Tuesday, July 28, 2015

A Real Evolutionary Psychology

Image Source: Flickr
Some estimates suggest that, compared to women, almost twice as many men cheat on their spouses.  Why?

The answer seems obvious.  The more sex a man has, the more likely he is to get a woman pregnant.  The more sex he has with more than one woman, the more likely he is to get more than one woman pregnant.  The more women he gets pregnant, the more children he will have, and the more of his genes that will live on in future generations.  Selfish genes in action, as it were.

The poor woman is not so lucky, as getting pregnant once removes her from the system for at least 9 months, if not longer, practically speaking.  But why, then, do at least some women cheat, too?  With a little imagination, we can make sense of this.  It turns out that cheating provides an opportunity for the woman to get better genes from fitter donors.  She can maximize her limited opportunities to get pregnant and give birth.  It is a matter of quality, then, not quantity.

Or is it?

Speculations of this kind about the evolutionary origins of human nature are popular in some circles and such thinking is formalized in the field of evolutionary psychology.  Evolutionary psychologists contend that much of our behavior has been hard-wired into our brains via natural selection.  The field has given off much heat but very little light, but it remains popular.  This is unfortunate, as the heat generated by EP stands to keep people away from the light produced by more rigorous analyses of human behavior rooted in science, not pseudoscience.

Evolutionary psychology is not a science, it is “just-so” psychology in which people are observed to behave in certain ways, some plausible story about the problems faced by our ancestors is formulated to explain why the observed behavior would have evolved, and then observations that people do indeed behave that way are offered in support of the explanatory story.  This is circular reasoning at its finest.  (This is to say nothing of the gross misuse of heritability estimates in research on psychological traits, but that is for another blog post.)

Invoking Darwin, selfish genes, and DNA cannot save it from the clutches of pseudoscience.  So what is the alternative?

It turns out that learning is a much more powerful and plausible explanation for much of what is attributed to genetics by evolutionary psychologists.  Our brains are not hardwired with many special purpose modules guiding our behavior.  What has evolved is an exquisite behavioral sensitivity to our environment, allowing our behavior to be shaped by the pressures we face rather than the pressures our ancestors faced.  More than 100 years of experimental evidence supplies us a rich framework with which to understand human behavior (and the behavior of other animals).  There are, indeed, parallels between biological evolution and behavioral evolution, just not the kind proposed by evolutionary psychologists.

There are three processes critical to biological evolution: variation, selection, and retention (via replication).  As it turns out, these three processes are common features of both biological and behavioral evolution. At the behavioral level, behavior varies from one situation to another and from one time to another in the same situation.  Certain types of behavior will produce beneficial outcomes and some will not.  The beneficial outcomes "select" the behavior that produced them, leading to that behavior being repeated (due to changes in the nervous system) in similar circumstances in the future.

B. F. Skinner
Image Source: Flickr
We know about this process of behavioral evolution, and the details of how it occurs, chiefly from the work of the late American psychologist B. F. Skinner and his many students and colleagues.  Skinner famously discovered operant conditioning, a term he used to collectively refer to the basic laws governing most of human and nonhuman behavior.  Operant conditioning encompasses a relatively small number of very powerful principles, including reinforcement, punishment, extinction, and stimulus control.  Like evolution by natural selection, operant conditioning is a simple mechanism that gives rise to great complexity.

In Skinner's terms, behavior that produces favorable outcomes (reinforcers) is more likely to occur in the future (an outcome called reinforcement). Additionally, the behavior is most likely to occur in the future in situations that are similar to situations in which the behavior has been reinforced in the past. This acquired effect of the environment is referred to as stimulus control, and the controlling stimuli are called discriminative stimuli.  In this way, the reinforcer selects a relationship between a set of environmental circumstances (discriminative stimuli) and certain types of behavior.  That is, the reinforcer "reinforces" the relationship between the circumstances and behavior, meaning that the behavior is more likely to occur when those circumstances present themselves.  The opposite effect is seen with punishment, whereby certain behaviors are less likely to occur in certain situations because, in the past, those behaviors have produced unfavorable outcomes (punishers) in those situations.

