Regarding the Stanford study:
The researchers caution that the study was conducted with a small group and needs to be replicated with girls as well as with more boys before they can be certain they have found a universal neurobiological marker for the disorder. They also reported that they tried the experiment on three other boys who had not been diagnosed with ADD but who had siblings with the diagnosis. Two of the three siblings of ADD boys showed some Ritalin enhancement of the striatum, as did their affected brothers.
Because of the widespread interest in ADD, the researchers also were careful to point out other limitations of their study....It also did not attempt to investigate possible changes in brain activation that might occur with the drug over time. The ADD boys did show a different brain activation level when not on the drug, but all of them had been taking Ritalin previously as part of their ongoing treatment.
So in other words, the guy in the Times pointed to exactly these limitations as being potentially serious flaws in the logic around Ritalin use.
I'm a scientist and one my greatest frustrations with my field is the tendency to draw conclusions from too little data. It has been proven again and again that a too-small sample size can lead to false-positive and false-negative findings. IMO, a study like this shouldn't be published until the authors have tested a much larger sample size.
Put another way, this is why the FDA insists that drug and device companies run Phase III clinical trials with thousands of people.
