From Google Scholar, or links from these websites: https://medlineplus.gov https://www.mayoclinic.org https://www.drugabuse.gov https://www.psychdb.com/addictions/home https://psy

2. Uniform random number distributions are not terribly interesting in many applications, but they turn out to be about the only kind you can generate easily by numerical methods. Of course, as everyone is aware, it is impossible to actually generate random numbers on the computer, merely sequences that are sufficiently random for our purposes. A more generally applicable distribution is the Normal (`Gaussian’, `bell-shaped’) distribution. It is relatively easy to generate a normally distributed random variable from a uniformly distributed one by a number of means. The simplest way is to add up 12 uniformly distributed random variables, wave your hands, invoke the Central Limit Theorem and claim that a normal distribution is shown by the sum. (An early version of the IBM scientific subroutine package used this approach to generating normal distributions. Write a program that will return normally distributed random variables. Generate a histogram of values returned by the function. (Remember! These values are no longer necessarily between 0 and 1. You will need more than 10 bins on your histogram and you may have some extreme values.) Compare the values with what you expect from a normal distribution. The exact form, tables of values, and copious commentary about the normal distribution is found in any elementary statistics textbook.