symbol:
-critical ideas from the Emotion Machine
""quotation
?questions
()my ideas
Chapter 1 Falling in Love
-"suit-case" word are those psychological words which are actually a diverse
collection of mental states
-a certain emotion is one way of thinking, or a mental states
"But why do we pack such dissimilar things into a single suitcase-like word?
It¡¯s the same for our other ¡®emotional¡¯ terms; each of them abbreviates a
diverse collection of mental states. Thus Anger may change our ways to perceive,
so that innocent gestures get turned into threats, and alters the manners in
which we react, to make us more inclined to attack. Fear too affects the ways we
react, but makes us retreat from dangerous things (as well as from some that
might please us too much)."
-"self-model" includes several structures and models
-there is a machinery for absorbing "goals" from environment.
?If there are multiple mental states behind a single suit-case word, why do we
pack them together? Does the reason for this packing is that a single underlying
mechanism behind that word?
Chapter 2 Attachments and Goals
-This chapter tries to answer two questions close related to human behaviors:
1. How do we learn to achieve a goal
2. Where do goals come from?
How do we learn to achieve a goal
-Learn to achieve a goal as well as avoiding a mistake.
-Learn the same things in different ways
-There is no use to learning something new unless one also has ways to retrieve
it when it is relevant
"they may tend to form roughly hierarchical structures, because each fragment of
newly acquired knowledge is built upon things that we¡¯ve learned before" (this
is open to debate. As Minsky himself put, knowledge is also a 'suit-case word'.
We can't hope to describe "knowledge" by one structure or model)
-Learning is not just a matter of making connections but of constructing the
structures that those connections connect
Where do goals come from?
(Minsky's theory inherits many parts from Freudian psychology, e.g., relation
between ID, ego and superego)
-He described several phenomena of attachment between child and its imprimers
(including some animals)
-imprimer->self-image(described in terms of sets of traits)->prediction->goal
-attachment-based learning
-¡®internal impriming¡¯ by using "mental models" of imprimters
?what or who gernerates a goal?(is the minsky's theory enough?)
?why we tend to stick to one moral pattern?
(Minsky's discussion about "goal" is more about our objectives on higher level.
However, if we think every human behaviors are goal-driven, how about the
motivation behind of everyday routine or every decision human made)
Chapter 3 FROM PAIN TO SUFFERING
-Pain raise a goal to 'get rid of it', if it causes large-scale 'cascade', then
there won¡¯t be much (other goals) left of the rest of your mind.
-"both pleasure and pain engage some of the same kinds of machinery"
-Our concept of pain has to do with our ethical intuitions, our senses of
suffering, obligation, and evil.
-A physical pain vs. a mental pain
"I think this shows how much higher-level activities can help to set off
large-scale cascade of suffering."
-Prolonged and Chronic Suffering
"chronic pain did not evolve from selection, it is a bug"
-Correctors, Suppressors, and Censors.
"knowing which actions not to take" by a group of Critics (three kinds: A
Corrector, A Suppressor, and A Censor)
?Do we really have such explicit mechanism to "knowing not to do sth"? Once you
want to reach some goal, there are endless things you can't do. Can we think it
is just implied in our understanding of the constraints of the world and in the
goal.
Interactions between them
-Critics activate selectors
-Critics switch
-positive and negative learning--suppress or select--not only positive
reinforcement learning
-consciousness and unconsciousness and preconscious--a doorkeeper by Freud
(the real issue is if there is the well-defined boundary between conscious part
and unconscious part of minds?)
Self control=Critics switch moods, suppress one's Suppressors
"One answer is that the concept of ¡®rational¡¯ itself is a kind of
fantasy¡ªbecause our thinking never is entirely based on purely logical
reasoning." instead, we "exploit an emotion to solve a problem"
(in my opinion, emotion involvement for rational thinking is most probably in
choosing which logical interence should be taken)
"For, while some parts of your brain get information from the outer world, most
of them are reacting to information they get other places inside your brain."
