Chapter 1 Notes: The Science of Psychology
    Why does psychology insist upon calling itself a science when so many REAL scientists (chemists, astronomers, physicists) believe that anything involving creatures as quirky as humans cannot possibly be scientific?  Mankind  has been speculating about the nature of man for thousands of years.  The ancient Greeks and Romans were famous for their theories about the nature of man.  Philosophy is old and venerable; psychology is quite young; just slightly over 100 years old.  Psychology became a science when various thinkers not only had ideas about how man "works," but figured out ways to test these ideas.  According to most psychologists, the goals of psychology are description, explanation, prediction, and control.  In order to achieve these four goals, we have to decide exactly what we are looking for and then study it in as precise a manner as we can.  Your text discusses research methods. There are  four research methods because correlation is merely a mathematical tool for data gathered using the first three methods.  Going from least precise to most precise, the research methods are:
            1. Naturalistic observation
            2. Case studies
            3. Surveys and tests
            4. Experimental method.
 
    Naturalistic observation involves looking at some behavior in a natural environment and not interfering in any way.  We use naturalistic observation as a starting point in some cases, such as videotaping a mother interacting with her baby.  After we watch a number of mother-baby combinations, we might try to set up an experiment of some type to look at a small piece of behavior that we wanted to understand better. In other cases, naturalistic observation is the only safe way to conduct research.  If you were studying riot behavior, it would be better for you to film it from a distance than get down in the street and administer questionnaires to people hauling TVs out of store windows.
    Case studies are helpful within limits, but the worst problem with case studies is that they can be flawed or skewed by the interests of the person interacting with the subject.  Freud based almost all of his theories on his own case studies, and since he was interested in unconscious processes and motivations, he ignored all "obvious" explanations when his patients presented with their problems.  I am reminded in particular of the case of a female patient who was "hallucinating" cigar smoke.  The young woman was most probably smelling cigar smoke on her own hair and clothes, which was caused by living in  a household with cigar smokers.  Freud himself would not have been able to realize that she was reeking because he smoked almost continuously himself.
    Surveys have one magnificent advantage and one horrific disadvantage.  Their advantage is that we can get massive amounts of data from captive or semi-captive audiences for free.  One example of this was the amount of data collected during World War II from our young men and women, whether drafted or volunteers.  All people going into the military had to take all manner of tests, including IQ tests.  For the first time in our history, we were able to take a "snapshot" of a large segment of our population.  The disadvantage of surveys is that people lie.  Sometimes they lie accidentally, sometimes they do it on purpose, but a certain percentage of survey results will always be inaccurate.  It is easy to give an example of this because I do it myself.  Every two or three months, I get a "consumer's survey" in a big envelope with grocery store coupons.  I am as truthful as I can be about the products that I buy.  But when I get to the question that says " I have contributed to the following by mail in the past six months....religious, humanitarian, political, etc,"  I always say " NO."  In fact, I contribute to many causes by mail; but if I tell the survey takers about my charitable nature, I have this feeling that I will get 20 solicitations per week instead of the 5-10  I  now receive.  Another problem with surveys rests with the survey givers.  Are the people doing the survey being careful to get either a representative sample or a random sample?  It is not terribly hard to get a representative sample; you ask a few qualifying questions or you pick a location or demographic group and just gather your data.  A random sample is a genuine challenge because even those of us who are scientifically trained  are not able to pick a  random group of people.  There are randomization techniques used by the major opinion poll designers to prevent selection bias.  As an example, I worked for Louis Harris (political pollster) in college.  I was told what house to start with in a given neighborhood, and then I had to go to every third house.  Even if someone was obviously home in house #2, I was obliged to skip him and go to the house next door and try to talk with a shotgun-toting  drunk.
    Correlation is a mathematical technique for dealing with data gathered  from naturalistic observation, case studies, and surveys/tests.  Correlation tells you if a relationship exists between A and B.  To explain correlation, I have to do a number line:
            -1________________________0___________________+1
 