In the 20th century, Skinner was the most notable champion of such selectionist principles as the basic mechanism of learning, but he was by no means the only psychologist to suggest as much. For example, at the turn of the 20th century, one of the most famous and influential psychologists of all time, Edward L. Thorndike, was conducting his research on animal intelligence that resulted in his formulation of the Law of Effect—one of the very few "laws" in psychology.  Thorndike placed  hungry cats in a handmade wooden "puzzle box," with a dish of food placed outside and in view.  The  boxes differed from one to another, but all had some way that the cat could open the box and escape, such as a lever that opened a door when pressed.  Thorndike observed that, at first, the cats took quite a while to accidentally make the response that opened the box, essentially stumbling on the right solution by trial-and-error.  After a cat finally did escape, it was placed back in the box and process was repeated many times, across many days.  Over time, and as a result of their experiences, the cats were quicker and quicker to escape, ultimately pressing the lever as soon as they were put into the box.

E. L. Thorndike's "Puzzle Boxes"
Image Source: Yale University Archives
This work is transparently selectionist, insofar as Thorndike reported an ever-decreasing range of behaviors exhibited by the cats as they were repeatedly exposed to the puzzle box.  Over time, the wide range of behavior originally seen (e.g., scratching, sniffing, whining) narrowed so as to be dominated by the lever press, which most reliably opened the box and resulted in escape and access to food.  In this case, the selecting aspect of the environment for the behavior of the hungry cat was escape from the box and access to the food.  This beneficial outcome selected lever pressing as the dominant behavior, while other varieties of behavior went extinct because they were not beneficial under these circumstances.  The cats still did all those other things (e.g., scratching, sniffing, whining), but not when they were in the puzzle box.

In the decades following Thorndike’s work, other psychologists arrived at similar conclusions about the importance of environmental selection as a causal mechanism of learning.  Again, in the language of Skinner’s operant conditioning, those events that follow behavior and result in the behavior occurring more often in the future under similar conditions are called reinforcers; the increase in future behavior is the result of reinforcement. In short, reinforcement selects behavioral characteristics of the individual in much the same way survival selects other (e.g., physical) characteristics of a species.

A chief benefit of learning over natural selection is that when environment change rapidly, the individual can change, too.  With both Darwinian and Skinnerian evolution, a selected characteristic is beneficial only in environments similar in some critical ways to the environment prevailing at the time of selection.  At the Darwinian level, a heightened sensitivity to fatty foods as reinforcers serves the species well in times when fats are in short supply, because we need fat in our diets to survive.  Across many generations living under such conditions, the numbers game favors those individuals most motivated to seek out and eat fatty foods, as they are the most likely to survive long enough to reproduce.  Should conditions change (as they most certainly have) with respect to the availability of fatty foods, this sensitivity to fat as a reinforcer loses its adaptive value and can even become a hindrance.  However, it can take quite a long time for evolution to fix the problem, which is where learning steps in.

Of course, learning requires, at the very least, a behaving organism that is the product of natural selection.  Even more than this can be granted to evolution, however.  Our genes influence our sensitivity to certain aspects of our environment and thereby play a crucial role in determining the extent to which those things influence our behavior.  In this way, our evolutionary past most certainly plays a role in determining our behavior, but not in any direct way, at least in most circumstances.

(The role of genes is undoubtedly even more pervasive in some circumstances.  The presence of seemingly innate aggressive responding to painful or otherwise aversive stimulation has been documented in non-humans and has led many to conclude that the same might hold true for humans.  Though the jury is still out, the claim is not entirely unfounded nor is it especially improbable.)

It is obvious that our genes play a crucial role in the ability of the organism to learn and perform certain behaviors, but the genetic role is one of proximate causation driven by the environment as the ultimate source of causation.  Even our genes are changed as a result of our experiences.  To quote science journalist Sharon Begley, "Evolution indeed sculpted the human brain. But it worked in malleable plastic, not stone, bequeathing us flexible minds that can take stock of the world and adapt to it."

We have inherited an ability to learn from our experiences, because doing so allows us to survive in an ever-changing world.

Monday, July 13, 2015

Why Psychologists Study Behavior, Not the Mind

Image source: Flickr
For psychology, behavior is always the subject matter, despite what some might claim to the contrary. At the end of the day, all that psychologists can measure is behavior. Even the psychologist who claims to study the mind is, in practice, studying behavior. There is behavior and there is the nervous system, but everything else is an inference. To infer something between the nervous system and behavior is dualistic, and so much the worse for the progress of psychology as a science.

Consider the following example. We want to know if understanding a question written in all capital letters without any punctuation is more difficult than understanding a question written normally. We hypothesize that it will take longer for subjects to answer if the question is more difficult to understand. We present the following question to half of our subjects:

"JIM HAS FIVE DOLLARS JOHN HAS TWO DOLLARS HOW MANY MORE DOLLARS THAN JOHN DOES JIM HAVE"

We present the following question to the other half of our subjects :

"Jim has five dollars. John has two dollars. How many more dollars than John does Jim have?"

