Thermo-psychonomics - brain, mind, & emotion

[JAK]

The issue has arisen of whether all facets of life, including the mind and emotions, can be described logically in terms of standard scientific terms. This would be both reductionistic and materialistic.

The benefit of this discussion is for artificial intelligence (AI). If an architecture can be defined which could simulate human characteristics including emotions and "consciousness," this would facilitate our progress toward working robots in the future.

This thread is not intended to support or refute the existence of reality, Descartes dualism, the existence of God, or any other spiritual question.

The ideas postulated will be thermo-psychonomic theories - theories which predominantly incorporate thermodynamics, psychology, and economics.

This approach was first attempted by Elizabeth Duffy in the first half of the 20th century. Being a woman, she was ostracized by the male dominated field of psychology. As a result, she faded from most references. (This was unfortunate and very costly.)

Unknowingly, two researchers traveled separately down Duffy's logic and came to similar conclusions. The resultant theories were Behavioral Investment Theory (BIT) by Dr. Gregg Henriques, and Emotive Energy by me - J.A.Keeran. Only by accident did we become aware of each other and find our work fundamentally the same.

My material is available at http://www.theoryofmind.org. I can get copies of Dr. Henrique's work for anyone who requests it. I will be presenting my flavor because I, naturally, understand my version best.

In general, the basic theorectical concept is that natural selection demands efficiency from all life forms. Energy is the currency of life, and energy resources must be managed wisely. (Time and money reduce to an energy basis.) You can see efficiency in the sleekness of birds and fish. You see it with every corporation and government trying to maintain a budget. You can see it the next time you go into a car dealer and tell the saleperson you wish to pay 5 times the asking price for a brand new Lamborghini. You can see it the next time you need 35 hours in a day. Resources are limited, and each of us must choose how and where to use them.

For anyone wishing to participate, I encourage you to read the "conclusion" of my website for a quick overview and synopsis of where we are headed:
http://www.theoryofmind.org/unified/emotive/conclusion.html

 

[Dymanic]

I visited your site on about the first day you showed up, after obtaining the url from your profile, so I had some idea of what you might be presenting, but I never really did go over it all in detail. I'll have to do that before I'll be ready to begin ripping you to shreds.

I guess the 'global workspace theory' is the first thing that piques my interest. It sounds a lot like something in a book by Douglas Hofstadter, "Fluid Concepts And Creative Analogies", where he described several 'bottom-up' approaches to AI within narrow (but very interesting) domains.

Something I've given a lot of thought to without ever feeling like I had made much progress is what type of approach might be used to simulate emotion in a computer program. Maybe that would require specialized new hardware. Or something. Some of the programs Hofstadter described utilize a concept called 'temperature', which might be thought of as a measure of the degree of confidence the agents express in the sub-solutions they submit to the global workspace; i.e., more confidence lowers the temperature, causing the agents to become less active in their searches, and makes access to the global workspace more restricted.

I have reading to do. I'll be back.

 

[JAK]

...
Some of the programs Hofstadter described utilize a concept called 'temperature', which might be thought of as a measure of the degree of confidence the agents express in the sub-solutions they submit to the global workspace; i.e., more confidence lowers the temperature, causing the agents to become less active in their searches, and makes access to the global workspace more restricted.

I have reading to do. I'll be back.

I like the "temperature" idea. I'll do some follow-up on Hofstadter.

 

[JAK]

...
Something I've given a lot of thought to without ever feeling like I had made much progress is what type of approach might be used to simulate emotion in a computer program. Maybe that would require specialized new hardware. Or something. Some of the programs Hofstadter described utilize a concept called 'temperature', which might be thought of as a measure of the degree of confidence the agents express in the sub-solutions they submit to the global workspace; i.e., more confidence lowers the temperature, causing the agents to become less active in their searches, and makes access to the global workspace more restricted.
...

If I am correct about how the biological system works, then emotion is fundamentally tied to whether the brain perceives possible danger or possible opportunity. (This is based largely from the work of Robert Plutchik.) Given the assessment by the brain of "good" or "bad," the brain then must redistribute the organism's energy resources to either "seek" or "avoid" the opportunity or hazard. For dangerous hazards, according to Walter Cannon (1920s), energy (blood flow) is diverted to the extremities to facilitate "fight or flight." This would give the classic "sinking feeling in the gut as it is drained to maximize resources for use by the extremities. The reciprocal is diverting resources back into the torso to facilitate digestion. Mixtures of hazard and opportunity cause mixed emotions (what writers often call "bittersweet") where both approach and avoidance behaviors are concurrently active.

