnology program at the University of Queensland, put it during a discussion on the UQ Molecular Biotechnology Student Club forum-
“Companies have always been able to protect plant varieties with plant breeder’s rights, whether produced by conventional or the new, less hit-and-miss genetic technologies… If you create something new and make a large investment in developing it, other people should not be able to steal it. If you create a piece of music or a work of art, you expect it to be protected. If you want companies to stop investing in production of new plant varieties, then sure, stop them protecting their inventions. But then, please expect your own inventions to be stolen without payment”.
Most of us have excellent hindsight, I mean, who looks back on a mistake and says “I would’ve done that the same way or worse”? We tend to look back and reflect at how we could have done better. Hindsight bias is where we tend to look over past events as being predictable, particularly if we are looking over what someone other than ourselves has done. It is very easy to say how you would have done something differently following someone else’s failure. But think about it; how do you know that you would have done it differently, and what makes your methods better?
An example of this memory error in full effect can be seen during medical malpractice trials. The plaintiffs in such trials are often favoured due to hindsight bias. A study mentioned by Glauser in 2010 of a 1976 Californian case were a man disclosed threats made against a woman named Tatiana Tarasoff to his therapist. The therapist notified the police, but not Ms. Tarasoff, who was subsequently murdered. The study found that those who, 20 years later, knew about the murder after the man spoke to his therapist were likely to put some degree of blame on said therapist for the outcome (despite the notification she gave police).
When doing your own research it is always good to keep hindsight bias in mind, either when you are reading over critical literature/reports or talking to colleagues who have made a potential mistake in the lab (or wherever their work environment is). However it is a fine line; for example, it wouldn’t be hindsight bias if somebody did something a different way to you, without good justification as to why, and it turns out a poor result. If you have experience in an area, you can look at what they did and say, “I would have done it differently”. Knowing whether or not you’re experiencing hindsight bias can be the difference in coming off as arrogant and giving legitimate advice. More importantly, as the case above demonstrates, it can have dire consequences on another’s life.
Illusion of Truth Effect:
In general, people tend to identify statements they have heard previously as being more truthful, irrespective of the validity of the statement. This is termed the illusion of truth effect, and it’s one of my pet hates. A lot of nonsense gets spread around and handed down over the ages, and the internet makes the spread of misinformation so much easier. This can range from the relatively innocuous such as the old wives tale that eating carrots improves your vision (it doesn’t, but the vitamin A can help maintain healthy eyes), to pseudoscience nonsense (I could use the teaching of Creationism as a science for the example, but I don’t want to offend anyone), all the way to the down right dangerous.
An example of something many people hear frequently that has led to the spread of dangerous misinformation is that vaccines contain deadly mercury. Let me make two things very clear; firstly, elemental mercury and organic methyl mercury (these can be harmful to the body, depending on the dose and frequency of exposure) are different to the molecular mercury present in a handful of vaccines. Secondly, due to the illusion of truth effect causing a public outcry, thiomerosol, the type of molecular mercury compound found in some vaccines to prevent microbe growth, has been removed from many vaccines despite the lack of evidence that it is harmful (in fact removing it may be a worse option), and at great cost to the companies who have to reformulate their products.
Here you can see that thiomerosol on the top possesses a distinct
chemical structure compared to methylmercury on the bottom.
This also serves as a good example of a situation where multiple logical fallacies and cognitive biases come into play, including the bizarreness effect, illusion of truth, the Dunning-Kruger and anchoring effects, in addition to attentional bias, among others. Dr Diekema last year wrote in the New England Journal of Medicine that the misinformation being spread about vaccines contributes to increased risk, and that in California during 2010, there were more than 9000 cases of pertussis (a vaccine-preventable infection), which was the highest figure in that state since 1947. I hope I have been able to succinctly demonstrate how deadly this memory error can be.
Don’t you just get frustrated when you see politicians, particularly in the run up to an election, touting all of the successful things they have done, but neglect to touch on all the broken promises and failed policies? Welcome to the self-serving bias, a type of cognitive memory bias I’m sure you have all been privy to in one way or another, whereby we perceive ourselves responsible for positive outcomes, but not negative ones. Now in politics it makes sense that the candidate up for re-election wants withholds the negative; they have a vested interest in getting you to vote for them. If they did otherwise, it would be campaign suicide (unless you are Silvio Berlusconi, I just cannot keep up with that man).
