Monday, 24 January 2011

“When discussing psychological research, what surprises every economist or physicist is that psychology has no theory.”

Psychology, as subjects go, tends to have a fairly bad reputation regarding its scientific credentials. For the most part I think that reputation is unfounded (perhaps unsurprisingly), but one persistent criticism stands out – the argument that psychology lacks any sort of unified theory, a necessity for a field to qualify as scientific under a Kuhnian viewpoint.

An article written by Gerd Gigerenzer was recently published in the journal Theory and Psychology discussing this state of affairs, and what he thinks can be done to solve it. In short, Gigerenzer’s view is that psychologists are currently failing to integrate their findings into overarching theories, and – perhaps more importantly – must learn how to construct such overarching theories in the first place.

To go into slightly more depth about the content of his paper, Gigerenzer bemoans how psychologists are generally “content working within the confines of their own small territories”; contemporary psychology is, he suggests, “a patchwork of small territories, resem­bling, to use a political metaphor, Italy or Germany before unification around 1870”. The insinuation is made that psychologists are skirting around this issue; Gigerenzer notes that he has written about the lack of over-arching theory in psychology before, and that those papers are amongst his least cited.

He makes the proposal that theory construction should be taught to students of psychology; not exactly an unreasonable suggestion. Many current theories in psychology, Gigerenzer argues, differ from those in other subjects on a number of dimensions – “The nature of parameters in a theory … formal versus verbal theories, optimization versus heuristic processes, as-if models versus process models… domain-specific versus domain-general theories...” – and that student psychologists need to be taught that the construction of good psychological theories should mirror the construction of good theories in other fields. The possibility of physics envy aside (and, perhaps, economics envy?), it’s hard to find fault with this proposal.

A large chunk of Gigerenzer’s paper is then dedicated to explicating two specific elements of teaching theory construction; the first being the study of the history of previously developed theories. One of the main methods of theory generation in psychology, Gigerenzer suggests, springs from the adaption of psychologists to a new tool; what he calls the tools-to-theories heuristic. Examples he gives of such theories include signal detection theory arising from the Neyman-Pearson hypotheses testing method, and Bayesian theories of cognition arising from Bayes’ theorem. The second element that he then discusses is the art of spotting things which claim to be theories, yet are not. Gigerenzer’s most prominent example of such “surrogate theories” is the classic “Why does opium make us sleepy? Because of its dormative properties”; the theory doesn’t actually explain anything about opium’s effect, it merely redescribes the effect using different terminology.

Gigerenzer’s proposal is certainly an interesting one, although I’m not sure optimism is warranted when wondering whether psychologists as a group will listen. Perhaps the most pertinent question that it raises to my mind, though, is; why does psychology seem so averse to theory unification? At what point did that become a quality of the field? Gigerenzer certainly isn’t the first to notice it; Hans Eysenck, in his last submitted paper (published after his death) argued that the unification of disparate psychological subfields was necessary in order to produce theories with greater explanatory power. Howard Gardner has noted that psychology seems to be fracturing, with some aspects of the field apparently merging with cognitive science and neuroscience (those aspects probably being the ones I fall into), and some blending into cultural studies.

The theoretical unification of psychology still seems like a work in progress, which is… disappointing, I suppose. Despite the possibility of further fragmentation (the myriad of post-cognitivist views springing to mind), though, I can’t help but be optimistic about the future of the field.


Esyenck, H.J. (1997), Personality and Experimental Psychology: The Unification of Psychology and the Possibility of a Paradigm, Journal of Personality and Social Psychology, 73(6), 1224-1237

Gardner, H. (1992), Scientific Psychology: Should we Bury it or Praise it?, New Ideas in Psychology, 10(2), 179-190

Gigerenzer, G. (2010), Personal Reflections on Theory and Psychology, Theory & Psychology, 20(6), 733-743

No comments:

Post a Comment