Friday, April 8, 2011

Psychology, Language, Science and Nomenclature

I have spent the past few weeks in the (infinitely long) process of making edits to my thesis.  And I have come to the following conclusion:

Research is annoying.

And, might I add, research in psychology is especially annoying.  Don’t get me wrong.  I love what I do, and don’t think I would find another field more fascinating.  But research in psychology has one big (very very big) shortcoming: our nomenclature stinks.

The types of models we work with in medical psychology are complicated.  They demand a combination of psychological and physiological constructs, the validity of which is often fuzzy.

Take depression, for example.  Depression is a construct that we spend a lot of time on in medical psychology, because, well 1) it is widespread, and 2) it happens a lot when people get sick.  In studies, it is typically assessed using self-reported symptom checklists that essentially ask people if they are feeling sad and hopeless, and if they have a variety of other symptoms of depression.

But are these symptom checklists really reflective of what we are looking for?  Or maybe that’s the wrong question.  Maybe the question should be, what do we mean by “depression”? 

To get a clinical diagnosis of major depression you can be either 1) sad, and 2) have no appetite, not be able to sleep, be restless, and cry all the time OR 1) have lost interest in things, and 2) have excessive appetite, sleep all the time, be lethargic, and never cry. (Or at least this was the case in DSM-IV...)  But that raises a question: are those two depressions the same issue?

Even worse, you can have depressive symptoms secondary to a negative event in your life (e.g., loss of job) or secondary to unresolved childhood abuse and PTSD.  Now those two things are definitely NOT the same thing.  The first is generally self-resolving as long as the individual has the personal and social resources necessary, and may or may not have startling health consequences.  The second implies changes in the CNS from a young age (Shea et al., 2005), as well as a different constellation of personal and social resources throughout the lifecourse, and is almost certain to have long-term mental and physical health consequences (Dong et al., 2004; Dube et al., 2009; Ramiro et al., 2010).

To make matters worse, we know that there are neural correlates to constructs such as depression.  There is no clean split between the physiological and the psychological constructs in medical psychological research.  Human psychology may be an epiphenomenon of biochemistry (but that gets us into a whole ‘nother philosophical debate).

So depression is complicated and comes in many forms.  But we often don’t make these distinctions in research.

The problem is that they all look like what you and I, and everybody and their brother calls “depression”.  Psychology does not have a periodic table of the elements to organise itself.  We rely on human languages.  And the problem with language is that it means whatever the people using it decide it means at any given point in time.

(Fun fact: Many languages do not have a word for “depression”.  What implications does this have for the study of such constructs cross-culturally?  Or again, maybe the better question is: Do these constructs exist cross-culturally? For an interesting discussion see Urban Semiotic.)

To be clear, I am not advocating reducing psychology to its neurological bases.  I am not a person who believes that you cannot scientifically examine subjective phenomenon.  After all, I am studying pain – it’s hard to get more subjective than that. (Although I expect there is a word for pain in most languages, which might make it slightly superior a construct to depression… Must look into that.)  But I think the question of how we in the human sciences create our constructs deserves more attention than it receives. 

If we are not going to invent new words to describe our constructs, then we need to know their limitations.  Perhaps depression as we know it in the West does not exist in all cultures.  Perhaps depression is a social construct that exists here that does not exist everywhere.  Alternatively, maybe depression as we understand it in the medical sense exists everywhere in different forms, and we have just decided to label it this way and need to be creative when studying it cross-culturally.  (Kind of like carbon atoms exist everywhere regardless of whether or not there is a term for them.)

We need to be less reliant on our languages.  Those types of depression described above may be qualitatively different as regards their physiological correlates, and if so, they may be responsive to different types of treatments.  Just because they all look like what the English language calls “depression” does not mean they are clinically identical or on the same continuum.

These are challenges in our field that many in the “hard” sciences hold against us.  And for good reason.  Until we can be quite clear about what we mean by a given construct and what we intend to pick up on when we measure it, we cannot pretend to understand its health implications either on the level of the individual, the group, or society.