Correlations are always expressed as a number below 1, such as .7 or -.8.  The reason why you never see a "perfect" correlation of -1 or +1 is that -1 means that when A happens, B never happens and +1 means that when A happens, B always happens.  Nothing in nature or in the study of behavior is that perfect.  The closer you get to -1 or  +1, the stronger the correlation is.  For instance, -.97 is a stronger relationship than .88.  It is an INVERSE relationship, but it is a strong relationship anyway. The closer you get to the middle of the number line, the greater is the likelihood that A and B are not related at all.  For instance,  if  I gather some data and wind up with a correlation of .4, my research goes in the trashcan rather than  in a professional journal.  We have one rule regarding correlation that we are never allowed to break, no matter how strongly we feel about the correlation: we are never allowed to make a cause-effect statement based on a correlation.  The relationship may in fact BE a cause-effect relationship, but we are only permitted to say that there is a strong relationship between A and B.  A few years ago, there was a heart conference here and a woman doctor presented a piece of correlational research: the higher the educational level of the wife, the stronger the tendency of the husband to have heart problems.  I asked my students to explain that one, and most of them tried to pin the "blame" on the wife by suggesting that the man's ego was ruined or she didn't cook him enough meals so he ate too much "fast food."  Since I know the limits of correlation, I suggested that she did absolutely nothing to him except marry him.  To me, it was obvious that if a woman goes to the trouble of getting either a four-year or advanced degree, she is not going to marry someone who is  going to lie around all day; she's going to marry someone who is "Type A."  Maybe the man would have had heart problems if he had never married at all.  Speaking of "Type A," this was a piece of correlational research that wound up in all of the psychology textbooks without any testing.  All textbook authors broke all the rules on that theory.  Now there is some research to back up the "Type A" theory,  but it is a good example of rushing to support an idea that is intuitively obvious but unproven.
    The only way  that we are allowed to make a cause-effect statement in psychology is to use the experimental method.  Sometimes we stumble through the first three methods before we arrive at an experiment, but if we have a strong correlation, we feel compelled to follow it through and see what happens.  My example is from medicine, but something similar has probably happened in the psychological arena as well.  Once upon a time, not so very long ago, people in this country thought smoking cigarettes was not just okay, but GOOD for you.  During the second World War, every military person got two packs a day, unfiltered of course.  Even military personnel in hospital beds with fairly grievous wounds got their two packs a day.  My father had one lung and a huge tube sticking out of his back, and he still got his ration. (He said he and the other "chest cases" used to impress the USO girls at parties by inhaling and then blowing the smoke out of their tubes instead of the usual route. Don't ask me how impressed the girls were.) Around the time of the second World War, a doctor in New Orleans named Oschner made the naturalistic observation that his patients who smoked were less healthy than his non-smokers. Dr. Oschner also had file drawers full of case studies to back this up. He wrote articles about the dangers of smoking and was almost ridiculed out of the medical profession by his co-workers. All of his doctor friends acknowledged that he was brilliant, but he just had this crazy little flaw that he couldn't be talked out of. Quite a few years passed before someone decided to use the third method, surveys, to see just how crazy Dr. Oschner was. It was very easy to gather the data; for every lung cancer hospital admission, one simply asked, "Do you smoke?  How many years?  How many packs a day?"  As a control,  the researchers would have  also surveyed a selected group of  non-cancer patients also.  Within a short period of time, the correlation between smoking and serious disease was made obvious.  Since it was only a correlation,  science had to go to the final step, the experiment.  We gathered up some of our biological close relatives, apes, and taught them to smoke.  When they died, we autopsied them.  Lung cancer.  End of story, except for the warnings on the cigarette packs.
    This example of the progress of science from naturalistic observation to the experiment brings us to another issue: the use of animals in psychological research. We are obliged to use animals for five reasons:
        1. Ethical considerations; some things we just can't do to humans. In my example above, it would have been scientifically precise if we could have rounded up some 12-year olds, made them smoke four packs a day, and then waited. But we aren't allowed to do things like that; it's against the law. (It's also immoral.)
        2. Shorter life spans; as a researcher, you can get your results quite rapidly. If you are doing a study with human subjects and the effect takes a long time to manifest, then you could be dead before your study is finished. There is a study currently being conducted by Stanford that began in the 1920's and is still going on; they are looking at a selection of humans from birth to death. It was (and is) a very ambitious project in the field of human development, but the study is now on its third generation of researchers.
        3. Better experimental control; when you use animals, you know where they sleep, what they ate, and they are always available when you need them. When you do human research, your subject could be a) mad at his girlfriend, b) stoned, c) hung-over, d) upset over his grades, e) you use your imagination to fill in whatever else could be on this person's mind to interfere with his full concentration on your survey or experiment.
        4. They are similar to us physically and psychologically. Right, you're saying, I have NOTHING in common (especially psychologically) with a white rat. Wrong; I can put a rat on a VR schedule of pressing a bar (see Chapter 5) and watch any human playing a slot machine and the behavior is identical.
        5. No subject bias. When we do human research, we often have to lie (we call it using deception) about the nature of the research. We do this because we want natural, uncontrived behavior. Humans, if they know what you are looking at or looking for, will try to help you. This is called subject bias. We don't want help; we want ordinary behavior. Rats, pigeons, and chimps genuinely do NOT care what you are looking for; they are just themselves.  This means that we don't have to sit up nights making up some plausible reason for putting you on a roller coaster or asking you to shock someone.
    Please note that we do not try to be unnecessarily cruel to either people or animals; your text discusses ethical controls, so please look it over.
 

