Wednesday, July 04, 2012

RS encore: Complex decision? Don't think about it!


chamonixvue.files.wordpress.com
by Massimo Pigliucci

The more complex a decision we have to make, common sense would suggest, the more we ought to think about it. Be that buying a car, deciding on a career choice, or getting married, you surely don't want to make impulsive decisions. Well, not exactly, according to a recent study by Ap Dijksterhuis and colleagues at the University of Amsterdam (Science, vol. 311, p. 1005, 17 February 2006).

The Dutch researchers started with a fairly simple laboratory situation, in which they asked several subjects to pick among four cars to buy, based on only four attributes. After four minutes of pondering (I don't know why Dijksterhuis and collaborators picked four as a magic number, but that's another issue), most subjects chose the right car – i.e., the one that objectively scored better than the other ones on the four attributes considered.

Good, but things got interesting when the task was made more difficult by increasing the number of attributes to 12, still with four cars in the running. Now people picked the best car in about 25% of the cases, i.e., consistent with a random choice. And here comes the real twist: researchers posed the 12-attribute problem again, but this time they distracted the subjects during the four minutes of decision making: surprisingly, more than half picked the right car!

What are we to make of this? For someone writing a blog called “rationally speaking” it would seem that psychological research is dealing a blow to the whole idea that rational, conscious thinking is all that useful precisely in those areas where we think it ought to be: to tackle complex problems. Indeed, Dijksterhuis comments in Science that conscious thought can only handle so many things at one time, and that the evaluation of many factors may best be left to the unconscious processing of information.

Yet, not all is lost for the rational thinker. According to Jonathan Schooler, a psychologist at the University of British Columbia commenting on the findings of the Dutch group, “What I think may be really critical is to engage in [conscious] reflection but not make a decision right away.” Indeed, Dijksterhuis himself says that whenever he is faced with an important decision, he does collect the relevant evidence and consciously thinks about it, but then postpones decision until his unconscious processing (also known as “intuition”) seems to reach an equilibrium. As he put it, “I sit on things and rely on my gut.” Of course, the question naturally arises of how to decide how long to sit on things, which of course often is a complex decision in itself, influenced by many factors...

30 comments:

  1. This comment has been removed by the author.

    ReplyDelete
    Replies
    1. Massimo, before trying to make sense of this and similar research outcome, it may be worth to wait until results have been thoroughly scrutinized and independently confirmed, in addition to taking a critical look at the actual research questions and methodology.

      It is not specific for this research, but the discipline of Psychology has been littered with poorly designed and controlled research most of which are difficult to replicate [1], not to speak about the many cases of scientific misconduct lately (e.g. Stapel, Smeesters [2,3]). Not surprisingly, cases of recent misconduct predominantly involve psychologists. We then have not even touched upon the eagerness in Psychology and Psychiatry (compared to other disciplines) to report positive results particularly on 'exciting' or 'eye-catching' topics [4].

      But let us indeed be fair: the research should speak for itself. Nevertheless reasons enough to be extremely careful in too eagerly embracing novel findings in psychological research, even if they have been published in high level journals.

      Marc.

      [1] https://www.sciencemag.org/content/335/6076/1558.full

      [2] http://news.sciencemag.org/scienceinsider/2012/06/rotterdam-marketing-psychologist.html

      [3] http://www.nature.com/news/2011/111101/full/479015a.html

      [4] Fanelli D (2010) "Positive" Results Increase Down the Hierarchy of the Sciences. PLoS ONE 5(4)
      http://www.plosone.org/article/info:doi%2F10.1371%2Fjournal.pone.0010068

      Delete
  2. For instance, see http://www.sciencedirect.com/science/article/pii/S0749597811001385. Citing the abstract: "First, an exact replication of Dijksterhuis et al. (2006a) study should indicate that unconscious decisions are superior to conscious decisions. Second, decisions should improve with duration of conscious thought. Third, unconscious decisions should be superior to conscious decisions, even if unconscious decisions are deliberated while having access to information. Fourth, unconscious decisions should be based on a weighting strategy. We report results of four studies, featuring 480 participants, that yield no evidence in favor of these predictions."

    ReplyDelete
  3. This comment has been removed by the author.

    ReplyDelete
  4. So, the research suggests that unconscious processing would be somehow superior to conscious. I would have no problem with processing at subtle levels concurrent with obvious levels, as I would see consciousness as a process of building sufficiency in the brain by gross anatomical input and ouput cycles. But my commonsense view would be of the subtle as being continually supportive of the obvious as a peak, rather than superior.

    An alternative would be to say, and this is very improvisational for this argument, that there is a limit to calculation by biological processes, somewhat like Braess' Paradox in the Measure of Doubt blog. Physically quantifiable examples of shortcuts that only create delays / stresses, noted in the blog, could apply if we allow less time / concentration in some analyses where a squinted view provides an outline of main roads and allowing for extra permutations & combinations by making calculations terminally clogs the process. The trick is knowing where that principle applies across nature and to what extent.

    PS. I have now checked back to the original 2006 blog as a cheat, and one answer rings very true, that we can fall into the trap of thinking too much. This would get back to Braess' Paradox. It would be a conscious process supported by the subconscious, but prone to fall apart with too many perms & combs...of the type measured in the study. I would take it one case at a time to outline the general application of the principle in psychology. Another research idea for you.

    ReplyDelete
    Replies
    1. You might want to research the difference between a logical paradox and a mathematical one first.

