[extropy-chat] Game theory of common cold

[Anders Sandberg]

Rafal Smigrodzki said:
> I think we need to examine the consequences of improving the prevention of
> cold transmission over longer periods of time. It appears that a cold
> epidemic ends only when the fraction of never-infected, and therefore
> susceptible hosts, drops to some low level. Diminishing the transmission
> of
> the cold virus by changing behavior of hosts, rather than vaccination,
> would
> result in maintenance of large numbers of susceptible hosts, and over many
> years the number of multiple-susceptible hosts would increase.

Although I think Robert is right about colds, your argument seems to have
merit against some other diseases like Measels in children.

I ran some simulations and introducing immunity seems to make the system
more altruistic! The reason is partly that disease incidence is reduced,
so there are fewer epidemics. This in turn makes the benefit of being a
worker less, and there are more people staying at home. Increasing the
infectiousness of the disease still requires quite extreme levels before
we get an altruism breakdown. Increasing the base probability of the
disease had little effect, since immune people still don't get it.

> If so, the system might become
> metastable - a large population of susceptible hosts and many versions of
> viruses, kept apart only by the behavioral modification. What if at some
> point a critical mass is reached, and a pandemic develops even in the
> presence of the behavioral modification?

This would require a behavioral modification that was extremely efficient,
able to maintain isolation against multiple flu viruses or something
similar over the span of many years. This is unlikely to work perfectly,
and we get enough local breakdown to introduce immunity in the population.

Rare and serious diseases still can cause wipeouts, but they are unlikely
to be manageable by the system I described. People should certainly stay
away from work if they have the bubonic plague, but there is little social
and game-theoretic evolution of the behavior (but likely a bit of common
sense limiting work while infected by the plague).


> Also, the idea that specific persons can be punished for infecting others
> implies the ability to detect these persons, either beforehand, or after
> the
> deed - and therefore, civil liability, the law of torts, contract law,
> private exclusionary practices, and other perfectly non-coercive means can
> be fruitfully employed. That they are not used, is IMO due to the marginal
> disutility of the common cold (compared to what I see on the ward
> sometimes), and the possibly substantial costs of preventing its
> transmission (aside from vaccinations).

Within a few years I predict we will have bacterial/viral detector chips
in hospitals, then in the home and finally in the cellphone. Imagine the
public health effects of being able to scan the environment for flu. Or to
prove with a high likeliehood that Mr Smith was the source of your cold
given strain similarities.


-- 
Anders Sandberg
http://www.nada.kth.se/~asa
http://www.aleph.se/andart/

The sum of human knowledge sounds nice. But I want more.