Be careful about bringing up the old "nature versus nurture" argument; it's a fallacious one.
There is nothing that is entirely genetically determined or environmentally deterimined. And this is not a trivial position, either--it's very easy to start arguing that one thing or the other dominates.
It is guaranteed that certain neural features are structured, shaped by genetic features, but those genetic features can only be expressed within the context of other genetic, epigenetic, and environmental features.
Yes, there are studies showing that infants can imitate facial expressions. Does that mean that the ability to imitate is part of the genetic architecture of a human being? Yes and no. Yes, there are genetic features that allow for imitative behavior--e.g., beetles don't run around copying one another, but chimps and humans do. Yet at the same time, there are almost certainly environmental or developmental features that also must be present for imitation to take place. Brains are not like machines that are built according to some blueprint and then stuck out into the world; part of the power of the brain is the fact that its structure is determined dynamically with the environment it is processing.
There is no "genes or environment", just which genes and which environment.
Really, it's not just about repeated playing with the same individual, but whether information about your past moves is transmitted on to future competitors.
Novak et al.(?) have a series of papers in Science/Nature (? they blur together at some point) where they demonstrate that "gossip" increases the likelihood of cooperation tremendously. The efficiency with which information on any given individual's past moves is transmitted to others is related to the rate of cooperation in a system. Other papers (I can't remember who--Olsson?) have suggested similar effects vis-a-vis ostracism, etc. Basically, when everyone knows who you are, it's in your best interest to be nice.
In fact, some talk about "second-order" cooperation--whether or not you pass on information you have about a defector.
Novak et al. introduce prisoner's dilemma with communication as a way of showing how altruism develops in a social population. In fact, there are probably some good arguments to be made that language evolved jointly with altruism--one drove the other to stability, so to speak.
Now, about *games*: I've always thought one of the biggest appeals of adventure games is their non-zero-sum nature. In adventure games/IF the point is all about exploration and discovery. You don't want to keep playing to gain resources of some sort (kills,cities,etc.), or to avoid resource loss of some sort (kills,cities,etc.), but rather to see what's behind the door or down the hole.
It's entirely possible that the attribution of a "mirror" function to particular neurons is specious.
There is a more general question, however, of whether or not there is some localization of imitative function in the brain. Imitation in itself is a very complex phenomenon, and one that now has been documented quite extensively in the cognitive psych literature. Think about it: how the heck do infants imitate facial expressions?
There has to be some self-representation in place and a way of mapping that representation onto the other representation. Perhaps learning to segment facial features coincides with coordination of self features (joint unsupervised learning of two sets of features simultaneously, for example).
If you accept imitation as a distinct cognitive process, you almost certainly have to argue that it has neural substrates. Now, those neural substrates will take a certain form, and cannot be completely localized to a particular region--nothing is. However, if imitation is like anything else, there probably are certain regions that are consistently involved more than others.
There have been some imaging studies on this already, and some consensus is taking shape that there are, in fact, regions of the brain involved in imitation more than others (the left occipital/temporal region, is one potential culprit, for example, if I recall correctly; there is also every reason to expect that the "error checking" function of the prefrontal cortex is also involved).
The idea of "mirror cells" may or may not be accurate--the more important issue is how imitation is executed neuronally. As it is a given that imitation does have some neural substrate, the interesting question is not *whether* there are brain features involved in copying others, but what those features are.
As a number of posts have indicated, there are issues to be resolved in this study.
The best way to resolve them, though, is *empirically*:
(1) IQ or motivation? Just give items to assess both! Just ask them "on a scale of 1 to 7, how motivated were you to succeed in this game?" Ask this a couple of times different ways to increase reliability. Then you can statistically control for motivation in calculating performance/achievement correlation. Do the same for experience with a given game, etc... Just watch out for ceiling effects--many students probably play games on some basis.
