Mind Scans to Map Decision Making Mechanics 218
rrangel writes "Newsweek is running an article on the fMRI, which tracks brain function by measuring blood flow, and using it for watching the mechanics of economics and choice. Best quote on economic choice: '... there is no quantity of juice sufficient to get a male monkey to look away from the hindquarters of a female in estrus.' H. Hefner has known that all along."
Two things... (Score:5, Interesting)
Second, juice may not get him. but cocaine will. I saw a study that showed a monkey will give up everything, including food and sex, for cocaine.
Rationality and expected value (Score:5, Interesting)
Dropping $20 on an array of Mega Millions tickets is mathematically irrational, but with or without that $20, my life for the next two weeks will be about the same. If I were to win, however, even the second-best prize, it would enable me to purchase a nice house.
When it's a matter of playing a game where the expected value of my dollar is $0.95, but I'm more likely to win $2 or $3, why bother? But even if the expected value of my dollar is $0.75 or less with a prize of many million and many over $100k, despite the miniscule chances of winning, it would change my life.
Of course, if I had an expected value of $1.05 for my dollar, I'm smart enough to play consistently even if my dollar only wins a little at a time.
-PM
Re:Rationality and expected value (Score:3, Interesting)
This isn't strictly relevant, but has anyone figured out why most people get the probablities wrong in Don't Get The Goat [grand-illusions.com] (no relation to goatse). Even intelligent people often get it wrong. I remember spending ages trying to explain it to an intelligent person with good maths skills - and they still didn't understand.
Re:Two things... (Score:5, Interesting)
One can see motor movements in the brain. I tell you to move your finger (or think about moving your finger ) and I can see in the brain the area that: hears me say "move your finger" then the language area that interprets "move your finger" and the pre-motor area firing, then the motor area firing.
There are a million tests that can be given in the MR scanner. Some of them can be really funny.
Examples on request.
'there is no quantity of juice sufficient' (Score:5, Interesting)
Re:Two things... (Score:2, Interesting)
Next take a patient in an fMRI study. A very typical normalization task is to simply use a soft brush to rub say the left hand. Neuroanatomists have known for pushing a hundred years where in the brain (specifically where on the humonculus) this will be registered. (By reverse engineering: damage to some part of the brain leaves the patient insensitive.) Even trivial analysis of the fMRI results will show activation in the appropriate area.
On a more fundamental level, there is no causation EVER. Any first year graduate student in physics can tell you that. At the micro-scale all things are probabilistic. At the macro-scale the incoherence of things across the micro-scale leads to well defined and narrow probability distributions further leading to the illusion of concrete objects. The very idea that "causation" exists is due to the completeness of the illusion and the underlying incoherency.
Consciousness Theory (Score:5, Interesting)
In this book he uses multi-dimensional scaling analysis of fMRI scans to predict past and future states of the same brain, as well as doing the same thing with artificial networks.
It then uses the evidence from this research to propose what (to me, at least) is the first really solid explanation for what consciousness may actually "be".
The book is written in 2 parts... the first one is a detective novel where the main character is a Phenomenologist and in the process of solving a murder finds a theory of consciousness. The 2nd part of the book is a factual appendix describing the work.
Awesome stuff, and I highly recommend it to anyone interested in neural nets and AI.
Re:Two things... (Score:5, Interesting)
Second test: Stroop. Never seen so many smart people get so frustrated. A word is presented: "RED" It is written in green ink. What color is the ink? Then, just as you get the hang of it, what is the word?
Third: Nicotine addictions. Drop a bolus of nicotine into a willing research subject. I've heard "That's better than sex" to "Ohhhhhhh" to "I think I wet myself"
More later.