To introduce "emotion" into a computer, "feeling all gushy in my heart" is not useful nor likely to happen. However, redirecting the computer toward "seek" or "avoid" routines would be appropriate. Though the machine may not "feel" as we do, it will respond similarly because it truly does have an inherent "value" system.

**NOTE TO ALL:***
I am delaying a bit in presenting my material to allow time for others to read the "conclusion" of my website, www.theoryofmind.org. My first entry into this thread will be the 3 Laws of Difference (differences exist - nouns & numbers; differences are dynamic - verbs, processes, & time; and differences are relative - adjectives, adverbs, magnitude & dimension).

 

[Dymanic]

Originally posted by JAK

Given the assessment by the brain of "good" or "bad," the brain then must redistribute the organism's energy resources to either "seek" or "avoid" the opportunity or hazard.

It also seems to have much broader implications in the form of system-wide reconfiguration of the way information is aquired and prioritized. That's what I like about the 'temperature' sort of idea.

I am delaying a bit in presenting my material to allow time for others to read the "conclusion" of my website.

I think we're going to need about six months to get through the whole thing.

 

[JAK]

It also seems to have much broader implications in the form of system-wide reconfiguration of the way information is aquired and prioritized. That's what I like about the 'temperature' sort of idea.
I think we're going to need about six months to get through the whole thing.

The "temperature" is a keystone of the hypothalamus which is on our itinerary.

(Six months? Let six months or less be our target!)

It’s time to start ...

A group of a priori statements or premises are necessary in order to provide a rigorous and structured path:

TRUTH – Truth is a tool created by humans. If truth is to have any use, any utility, it must be reliable. Therefore, the best truths are ubiquitous – omnipotent and omnipresent.

The following three laws appear to be truth:

1st Law of Difference – DIFFERENCES EXIST. Using “negative” terms (“no,” “not,” “nil,” etc.) grounded in sentential logic, we can separate any object/aspect of the universe from another. A bus is not a rose, a rose is not the sea, and the sea is not a 4-speed transmission. For convenience, this law can also be termed the Law of Nouns & Numbers, or the Law of Objects & Patterns. However, it also encompasses the following 2 laws of difference:

2nd Law of Difference – DIFFERENCES ARE DYNAMIC. Objects appear to change over time. A flower grows, a ball bounces, and a volcano erupts. For convenience, this law can be called the Law of Verbs & Time, or the Law of Processes. (Processes are patterns of change across time.”)

3rd Law of Difference – DIFFERENCES ARE RELATIVE. Many things have positional significance to us: “on top of,” “on your right,” “farther,” “louder,” “rougher,” etc. They also provide measures of change: “quickly,” “silently,” “salty,” etc. Through this law we determine dimension and magnitude. For convenience, this law can be called the Law of Adjectives & Adverbs, or the Law of Scale & Comparison. (All relations are determined through comparison – a process.)

SET THEORY – By accepting the above three laws, a collection or “set” has been created with three components or “subsets.” With set theory established, the foundation for formal logic (syllogisms, etc.) is established.

Agreements? Disagreements? Changes/additions?

 

[drkitten]

Agreements? Disagreements? Changes/additions?

Yeah. The following paragraph is mathematically meaningless:

SET THEORY – By accepting the above three laws, a collection or “set” has been created with three components or “subsets.” With set theory established, the foundation for formal logic (syllogisms, etc.) is established.

The three laws given do not fit into any of the frameworks of set theory with which I'm familiar, nor can I see any way in which they could serve as an independent foundation for a meaningful scheme of formal logic. You are (ahem) invited to explain this paragraph further.

If you really want to do this connection properly, the easiest way to work would be to use a standard set theory framework (I suggest ZFC -- look up "Zermelo-Fraenkel" on your search engine of choice) and show how the axioms of ZFC can be derived within your difference framework. There's not a lot of them, and it would put your work on a much sounder footing.

 

[Dymanic]

Originally posted by JAK

Truth is a tool created by humans. If truth is to have any use, any utility, it must be reliable.

If we insist on exploring every possible tangent, this discussion may get bogged down rather quickly. Therefore, I (for one) am prepared to accept this provisionally. We might consider how truth can be a product of utility, that is, a property of internal representations by some system of an (alleged) external reality, essentially inseparable from the perspective of its own goals. But maybe we don't need to go there.

Any act of perception through our senses (perceiving colors, sounds, smells, etc.) is a process. Perceiving a wave of light or a soundwave is an animated process just as determining odors and tastes are animated processes. Any sensation of our senses and even our thoughts exhibit process.

