Another excerpt from the upcoming new edition of my Med Guide

Stimulants
The psychostimulants, as a class, have been around for a long time. These were some of the first drugs to be used in psychiatry, and so physicians feel fairly comfortable with these medications--some would say too comfortable. In children, stimulants are by far the most commonly prescribed psychoactive medications, primarily for the treatment of attention deficit/hyperactivity disorder (ADHD). Stimulants are also used, with questionable effectiveness, as appetite suppressants to promote weight loss. They can be valuable for people with narcolepsy, since their stimulating properties can help narcoleptics stay awake and function through the day. Some adults with ADHD also seem to benefit from stimulant medication.
Currently, there is a bit of controversy about the frequency with which stimulants are used; many people (this author included) feel that these medications are overprescribed and that ADHD is overdiagnosed. It is likely that the ability to attend, as a biological trait, exists on a continuum: some individuals are designed to focus narrowly on the task at hand, while others are always open to input from the environment, and are thus "easily distractable". It is also likely that many children and adults nowadays are labeled as having ADHD when they simply are normal individuals at one end of the spectrum. Viewed this way, medicating these individuals constitutes cosmetic psychopharmacology.
This argument certainly does not apply to Fragile X, however. Children with Fragile X have a well-defined, single-gene disorder which causes attentional deficits along with other characteristic symptoms. They (along with a fair number of children who really do have neurologically-based ADHD) have symptoms which offer a clear-cut rationale for the use of psychostimulants to enhance attention. The only reasonable question is whether these medications work for children with Fragile X.
To answer this question we must first consider how these drugs work. Although the mechanism of action is by no means fully understood, the conventional wisdom is that psychostimulants work by promoting release of certain neurotransmitters, especially dopamine (but also norepinephrine, as well as other things secreted along with them). The primary effect of psychostimulants, the enhancement of attention and concentration, is thought to result from the increased release of dopamine in the frontal areas of the brain. But, of course, the drug is present in other areas, too, and exerts effects there as well. The areas of the brain which regulate level of arousal, blood pressure, heart rate and other "autonomic" functions are also stimulated, while the area controlling appetite is inhibited. Under normal circumstances, most people are much more sensitive to the primary effect of facilitating dopaminergic transmission in the frontal lobes, and at most therapeutic doses will experience an enhancement of attention, concentration, and overall cognitive performance (which is why these medications were initially touted as "smart drugs").
As attention and concentration (referred to by some as "focus") increase with increasing doses of a stimulant, physical activity tends to decline, accounting for the paradoxical decrease in hyperactivity (which is ususally the objective of treatment). However, the primary and secondary effects of stimulants, enhancement of attention and reduction of hyperactivity, are known to occur at different dosages. Lower doses of a stimulant are likely to enhance attention optimally and improve cognitive performance, but may not control hyperactivity. Higher doses are likely to reduce hyperactivity, but may actually result in "overfocus", in which attention is focused so narrowly that actual cognitive performance declines. Still higher doses will cause psychiatric symptoms in virtually anyone, including irritability, aggression, anxiety, agitation, paranoia, or hallucinations. Fortunately, for most people the dose required to cause trouble is much higher than usual therapeutic doses. However, the situation is somewhat different for Fragile X individuals.
Fragile X predisposes one to anxiety, aggression, and agitation. On an intuitive level, it seems obvious that care should be taken with any substance which could aggravate these. On a biochemical level, psychostimulants are "sympathomimetic": they mimic the effects of adrenaline in the central nervous system, heightening arousal as well as increasing heart rate and blood pressure. Since Fragile X individuals often have problems with hyperarousal, stimulants may make matters worse in some cases. Many Fragile X individuals are able to achieve significant improvements in attention and cognitive performance with low doses of stimulants, though, and any potential adverse effects are readily reversible should they arise. Therefore, a trial of a stimulant is rational and safe for a Fragile X individual with particular attentional problems, but should be done with caution. Dosages should be relatively low, and it cannot be expected that significant reduction of hyperactivity will occur, at least compared to the sometimes dramatic response seen in "garden variety" ADHD. Careful monitoring for emergence or exacerbation of anxiety or aggression must occur throughout treatment. Many Fragile X parents are not informed of this risk, are unaware of the connection between stimulants and worsening of aggression or anxiety, and therefore continue administering the medication even when adverse psychiatric side-effects occur--despite the fact that these effects readily reverse upon discontinuation of the drug.
Psychostimulant medications can cause uncommon, but serious, medical problems. Most worrisome is the development of motor tics. This can start as a subtle, almost undetectable twitch, and progress to severe involuntary muscle movement. It usually stops soon after the stimulant is discontinued or the dosage decreased, but sometimes is frighteningly persistent. The key is to catch the tics early on; more persistent tics usually occur following longer treatment in which early signs were ignored. This side-effect is usually dose related, so reducing the dose can be helpful and allow for uninterupted treatment. Also, since most Fragile X children are treated with lower doses of stimulants, this may be less likely to occur in the first place (there are no reliable statistics on the frequency of this side-effect in Fragile X). Tics can also be treated with clonidine if they persist, or if the clinical judgment is made not to interrupt stimulant therapy.
Since they can be potent appetite suppressants, psychostimulants can cause some growth delays during long-term administration. Several studies have shown, however, that children will eventually catch up, even if the medication is continued. Often, "drug holidays" are taken during non-critical times (such as summer vacation) to expedite this process. In any case, growth charts should be carefully monitored for all children on stimulant medications, and significant growth delay is an appropriate reason for discontinuing the medication.
Specific side effects and their medical management are discussed in the individual reviews of medications.

Update 2008: The caveats concerning stimulants noted in this book have proven to be well founded, and just as frequently ignored as ever. Stimulants are nearly irresistable for all concerned, offering the promise of an instant fix for the disabling inattentiveness and the disruptive hyperactivity seen in nearly all children with fragile X. Teachers love stimulants because they help so many kids participate in class. Parents love them because they start working right away, and because they want their kids to get the most out of school. Pediatricians love them because they’re easy to prescribe and have a long safety record. The only problem is that they don’t work very well in kids with fragile X! At least half of all stimulant trials in fragile X kids end abruptly because of immediate psychiatric side effects---usually extreme irritability or increased aggression. Seizures and tics are also seen with alarming frequency in this population following the start of a new stimulant medication. However, these are all fairly obvious and easily recognized, and when the drug is discontinued the adverse effects dissipate rapidly. Perhaps more concerning are the psychiatric side effects which develop insidiously over much longer time frames; I have consulted on many cases in which a fragile X child has an excellent initial response to a stimulant with no apparent side effect, only to have multiples problem emerge months or years later. Most commonly, obsessive-compulsive symptoms worsen over time, as mood deteriorates and tantrums increase. Because stimulants cause significant physiological dependence, everything gets worse if any attempt is made to discontinue the drug, and this is often seen as evidence the stimulant is not to blame. In fact, this is only evidence that stimulants are potentially addictive, and demonstrates why they are highly controlled substances. The appropriate response is a gradual taper of the stimulant dose over the course of many months.
The entire field of Child Psychiatry appears to have come to the realization that excessive and indescriminate use of stimulants has led to an increase in childhood Bipolar Disorder, Anxiety Disorders, and psychoses. Indeed, the current fad is for overdiagnosis of Bipolar Disorder, after ignoring for decades that this condition can present in childhood. So, the most important lesson is to be vigilant over the long term; most medications which work right away have serious side effects which develop much later.