Bias

“Hindsight is always twenty-twenty.” Billy Wilder

 

TASK AT HAND: This week I’m thinking about hindsight bias. In my opinion, this is one of the more comical fallacies for the degree of rationalization and inaccuracy inherent in its fabric. The hindsight bias, also known as the “knew it all along” effect, occurs when after an event has happened, we purport that we correctly predicted the outcome. However, the fallacy exists because there is no objective evidence for us having predicted the event.

Hindsight bias, commonly referred to as creeping determinism, is the basis for that feeling we get, “I knew it all long…”. For example, a patient presents with loss of vision and the eye physician diagnoses a retinal detachment. In summary, the doctor concludes, “I knew it! I had a feeling it was a retinal detachment”. Or take last year’s dramatic Super Bowl comeback win by the New England Patriots over the Atlanta Falcons. In various debriefs, you heard fans, players and commentators alike with phrases like: “I knew we were going to come back and win” or “I knew we could do it”. In reality, there was no way possible to predict this. Sports and medical diagnoses are areas where the hindsight bias has been extensively studied. You can find a nice summary here by Neal J. Roese of Northwestern University. Hindsight bias is a decision trap because it falsely supports our ability to predict events that cannot be predicted.

In an excellent commentary, from the September 2012 issue of Perspectives on Psychological Science, psychological scientists Neal Roese and Kathleen Vohs (you can find the article here) show that we bias and selectively recall information that confirms what we know to be true; then, we synthesize a narrative to describe this truth. If our brain has an easy time creating this narrative, then we interpret it to mean that the outcome must have been predictable and we identify with having correctly predicted it.

How does hindsight bias hurt us?

Hindsight bias is troublesome because it limits our ability for introspection. We all have a need for closure and a strong innate desire to make sense of events. Whether they be close relationships, world politics or natural disasters, we strongly want to establish some order and cause to events. This benefits our view of ourselves and the world. However, much – if not everything – lacks any sense or logic. Randomness runs rampant! Hindsight bias limits our ability to learn from events because, if we feel we correctly predicted them, then it follows that we must be in tune with the decision-making process  – which is untrue if this fallacy is present. Contrastingly, with honesty, as witnesses to our errors and miscalculations, we gain valuable insight and maturity in how we come to terms with external stimuli.

 

MEDICINE & MACULA: I was in Barcelona, Spain this week for the EURETINA 2017 Congress.

On Friday (8 September 2017), I presented a talk on our recent findings and techniques for proliferative vitreoretinopathy (PVR). My talk, Predictive factors for proliferative vitreoretinopathy formation after uncomplicated primary retinal detachment repair (David RP Almeida, Kunyong Xu, Eric K Chin & D Wilkin Parke III) looked at predictive tools for patients who develop this complex condition.

Many thanks for all the interest and international support!

 

GRATIS: “You can’t operate by hindsight.” -Max Baucus

 

My best to you,

David Almeida

david@davidalmeidamd.com

Read more

sss-bias-opinion

“Every piece of data is biased. Every argument has opinion.”

 

TASK AT HAND: This week I’m thinking about the effects of bias and opinion. Analysis after the US election show that “fake news” – stories that are false but presented in a truthful manner (e.g., newspaper article format) so as meant to deceive – outperformed legitimate news stories on social media. We now occupy the post-truth economy of thought. In this state, opinion and argument are given the same credence as fact and truth.

Recently, the Oxford Dictionary announced that “post-truth” is its 2016 word of the year. It defined it as ‘relating to or denoting circumstances in which objective facts are less influential in shaping public opinion than appeals to emotion and personal belief.’ Simply put, If you believe something strongly enough, talk loudly enough about it, and can connect with someone emotionally with your argument, then it impacts others like truth or fact.

As a scientist, I am vehemently against this idea that opinion can be substantiated for fact. Truth requires evidence and logic. Truth should be void of bias and opinion.

“Truth requires evidence and logic. Truth should be void of bias and opinion.”

I have heard many times, “what’s the problem with voicing your opinion?” There is nothing wrong with expressing your opinion – I encourage this! However, please don’t confuse the expression of worthy words versus baseless chatter. Your argument should still be based on truth and constructed in a logical manner.

The problem of repeating nonsense over and over is related to how our brains form memories. Studies show that the more often a message is repeated, the more likely we are to remember it. This effect is called fluent retrieval. However, our brains then erroneously extrapolate that, what we can remember easily, must be true (Inferring facts from fiction: reading correct and incorrect information affects memory for related information. Memory 2012 Jul;20(5):487-98; you can find the full study here). The effect is that if you repeat a lie often enough, it starts to feel like truth.

But how to uncover bias and opinion? First, assume every piece of data is biased and every argument, whether it be in a newspaper article, social media post or formal communication, has opinion. You are a detective and must identify the bias and opinion in everything you consume.

“You must identify the bias and opinion in everything you consume.”

You can minimize bias and opinion by sticking to trusted reporting. However, this is not enough. In science and medicine, we have peer-reviewed literature which is considered the benchmark for bias-free communication. Peer-reviewed studies have experts and thought leaders review the work in question to ensure it is scientifically sound before being published. Having personally published over 100 papers, I can tell you that even this process can have bias. Reviewers have personal and professional biases and humans operate poorly in recognizing their own biases.

Second, when developing an argument, use multiple sources. Never stick to one reference and never rely solely on one authority. Attempt to survey as many respectable sources as possible when trying to come to a conclusion. This synthesis of thought is the crux of thinking for yourself because it forces you to take multiple vantage points and create a unique one for yourself. This is hard work and the main hurdle to overcoming herd mentality.

Finally, refute and reject frequently.Blind belief in authority is the greatest enemy of truth” (Albert Einstein). This is not a call for anarchy but a reprisal for individual thought. Authority, or that which is accepted as truth, needs to be questioned. Force yourself to formulate independent thoughts as often as possible. This is not your brain’s default mode so you have to work at it. The goal is to have a society of independent thinkers that question truisms and myths alike. A culture that challenges arguments without fact and calls out opinions lacking logic.

How to uncover bias and opinion:

1.    Assume every piece of data is biased and every argument has opinion.

2.    Never stick to only one reference. Use multiple sources.

3.    Refute and reject regularly.

 

“The goal is to have a society of independent thinkers that question truisms and myths alike. A culture that challenges arguments without fact and calls out opinions lacking logic.”

 

MEDICINE & MACULA: One of our recently featured publications on the ongoing debate of auto-antibodies.

osli-auto-ab

Check out the paper, Positive Auto-Antibody Activity With Retina and Optic Nerve in Smokers and Non-Smokers: The Controversy Continues, published in Ophthalmic Surgery, Lasers and Imaging Retina (OSLI Retina). You can find the study here.

osli-retina

 

GRATIS: “If you tell the truth, you don’t have to remember anything.” ―Mark Twain

 

My best to you,

David Almeida

david@davidalmeidamd.com

Read more