This study however does not stand by itself. There will be more, if there aren't already, and all they are doing is backing up the conventional wisdom about the effects of marijuana.
It definitely makes you forget. I grew up with two parents who smoked it, DAILY. They remembered nothing. Seriously.
In the absence of a randomized trial, the dose-response relationship is one of the most important tools available for teasing out causality. If you can't show that increased dose strongly correlates with increased response, you likely are looking at population variables, not drug/lifestyle effects.
It turns out that almost everything that gets people's hands wringing from fatty foods to chemicals with complicated names have almost flat dose-response curves. The ones that stand out as having strong dose-response curves are alcohol and especially smoking, which meets all of the causal criteria you want to see in a correlative relationship: http://www.skepticalraptor.com/skepticalraptorblog.php/corre...
However, I think that studies like this are particularly vulnerable to this critique of methodology. Any time you assess the performance of a large group and then divide that group into subgroups, you are going to find trends of varying levels of statistical significance.
It is incredibly easy to do. http://www.tylervigen.com/
That is why, in every scope-and-methods class in every college across the country, the various methods for ascertaining causality are demonstrated and taught.
Reading only the linked article and not the study itself, I so far have the impression that none of these were seriously considered - presumably on the basis of ethical treatment of humans subjects. You can't just take a group of people with a particular set of desirable control features (perhaps, in this case, a typically shaped hippocampus) and force-feed them cannabis smoke for three years.
On the other hand, it's just as easy to read this study as meaning that people who performed poorly on the relatively specific requirements of this study seemed to prefer to smoke marijuana. If this study were about serious science, the conclusion might be the utterly uninteresting sky-is-blue headline, "absent-minded people likely to smoke weed."
user has made no other comments thus far.
Wonderful. Another study with no basis for distinguishing causality from correlation.
I bet you can randomly select a subgroup of 18-percent poorer performers and find that they have all sorts of things in common, from diet to exercise to presence or absence of parents - as well as bizarre factors like who was born on a Tuesday.
There are probably dozens of markers that both correlate with these kinds of results and also plausibly suggest an explanation.
> “It is possible that the abnormal brain structures reveal a pre-existing vulnerability to marijuana abuse,” Smith said. “But evidence that the longer the participants were abusing marijuana, the greater the differences in hippocampus shape suggests marijuana may be the cause.”
What?! No it doesn't. It suggests quite the opposite: that people with greater abnormality tend to use marijuana longer.
I suspect that, now that medical cannabis is widely available and somewhat widely used in the treatment of childhood maladies, we may begin to see, in the next decade or so, results based on much sounder science simply because of the availability of higher quality sample groups.