Subjective beliefs can offer a path to decreasing prediction error. This is the power of Bayes' Theorem. Plus, our flawed brains may be natural Bayesian operators.
If anyone is interested in looking at the next layer in how our brains may be functioning, I would highly recommend taking a look at Karl Friston's Free Energy Principle / Active Inference. It is not only a theory of brain function, but possibly even that of all of life. I have written a quick intro to it here: https://medium.com/@vinbhalerao/the-free-energy-principle-the-ultimate-theory-of-life-cade09130a06
Thanks for that review. I think that the idea that the brain act like a Bayesian optimal decision maker in many situations is a useful working assumption. More useful in my opinion that lots of alternatives. Having said that, a purely Bayesian approach would be difficult because the world contains surprises. Savage's foundations for Bayesianism were explicitly for a "small world" where there are no unknown unknowns.
I liked the book. History isn't that relevant to me, but I suppose it was interesting and makes for a good book. I am not sure about the Bayesian brain stuff. It sounded interesting. I can't remember but it seems like people have evolutionarily wired cognitive biases that don't seem to update with their exposure to the world--did he talk about that?
I still don't feel like I have been persuaded of abandoning the frequentist paradigm (NHST). Maybe that argument would be too technical. They seem fine with me when used correctly. The real problem is researches and researcher degrees of freedom/p-hacking etc. If researchers are willing to be misleading (intentionally or unintentionally), we still have major issues, no? I can see p-values and effect sizes and analyze this in a Bayesian way considering my prior knowledge. Not sure. What do you think?
So I think Chivers does a good job of integrating a rationale brain hypothesis with received understanding from the social psych literature on cognitive biases. Chivers of course critiques some of the cognitive bias findings like those concerning priming, but acknowledges there are some biases. The biases that do exist tend to be worse in abstract or unrealistic scenarios. Sometimes they completely resolve in real-world settings. Other times the biases are likely adaptive and are in some respects still Bayesian.
He does acknowledge that a Bayesian framework for science is likely to create space for more disagreement over similar evidence because of the influence of priors. To be honest, I am all for that.
He doesn't advocate a complete abandonment of NHST. He just argues that Bayes would likely be the better default for a lot of our scientific questions. I think this is probably true for most clinical trials of drugs.
My review did gloss over the Freq vs Bayes debate. I think it's more of a case by case thing. If we can be confident about getting good priors then Bayes is likely the superior amway to go. If we're true in unknown territory and we can collect a lot of high quality data at once then Frequentist approaches are superior.
He talked Lakens and had his comments. Lakens seems perfectly reasonable as an "arch-frequentist." That's just been my impression. Not sure what I'm missing.
Very nice summary of important topic in an engaging way that all should have understanding and exposure to (not just those in medical and scientific world) Too bad many didn’t apply these concepts to risk benefit analysis to new platform Covid-19 injections. Over 7% had to seek medical attention after injection
after injection and there were real risks for serious AE and even death which did not meet R/B (risk/benefit) equation for a therapy that didn’t prevent infection OR transmission and yet much of the published “data” parroted “safe and effective.” Manipulation of scientific data is far too prevalent and many non educated person were able to better discern R/B than “more educated” ie indoctrinated persons. Sadly emotional manipulation and social pressure trumped basic mathematical formulations.
Nice intro to a very important topic.
If anyone is interested in looking at the next layer in how our brains may be functioning, I would highly recommend taking a look at Karl Friston's Free Energy Principle / Active Inference. It is not only a theory of brain function, but possibly even that of all of life. I have written a quick intro to it here: https://medium.com/@vinbhalerao/the-free-energy-principle-the-ultimate-theory-of-life-cade09130a06
Thanks for sharing. I don't mention this in my review but Chivers' does heavily cite and discuss Friston in his chapter on Bayes and the brain.
Thanks for that review. I think that the idea that the brain act like a Bayesian optimal decision maker in many situations is a useful working assumption. More useful in my opinion that lots of alternatives. Having said that, a purely Bayesian approach would be difficult because the world contains surprises. Savage's foundations for Bayesianism were explicitly for a "small world" where there are no unknown unknowns.
I liked the book. History isn't that relevant to me, but I suppose it was interesting and makes for a good book. I am not sure about the Bayesian brain stuff. It sounded interesting. I can't remember but it seems like people have evolutionarily wired cognitive biases that don't seem to update with their exposure to the world--did he talk about that?
I still don't feel like I have been persuaded of abandoning the frequentist paradigm (NHST). Maybe that argument would be too technical. They seem fine with me when used correctly. The real problem is researches and researcher degrees of freedom/p-hacking etc. If researchers are willing to be misleading (intentionally or unintentionally), we still have major issues, no? I can see p-values and effect sizes and analyze this in a Bayesian way considering my prior knowledge. Not sure. What do you think?
So I think Chivers does a good job of integrating a rationale brain hypothesis with received understanding from the social psych literature on cognitive biases. Chivers of course critiques some of the cognitive bias findings like those concerning priming, but acknowledges there are some biases. The biases that do exist tend to be worse in abstract or unrealistic scenarios. Sometimes they completely resolve in real-world settings. Other times the biases are likely adaptive and are in some respects still Bayesian.
He does acknowledge that a Bayesian framework for science is likely to create space for more disagreement over similar evidence because of the influence of priors. To be honest, I am all for that.
He doesn't advocate a complete abandonment of NHST. He just argues that Bayes would likely be the better default for a lot of our scientific questions. I think this is probably true for most clinical trials of drugs.
My review did gloss over the Freq vs Bayes debate. I think it's more of a case by case thing. If we can be confident about getting good priors then Bayes is likely the superior amway to go. If we're true in unknown territory and we can collect a lot of high quality data at once then Frequentist approaches are superior.
He talked Lakens and had his comments. Lakens seems perfectly reasonable as an "arch-frequentist." That's just been my impression. Not sure what I'm missing.
Very nice summary of important topic in an engaging way that all should have understanding and exposure to (not just those in medical and scientific world) Too bad many didn’t apply these concepts to risk benefit analysis to new platform Covid-19 injections. Over 7% had to seek medical attention after injection
https://www.prnewswire.com/news-releases/cdcs-covid-19-vaccine-v-safe-data-released-pursuant-to-court-order-301639584.html
after injection and there were real risks for serious AE and even death which did not meet R/B (risk/benefit) equation for a therapy that didn’t prevent infection OR transmission and yet much of the published “data” parroted “safe and effective.” Manipulation of scientific data is far too prevalent and many non educated person were able to better discern R/B than “more educated” ie indoctrinated persons. Sadly emotional manipulation and social pressure trumped basic mathematical formulations.
One minor criticism-I wouldn’t recommend using google and Wikipedia as an authoritative source with biased algorithms