      Delete
    2. You will see if you investigate the Measure of Doubt blog I am referring to, that "Paradox" is their description, and I make a brief comment diagreeing with that description, but the word is a red herring to the argument, so I haven't persisted. I avoid red herrings, its a good time saving habit.

      Delete
    3. You're right and wrong at the same time. Braess' situation is not really paradoxical, just counterintuitive. But the ways our thought processes and functions have evolved to complement each of the others is nothing like evolving traffic routing problems.

      Delete
    4. Yep counterintuitive, but we do not know how thought processes have evolved except by the usual means of tracking evolution, which is open to the usual commonsense interpretations you may be applying, which is fine. Neuroscience has far to go, and we need to link genetic evolution to that model. "Nothing like"? is a bit heavy, I would say when we have a squinted conscious view (distracted perhaps) we may obviate alternatives other than basic points, as a process that reflects simply taking main roads rather than being enticed by "shortcuts". Nothing like? A matter of opinion and an open or closed mind when no explanations currently exist, worth further exploration. The brain is as physical a process as moving cars.

      Delete
    5. The brain, over the millennia has evolved itself by the intelligent trial and error process, whether helped by the stochastic or adaptive methods of mutation. And for an entirely different purpose than for routing the traffic of mechanical objects. (Which in the trial process, discovered an error.)
      In either event, time is a factor, and speed by which the brain transfers its signals has not been a problem. The increasing size and "traffic" of our prefrontal cortex have not slowed down our metal processes - quite the contrary.
      Without further exploration you came up with a bad analogy. Bad analogies are not considered logical.

      Delete
    6. You facts are dodgy there, and I doubt your logic, but you are having a try, which is good.

      Delete
    7. Your brain is as physical a process as moving cars.

      Delete
    8. Agreed, it is a routing process.

      Delete
    9. If you go ad hominem and it backfires in logic, own it.

      Delete
    10. I don't believe your brain is as physical a process as moving cars. You said it was, and I was mocking you by agreeing. Is that ad hominem? No more than telling me my facts are dodgy.
      I could have said it's really dumb to think your brain is as physical a process as moving cars, but I didn't.

      Delete
    11. A non-physical process? Good luck in the unforseeable future. In each case there is a causal sequence of physical events (a route), no point in trying to quibble that basic description except by reference to something non-physical or something that is not a process. You have had ample opportunity, so case closed.

      Delete
    12. AS physical a process AS moving cars. Where did you get a non-physical process from that? Case closed? You never had a case to close to start with.
      As I said, analogies are something you have a problem with, and it's not a physical one.

      Delete
    13. The problem is you don't understand working from general principles to particulars. The brain is entirely physical and entirely a process, as is the routing of cars. That's how to build to an analogy, which remains to be proven as I have only stated a broad starting principle. If you disagree, you need to show why the brain is not so. You have not done that because you fail to understand the process of reasoning, and have no relevant facts at your disposal to present anyway. So case closed, ample opportunity wasted in making irrelevant quibbles.

      Delete
    14. The brain is routing information that conveys meaning. It's thinking. Your road system isn't thinking at all. The cars are not thinking that get jammed up. Only the drivers are thinking, and when they find out that the other thinkers who developed the roads made a mistake, they will sooner or later correct it. On one side of your cockamamie analogy you have the thing that thinks and on the other side you have the thing that's thought about. Worst form of analogy ever.
      These opportunities that you think you have to offer don't exist. We all know you're trying to sell a book. If it makes no more sense than you're making here, you're unselling it.

      Delete
    15. Total obfuscation in your post. Always the incorrect facts too. You try too hard to hurt when its pointless as you are just a series of badly strung words in a post. I am not trying to sell my book, I am giving it away. Meaning has nothing to do with the issue, I am only looking at the process level. Anyone can read these posts, your approach, facts, and logic, and judge for themsleves. You are basically, all over the place, ad hominem, and to be avoided by me in future.

      Delete
    16. Meaning has nothing to do with the issue? Wow.

      I noted that your book was on sale for 20 Australian dollars. I must have misread that.

      But please do avoid me in the future.

      Delete
    17. Sadly I have to post to correct misinformation. Do go to the site where it explains I don't want to deal in books so you can have the PDF free, but if you want a book it is $20 to cover costs. You are dangerously bad Roy, and maybe more.

      Delete
    18. Bad? You should have Massimo do a book review. Or anybody for that matter.

      Delete
    19. Don't evade Roy, there is a lesson here for readers about impugning motive as a last deperate resort, which borders on and becomes ad hominem if tendentious and distorted, Roy. Case closed, Roy.

      Delete
  5. Minor point: Jonathan Schooler *was* at UBC in 2006, but he is now at UCSB:

    ReplyDelete
    Replies
    1. http://www.psych.ucsb.edu/people/faculty/schooler/

      Delete
  6. I'm agree to some extent as I make my decisions of many sort in just seconds. But I can't ignore the unconscious processing of mind

    ReplyDelete
  7. This comment has been removed by the author.

    ReplyDelete
    Replies
    1. Further thoughts as to the physical routing. If a connection, like the access road in the Measure of Doubt Blog, opens up for efficient processing of some connection creating an aspect of thought, it might not necessarily facilitate another aspect. In my model, this would be a zero sum game, with limited resources to share between connections for priority, and perhaps for levels building.

      The idea would be that some connections may be usefully opened for comprehensive or conclusive thought, whilst they are unnecessary for the squinted view. If main roads and access roads exist in the brain across prioritizations, then the principle suggested by Braess might apply. I keep an open mind for the time being, as it may be so useful if true that it is worth persisting with passing thoughts.

      Delete

Note: Only a member of this blog may post a comment.