Adaptron Test Run-1 4th Jan, 2005
I’m having a problem with repeats like AB becoming permanent before either A or B become permanent. It’s because with thinking=true it thinks about the next resulting stimulus B in the S-Habit rather than thinking about the resulting recognition of a 1. See below.
Here it thinks about the S-Habit 1 and the result is a reflex reaction of ‘a’ to the A. So what I really need to decide is what is the purpose of thinking. It is to simulate the real world a step ahead of it happening in order to make the next response (or no response) in order to achieve the goal. In this case the goal is novelty. When it receives the 3rd A it could think
1/ the next input will be a B, is there anything that follows it that is novel - yes the B1 combination (S-habit 2) or
2/ the next input is a B meaning I will have recognized an AB (S-Habit 1) and this I’m familiar with, will it result in anything novel, I don’t know, it’s at the end of LTM. Or
3/ maybe we don’t think ahead so far when it comes to S-habits. See below
Maybe thought is only necessary for A-Habits - thinking about the next result not about what is going to be recognized if no action is taken. Do we think ahead when listening to music or just take it in as fast as it occurs? Do we think ahead about what someone is about to say? Yes to an extent that we have an expectation of their next words but not of what will follow those words if we continuing listening. We are more likely to wait for those words so we can recognize/interpret the whole sentence (S-Habit). Maybe we think about the subsequent result of an action because once performed it can’t be taken back whereas we can always listen and act afterwards thus no need to think ahead on listening. Thinking is necessary for problem solving which is goal directed behaviour so as to put together two actions in sequence but not necessary to put two stimuli together in sequence.
Adaptron Test Run-2 4th Jan, 2005
In this sequence A & B repeats forever. It is in the process of recognizing S-Habit 2 = BAB as the result of the action a at 4. This seems a long time to wait for the result just because the 1st stimulus of the result (B) was familiar. Another strategy might be to say BA should be recognized before BAB and on receiving the latest A the B and A should be stored in LTM at locations 5 and 6. Or to start a new A-Habit based on the A matching at 4 - it does have an interest level of 19 which is greater than the current interest level. This would store the B at 5 and the A at 6. See below
Here the action habit at 4 is repeated but the ongoing S-Habit has not been stopped and the result is the success of the A-Habit and the recognition of a 2 S-Habit. What should the criteria be for stopping ongoing S-Habits when starting an A-Habit? Stop them all? Stop only ones with lower concentration levels so that we can listen to the music we are playing (subconscious actions but conscious self entertainment)?