In addition to this thread's assigned reading, I am also working on a new inter-library loan arrival, Dennett's "Brainstorms" (one of the few by him I have not read). Since my practice is to alternate between various things I am reading at any one time, my encountering the above passage happened to roughly coincide with this, from Dennett:

------------
"Any time a theory builder proposes to call any event, state, structure, etc., in any system (say the brain of an organism) a signal, or message or command or otherwise endows it with content, he takes out a loan on intelligence. He implicitly posits along with his signals, messages, or commands, something that can serve as a signal-reader, message-understanderer, or commander, else his "signals" will be for naught, will decay unrecieved, uncomprehended. This loan must be repaid eventually by finding and analyzing away these readers or comprehenders; for, failing this, the theory will have among its elements unanalyzed man-analogues endowed with enough intelligence to read the signals, etc., and thus the theory will postpone answering the major question: what makes for intelligence?"
------------

I'm not suggesting any real significance to this, it's partly just a way of letting you know I'm still interested, and partly a delaying tactic intended to buy me some more reading time.

 

[JAK]

SET THEORY – By accepting the above three laws, a collection or “set” has been created with three components or “subsets.” With set theory established, the foundation for formal logic (syllogisms, etc.) is established.

...
The three laws given do not fit into any of the frameworks of set theory with which I'm familiar,
...

Agreed. Looking back, I can see where my material could easily be seen that way. However, the three laws are not intended to be a part of set theory. Once presented, they can be described as a set, just as all of the stuff in a closet can be deemed part of a set named “closet.” The three laws are merely presented as elements of a set. The elements, collectively, are named “the 3 laws of difference.”

...
nor can I see any way in which they could serve as an independent foundation for a meaningful scheme of formal logic.
...

Actually, the first law is a variation of Aristotle’s law of non-contradiction: “for any proposition P, it is not both the case that P and not-P” (http://www.campusprogram.com/reference/en/wikipedia/l/la/law_of_non_contradiction.html). This is one of the foundations of sentential logic.

I present these laws up front for a number of reasons:

First, and foremost, difference and change are ubiquitous, and all living creatures must be sensitive to this. To create any form of robotics, processing “differences” will be a critical function.

Additionally, the laws are useful in clarifying definitions. Dancing David, in the “Wanted: non-magical observer” thread, noted that many things are “fuzzy,” and I agree. Nevertheless, there needs to be some way of separating objects and concepts. Sentential logic holds benefits for doing that.

The 3 laws can also be useful in problem-solving by providing a “back to basics” perspective. I have used it many times in creativity seminars to shake people out of “rut thinking.” It has also been “part and parcel” for any and all trouble-shooting I’ve done for Oracle, HP, MCI, and a variety of other corporations and governmental departments.

Lastly, as long as we use nouns, verbs, and modifiers (adjectives and adverbs) to communicate, we should acknowledge them as a premise, as a priori knowledge. The three laws encompass these foundational elements of language.

If you really want to do this connection properly, the easiest way to work would be to use a standard set theory framework (I suggest ZFC -- look up "Zermelo-Fraenkel" on your search engine of choice) and show how the axioms of ZFC can be derived within your difference framework. There's not a lot of them, and it would put your work on a much sounder footing.

ZFC and other frameworks are very elegant, and I would venture that they would, as you noted, “... do this connection properly.” And we may eventually go that route, too. Yet, learning ZFC and making it required reading for the participants in this thread may diminish participation. Further, the symbols are not convenient for entering into the text of this thread. I would prefer to press forward with as little structure as we can get away with. It’s like the wings of an aircraft. They are designed to provide sufficient strength with minimum weight. I would like to continue in that spirit. If we have to return to infuse more structure, then we’ll do so.

Even so, putting this “on a much sounder footing” is very attractive.

If we insist on exploring every possible tangent, this discussion may get bogged down rather quickly. Therefore, I (for one) am prepared to accept this provisionally.
...

I agree. Let’s press forward. If the future holds a quagmire that requires us to return and clean this up, then let’s do it at that time.

...
"Any time a theory builder proposes to call any event, state, structure, etc., in any system (say the brain of an organism) a signal, or message or command or otherwise endows it with content, he takes out a loan on intelligence.
...
the theory will postpone answering the major question: what makes for intelligence?"
...

- Very interesting comment. If an atom “wants” 8 electrons in its outer shell, is this intelligence? (This is orderly, but then all natural laws tend to be so.) If a theory builder proposes an event, state, structure, etc. that relies upon natural laws, is that taking out a loan on intelligence?