(an explanation of information completeness, but I think we don't generally do
the completion unless we need to, i.e., we have a motivation to complete
information. When is the time to retrieve the knowledge? Why do we retrieve it?
How do we retrieve it)
Chapter 4 CONSCIOUSNESS
A-Brain and B-Brain
-"For there is no significance in those (sensing) signals themselves: their
meanings to those lovers lie in how they represent and process them in the
higher levels of their minds"
-B-Brain (higher level brain) can only react to signal (concept?) from A-Brain
(lower level), C-Brain, D-Brain,......
-higher level didn't know the details of lower level.
?Is thinking pattern a good way to distinguish various Brains? What is the
meaning that the upper level brain can only get information from lower level in
this separation?
(this model interprets how minds interact with external world but no
interpretation of internal information processing)
Subjective Experience
(Mostly agree with his ideas. Subjective experience may be related to our
concepts.
Think about whether one first knows the word "red" or one first know oneself
have the feeling of red?
Some patients of color blindness who can not discriminate between red and green
may have the same feeling of green as us which actually is red for most people.
Thus, we can safely say that it is just a brain mechanism of subjective
feelings.)
(Moreover, we may assume that in fact no two people have exactly the same
feeling of others like they claim to be because no two people have the same
organism! Though in concept level they have ¡°the same¡± feeling.)
why do we pack the word?
"Well, we can see a clue by looking inside someone¡¯s travel bag: you don¡¯t need
to assume that those objects themselves have any common features¡ªexcept that
each of them serves some of the goals of the person who packed them into that
bag!"
(If the goal is to describe sth--this is why there are words, what do they
describe? Why do they describe total different things in one word? There should
be some common elements in the word to permit this description.)
Chapter 5 LEVELS OF MENTAL ACTIVITIES
-Instinctive Reactions: if-do-then rules (we can predict or imagine)
react to events in outer world and in our brain
-Deliberation
find island when searching by using commons sense (analogy)
"there is a difference between the rigid methods of Logic and the seemingly
similar chainlike forms of everyday commonsense reasoning"
"So using Logic is somewhat like walking a plank; it assumes that each separate
step is correct¡ªwhereas commonsense thinking demands more support; one must add
evidence after every few steps. And those frailties grow exponentially with
increasingly longer chains, because every additional inference-step may give the
chain more ways to break. This is why, when people present their arguments, they
frequently interrupt themselves to add more evidence or analogies; they sense
the need to further support the present step before they proceed to the next
one."
-Reflection-react to our mind, not in our brain (recent change)
it seems the same as react to outer world
-Self-reflective thinking (self-model)
"The reflective systems we just described can think about some of their recent
deliberations. Self-reflection does just a little more: it considers not only
its recent thoughts, but also the entity that had those thoughts."
"...when we engage higher levels of thinking: it is when our usual systems fail
that reflective thinking gets engaged."
-Self-conscious: thinking of self behavior compared to one's own value, ideals,
censor and taboo
?It is suspicious that whether the distinct layers exist. Does a specific "way
of thinking" emerge at a certain circumstance? Does it need information from
other levels?
-Then, what is "the self" and who will control and manage the system?
ID, ego, superego->Then a person¡¯s ego may represent itself as being in control
of things
?value, ideals->a process or something else? no specific description here
?Those levels are thinking pattern rather than the reason to the thinking
pattern. Why place ID, ego and superego which state the reason to a thinking
onto those thinking levels?
-perceptual parts interaction, perceptual process and plan process is intervened
(action), visual process is not just down-up (includes knowledge).
(Actually, there are two competitive perception models, one is a top-down model
by Gregory (1970), the other is a bottom-up model by Gibson (1966). Also, there
are combination of these two models, say, perception cycle)
-imagination=envisioning a change at any perceptual stage
"Imagining is somewhat like seeing, except that when we alter internal
representations"
-levels of changing: low (pixel), intermediate (part), high (symbolic)
-prediction: in If-do-then rules
upper thinking pattern
-------------------------------------------------------------
lower knowledge organization
Chapter 6 COMMON SENSE
why common sense
-knowledge retrieval--but also from how those visual clues lead you to retrieve
other knowledge.
