N=1 Intoxication Quantification

As I’m sipping a glass of champagne while finishing up some coding for the night (yes, I’m one of those people that don’t think you need a special occasion to drink champagne) an interesting self experiment came to me.  I’ve heard of and seen video of people who are put in driving simulators to show the difference between difference levels of blood alcohol levels and the impact on driving performance.  I’ve sadly seen the direct effect on people as well.  Wouldn’t it be interesting to try to do a direct measurement of this in a safe way?

Alcohol dulls the senses in terms of cognitive ability, response time, and coordination.  Sure there is potentially a “sweet spot” where it can reduce over-thinking without diminishing things to the point that you can have some creative progress or enhanced abilities but really it’s mostly a deadening effect not an amplification effect.  If we had a mechanism for measuring those three things in a safe environment it’d be possible to measure for an individual their BAC to impairment function.  Is there such a thing we could use? In fact there is!

While I have totally fallen off the wagon of my five metrics, yet I’m still tracking it, there are two things that I have started doing daily very consistently: language training in DuoLingo and brain training using Lumosity.  While the efficacy of either of these systems with respect to their marketing schtick is still debatable there is no question that it provides a direct scoring mechanism across a range of mental metrics in a repeatable way.

My thought is that an experiment of this sort would take on the following composition.  Over a few days there would be some benchmark runs where the same set of “games” are played over 3-4 hours every 30 minutes.  There will be no alcohol consumed during these control experiment so that we can test whether there is an increase in performance over the sampling period.  With these benchmarks we can then go on to the impairment test.  With a breathalyser in hand a person would start the sample period with a BAC of 0.0 and then gradually consume beverages over the 3-4 hours repeating the 30 minute test intervals but this time measuring BAC right before each test.  We should probably perform this several times as well.

We then can compare which BAC that true impairment occurs and on which metric for the individual.  This probably will show that while people think they can function well intoxicated at or a bit above the legal limit that in reality they are substantially impaired well below that limit.  I’m a lightweight and I’ve done something notionally like this when I was just out of college.  I found that I noticeably off as low as 0.04, say it felt like if you had a head cold level of impairment.  This test could also be used to calibrate people’s impairment from other things that aren’t otherwise testable like marijuana or sleep deprivation, but I don’t know how to quantify the effective levels in the system for something like that.

Interesting Gedanken experiment, maybe an actual one some day…