Let’s revisit this shortly. My next stop will be “systems” and “identity.” I think the issue of “intelligence,” “information,” and “knowledge” will be an issue there.

 

[JAK]

Okay, a couple of weeks, 154 views and 8 posts ‘til now. I suspect that those who would read the conclusion of Emotive Energy on www.theoryofmind.org have done so.

new drkitten suggested that other formal approaches are likely better than the one being used. Dymanic suggested accepting my points “provisionally” until they become a problem in the deductions which ensue. I agreed with both. In deference to time, I will proceed with my approach until it becomes unwieldy or too ambiguous. At which point, following new drkitten’s suggestion would be the next recourse.

Onward ...

SYSTEMS
If any object lasts for any duration of time, something must maintain its existence. This may be from internal or external forces. For instance, a crystal of quartz (SiO2) retains its shape and existence over time by virtue of powerful bonds between the silicon and oxygen atoms. Thus, internal constraints, or processes within the object, hold the object together. As for external constraints, a river retains its identity over time by following its river bed. The water is constrained into a regular channel by the river’s banks and gravity. The banks and gravity are the external forces operating upon the water. Should either of them fail, the water would run helter-skelter outside of the original channel.

Of course, as Heraclitus said, you never put your foot in the same river twice. And quartz can be broken down by chemical processes or extreme heating or other extreme measures. But these objects retain enough of their identity for long enough to allow us to name them – quartz, river. In essence, their identities repeat from nanosecond to nanosecond to nanosecond – even to millennia. This repetition is the heart of what we call “systematic.” Some repeated process is at work for the duration of the time involved.

In fact, it is likely the “system” rather than the molecules and atoms which acquires our attention and naming. As noted, the water of a river is constrained but eventually spills into a lake or ocean (or evaporates on the way). Though every droplet is considered to be “part of the river,” once it hits the lake or ocean, that water is no longer considered part of the river. Similarly, if you have a four-legged table and replace one leg, it is still a table. Even after replacing all of the legs AND the table top, it is still a table.

How about looking in the mirror? Does your face look different from when you were a child? Have a grey hair or two? Maybe a few new wrinkles? Knowing that your body is replacing skin and tissue constantly, and watching the face change in the mirror over the years, suddenly you may be just like the table with all of the legs and top replaced. But despite the changes, it’s still “you,” right? Okay, other than you the “system,” what part of you is still here from the “you” that began this journey years ago?

I suggest that a systematic nature is key to our process of identification. Existence of anything long enough to be perceived or recorded is long enough for us to name it – give it identity. To me, repetition across time is the hallmark of a system and the basis of identity.

Finally, if something is “named,” is that information or knowledge or both? How about before we name it and just “perceive” it. Is it information or knowledge or both? (How do we want this robot to operate?) If a robot is to function similarly to a human, it needs to be sensitive to “differences” and to “systems.”

Next stop: “internal and external” – the eventual causes of the sympathetic and parasympathetic nervous systems.

 

[number9]

Multiple systems, one universe

Emotion comes from a non-thinking awareness of our environment, condition, and situation.

There is a logic to emotion, but it is not computed in the same place that thought is computed; the means by which an emotional decision is made are not visible to our logical processing centers.

In order to create emotional behaviors in a computing device, all that is necessary is to perturb it with commands from a separate computing device programmed to make decisions in emotional ways.

Then program the logical portion to determine whether to succumb to or fend off the suggestion.

Which is all sort of a canard, because our "logical" brain is built from an infrastructure of visceral reaction mechanisms, so a logical decision is predicated on an emotional decision to act logically.

(N.B.: Those who agree with me on this are thinking logically. Those who do not are just being emotional.)

 

[Dymanic]

Originally posted by number9


There is a logic to emotion, but it is not computed in the same place that thought is computed; the means by which an emotional decision is made are not visible to our logical processing centers.

That involves so many assumptions that it leaves me at a loss as to where to even begin.

In order to create emotional behaviors in a computing device, all that is necessary is to perturb it with commands from a separate computing device programmed to make decisions in emotional ways.

That seems a bit like offerring an explanation for the origin of life on earth based the idea that it arrived from some extraterrestrial source; i.e., it doesn't really answer any fundamental questions, it just relocates them. I agree that there must be a logic to emotion, and I guess I'm really more interested in exactly what specific qualities the logic of those 'visceral reaction mechanisms' might employ that would distinguish it from the logic of... logical thinking.

 

[number9]

That involves so many assumptions that it leaves me at a loss as to where to even begin.

Mu.

