James and the Giant Corn Rotating Header Image

Some thoughts on open science and the power of defaults

Tomorrow I’m driving out to St. Louis for four back-to-back related meetings. Genomes to Fields, Corn Breeding, MaizeGDB, and Maize Genetics Conference.

For the 3rd year in a row the Maize Genetics Conference is going to operate under an “opt in” social media policy. Unless people explicitly opt in, attendees are forbidden from discussing talks or posters on social media (presumably this include blogs). Seven years ago, at my second maize genetics conference ever, I would have been in violation of this policy (if it had existed at the time) because I wrote these two posts. I know one of the authors well and he’s never expressed any concern over that post, and, while I’ve only met the second author in passing, I’m guessing she wasn’t bothered by my post since she cited it in her masters defense announcement.

In principle opt-in and opt-out should give identical results, but we know from a number of natural experiments that this is not the case, and that changing between these two can be used as a small nudge to produce socially desirable outcomes.

The first example is organ donation where there is clearly a net benefit to society if people donate their organs, but some fraction of people have strong religious or other objections to doing so.

The researchers discovered that Americans view organ donation in opt-in countries as extraordinary altruism—more like leaving 50% of your estate to charity than leaving 5%. Yet in opt-out countries, what’s extraordinary is not donating your organs—more like skipping your child’s graduation than skipping your child’s baseball game. Americans also liken organ donation in opt-in countries to costly acts like going on a hunger strike, but see organ donation in opt-out countries as less consequential—more like letting someone go ahead in line.

Opt-out countries can achieve donation rates on the order of ~90%, while opt out countries (like the USA) have donation rates closer to 15%. That’s an awful lot of people who aren’t strongly opposed to organ donation but don’t go the extra mile to opt in.

The second example is retirement savings. In the USA many employers offer 401k or 403b plans that let people save for their own retirements in a tax favorable way. Generally people have to file a one page form to start saving money for their own retirement (opt in). But a few companies have started to experiment with opt-out (people have to file the same one page form to avoid saving money for retirement). Again, one can envision good reasons a person might not want to save for retirement: paying off high interest rate debt, family emergencies, medical or genetic conditions that suggest they won’t live long enough to retire, all sorts of things. And those folks are definitely going to opt out. So again, how big a difference does opt-in vs opt-out make in saving for retirement?

Fidelity … says 76 percent of 20- to 24-year-old workers stay in its opt-out plans, compared with 20 percent who sign up for opt-in plans.

That’s 56% of people whose decision about whether to save for their own retirements was determined by the default setting.

Now one could still argue that we don’t want scientific ideas and results from the maize genetics conference to be discussed on social media. (I’d be curious what the arguments for this position might be).

But given the assumption that, unless there are particular circumstances that dictate otherwise, it is to the benefit of our community for ideas to be disseminated and discussed widely, it is clear that opt-in policies leave a lot of potential community benefit unrealized, simple because lots of people who don’t scare strongly one way or another won’t bother to opt in (even though they wouldn’t bother to opt out either).

P.S. Here is the letter which the organizers of the maize genetics conference are currently discussing whether they will allow to be sent to speakers (they sent it out, thanks!), making the case for the benefits of opting in to social media based discussions of their talks during the conference.

P.P.S. There will always be some bad actors in any scientific community. One of my good buddies during my postdoc had a big part of his PhD thesis lifted off a poster and published as an independent paper in another journal before his own paper came out. But it’s important to keep in mind that bad actors aren’t bound by social media use policies. They’re the same folks you see photographing whole posters without permission (a big no-no in my scientific field), or photographing every slide of a talk with their smartphones or tablets (also not okay, and seriously you cannot even be bothered to be more subtle than raising your tablet above the heads of the audience each time a new slide comes up?!)

Leave a Reply

Your email address will not be published. Required fields are marked *