Notices
Results 1 to 7 of 7
Like Tree1Likes
  • 1 Post By Strange

Thread: Ethical question about behaviour study

  1. #1 Ethical question about behaviour study 
    New Member
    Join Date
    Mar 2014
    Posts
    3
    My name is Peter Gerber and I'm at the moment included in a research project about the predictability of human behaviour. Our researches are based on former research summarised at: northeastern.edu/news/stories/2010/02/network_science.html

    This study showed a 93 percent predictability of human travel behaviour. Since travel destinations can change through outside influences like cancelled flights and weather, we analysed data that is created inside the human brain without outside influences. We examined a television series for nontrivial patterns and found one that matched for every episode. The probability that this pattern is only caused by chance could be calculated to a one in a million chance. So we were able to verify the former results.
    My question to this topic is, if we discover that human behaviour is fully predictable by knowing the right pattern and as long as current activities are not interrupted through random events, is it ethical defensible to publish this research? I wouldn't liked to be involved in a research project that made everyone calculable. We already achieved some significant progress and I'm unsure whether to further participate or not. I would like to hear an other opinion.

    Our current progress is temporary stored at: vixra.org/pdf/1403.0082v1.pdf


    Reply With Quote  
     

  2.  
     

  3. #2  
    Genius Duck Moderator Dywyddyr's Avatar
    Join Date
    Jan 2013
    Location
    Scunthorpe, UK
    Posts
    11,765
    Quote Originally Posted by pege View Post
    We examined a television series for nontrivial patterns and found one that matched for every episode.
    Really?
    Not according to your own paper.
    We tested this pattern on data it was not designed for. We assumed for the first season the episodes would start after the opening credits. 24 times tested, the pattern fit 12 times and did not fit 12 times,
    I'm highly sceptical.


    "[Dywyddyr] makes a grumpy bastard like me seem like a happy go lucky scamp" - PhDemon
    Reply With Quote  
     

  4. #3  
    New Member
    Join Date
    Mar 2014
    Posts
    3
    Quote Originally Posted by Dywyddyr View Post
    Quote Originally Posted by pege View Post
    We examined a television series for nontrivial patterns and found one that matched for every episode.
    Really?
    Not according to your own paper.
    True. As we tested it for the first time on the fifth season it didn't fit for one episode. After a small adjustment the pattern fit for this one too and still for all other episodes.

    Quote Originally Posted by Dywyddyr View Post
    We tested this pattern on data it was not designed for. We assumed for the first season the episodes would start after the opening credits. 24 times tested, the pattern fit 12 times and did not fit 12 times,
    I'm highly sceptical.
    For mentioning three examples: In 1x18 five persons appear together after the opening credit. This combination is not allowed at the beginning so the pattern don't fit. In 1x11 a person is affected negatively, but this don't fit with the pattern, because the previous movement through the pattern is fixed through two persons appear at the same time. In 1x23 the pattern fits nearly until the end, but then there is nothing new for longer than 3 minutes. This is described in detail in Appendix B.
    Reply With Quote  
     

  5. #4  
    Genius Duck Moderator Dywyddyr's Avatar
    Join Date
    Jan 2013
    Location
    Scunthorpe, UK
    Posts
    11,765
    Quote Originally Posted by pege View Post
    True. As we tested it for the first time on the fifth season it didn't fit for one episode. After a small adjustment the pattern fit for this one too and still for all other episodes.
    Really?
    I'll quote your paper again:
    did not fit 12 times
    For mentioning three examples: In 1x18 five persons appear together after the opening credit. This combination is not allowed at the beginning so the pattern don't fit. In 1x11 a person is affected negatively, but this don't fit with the pattern, because the previous movement through the pattern is fixed through two persons appear at the same time. In 1x23 the pattern fits nearly until the end, but then there is nothing new for longer than 3 minutes. This is described in detail in Appendix B.
    Yeah.
    The fact that you're working on a scripted, time-limited, budget-limited TV programme (that has a particular story to tell) may have something to do with me being dubious as to how - or if - your results read across to real life.
    "[Dywyddyr] makes a grumpy bastard like me seem like a happy go lucky scamp" - PhDemon
    Reply With Quote  
     

  6. #5  
    New Member
    Join Date
    Mar 2014
    Posts
    3
    Random number generators are deterministic (grasshopper3d.com/forum/topics/what-are-random-seed-values). Creating 3 times 20 random numbers with the same seed/initial values lead to the same random numbers:

    First try:      "4 8 1 2 8 5 1 2 3 2 8 5 0 5 2 9 7 3 6 6"
    Second try: "4 8 1 2 8 5 1 2 3 2 8 5 0 5 2 9 7 3 6 6"
    Third try:     "4 8 1 2 8 5 1 2 3 2 8 5 0 5 2 9 7 3 6 6"

    Different initial values lead to different random numbers. Biological computers have their own random generating algorithms. The stoats hunting behaviour is composed out of the most possible random behaviour to approach to its prey but not getting recognized as threat (video.nationalgeographic.com/video/worlds-deadliest-ngs/deadliest-stoat). The more random the stoats behaviour is, the likelier the stoat survives. We had no videos of stoats chasing rabbits to analyse this behaviour, for that reason we analysed random data created by humans. To create different plots for a television series under the same circumstances (same persons, same places, science fiction) appeared to be similar to equal initial values for random number generators and so we looked if there is a pattern that proves similarities.