(2) Which game? Try a couple and find out which works best! Different IQ tests load on g to different degrees, and load on different subtests to different degrees; it would be interesting to know which test loads on which aspect of intelligence. It would be interesting to know that performance on certain games was correlated with performance on certain other games--you might be able to empirically identify game proficiency "factors". Game developers would probably be very interested in such info, actually.
(3) Don't use multiplayer games. This only complicates the interpretation of results, as the opponent changes from individual to individual. With AI/computer, opponent/difficulty is constant for everyone. Then you can just partial out experience with a game statistically (see 1 above).
(4) Don't worry too much about sampling--participants in these psych students tend to be from all over the place. Bright, motivated students tend to participate in studies out of a sense of conscientiousness, doing the right thing, etc., and struggling students tend to particpate to get some sort of reward for something. Once your sample is reduced to college undergrads, anything else is sort of a secondary concern.
A discussion of languages on slashdot inevitably devolves into matters of preference and the recognition that different languages are different tools for different jobs.
Nevertheless, it seems Ruby has not recieved as much press as it should. I have yet to hear one bad thing about Ruby itself (ignoring observations about the relative size of its support base relative to Perl and Python, which may simply be the result of Ruby's age). This is more than I can say about Python or Perl.
I have been using Perl extensively until this point, but Ruby sounds very appealing. Is there anyone who has tried Ruby and not been impressed enough to switch? Why?
The defense department and a company bent on copyrighting the entire human genome teaming up to build the most powerful computer known to mankind.
That way, after Celera copyrights your genes, they can sell them to the NSA at a discount for "national security" research.
Working in the realm of genomics, this is something that has greatly troubled me for some time.
Part of the problem is the confusion that exists over exactly what the gene patents are for.
As some posts have pointed out, the patent filer usually claims that the genes themselves are not being patented, only the use of those genes for some technological benefit.
Excuse me? Where do you draw the line? At what point does my wanting to sequence someone for research on bipolar disorder become something that violates the patent? What the heck does it mean that I have to contact an "owner" to get the sequence of polymorphism of a human gene existing naturally in the population? Generally, it's not a problem, or only a problem of convenience, but sometimes it is.
Consider a patent for "human heart tissue". I could use heart tissue for some treatment at some point in the future. Does this mean it can be patented? Please!
Another problem is confusion over knowledge over function. Yes, patent law requires knowledge of function. However, this requirement is generally obviated by "predicting" function from sequence data. Thus, understanding of function is claimed when not actually known.
I have written many legislators about this, and I am convinced that a third problem is their (legislators' and patent officers') lack of understanding of this problem. They are truly well-meaning, and concerned when they find out about it, but often don't know enough about the subject to know what to do. Many of the existing gene patents would not have been issued under current restrictions, enacted after patent offices realized how they were being conned.
Of course, many gene patents are legitimate, and I don't want to detract from them. But the problem is knowing when one is, when one isn't, and how many existing patents aren't. It seems better in my mind not to allow patenting of any naturally occuring sequence variants. If this were the case, then why are so many filers being criticized for matching their sequences against existing databases to predict function?
Going to to cool off now...
The possibility of the U.S. being less hesitant to initiate military conflict is greatly disturbing.
However, arguments that somehow the development of unmanned aircraft is a bad thing are short-sighted.
Consider the alternative: manned aircraft. Unmanned and manned aircraft are the only options. No aircraft is not an option, nor will it be for some time; humans are too imperfect to count on not to ever be aggressive.
As both manned and unmanned aircraft result in horrible devastation to the targets, the only cost to reduce is that to the targeter.
Perhaps Americans are cowards. But if self-preservation is cowardice, everyone wants to be one, and, unfortunately, the biggest coward wins.
I think it is time to redefine--or revisit--what constitutes "good" software.
Is IE an acceptable product? Of course. A rather nice product, if defined in terms of speed and functionality. Maybe not the best, but certainly a contender, as one can see by the posts.
However, speed and functionality are not the only issues of relevance in defining a "good" product.
