For the moment.
The situation in Europe is very interesting. There is a strong atheist movement, yet catholics seem to be on a comeback lately, and the mostly muslim immigrant population growing too - and at an even faster pace.
If religion says "go forth and populate the earth", and if atheist have less than 2.1 children, how long until they get outnumber and forced to chance their allegiance by the religous - at gunpoint if necessary? [nobody expects the spanish inquisition :-)]
So I would not base any conclusion on a punctual observation when you have partial derivative pointing to different conclusion in a longer timeframe.
Counter example : if large-scale societies survive fine without religion, how come there are so few of them ? Why do they have a tendency disappear in history, and be replaced by religious societies?
Also, the method of the prisoner game is questionable. The article says the amount was not insubstantial - around several days of work, but does not explain how it was chosen or how it was tested.
I would be very interested to know about the price elasticity of the acceptance rate and the price elascticity of the percentage offered.
I mean, do people in this group have a constant acceptance (fully inelastic - they always accept) or can you get to a point where they refuse because they think it's unfair?
Here's a quick example - go to any fast food join, eat normally, then when you have to leave, prepare the exact amount of cash in one side, and the tip in the other side of the table, and say to server the food was not to your taste, therefore you are leaving only a one cent tip (or a dime)
See if they take the free money, or if they feel so insulted that they refuse. Rince and repeat until you figure the amount they will accept without feeling insulted.
So I really wonder if there could be a cash amount where the remote people from this tribe would react just the same as all of us - which would just point to a calibration problem.