|
Select values for the variables. Return to Expersim introduction.
Etiology of SCHIZOPHRENIA SimulationSchizophrenia is one of the most common, most serious, and most mysterious mental illnesses. Over 50 years of research has not produced a totally effective prevention or cure. Perhaps this is because research has also failed to uncover any definite cause. At present we have only a number of unproved and conflicting theories to go on.Many theories stress unfavorable social and emotional experiences as the probable cause. In particular the schizophrenic may have been the victim of harmful family influences or faulty child-rearing practices during his formative years. For example, some therapists have described the "double-bind" situation as critical: the patient's mother was always ready to criticize him for doing one thing and reject him for doing the opposite, resulting in constant frustration. Other researchers see schizophrenia as a physical disease of the nervous system. For example, the patient may be "intoxicated" by some chemical (possibly similar to LSD, DMT, or mescaline) which his body may be producing due to a genetically inherited error in metabolism. Certain forms of mental retardation are known to be passed on genetically. Perhaps the same thing could be true of schizophrenia. Your object in this simulation is to perform a series of experiments which determine if the first sort of theory (social) is right, the second (genetic), neither, or both, Your object is to set up experiments that provide the fairest and most decisive tests of these theories. Actually, we will be simulating a field survey rather than a lab experiment. In the Scandinavian country of Finland the population is relatively homogeneous and bureaucrats keep voluminous records. There are excellent records of adoptions (including the whereabouts of the adopted child's biological relatives), records that enable us to locate any desired relatives of hospitalized schizophrenics, etc. The person whose relatives we are looking up will be called the Index subject. The relatives will be our actual subjects. Having located a particular relative we can interview him, interview people who know him, give him psychological tests, and examine his medical records. In this way we will decide if a given subject is normal or schizophrenic. Naturally this is a time-consuming process. Therefore, no actual research could afford to use a vast number of subjects. You are to use no more than a total of 500 subjects in all groups in each experiment. The results of the experiment will be the number of schizophrenics diagnosed in each group you specify. These results will be roughly consistent with real-life research findings. When you have collected all your data you should interpret it and arrive at a conclusion about the validity of the two types of theories. A totally clear-cut decision may not be possible.
NOTES AND HINTSSuppose you compare two groups of subjects in hopes of finding a difference in the occurrence of schizophrenia. To what do you attribute this difference? Naturally it must have been caused by whichever independent variables you specified as different between the two groups. But it two variables are changed from one group to another, we can't tell if either or both are responsible for any change in the dependent variable. For example, if two groups differ both in family environment and genetic makeup, we don't know which to blame for an increased rate of illness in one group. We would say that the two variables are "confounded" with each other in such an experiment. For this reason it is often useful to ask yourself what sort of variables you want to change. There's sometimes a subtle problem with this. When you think you are changing one variable--such as genetic relationship--you may be implication be changing or specifying something about another factor such as social relationship too. It may help to ask yourself at each stage how both a psychoanalyst or learning theorist AND a geneticist might explain the same result in different ways.Because a difference between any two groups might be explained in two or more different ways, one comparison of two groups may not settle all the issues even if that comparison is very well designed. What to do? Take one of the alternative explanations, perhaps, and figure out what it would predict about some new experimental groups as compared with either one of the original groups or with another newly specified group. The result may test the validity of the alternative hypotheses. Another strategy is to make a series of pairwise comparisons between groups. If these comparisons are chosen well, the combined results may be much more logically watertight than those of any one comparison. At some point you may be satisfied with a definite conclusion. You are encouraged, however, to go on and do other experiments to refine and check you theory, to go on and do other experiments to refine and check you theory. You might also want to form a quantitative picture of the occurrence of the disease and in connection with various factors. Find out all you can from your subjects.
REFERENCES Heston, L.I. The genetics of schizophrenia and schizoid disease. Science, 1970, 67, 249.
Katy, S.S. Genetic-environmental interactions in the
schizophrenia syndrome. In Cancro, R. (Ed.), The Schizophrenia Reactions, 1970. Rosenthal, D. Genetic research in schizophrenia. Also in Cancro.
Stabenau, J.R.
Heredity and environment in schizophrenia. Archives of General Psychiatry, 1966,
18, 458. Select values for the variables. Return to the expersim page. Comments to: Gary.McClelland@Colorado.EDU Revised: 22 August 2001 File: expersim/introschizophrenia.html |