    How mentioned at the results we found a pattern that described the data set like this:

    First try:      "4 8 1 2 8 5 1 2 3 2 9 5 0 5 2 9 7 3 6 6"
    Second try: "4 2 9 7 1 9 1 6 0 5 9 6 7 3 4 8 8 3 5 4"
    Third try:     "8 5 9 2 4 8 3 5 9 3 9 1 9 0 2 8 8 1 7 6"

    pattern:        4 8 1 2 8 5 1 2 3 2 9 5 0 5 2 9 7 3 6 6
                     or 8 2 9 7 1 9 3 6 0 5    6 7 3 4 8 8 1 5 4
                     or    5       4 8    5 9 3    1 9 0             7

    This means the first number always has to be 4 or 8. The second number always has to be 2, 5 or 8. The third number always has to be 1 or 9. For real random data the pattern would be for trying enough: The first number can be 1, 2, ..., 9. The second number can be 1, 2, ..., 9. And so on. We found a pattern for 76 episodes that had a clear shape how persons had to appear and had to be affected. We tested this pattern at a later starting point for season 1. Random number generators with equal preconditions create the same random numbers, but if the first numbers getting erased the previous pattern doesn't fit anymore. For Example:

    First try:      "4 8 1 2 8 5 1 2 3 2 8 5 0 5 2 9 7 3 6 6"
    Second try: "4 8 1 2 8 5 1 2 3 2 8 5 0 5 2 9 7 3 6 6"
    Third try:     "4 8 1 2 8 5 1 2 3 2 8 5 0 5 2 9 7 3 6 6"

                         pattern:        4 8 1 2 8 5 1 2 3 2 9 5 0 5 2 9 7 3 6 6
    or

    First try:      "4 8 1 2 8 5 1 2 3 2 9 5 0 5 2 9 7 3 6 6"
    Second try: "4 2 9 7 1 9 1 6 0 5 9 6 7 3 4 8 8 3 5 4"
    Third try:     "8 5 9 2 4 8 3 5 9 3 9 1 9 0 2 8 8 1 7 6"

                         pattern:        4 8 1 2 8 5 1 2 3 2 9 5 0 5 2 9 7 3 6 6
                                          or 8 2 9 7 1 9 3 6 0 5    6 7 3 4 8 8 1 5 4
                                          or    5       4 8    5 9 3    1 9 0             7

    Searching for this pattern after this later starting point did fit 12 times and did not fit 12 times. So we can say this pattern can be created through chance with a 50% probability. We searched for this pattern at the fifth season. Before a small adjustment the pattern fit 23 times and did not fit 1 time. So it should be proved that it is unlikely that this pattern is created through chance and don't has an actually origin.

    Hopefully not to complicated
    Reply With Quote  
     

  7. #6  
    Forum Cosmic Wizard
    Join Date
    Dec 2013
    Posts
    2,408
    "This paper deals with the question what patterns in seeminglyrandom data tell us about their human creators."

    Are you trying to say that what you thought might be random, the order of appearance of characters in soap operas, is not really random after all?
    I really don't find that very surprising because there are certain orders imposed by the habits of the script writers.

    Random number generators are deterministic (grasshopper3d.com/forum/topics/what-are-random-seed-values). Creating 3 times 20 random numbers with the same seed/initial values lead to the same random numbers:
    Which is why random number generators often use random seed generators too. Gambling Casino slot machines usually rely on the internal timing clock signal at the moment the gambler pushes the play button to generate the seed number for the random number generator in the machine.


    Also of interest:Approaching the Limit of Predictability in Human MobilityXin Lu, et. al. Sept 2013.
    Approaching the Limit of Predictability in Human Mobility : Scientific Reports : Nature Publishing Group

    I note that predictability from both of the cellphone tracking projects reaches an upper limit, but it is still a reminder that we tend to be creatures of habit. It is not just habit at work though because there are environmental forces at work too that shape where people tend to go.

    I would say you have nothing to worry about as far as ethics are concerned.
    Reply With Quote  
     

  8. #7  
    Brassica oleracea Strange's Avatar
    Join Date
    Oct 2011
    Location
    喫茶店
    Posts
    16,965
    Quote Originally Posted by dan hunter View Post
    Gambling Casino slot machines usually rely on the internal timing clock signal at the moment the gambler pushes the play button to generate the seed number for the random number generator in the machine.
    And, of course, some systems use real random numbers, rather than pseudo-random sequences. You can get access to these online now, so there is no real reason not to.
    dan hunter likes this.
    ei incumbit probatio qui dicit, non qui negat
    Reply With Quote  
     

Similar Threads

  1. Replies: 20
    Last Post: June 10th, 2013, 10:53 AM
  2. are advances in healthcare ethical?
    By Rusty2406 in forum Health & Medicine
    Replies: 11
    Last Post: March 30th, 2010, 10:35 PM
  3. What flaws can be in a teleological suspension of ethical?
    By That Rascal Puff in forum Personal Theories & Alternative Ideas
    Replies: 3
    Last Post: January 4th, 2010, 07:51 PM
  4. Share your thoughts on your ethical beliefs
    By blacknamed in forum Philosophy
    Replies: 51
    Last Post: October 31st, 2009, 09:14 PM
  5. Ethical genebiology
    By eternal in forum Biology
    Replies: 10
    Last Post: February 6th, 2006, 05:26 PM
Bookmarks
Bookmarks
Posting Permissions
  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •