A Study To Remind Us Of What It Means To Be Human
/Generosity appears to be built into our brains.
It’s been a rough few months folks, I’m not going to lie. Over the past 12 weeks, 11 of these Impact Factor episodes have been devoted to Coronavirus stories.
Now, as the nation tries to grapple with a potentially rebounding epidemic as well as with the health and welfare-based sequela of four centuries of racism, my expectations for the future have been growing pretty bleak.
But I read this study, appearing in Science Advances, which gave me a bit of hope.
No, it doesn’t address the health implications of systematic racism. It doesn’t promise a new treatment for COVID-19. What it does is remind us that, fundamentally, humans are creatures who take care of each other.
This paper examines the mystery of prosocial behavior in humans. Prosocial behaviors are those that benefit someone else, even when it hurts ourselves. They are the behaviors that knit society together, that keep us from killing and beating each other, and that keep us wearing masks to protect others even when it is inconvenient to us.
The four major drivers of giving
Humans can be pro-social for a variety of reasons. The authors identify four major drivers:
You may give to someone else, because that person gave to you in the past: direct reciprocity.
You may give to someone else, because a third person had previously given to you: general reciprocity or “paying it forward”.
You may give to someone else because a third person is watching and you’re hoping to get something in the future: reputational giving.
Or you may give to someone else because you saw them give to a third person: that’s rewarding reputation.
Undiscussed but present in the paper is a fifth condition – you may give to someone else for no reason at all. Or maybe just because it feels good. Sometimes this is referred to as “warm glow” altruism.
Many studies have documented that these four conditions are associated with giving, but until now no one has really tried to integrate ALL of them, which is what actually happens in real-life.
Competing interests: Should you give to this man with a bad reputation if he gave to you?
For example, let’s say there is someone who is really mean and selfish – never giving to anyone. But he then gives you a gift. Will you reward that behavior? Will the direct reciprocity trump his bad general reputation?
The experimental design is pretty clever. Internet users from Amazon’s Mechanical Turk were assigned “tokens” that they could give to other individuals (although, in reality these other individuals were artificial to allow the researchers to vary their behavior). They were told that the more tokens they had in the end, the more they would be paid in real dollars – so there was a strong incentive to hoard, but also to make these other “people” happy so that they might give more to you.
The baseline. How much will you give to a random stranger? They will have no opportunity to give back.
Let’s start with the baseline condition. The participants were given ten tokens and asked to give to an anonymous stranger from zero to all 10 of them.
In this situation, there would be no reciprocity. After you gave away tokens, that’s it. On average, individuals gave 5 tokens.
I just want to point out that that’s nice. This is how we behave with NO expectation of payback. This is just warm-glow altruism.
What if the person you give to will then have the opportunity to give back to you?
What if you go second? Well, more to you = more back to them.
Well, in that case people tended to give a bit more, but not much. About 5 and a half tokens.
And if you went second, well the more they gave you, the more you gave back to them, though it was far from a 1:1 correlation.
To account for all the other reasons to give tokens, a large number of variations of the central experiment was performed. In this example, you can see the researchers throwing in permutations of general reciprocity and reputational giving.
The crazy thing is that your brain can actually figure this out.
The interesting finding was that virtually all the modes of altruism were robust to variation in the other modes. The only exception was in the example I gave you earlier – when you are giving to someone after they have given to you, it doesn’t matter what they’ve done for other people.
You could look at these findings cynically – that all this giving is really just to get something back in return, but I don’t see it that way. First of all, we know that even with no expectation of return, individuals were willing to give to random strangers on the internet.
Moreover, the fact that all these modes of giving are operating largely independently tells us something about what it means to be human. We are highly social creatures, capable of integrating vast amounts of social information effortlessly before we engage in a behavior.
This study is missing one thing though. Since the participants were, nominally, giving to strangers on the internet, there was no “in-group” phenomenon. Imagine if the participants were told that the random strangers were republicans or democrats, or black or white. Would the results be the same? Does our drive to help each other translate across the boundaries we erect around ourselves?
I hope it does. A natural experiment to that effect is taking place all around us right now.
This commentary first appeared on medscape.com.