Consider Office 2000, for example. I had always believed that Office was the only MS product that was deservedly superior to its competitors, and felt comfortable using it. I was disgusted, however, when receiving Office 2000, to find myself being required to send all sorts of information about myself to MS before I could use it. Not incredibly important info, but info nonetheless.
Is this sort of unnecessary intrusion into my life somehow less of an issue than speed or performance? No, of course not. When I have a choice between two products, such as Star Office and MS Office, that are of comparable performance, ridiculous requests for info about me become very relevant to my choice of product.
It is important to remember that privacy, security, and freedom are just as important as issues of speed and function in defining good software. All other things being approximately equal, which they increasingly are, I choose the product that respects me and my choice more.
Open source was never supposed to be only about functionality. It is one large part of its appeal and utility, but not the only one.
I think sometimes arguments about Netscape, IE, Mozilla, can get bogged down in technical details about speed, etc., which matter less and less anyway. Mozilla and IE are comparable now, will be even more so soon, and we should shift discussion back to other issues that distinguish products more.
However, you have to be careful about claiming that self-reported information is suspect. It all depends on how you operationalize constructs.
For example, consider "harmful". How do you define that? If in terms of objective consequences, perhaps self-reported data is suspect.
However, for emotions and other subjective or perceptual phenomena, self-report is probably the best source of information that we have. If people say they like the net, they almost certainly like the net.
The analogy with drug abusers posted elsewhere isn't completely accurate, either. Most drug abusers don't feel good. They might not understand why they feel bad, but they feel bad about a lot of things. My guess is that most abusers do not, in fact, consider their habit a positive thing. The question is how bad they see it as being.
Moreover, regardless of the validity of self-report, there apparently IS a source of observer information in this case: parents about their children. In this case, we have informants reporting about the effects of use on users other than themselves. So self-report validity sort of becomes irrelevant with that data.
I agree that we have to be careful about interpreting self-report. But it is one of the richest sources of information we have, and it is often quite valid.
Slashdot Mirror
Be careful about bringing up the old "nature versus nurture" argument; it's a fallacious one.
There is nothing that is entirely genetically determined or environmentally deterimined. And this is not a trivial position, either--it's very easy to start arguing that one thing or the other dominates.
It is guaranteed that certain neural features are structured, shaped by genetic features, but those genetic features can only be expressed within the context of other genetic, epigenetic, and environmental features.
Yes, there are studies showing that infants can imitate facial expressions. Does that mean that the ability to imitate is part of the genetic architecture of a human being? Yes and no. Yes, there are genetic features that allow for imitative behavior--e.g., beetles don't run around copying one another, but chimps and humans do. Yet at the same time, there are almost certainly environmental or developmental features that also must be present for imitation to take place. Brains are not like machines that are built according to some blueprint and then stuck out into the world; part of the power of the brain is the fact that its structure is determined dynamically with the environment it is processing.
There is no "genes or environment", just which genes and which environment.
Really, it's not just about repeated playing with the same individual, but whether information about your past moves is transmitted on to future competitors.
Novak et al.(?) have a series of papers in Science/Nature (? they blur together at some point) where they demonstrate that "gossip" increases the likelihood of cooperation tremendously. The efficiency with which information on any given individual's past moves is transmitted to others is related to the rate of cooperation in a system. Other papers (I can't remember who--Olsson?) have suggested similar effects vis-a-vis ostracism, etc. Basically, when everyone knows who you are, it's in your best interest to be nice.
In fact, some talk about "second-order" cooperation--whether or not you pass on information you have about a defector.
Novak et al. introduce prisoner's dilemma with communication as a way of showing how altruism develops in a social population. In fact, there are probably some good arguments to be made that language evolved jointly with altruism--one drove the other to stability, so to speak.
Now, about *games*: I've always thought one of the biggest appeals of adventure games is their non-zero-sum nature. In adventure games/IF the point is all about exploration and discovery. You don't want to keep playing to gain resources of some sort (kills,cities,etc.), or to avoid resource loss of some sort (kills,cities,etc.), but rather to see what's behind the door or down the hole.