-Claim: "Panalogy"=the way we use to retrieve related knowledge+the way we
switch "realms of thinking"
"However, when these are linked into panalogies, then when situations don¡¯t seem
to make sense, we can think about them in alternative realms¡ªwithout the need to
start over again."
-pending questions:
Student: How many different realms can a person use for any particular concept
or object? How many of them can we handle at once? How does one know when it¡¯s
time to switch? To what extent do different persons partition their worlds into
similar realms?
(Can we find sufficient realms? Are there realms at all?)
-a plan occurs before reaction
Commonsense Knowledge and Reasoning
"none of our knowledge would have any use unless we also had effective ways to
apply that knowledge to solving problems."
-System notion of Change: The Optimization Paradox; The Investment Principle;
The Parallel Processing Paradox.
?Are there some substrates underlying our knowledge-retrieval process?
"This may seem remarkable, but it might not be truly exceptional, because, in
1986, Thomas Landauer concluded that, during any extended interval, none of his
subjects could learn at a rate of more than about 2 bits per second, whether the
realm be visual, verbal, musical, or whatever."
(this is open to debate. It depends on how do we measure information. If we
consider the meaning of information. This measurement has to do with how is
knowledge represented.)
"long-term memories are not so permanent as we used to think; it seems that they
can be altered by suggestions and other experiences."
(long-term memory is not a single memory. At least until now, there are several
different memories called long-term memory: procedural memory/Implicit memory,
declarative memory/Explicit memory (Episodic memory, Semantic memory) and so on.
Is it reasonable to discuss memory without considering those different sorts.)
(knowledge retrieval) "links to the goals that each fragment of knowledge might
help to achieve"
Intentions and Goals
"there is no reason to expect all of one¡¯s goals to be consistent"
(is this where the irrational come from?)
"both forces and goals share some features"
"when some resources make choices in ways that the rest of your mind cannot
control, as when you do something ¡°in spite of yourself.¡±"
Difference-Engines
"such instinctive reactions begin at very low cognitive levels"
(But not all are beginning at the low level. How about those beginning at the
higher levels?
-the result of goal achievement or not can induce "cascades of feelings"
"On the positive side, the General Problem Solver was equipped with several ways
to reduce each kind of difference, and it even included a place for ways to
introduce new kinds of representations."
?How to represent goals/purposes in "different network". Is it represented as
target situation description? How can we have the insight to describe the
situation? How do we retrieve knowledge to find appropriate actions to resolve
the difference. What are the criteria for choosing ways to resolve differences?
?what is the difference in the first place? If it means the difference between
current situation and one's goal. Then this 'difference' view says nothing
specific about problem solving mechanism. Because it is just a description of
our behaviors or a problem solving procedure. It does not give us the underlying
mechanism--how to resolve the difference
Free will and decision making
"¡°free will¡± is not a process we use to make a decision, but one that we use to
stop other processes!" (knowing not to do sth)
(In Society of Mind, he claims there is no 'free will' at all. The 'free will'
here has different meaning--a personal feeling reflects underlying mechanism.)
-knowing ways to analogies is a vital part of how people think
?What is the mechanism to realize effective analogy? something is so different
from goals but why human can reasoning so quickly? How can the difference engine
do analogy?
How analogy works...
"The problem that I am working on reminds me of a similar one that I solved in
the past¡ªbut the method that was successful then does not quite work on the
problem that I am facing now. However, if I can describe the differences between
that old problem and this new one, those differences might help me to change
that old method so that it will work for me now."
