[lutusp]

> "X causes Y" is separate and can be proven separately from "X causes Y by Z mechanism". The former seems to have been proven here.

The experimenters changed the environment and noted a change in behavior. This isn't enough to be able to say which aspect of the change induced the behavioral change, in particular because the study isn't testing a theory, or a cause. It's testing an effect -- the outcome of an environmental change.

This is something that Richard P. Feynman described in some detail in his now-famous talk entitled "Cargo Cult Science." He describes a number of social science studies that went off the rails because the experimenters were too quick to draw conclusions based on experiments the investigators didn't really understand. As a counterpoint he describes one outstanding experiment in which the experimenter went to great lengths to find out why a certain effect took place. Feynman's point was that science requires great care that we don't draw conclusions not sufficiently informed by skepticism.

http://calteches.library.caltech.edu/51/2/CargoCult.htm

To summarize, to be able to claim a cause -- effect relationship, we would need to understand the cause -- we would need a falsifiable theory about why a change took place, not merely that there was a change.

[vertex-four]

So science is incapable of saying, for example, that raising my cup to my lips allows me to drink coffee, since we can't yet describe that from quantum phenomena upwards? What a pointless method of study.

[lutusp]

> What a pointless method of study.

A pointless method of study that's responsible for building the world you see around you, including the computer you're using for this conversation.

In science the problem is not getting the answers, it's being sufficiently skeptical of those who think they already have the answers.

> ... since we can't yet describe that from quantum phenomena upwards?

But that's not true -- we can describe it, but we can't explain it. Literature describes, science explains. Good science resists premature explanation.