Does Creatine Supplementation Augment Cognition?

    I've found five short-term (six weeks or less), small-scale studies that address this question.

The 2003 University of Sydney Study of Creatine Supplementation in Vegetarians
    The first of these, contributed by Allan Jackson, is "Oral creatinine monohydrate supplementation improves brain performance: a double-blind, placebo-controlled, cross-over trial", performed in 2003 on 45 young adult vegetarians by faculty members at the University of Sydney and Macquarie University. 
The results: Five grams a day of creatine supplementation for six weeks produced dramatic improvements in backward digit span (with an average increase in backward digit span of about 1½ digits for the creatine-fed subjects), and scores on the Raven Advanced Progressive Matrices (with an average improvement of 50% in the number of correct answers for the creatine-fed subjects).
    We probably need to bear in mind that these vegetarians were suffering from a dietary deficiency of creatine.
Background:
    The researchers chose vegetarians because vegetarians get no creatine in their diets. (Dietary creatine is derived from animal sources.)
(1)  The study report observes that creatine in the brain is both synthesized and externally provided by food (Wyss & Kaddurah-Daouk 2000, "Creatinine and creatinine metabolism". Physiol. Rev. 80, 1107–1212) 
(2) It cites a previous study showing that ffve grams a day of oral creatine supplementation increased the creatine levels in the brains of non-vegetarians by 9% (Dechent, P., Pouwels, P. J. W., Wilken, B., Hanefeld, F. & Frahm, J. 1999 Increase of total creatine in human brain after oral supplementation of creatine-monohydrate. Am. J. Physiol. 277, R698–R704). 
(3) It had also been shown that the creatine levels in the brains of vegetarians is lower than it is in omnivores (non-vegetarians): (Delanghe, J., De Slypere, J. P., De Buyzere, M., Robbrecht, J., Wieme, R. & Vermeulen, A. 1989 Normal reference values for creatine, creatinine and carnitine are lower in vegetarians. Clin. Chem. 35, 1802–1803).
(4) The report mentions that supplementation "can only increase creatine levels to a saturated value, beyond which excess creatine is excreted".

The 2006 University College Chichester Study of Sleep Deprivation with Mild Exercise
    This short-term (7 days), small (10 subjects) study fed 20 grams of creatine a day to its experimental group, measuring its performance after 6, 12, and 24 hours of sleep deprivation, with intermittent exercise. The paper isn't available, so I can't quantify the results. However, creatine supplementation improved the mood and the performance of complex tasks that place a heavy load on the prefrontal cortex.

The 2007 University College Chichester Study of Sleep Deprivation, Cortisol, Melatonin and Behavior
    The same research team performs a similar study measuring cortisol and melatonin levels. The cortisol and melatonin levels in both the experimental and the control group were the same. The only thing that differed was task performance which was superior in the creatine-supplemented arm.

The 2007 University College Chichester Study of Creatine Supplementation and Cognitive Performance in Elderly Patients
    This third short-term (7 days), small (15 subjects) study focused on random number generation, forward and backward number and spatial recall, and long-term memory tasks among the elderly. The researchers conclude that 20 grams a day of creatine led to improvements in all tasks except backward digit recall. Here again, the paper isn't available. We don't know, for example, the ages of the elderly.

The 2008 Bloomsburg University Study That Concludes That Creatine Supplementation Does Not Improve Cognitive Function in Young Adults.
    This 6-week, small (11 subjects) group of 19-to-23-year-old subjects took 2 to 3 grams of creatine for six weeks in a placebo-controlled, double-blind experiment to see if creatine supplementation would benefit normal young adults. It didn't. The researchers conclude that creatine supplementation has no effect upon healthy young adults. Creatine supplementation, they conclude, only includes cognition in impaired individuals.

Safety Considerations
    The normal rate of creatine turnover is about 3 grams a day. A breakdown product of creatine is creatinine, which is the major measure of kidney function. High serum creatinine levels are indicative of impaired kidney function. Creatinine clearance levels drop linearly with age, eventually reaching, at age 125 or so, levels that would require dialysis. A study panel has concluded that 3 grams a day of supplementary creatine are probably safe. Twenty grams a day probably is not, especially for the elderly.
Conclusions:
    There are a number of indications that creatine supplementation has positive effects upon cognition, including a neuroprotective effect in the brain, (Wyss M., and Schulze, A., 2002 Health implications of creatine: can oral creatine supplementation protect against neurological and atherosclerotic disease? Neuroscience 112, 243–260.) The Bloomsburg University study employed a dosage level that is probably safe for long-term administration, and the study lasted six weeks, and employed a variety of cognitive assays. On the other hand, it would seem to me that with the small sample size and the relatively low doses, it might be difficult to register statistical significance. However, if we accept the results and conclusions as definitive, it's still the case that most of us aren't 19-to-23 for very long. We might still be interested in creatine supplementation. 
    My personal take on this is that, given all the other "brain boosters" on the market, given my age, and in the absence of further information on creatine, I'll probably skip creatine supplementation for now.
    Interestingly enough, one study suggests that memory training in the healthy elderly boosts natural creatine production and levels in their brains, (Valenzuela, M. J., Jones, M., Sachdev, P., Wen, W., Rae, C. & Scott, G. 2003 Memory training alters hippocampal neurochemistry in healthy elderly. NeuroReport 14, 1333–1337.)

Mechanisms of modafinil: A review of current research.

.