Suppose that subjects in the first group take, on average, 5 seconds to answer. And suppose that subjects in the second group take, on average, 1 second to answer. We conclude that the first question--the one in all capitals with no punctuation--was harder to understand. Further, we conclude that it took the subjects seeing the first question longer to "process" the question, with the "processing" presumably being some cognitive processing taking place in the brain.

We might, and psychologists often do, assume that we are measuring cognitive processing of some sort with our experiment. But are we?

What we really are measuring is the relationship between the way a question is printed and the length of time (termed "latency") it takes subjects to answer the question. Providing an answer to the question is behavior and that is what we see and measure. We haven't measured a cognitive process, then, we've measured behavior. We infer the cognitive process from the behavior we observed.

But someone else might infer a very different thing from the very same situation. Maybe another researcher would infer that seeing all capitals and no punctuation makes the subject first say to themselves, "I wonder why the question is written in all capitals with no punctuation." Saying this before answering the question is what accounts for the increased time it takes them to answer the question. Still another research might infer something very different.

There can be no real disagreement about the response latency observed. It is objective, well defined, and easily recorded. No inferences necessary. That is the great benefit of approaching psychology as a science of behavior: it is conservative and assumes no more than what can be established through direct empirical investigation.

Put simply, behavior is anything an organism does. More specifically, behavior is the interaction of the muscles and glands of a live organism with the environment. For scientific purposes, we need to add a caveat. Behavior is anything an organism does that is observable and measurable. If we can't in some way see it, we can't measure it. If we can't measure it, we can't study it.

Talking is behavior. Running is behavior. Pedaling a bike is behavior. Writing an answer on an exam is behavior. Taking notes in class is behavior.

Things like understanding, knowing, loving, and even learning, are not behavior mostly because they are too vague--they are just labels for various kinds of actual behavior. Loving, for example, doesn't specify anything in particular and, not surprisingly, there is a lot of disagreement about what love is and how you would go about studying it. We can simplify things and get a foothold on studying something like "love" if we just identify some actual behavior that we believe to be relevant. We might decide to study how often a husband compliments his wife (and vice versa). We might study how often a wife says nice things about her husband to other people (and vice versa). We might study how close a husband and wife sit to one another at dinner or at a party.

Really, we'd have to identify a whole host of behaviors like those above and study them in concert--no easy feat, to be sure. Absent a clear definition of what we're studying and a way to record what we're studying, we'll be spinning our wheels.

Science is mostly a slow incremental endeavor. You can't start with all the answers to the most important questions. First you have to find out what questions to ask. For psychology, you have to figure out what behavior is relevant and how to measure it.

Monday, July 06, 2015

Bigger Is Not Always Better: Why Psychology Needs Smaller Studies

Image source: Flickr
Psychology has been embroiled in a professional crisis as of late, and deservedly so. The research methods commonly used by psychologists, especially the statistical analyses used to analyze experimental data, have come under scrutiny—again, deservedly so. Although it is encouraging that so many people are becoming aware of the many problems evident in mainstream psychology research, one fundamental problem has received almost no attention. Namely, the focus on studying large groups of people has gone unquestioned. However, focusing on between-group comparisons is, in my estimation, THE problem, especially because those designs are exactly the kind that lead to the very statistical analyses at the center of psychology's professional crisis.

Smaller within-subject studies are much more appropriate for the kinds of questions most psychologists are asking. Smaller studies also tend to produce data that can be understood without complicated statistics. Moreover, and contrary to popular belief, within-subject studies actually tell us more about each subject studied and, therefore, provide us with more information about when and where the findings are likely to be useful.

Generally speaking, within-subject research allows each subject to be studied very intensively and over a prolonged period of time. Also, because there are fewer subjects, within-subject research often can be conducted under very controlled conditions, unlike studies of large groups of people, which have to rely on one or a few measures of each subject. Not to mention that those few measures typically are measures of what the subjects say the will do, rather than what they actually do. (Unfortunately, what we say we will do rarely matches what we actually do.)

Moreover, because large group designs focus on the average performance of a large group of subjects, they don't tell us about any real effects on any real person. The "average" effect doesn't exist, and an individual subject almost never responds like the mythical "average." This makes it very difficult to translate research findings into practice, because we will never meet the average person, we will only meet a real person.

What we get from studies of large groups of people typically amounts to very little information about any actual person or persons, few and possibly invalid measures of subject performance, and findings that might be "statistically significant" but have no practical implications in the real world.