     The following is an exercise on the experimental method. It is like a big matching question. Every blank or underlined sentence has only one mate in the list.  The answers are farther down on the page.

        dependent variable            reliability                 control              experimenter bias
         independent variable        random                    data (numbers)                          hypothesis                      validity

"THE CASE OF THE SNORED-UPON SLEEPERS"

    For this experiment, we initially made an (1)educated guess that persons awakened many times during the night by snoring would tend to be in a bad mood the next day.  Therefore, the experiment was called (2)the effect of snore-caused awakening on (3)the mood of the awakened subjects the next day.  So that mood  changes could not be attributed to loss of sleep alone, we awakened (4)another group of sleepers using blasts of white noise equal in duration to the snore.  We operationally defined  a "bad  mood" as a positive response to (5)50 or more items on the California Mean Test (CMT) the next morning.  (6)We repeated the experiment five times to see if we would get the same result every time. (7)We later discovered that we were measuring the wrong thing!  When all groups were given the CMT after breakfast, no-one scored above 50.  We discovered that the person who designed this experiment is (8)not a breakfast-eater so he did not think of this as a possible confounding variable.  Another problem was that all of the subjects were college sophomores who had learned to sleep under adverse conditions; therefore the sample was not (9)__________.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Answers:
1. hypothesis     2. independent variable     3. dependent variable     4. control     5. data     6. reliability     7. validity     8. experimenter bias   9. random

    How did you do?  If this was easy or obvious, you understand the experimental method.  If  you made mistakes, you need to review that part of the chapter again.  The two areas of misunderstanding are usually about independent variable/dependent variable and reliability/validity. The independent variable is always what the experimenter DOES and the dependent variable is WHAT HAPPENS AFTER SHE DOES IT.  Reliability means repeatability. Can you do this experiment many times in a variety of labs with different people in charge and get the same results? In psychology, this is mandatory; otherwise, all of the information about learning theory, memory, and perception would not be in the textbook. This is the only way that we are allowed to call ourselves a science...we repeat various experiments until we are sure that the results are genuine.
     Validity is somewhat harder to explain; you have to answer the question, "Are you measuring what you THINK you're measuring?" If you design a mistake into the experiment, which is a form of experimenter bias, then the validity is gone. I am, unfortunately, an expert on validity because my Master's experiment turned up some very odd results.  I wanted to measure how long baby rats played with a "curiosity object" after either being dosed with antidepressants prenatally or NOT dosed prenatally. I tried one or two types of toys and none of the animals showed any interest at all. Finally, after I had made an adequate number of mistakes on my own, I was allowed to talk to Dick, the resident "rat man."  He laced his fingers behind his head, leaned back, and said, "Let's talk about what rats LIKE."  We established that rats have little hands and they liked to play with things that moved, but only if the rat moved the object.  Toy robots were out of the picture, of course.  After about three minutes of conversation with Dick, I went home and designed a "curiosity object;" a bunch of embroidery thread hanging from a string suspended over a 3'x3'  "open field" (psychology's name for a big box).  The "curiosity object" was a raging success with some of my rats; with only 180 seconds of exposure, some animals were spending 160 seconds with it by the last day of the experiment.  Some of the rats spent about 10-20 seconds with it regardless of the number of exposures.  Since my data varied wildly (10 seconds to 160 seconds), my statistical technique indicated that my drugged versus nondrugged groups were not significantly different from each other.
    The experiment, in other words, failed to support my hypothesis about behavioral differences in the drugged versus nondrugged babies.  Before I did my oral defense of the experiment, I went back to the original data sheets to see if I could find any flaws.  While I tested my animals before sex differences kicked in, I still indicated the sex of the animal on the original data sheet.  The answer was there: the Minnies were spending a lot of time with the curiosity object while the Mickeys were clearly bored by it.  It was some consolation when Dick, the rat man, said that he would have made the same mistake.  I was not measuring curiosity; I accidentally measured sex differences in pre-sexual rats.  Dick speculated that the Minnies were attracted to the embroidery thread because it reminded them of nesting material.  I wonder how things would have turned out if I had suspended a tiny pick-up truck from the string.  It is possible to be very careful in experimental design and wind up with unusual results anyway.  My story also illustrates something about scientific psychology in graduate school. When working for a Master's degree, one is allowed to make mistakes ONLY if one is intelligent enough to discover why things went wrong and how one would remedy the problem in a future experiment. When working for a PhD, the final product is not allowed to be flawed in any way. If you have to re-design your experiment five or ten times, well, that's your problem.