Douglas Lenat: "Analogy works because there is a lot of common causality in the
world, common causes which lead to an overlap between two systems, between two
phenomena or whatever. We, as human beings, can only observe a tiny bit of that
overlap; a tiny bit of what is going on at this level of the world. ¡ [So]
whenever we find an overlap at this level, it is worth seeing if in fact there
are additional overlap features, even though we do not understand the cause or
causality behind it.[35]"
(it's true analogy is a good and common way to thinking. But instead of asking
how to analogize, shall we answer first the reason for the analogy. Why, at a
certain situation, we want or try to use analogy?)
-Knowledge needs Multiple Representations
(It does not if we have multiple representations matter, it's how to choose the
appropriate representation matters)
Chapter 7 Thinking
"It took each of us many years of hard work to develop our more mature ways to
think¡ªbut whatever records remain of this have somehow become inaccessible. What
could have made us all victims to that ¡°amnesia of infancy?¡± I don¡¯t think this
is simply because we ¡®forgot.¡¯ Instead, I suspect that it¡¯s largely because we
kept developing new, better techniques for representing both physical and mental
events¡ªand some of these methods became so effective that we abandoned the use
of our previous ones. Now, even if those old records still exist, we no longer
can make any sense of them."
(Can we explain it this way: it is not because we forget or abandon the old
thinking,it's because we haven't developed the ability to memorize "thinking" in
our infant time.)
"What schedules our large-scale plans?
What reminds us of things that we promised to do?
How do we choose among conflicting goals?
What decides when we should quit or persist?"
"because he could no longer could exploit those Critics to choose which
emotional states to use."
?if emotion is a state to choose or is just a state of mind? In other words, we
have a mechanism like Critics to decide what kind of emotional state we should
have or we just describe a certain state as some emotion and the reason to this
state may not be single or homogeneous.
-Critics=ways to think
"we use our most common emotion words emotion words; they refer to classes of
mental conditions that produce external signs which make our behaviors more
predictable to the persons with whom we are dealing."
-Cognitive Contexts (working memory)
-What Controls the Persistence of Processes?
"each such concern would seem to suggest that we have some Critics that detect
when that particular element may be getting into short supply, and then suggest
ways conserve or replenish it."
Chapter 8 Resourcefulness
Panalogy:
"However, it would not help us very much to have so many methods (ways to think)
available, unless we also had some way to switch among them almost
effortlessly."
?Do we learn a model from instances or just store all instances?
-panalogy=multiply ways to see a single object and link them together
Learning
-a model or a representation=any structure inside one¡¯s brain that one can use
to answer some questions about some subject
"Today, we still know little about how our brains make those memory traces or
how they later retrieve and ¡®replay¡¯ them (all related representation)."
-contextual features-use these features to retrieve context
"the term ¡°micronemes¡± for the myriad of nameless clues that color and shade how
we think about things"
"On the input side, we shall assume that many of your mental resources¡ªsuch as
K-lines, Frame-slots, or If-Do-Then rules¡ª can alter the states of some
micronemes. Then the present state of your micronemes could represent much of
your current mental context¡ªand the states of those fibers are changed, your
far-reaching bundle of micronemes will broadcast that information to many other
mental resources"
"Indeed, a substantial volume of a human brain consists of bundles of nerves
that interconnect regions that are quite far apart."
"As Immanuel Kant pointed out long ago, learning to learn is one of the things
that we cannot learn from experience!"
"How should a brain proceed to select which representation to use?"
(In my opinion, it is a thinking routine for us, we normally choose a
representation that was used in a particular situation in the past. But if it
doesn't work, we may try other representation. This change is also a thinking
process.)
Chapter 9 The Self
"What could make something so hard to describe that we can only refer to
comparisons? Clearly, this is likely to happen when we don¡¯t have a way to
divide that thing¡ªbe it an object, a process, or a mental state¡ªinto several
parts, or layers, or phases."
"We began by portraying the mind (or brain) as being based on a scheme that
deals with various situations by activating certain sets of resources¡ªso that
each such selection will function as a somewhat different ¡°Way to Think.¡±"
¡¡