It's entirely possible that the attribution of a "mirror" function to particular neurons is specious.
There is a more general question, however, of whether or not there is some localization of imitative function in the brain. Imitation in itself is a very complex phenomenon, and one that now has been documented quite extensively in the cognitive psych literature. Think about it: how the heck do infants imitate facial expressions?
There has to be some self-representation in place and a way of mapping that representation onto the other representation. Perhaps learning to segment facial features coincides with coordination of self features (joint unsupervised learning of two sets of features simultaneously, for example).
If you accept imitation as a distinct cognitive process, you almost certainly have to argue that it has neural substrates. Now, those neural substrates will take a certain form, and cannot be completely localized to a particular region--nothing is. However, if imitation is like anything else, there probably are certain regions that are consistently involved more than others.
There have been some imaging studies on this already, and some consensus is taking shape that there are, in fact, regions of the brain involved in imitation more than others (the left occipital/temporal region, is one potential culprit, for example, if I recall correctly; there is also every reason to expect that the "error checking" function of the prefrontal cortex is also involved).
The idea of "mirror cells" may or may not be accurate--the more important issue is how imitation is executed neuronally. As it is a given that imitation does have some neural substrate, the interesting question is not *whether* there are brain features involved in copying others, but what those features are.
As a number of posts have indicated, there are issues to be resolved in this study.
The best way to resolve them, though, is *empirically*:
(1) IQ or motivation? Just give items to assess both! Just ask them "on a scale of 1 to 7, how motivated were you to succeed in this game?" Ask this a couple of times different ways to increase reliability. Then you can statistically control for motivation in calculating performance/achievement correlation. Do the same for experience with a given game, etc... Just watch out for ceiling effects--many students probably play games on some basis.
(2) Which game? Try a couple and find out which works best! Different IQ tests load on g to different degrees, and load on different subtests to different degrees; it would be interesting to know which test loads on which aspect of intelligence. It would be interesting to know that performance on certain games was correlated with performance on certain other games--you might be able to empirically identify game proficiency "factors". Game developers would probably be very interested in such info, actually.
(3) Don't use multiplayer games. This only complicates the interpretation of results, as the opponent changes from individual to individual. With AI/computer, opponent/difficulty is constant for everyone. Then you can just partial out experience with a game statistically (see 1 above).
(4) Don't worry too much about sampling--participants in these psych students tend to be from all over the place. Bright, motivated students tend to participate in studies out of a sense of conscientiousness, doing the right thing, etc., and struggling students tend to particpate to get some sort of reward for something. Once your sample is reduced to college undergrads, anything else is sort of a secondary concern.
Have fun!
A discussion of languages on slashdot inevitably devolves into matters of preference and the recognition that different languages are different tools for different jobs.
Nevertheless, it seems Ruby has not recieved as much press as it should. I have yet to hear one bad thing about Ruby itself (ignoring observations about the relative size of its support base relative to Perl and Python, which may simply be the result of Ruby's age). This is more than I can say about Python or Perl.
I have been using Perl extensively until this point, but Ruby sounds very appealing. Is there anyone who has tried Ruby and not been impressed enough to switch? Why?
The defense department and a company bent on copyrighting the entire human genome teaming up to build the most powerful computer known to mankind. That way, after Celera copyrights your genes, they can sell them to the NSA at a discount for "national security" research.