The best example of self-serving bias in medicine that comes to mind is from the hugely discredited former doctor, Andrew Wakefield (who I have briefly mentioned before and winner of the lifetime achievement in quackery award. ). Despite being shamed in most of the medical, scientific, journalistic and ration-thinking community, he still insists on standing up and saying his scaremongering about MMR vaccines is not to blame for the recent measles outbreak in Swansea, UK. He appears to refuse to acknowledge his role in the decline of childhood vaccinations given, and grandstands and gloats wherever possible.
The Wakefield example shows the danger of the self-serving bias when not quickly called out by others. When you are going through online information and you see people making claims about what they are and are not responsible for, try to assess if there are any conflicts of interest (such as Wakefield trying to get a single dose vaccine out into the market whilst discrediting the triple shot MMR vaccine). Doing this will help you be aware of any self-serving bias present.
Von Restorff Effect:
The Von Restorff effect was named after the German paediatrician and psychiatrist, Dr Hedwig von Restorff, and tells us that items that stand out are more likely to be remembered than other items (which of course makes sense as an evolutionary adaptation). Originally she described that when you isolate an item from a homogenous background, it facilitates the learning of that item. You can think of this by understanding that it would be easier to read and understand one book at a time, rather than multiple.
In terms of a memory bias, and how this can make people remember inaccurate information, I will use the example of sensational scientific headlines in the media. When members of the public (or lazy students) read headlines like “Scientists closer to declaring discovery the ‘God Particle” or “Temporary tattoos could make electronic telepathy and telekinesis possible”, and do not apply a healthy dose of skepticism or further reading, they tend to come away with the message of the headline rather than the actual story (although, many times in the media the story has been sensationalized). Also, people who then find out that the big claims in these stories are either far away from being reality (how many times have we seen this with potential cancer cure stories?) or that they are plain false, they are disappointed, even though they should be happy that progress is being made via research.
Given that Dr Steve Novella has already gone into detail about the about tattoo-telepathy news story in his neurologica blog (see references), I thought I would clear up the misconceptions about the so-called God particle, the Higgs-Boson. The amazing discovery by Aad et al. in 2012 was pretty much superseded by all the nonsense being reported in the media, causing mass public confusion and misunderstanding. In fact there was so much nonsense, Cracked even ran an article about the misunderstandings. The Higgs-Boson is a subatomic particle that helps gives mass to other particles (this is the simplest explanation). It does not prove anything about a God or God-like being. Some people were even trying to quote scripture (Colossians 1:17) as proof of the Higgs particle in the Bible. No, the term came about because of the combination of a misunderstanding, trying to express how important this particle is to particle physics, and book sales to the public who generally know very little physics beyond “if I jump off a cliff I fall due to gravity”. But due to the Von Restorff effect (and stupidity), people are still running with bunkum information about the Higgs particle, with one Christian website blogger using the headline “God Particle: Proof of God in the Smallest Parts”, despite not actually saying anything about how this could be the case. So please, when you’re doing your own research, allow alarm bells to ring when you read any sensational headline and have a healthy dose of Ol’ Doc Skeptics anti-snakeoil.
So that’s it for the three part series on cognitive biases to look out for when doing your own research. There are many more of course, and do not be so naïve to think that you can avoid being guilty of them; just try to limit their occurrence and try to spot them around you (it won’t be too hard now I hope). Next time I will probably have a chat about malaria, or fraud in science and medicine or even discuss some of the potentially dangerous pseudoscience around. Happy hunting.
Edited by Stephanie Haggarty
McDaniel MA, and Einstein GO. Bizarre imagery as an effective memory: The importance of distinctiveness. J Exp Psychol Learn Mem Cogn 1986, 12:54-65.
McDaniel MA et al. The bizarreness effect: it’s not surprising, it’s complex. J Exp Psychol Learn Mem Cogn 1995, 21:2:422-35.
Glauser J. Legal Notes: Of Hindsight, Juries, and Expert Witnesses. Emergency Med News 2010, 32:8:17-22.
Gilbert C and Foster A. Childhood blindness in the context of VISION 2020 — The Right to Sight. Bull World Health Organ 2001, 79:3.
Diekema DS. Improving childhood vaccination rates. N Engl J Med 2012, 366:391-93.
Aad G et al. Observation of a new particle in the search for the Standard Model Higgs boson with the ATLAS detector at the LHC. Physics Letters B 2012, 716:1:1-29