I would like to get an evaluation of this demo that I call “Detection of Life”. I’m putting it in this AI thread because I created it with AI. I gave AI the specs and we (Chat GPT and me) iteratively worked toward this demo, which I thought might eventually work as a game. So I’m distributing it here to see if anyone is willing to try it out and give me some ideas about how to make it work better as a demo and eventually as a game.
This demo/game challenges the player to find the live “bug” among a set of unlive ones. The live bug (there is always only one) is a control system and the unlive bugs are are dead ones. The live bug is moving around because it is controlling for following a desired path; the dead bugs are moving because they are being pushed around by different randomly varying disturbances.
By moving a mouse you can apply your own disturbance to the paths taken by all of the bugs. You can detect the live bug by noting which bug is least affected by your disturbances. When you detect that bug you can use the mouse to place the crosshairs over it and clicking the mouse. If it is, indeed, the live one, it disappears and one of the other bugs becomes live for the next trial. If it is not the live one then a new bug is added to the set of bugs on the screen and a different bug becomes the live one.
There are three pull down menus that let you vary aspects of the situation that should affect how hard it is to detect the live bug. The “Starting number” of bugs on the screen, the “Liveliness” of the live bug – a high liveliness bug controls with higher gain than a low liveliness one, and the “Speed” of movement of the bugs. When you change the settings of the se menus you must press “Start/ Reset” to set the demo to these new values. Difficulty should be increased by increasing the starting number of bugs, decreasing the liveliness of the live bug and/or increasing the speed of all the bugs.
This demo demonstrates the fundamental difference between the behavior of living and non-living systems, per PCT. It also demonstrates what can be done using AI as a tool. Somehow I think these things might be related.
But, again, I hope some (many) of you will tryout the demo and five me some “feedback” in the form of questions, comments and/or suggestions about what would make it more interesting (and, possibly, fun) for you.
I don’t understand why “decreasing the liveliness of the live bug” makes it easier. I understand that its regaining of control is easy to perceive visually because it is slower to regain control. But if gain is at maximum it would also be easy to perceive because it differs clearly from bugs that do not resist being diverted to a new path.
I can’t look to see perhaps more clearly what you mean, because I get this SSL protocol error and the Reload button does nothing:
Hi Bruce. Thanks for looking at it! And I’m glad you liked it but sorry you had trouble with it. I don’t have that problem and I see that Richard K. didn’t either. My mindreadings.com site is supposed to be https but if the problem persists let me now I’ll check my website provider and see what’s going on. But I do know that the version of the game I sent you to in my earlier post is slightly obsolete and I have a new version up at me site that has a new file name. Try going to it from my demos page, Click on “Detection of Life” and see if you have better luck with it. I really appreciate the comments!
You are absolutely right. Decreasing the liveliness of the live bug (which is just a matter of decreasing the gain of the control system) makes it harder to detect the live one. I think you just misunderstood my arcane little sentence where I said as much: "Difficulty should be increased by increasing the starting number of bugs, decreasing the liveliness of the live bug and/or increasing the speed of all the bugs. Sorry for the confusion but good on ya for understanding how gain would affect the difficulty of detecting the live one.
I hope things get fixed by going to the demo from my demos page. But if you still have trouble let me know and I will talk with my website provider.
Hooray. Thanks Richard. But if you used the old version, it is no longer up one te server. So if you want to try it again go to this one, the one that is now pointed to from my demos page as “Detection of Life”.
I am so glad to hear that! I never really say what is being controlled. Since I do know, it’s hard for me to evaluate how difficult the game might be (at the different settings for number, liveliness and speed of the bugs) for new players.
By the way, do you (or anyone) know to move discussion about the game out of this AI thread and in to a new thread – perhaps called “PCT inspired games”?
My experience was different from Richard’s: I did not found what the live bug is controlling. Only that I could not control it as easy as the others. By making larges movements by mouse I see what bugs started to follow and the just clicked the reluctant one.
OK, the level of protection I have running on my PC still complains about not seeing an SSL certificate. When I click the link directly to the demo, it’s the form that I posted which I can’t get past. When I go to the top level of your website, I get an “are you sure?” kind of message that I can get past.
I’d be disturbing one agent at a time and finding out which one resists disturbance. Disturbing the whole bunch at once is I guess a bit like a school of fish, except that they don’t move simultaneously, and just waving my mouse cursor around and having them all move but one is kind of odd. With just a few and moving slowly I can make it work. With too many in the space the autonomous one easily gets obscured.
Can anyone think of an actual situation with one organism among non-living but moving dummies? I think it would be heaps better as a ‘learn PCT’ demos if it could be described as modeling something that we might actually encounter.
Thanks for trying it! You make a very interesting point. It is related to the point Bill made in the Experimental Methods chapter of B:CP where he said that, when doing the Test for the Controlled Variable, the Subject (S) might be controlling for a “zigzag” pattern of coins while the Experimenter (E) might conclude that the Subject was controlling for an “N” or “Z” pattern. As BIll noted, S and E might argue about what to call the pattern being controlled “…forgetting that E has discovered what S is controlling, whatever either of them likes to all it.” I think Richard discovered that the “live” bug is controlling for moving in a particular path while you have discovered that the the live one was “reluctant” to be moved by you movements, indicating that it wanted to move differently than you wanted it to. So you both discovered the same controlled variable – movement – but described it differently: as either a movement along a particular path or movement along a path different than the one being imposed on it.
I’ll try to get it working on a phone but it may have to change quite a bit about it. I’ll check with my programmer (AI) to see if it can be done effectively;-)
Thanks Rupert. It’s especially nice to hear that coming from you.
Well, it’s not quite “actual” (yet) but while I was developing this demo I happened to go see the movie Project Hail Mary and I was blown away by a brief sequence fairly early in the movie where scientists were looking at a computer screen with dots moving around it and trying to see if any of them were “alive”. I was watching a movie that had almost the exact same display (dots moving around a screen) and goals (detecting whether any were alive) as the PCT game had just developed.
I’ve searched YouTube for what sounded like they might be the scene I’m looking for but I haven’t been able to find it. Although it’s a scene from a science fiction movie, it is about a situation that may soon be a real possibility as we search for extraterrestrial life. And it certainly couldn’t hurt efforts to promulgate PCT if we have a demo of an approach to doing something that was an important part of a hugely successful motion picture!
So if anyone out there can find a clip of the scene in Project Hail Mary that I’m talking about, please let me know!! Hint: it’s a scene from the flashback part of the movie, on earth, before Ryan Gosling gets launched on the mission. And, as I recall, they are looking at some recovered space particles to determine whether they (or some of them) are alive or not. It’s a very short scene.
Done. We all need to consider “Is this a germane reply, or am I actually forking a new topic?” If it’s a new topic, find the right category or subcategory for it and put your post there. If nothing fits, put it in “Uncategorized”. (We should poke through that and try to find homes for the topics posted there.)
It’s more involved to change the category after the fact.
This is great Bruce. I’ll continue the discussion when I have a chance. But I’ll just quickly mention that my 12 year old granddaughter tried my little “Detection of Life” game and she liked it!! She says it takes a while to figure out how to do it but once you do it’s too easy so I should add some harder levels.
I’ll be back as soon as I am done trying to keep up with Alaia (the granddaughter;-)
