Lecture � Society of mind

Greg Detre

Wednesday, March 12, 2003

 

do spicy foods activate your pain or your temperature receptors?

 

why do we have feelings at all, as opposed to being zombie machines?

just because you can imagine a person without feelings, doesn�t mean that that�s evidence for something

after all, how can you really imagine that, when you have no idea how (and can�t imagine how) a person works at all?

what about the people who think they don�t have any feelings?

called alexathemia(sp???)

and if you can�t imagine something, that�s just because you have too weak a imagination

when he imagines people with totally flat emotional affect, he thinks their cascades don�t work

he thinks feelings are how your various self-models describe your activities � they�re not primitive

 

pain, rather than being atomic, is difficult to talk about because it�s a linguistic shortcut for something very complicated spread over so many different brains systems

 

apparently, there�s a narrow band of frequency of laughing, and if you laugh at that frequency you start to find things funny

different theories of emotion:

james-lange � I�m happy because I smile etc.

emotions as the valencies (presumably of reward) attached to cognitive things

emotions as the kind of thinking that you use when you�re facing a conflict that needs to be resolved

reference: How to Solve it, Induction and analogy in mathematics

Poincare (Science and hypothesis, late 19th century, pre-Freud) � mechanics, topology � gets mentioned in ch 7

Poincare discusses preparation, unconscious, revelation (looking for patterns that indicate a solution, set up in preparation phase), criticism/evaluation

seems to ascribe importance to the unconscious process that works in the background � but he then debunks it on the next page

what should they teach children in elementary school about thinking???

doesn�t know � some skills you practice little by little �

give them really hard problems, then give them hints � rather than problems that are carefully-graded in difficulty

the most important thing to teach is enjoying getting a wrong solution and finding out exactly why it�s wrong

two alternative types of school

montessouri � start early with mathematical types of problem

kid started to build with these arithmetic blocks and the teacher said �no, those aren�t toys� J

waldorf � less structured, explore creativity

synectics � brainstorming company in Boston

start by just freewheeling crazy ideas, then critical, then split into groups

most people may go through these in microphases, a few every second

his theory is that manic-depressives� cycles of these are skewed

some psychologists try to get people to talk out loud about their thought processes, e.g. while packing clothes into a suitcase

but getting people to try and speak out loud about their unconscious processes is a bad method of testing people

where do people have their best ideas?

intoxicated in a bar, and on the toilet

locked up in a room, and in social isolation

Minsky: gets ideas in the shower, perhaps because of sensory deprivation (noise of the shower)

you need to brainstorm at different levels � spend at least a few minutes a day looking at the big picture

Don Norman

fired by Apple � why???

finding things wrong with OS 9 � Minsky thinks they�ve been made worse

he doesn�t like the dock

aliasmenus, even better than fruit menus

in hard sci-fi, you�re only allowed one invention, and then you spend the time working out its implications � you�re more likely to get something useful if you only break one rule

Inspiration � mind-mapping package

Fitzgerald-Lorenz contraction came before the Theory of Relativity � but they considered it just a fudge factor whereas he took it seriously

Clarke eventually gave up on the space elevator because of terrorism

the reason that Godel�s theorem isn�t a problem is because it only applies to consistent formal systems � so if the brain is algorithmic, but inconsistent

certainty vs truth

maths as consensus until someone points out an error

an algorithmic mind/machine throws out lots of true statements about mathematics, and a few false ones

defines a formal system as set of expressions and a set of rules to transform them

all of maths as proving tautologies

build an inconsistent formal system on top of a consistent system

he says that there�s lots of bits of tEM that should be taken out because they�re only there to irritate people J

 

 

Questions

minsky's talk of pain-behaviour as not needing to feel painful is surely nonsense - it allows implicitly for the idea that there are some behaviours that are/aren't conscious

i feel as though his claim that consciousness is not one-dimensional is either very right, or very wrong

let�s say we did build this extra system that served the same function that our pain-system serves now � wouldn�t that be conscious too???

well, as long as the higher levels are able to reason about it, certainly

perhaps assume it�s kind of like pain asymbolia

are the people with pain asymbolia motivated to avoid the pain though???

he says apparently not

does he have a cutoff point for which levels are conscious and which aren�t???

almost certainly not, because he wants to say that consciousness is not only a continuum (presumably), but multi/high-dimensional, and that any cutoff point would be terminological

is there more to suffering??? is it a unitary feeling???

presumably he�d say that the nature of your suffering depends on the nature of the constriction placed upon your thought processes???

is it an epiphenomenon??? is it designed into the system???

how does he argue for our multiple selfs all feeling like a single self (I�ve forgotten)???

emotions are just ways of thinking � but there are lots of ways of thinking that aren�t emotions

question: emotions as ways of thinking that activate your autonomic system �

Minsky: but where does the autonomic system start?

reference: Damasio on anasognosics

apparently Christopher Reeve can breathe for 30-40 mins at a time

he�s currently interested in trying to classify things we don�t have names for

e.g. the principle 50 critics that we use for �

teach them AI so that you can pin down the different types of frustration, anger, and the particular critics

interesting that there�s no word for these critics

would you expect to find these kind of domain-general critics though???