    When discussing the history of psychology, we always say that it was invented by Wundt in 1879. Wundt had the first psychological laboratory and was the first person to attempt to measure human experiences in a scientific manner.  While using somewhat primitive techniques by today's standards, Wundt's students studied human sensation and perception. These students eventually evolved into the Gestalt school of psychology, which formulated "laws" of perception that are still considered valid today. In the United States, functionalism didn't amount to much because it was not adequately scientific. Please click on this sentence for a discussion of early American schools of psychology.  On the first test, I will be asking you questions about Poyen,  Clarke, Brigham, and Beard.  In this country, we can trace scientific psychology back to Watson in the 1920's.  Watson invented behaviorism, which says: "Don't talk about it unless you can observe it directly!"  That leaves out areas like memory, problem solving, and thinking.  In fact, the only  area left to study was learning, or as you will learn in Chapter 5, classical and operant conditioning.  Skinner came into the picture in the 1930's, and from then until the late 1960's dominated the field of scientific psychology. Anyone trained in scientific psychology between the 1930's and about 1970 was trained as a behaviorist because behaviorists ruled.  If you were NOT a behaviorist or happened to challenge a behaviorist principle, your work got ignored.  A good example of this was what became of Tolman's work. Tolman discovered some amazing things about learning (not behaviorism, but thinking-type learning) in the 1930's, but he only appeared in our introductory psychology textbooks within the past 10-15 years. Why did the intellectual dictatorship called behaviorism finally lose its choke-hold on our scientists?  We had to open up a field called cognitive psychology because we were obliged to study the human thinking process in order to program thinking machines, computers.  Please understand that we did not all wake up one day and say, "Gosh, operant conditioning has become a bore since we have proven and re-proven all of Skinner's ideas."  We got dragged into the cognitive era because of computers.  The other "hot" new areas in scientific psychology are the biological approach and the evolutionary approach.  We now have sophisticated machines such as the  modified EEG,  PET, and fMRI that enable us to see what happens when someone thinks, remembers, or reacts emotionally. Evolutionary psychology has also gotten some good press; it looks at all human behaviors, both adaptive and maladaptive, and asks how this contributed to our survival as a species over millions of years. James Brody is the best writer in this field, in my opinion.

    There have always been two major schools of psychology: the scientists and the healers. Freud founded the psychoanalytic school of psychology around the same time that Wundt started his laboratory.  Freud was incredibly influential, so much so that anyone trained as a therapist from the late 1800s to about the mid-1960s was a psychoanalyst.  His influence in the United States was fueled by the emigration of thousands of educated people who got out of central Europe just prior to and during the second World War.  Freud's theories are somewhat pessimistic because they emphasize that you are controlled not by your own free will but by your irrational unconscious mind.  The emphasis on early childhood events also implied that if you had a rough time before age five, through no fault of your own, you were doomed to live out your childhood problems until you either died or stayed in therapy for decades.  Since the "American character" is inherently optimistic, we could not live with Freudian theory indefinitely.
      Humanistic psychology focuses on the present, emphasizes that people usually act on conscious motives, and insists that every person is born inherently good and striving.  One of the basic ideas of humanism is that we should never make anyone feel badly about himself.  Humanistic psychology promotes a type of self-absorption; you are encouraged to "find yourself," whatever that means.  Humanism has impacted American life  in three large areas during my lifetime.  I can see it because I can remember what life was like before humanism.  The impact on public education was dramatic; since we didn't want Susie to feel badly about not doing her homework, we passed her using the concept of "social promotion."  What we got was thousands of Susies and Johns who graduated from high school without even minimal reading and mathematical skills.  The second area of impact that I observed at close range was the change in a variety of mainstream religions.  Churches no longer talked about  sin or hell because they didn't want you to feel badly about yourself.  They were afraid that if they made you feel guilty, you would stop coming to church.  Many churches evolved into "feel good"  places instead of  teachers and enforcers of moral behavior. The third area of impact was parent-child relationships.  In the "old days," these relationships were mostly authority-based; I am your parent, I know what's best for you, and that's the way things are.  In recent times, parents have wanted their children to be their friends and companions.  We have seen a hands-off approach to parenting that scares me...it says, "If we demand certain behaviors of our children, we are stunting  their self-esteem and creativity."  While over-strict parenting is not a good approach, letting small children and then teenagers "find themselves" has led to tragic results as expressed by rising illegitimacy rates, teen crime, teen suicide,  and school problems.  If you are young enough to remember only humanism, this paragraph has probably made no sense at all.  If you are old enough to have experienced all of these social changes, then you know that  in this country  we have a definite "before" and "after" in our lives. The following are your study questions for Test 1:

 If you have read the textbook and are ready to do the quizzes for the first chapter, please click here
celtic2