Thread by @DellAnnaLuca

PLACEBOS, NOCEBOS, AND EMERGENCY RESPONSE (thread)

1/ Placebos are not only sugar pills and acupuncture. They are any suggestion that it's safe to invest in getting better, paraphrasing @rorysutherland.

Why do we need them? Why can't we always get better without them?

2/ Case study: Humphrey studied hamsters injected with bacteria. Their immune system reacts more strongly if they're on an artificial day/night cycle that suggests it’s summer.

Why? In winter you need energy to survive, better not waste it on a immune response if unnecessary.

3/ Similarly, in a war, better keep the energy to fight even if you're ill.

Placebos are suggestions it's time to invest in getting better.
Nocebos are suggestions it's NOT time to do so (because of other immediate risks).

What's the relationships with the current epidemic?

4/ It was clear since January that governments needed to act, closing borders and restricting mass gatherings.

These are costly measures, and governments needed to know it was safe for them to deploy this "immune response".

Instead they met public opinion that's "just a flu".

5/ Governments needed a placebo, people telling them it's safe to act proactively.

Instead they go a nocebo, experts saying "not to panic", inhibiting the immune response the governments could have produced.

Result: we got sick.

6/ Summary so far: we needed a fast immune response from the government but it didn't happen, because experts acted as a nocebo, suggesting it's not politically safe for governments to take costly measures such as closing borders at least partially.

However, …

7/ As @naval correctly pointed out, nocebos are a necessary component. Otherwise, we would overreact to any small risk, which would be harmful as well. Autoimmune responses are bad.

How does nature solve this problem?

8/ In any group of people, personalities are always distributed across a wide spectrum. Braver and risk-averse. Introvert and extrovert.

Natural selection can move the average, but keeps the variance.

Because variance is good; it allows for broader adaptability.

9/ What went wrong of late?

As @EricRWeinstein said, for 30-50 years we systematically selected for the wrong kind of experts.

We only selected for nocebos (ppl blocking immune reactions to risks) and didn't select for placebos (ppl saying it's save to invest in risk mgmt).

10/ If we want to avoid future crises like this, we should erect statues as soon as the pandemic ends, while the scar is fresh.

Statues:
- In front of hospitals, to excess bed capacity
- In front of stock exchanges, against bailouts
- In squares, on tail risks

11/ It's easy to blame those who said "don't panic", but to solve the long-term problem, we need to create a system that selects for the right kinds of experts or, even better, for a system that does not need experts.

Rules that guarantee redundancy, for example.

12/ Also, any system should have circuit breakers – alternative regulations kicking in during emergencies.

Example:
- Stronger, slower regulations for medical approval in normal times.
- Faster regulations for medical approval in times of emergencies.

luca-dellanna.com/circuit-breake…

13/ The point is, it is an illusion that we can only be fully risk-averse or fully optimized. It's possible to (almost) optimize for normal times and still have systems in place for emergencies.

Just like you can have elevators and emergency staircases.

14/ More on placebos, nocebos and chronic pain in my essay
"Pain is not a signal of damage but of vulnerability", luca-dellanna.com/pain-signal-vu…

15/ An addition to tweet #12: every system should have alternative regulations for emergencies.

This is because adapting a system to a single point of the spectrum is necessarily inefficient and will lead to some people saying "let's optimize".

Instead…

16/ …instead circuit breakers allow for the systems to optimize during good times in a way that keeps some redundancy and ability to react to emergencies.

Avoiding the situation we got to today.