Re:Two things... (Score:2, Interesting)
True, I've read about a similar experiment with a monkey. The experiment with the monkey is a crude measurement of how addictive a substance is. Basically the monkey has to press a button a certain number of times to get a hit of some substance. Each time the monkey gets a hit it must make more presses than the previous time. By the end of the experiment the monkey was pressing the button more than 13,000 times for a single hit of cocaine. This is far above any other drug.
love-hate relationship of Science and Media (Score:5, Interesting)
I've actually started a blog devoted to megnetoic resonance imaging (http://refscan.blogspot.com/ [blogspot.com] and would like to invite anyone else interested in MRI to visit and comment. Our patron Saint is Magneto
Comment removed (Score:3, Interesting)
Re:Perils of an incomplete model (Score:3, Interesting)
It's economics, not game theory, that assumes human rationality. In 90% of circumstances, that assumption accurately predicts behavior. It's the other 10% when tribal mentalities (including trust, disgust, vengeance, anger, jealousy, etc.) all kick in that the axioms need to be reexamined.
Comment removed (Score:4, Interesting)
booorrrrrrinnnnnnggggg (Score:1, Interesting)
Re:what advertisers won't do (Score:3, Interesting)
This is a fair question. I'm in one of the labs mentioned in this article, so I'll try giving it a shot.
Most basic research is often a number of steps removed from applicability. Most non-scientists do not think research is useful unless it has clear applicability. One could make the subtle argument that an increase in human knowledge, especially an increase in knowledge about ourselves, is an intrinsic good and elevates us as a society. I'm not going to make this argument alone, but I would like to throw it out as one "pre-emptive" rationale.
One could argue that that glue of human behavior is decision-making. Every voluntary action is preceded with a decision to make that action. The decisions we make determine much of the course of our lives, and the amalgam of decisions determine the course of society. It's therefore in our interest to undetstand the basis of decision making. That basis is a neural one, as the decisions you make are the result of an interplay of mechanisms in your head.
Many decisions we make are flawed. Decision-researchers (like Nobel laureate Daniel Kahneman) have demonstrated that we often fail to make rational choices, and that these failures are systematic. This is counter to the standard economic model of decision-making which assumes that players in a market are rational actors seeking to maximize their interests. If this is not true, then standard economic models are not taking the realities of human psychology into account. In aggregate, these irrational choices and persistent failures to account for them may have massive impact on economies. We are now in a position to actually elucidate not only the systematicity of our irrationalities, but also their very basis in the brain: to understand actually what specialized neural subsystems color and bias our decisions. Hopefully, one future result will be a more accurate (if less precise) calculus of decision making.
The potential for clinical impact is also tremendous. One of the chief burdens of most psychopathologies is a profound impairment in decision-making. Schizophrenics, autistics, sociopaths, anxiety and OCD patients, attention-deficit patients, depressives...literally any psychopathology you name has as one hallmark a particular sort of failure in the decisions made by patients. The flaw may be different in each, of course: e.g., autistics fail to consider the mental state of others in making their choices, schizophrenics may perseverate on a type of decision, while phobics greatly overestimate the impact of a particular alternative in certain choices. Nevertheless, knowing just this has very limited chance for helping us come up with effective therapies. Unless we understand how the "normal" brain makes decisions (what systems, what mechanisms, what areas, what neurochemistry), we will have very little to say about how to fix it when things go wrong. As it is, much psychopharmacology is a guessing game. We know (often by serendipity) that certain drugs are effective in certain in clinical conditions, but for many we still have very, very little understanding as to why the drugs are effective or even how they work.
Finally, you say that this sort of research is ripe for abuse by advertisers. Well, all knowledge is subject to abuse. But, in point of fact, this sort of research could be a huge boon for both consumers and advertisers (or at least for corporations that adverstise) alike. Advertising is, on the whole, annoying to consumers (I call us "consumers" because we're in a commercial context here). The dominant model of advertising is saturation: The more impressions a brand makes on you, the better. This is a very simplisitic model that is grossly ignorant of human psychology and neuroscience. It is true that companies see an uptick in sales after an advertising "barrage," but this completely ignores the infringement and frustration most of us feel being bombarded with images and force-fed advertisin
Jonathan Edwards (Score:5, Interesting)
Jonathan Edwards said that Free Will consists of the mind choosing that which it finds most pleasing or agreeable based on what it knows at that moment. I think considerations like this drove Soren Kierkegaard mad choosing to make himself miserable because it pleased him to exercize his will so.
It would be interesting to know what this continent's most thoughtful Calvinist would think about these experiments. I think he'd be pleased, but he might differ on the interpretations of the findings.