Working in the realm of genomics, this is something that has greatly troubled me for some time. Part of the problem is the confusion that exists over exactly what the gene patents are for. As some posts have pointed out, the patent filer usually claims that the genes themselves are not being patented, only the use of those genes for some technological benefit. Excuse me? Where do you draw the line? At what point does my wanting to sequence someone for research on bipolar disorder become something that violates the patent? What the heck does it mean that I have to contact an "owner" to get the sequence of polymorphism of a human gene existing naturally in the population? Generally, it's not a problem, or only a problem of convenience, but sometimes it is. Consider a patent for "human heart tissue". I could use heart tissue for some treatment at some point in the future. Does this mean it can be patented? Please! Another problem is confusion over knowledge over function. Yes, patent law requires knowledge of function. However, this requirement is generally obviated by "predicting" function from sequence data. Thus, understanding of function is claimed when not actually known. I have written many legislators about this, and I am convinced that a third problem is their (legislators' and patent officers') lack of understanding of this problem. They are truly well-meaning, and concerned when they find out about it, but often don't know enough about the subject to know what to do. Many of the existing gene patents would not have been issued under current restrictions, enacted after patent offices realized how they were being conned. Of course, many gene patents are legitimate, and I don't want to detract from them. But the problem is knowing when one is, when one isn't, and how many existing patents aren't. It seems better in my mind not to allow patenting of any naturally occuring sequence variants. If this were the case, then why are so many filers being criticized for matching their sequences against existing databases to predict function? Going to to cool off now...
The possibility of the U.S. being less hesitant to initiate military conflict is greatly disturbing.
However, arguments that somehow the development of unmanned aircraft is a bad thing are short-sighted.
Consider the alternative: manned aircraft. Unmanned and manned aircraft are the only options. No aircraft is not an option, nor will it be for some time; humans are too imperfect to count on not to ever be aggressive.
As both manned and unmanned aircraft result in horrible devastation to the targets, the only cost to reduce is that to the targeter.
Perhaps Americans are cowards. But if self-preservation is cowardice, everyone wants to be one, and, unfortunately, the biggest coward wins.
I think it is time to redefine--or revisit--what constitutes "good" software.
Is IE an acceptable product? Of course. A rather nice product, if defined in terms of speed and functionality. Maybe not the best, but certainly a contender, as one can see by the posts.
However, speed and functionality are not the only issues of relevance in defining a "good" product.
Consider Office 2000, for example. I had always believed that Office was the only MS product that was deservedly superior to its competitors, and felt comfortable using it. I was disgusted, however, when receiving Office 2000, to find myself being required to send all sorts of information about myself to MS before I could use it. Not incredibly important info, but info nonetheless.
Is this sort of unnecessary intrusion into my life somehow less of an issue than speed or performance? No, of course not. When I have a choice between two products, such as Star Office and MS Office, that are of comparable performance, ridiculous requests for info about me become very relevant to my choice of product.
It is important to remember that privacy, security, and freedom are just as important as issues of speed and function in defining good software. All other things being approximately equal, which they increasingly are, I choose the product that respects me and my choice more.
Open source was never supposed to be only about functionality. It is one large part of its appeal and utility, but not the only one.
I think sometimes arguments about Netscape, IE, Mozilla, can get bogged down in technical details about speed, etc., which matter less and less anyway. Mozilla and IE are comparable now, will be even more so soon, and we should shift discussion back to other issues that distinguish products more.
I have only read the Star-Tribune coverage.
However, you have to be careful about claiming that self-reported information is suspect. It all depends on how you operationalize constructs.
For example, consider "harmful". How do you define that? If in terms of objective consequences, perhaps self-reported data is suspect.
However, for emotions and other subjective or perceptual phenomena, self-report is probably the best source of information that we have. If people say they like the net, they almost certainly like the net.
The analogy with drug abusers posted elsewhere isn't completely accurate, either. Most drug abusers don't feel good. They might not understand why they feel bad, but they feel bad about a lot of things. My guess is that most abusers do not, in fact, consider their habit a positive thing. The question is how bad they see it as being.
Moreover, regardless of the validity of self-report, there apparently IS a source of observer information in this case: parents about their children. In this case, we have informants reporting about the effects of use on users other than themselves. So self-report validity sort of becomes irrelevant with that data.
I agree that we have to be careful about interpreting self-report. But it is one of the richest sources of information we have, and it is often quite valid.