Traditional Systems Lab

How the model works: improvement in each arm is natural recovery (all arms) plus the nonspecific effect of expectation and the treatment ritual (real and sham arms) plus the true specific effect (real arm only), each building over 12 weeks, plus random person to person variation (standard deviation 1.5 points). For the nature walk, the comparison is a city walk; people know where they walk, so only about half of the expectation effect is matched. Uncertainty ranges are 95% confidence intervals. Simplified for learning.

How the model works: a dose every 12 hours keeps the drug in a repeating pattern. The average level is proportional to the dosing rate divided by clearance. St. John's wort induces CYP3A4: clearance rises over about a week (3 day time constant) and falls back over about a week or two after stopping (4 day time constant). Grapefruit juice, a food shown for contrast, blocks CYP3A4 in the gut and works within a day. Ginkgo is modelled only as a possible added bleeding tendency; the evidence for that is mixed. Simplified for learning.

Make your prediction first. The picture then plays the result so you can watch whether you were right. The worked solution builds step by step. All values are rounded teaching numbers in points or relative units.