I guess I'm really more interested in exactly what specific qualities the logic of those 'visceral reaction mechanisms' might employ that would distinguish it from the logic of... logical thinking.

Well, there's behavior based on logical thinking, and then there's behavior based on impaired logical thinking (whether the impairment is willful or natural dysfunction, or simple ignorance), and then there's emotional behavior.

Depending on what kind of personality you want your "AI" to have, you can define its computable logic, and its emotional logic, yourself, and then set up a table by which it decides which will prevail.

Simple cases: It observes a condition it wants to attain, and seeks it without regard to other conditions you commanded it to attain. It observes a condition that may harm it, and avoids it despite your commands to approach it. Those demonstrate the emotional program overcoming the logical program (presuming its original logic includes following your commands).

I'd also have it "learn" about emotional situations and turn them into logical ones, the way we do, as well, or part of its personality will be that it is perceived as "stupid".

 

[number9]

it doesn't really answer any fundamental questions, it just relocates them

Oops. I relocated that one right out of my response.

Yes, that's the point. "Emotion" is a paradigm located outside of the paradigm of "formal logic." (That's a tautology based on a definition of the two as disjoint. N.B.: the logic of emotion may not be the same in every person.) And in fact, emotion and logic are disparate functional structures in our physical construction. Emotion can be purely reflexive and hormonal. Logic is purely cortical and can even be rendered symbolically outside the body entirely. The logical subsystem can observe the emotional inputs, and can even cause psychosomatic output to them (e.g., you obtain logical input when you perceive a story through symbols known as words, but your body can feel emotion as it interprets them), but the logical subsystem does not innately understand the original inputs that caused the emotional response.

"Learning about emotions" is one of the progressive accomplishments of our species throughout its history, and the fullness of our current knowledge requires wisdom from many more situations than one person can experience in order to provide a basic understanding of human nature. E.g., men still don't understand women, even when you explain them, because they're a moving target.

Unfortunately, as the political climate of this nation in recent years has proved, we have not progressed very far, and as a people may actually have regressed quite a bit. (N.B.: and I'm quite aware I'm being chauvinist in this thread, too.)

But the current state of politics (a violent slosh to the unthinking and emotional side) could also be an extreme statistical fluctuation in what is still a generally progressive trend towards reason over reactionary behavior.

 

[Dymanic]

Originally posted by number9

Mu

Couldn't have said it better myself.

Well, there's behavior based on logical thinking, and then there's behavior based on impaired logical thinking (whether the impairment is willful or natural dysfunction, or simple ignorance), and then there's emotional behavior.

I like your idea of 'impaired logical thinking', and I would go ahead and apply that toward attempting to explain 'emotion'. For instance, one thing that seems to characterize emotional thinking is rather less concern with accuracy in making connections between causes and effects. The logic seems to say: if two events occur at about the same time, just assume that they are causally connected. Flawed syllogisms are another possible example.

But I think it could be made even stronger. Some emotions appear to be not simply the result of application of a weakened form of logic, but an explicit and deliberate disabling of logical subsystems -- not simply for reasons of economy either, but for reasons having more to do with negotiating strategies between organisms with conflicting interests. In other words, on close examination, emotions often begin to look like tools for manipulating behavior in others, and are very dependent for their effectiveness on their not being the products of careful cost/benefit analysis, as well as their being beyond volitional control.

I also think that one reason that emotions play such a large role in making us the complicated creatures we are is that the same manipulative tools that are often effective in achieving behavior modification in others may also be effective in achieving the same effect on ourselves (or sub-selves or whatever). Our internal dialogues often resemble conversations with others, and when we are in conflict these dialogues can become quite neurotic.

Well, mine do anyway.

 

[number9]

emotional thinking is rather less concern with accuracy in making connections between causes and effects. The logic seems to say: if two events occur at about the same time, just assume that they are causally connected. Flawed syllogisms are another possible example.

Flawed syllogisms (for whatever reason the thinker allows the flaw to obtain) are a prime example of "impaired logic", but you're kind of missing the point.

Emotion happens outside of thought. "Emotional thinking" is like "basketball thinking". You can think about basketball, and basketball can provide an input to your thinking, but you can't think with a basketball, because it's not inside the part of your brain that's doing the thinking.

Logic is a means of manipulating symbolic information.

Emotion is semantic content inserted into logical equations as symbols from outside.

Lots of things become symbols to be operated on by logic. Emotion is just one of them.

When the logic is impaired, thinking stops, and reactions to emotion become atavistic and reflexive. And whether they're consistent or random I'm not entirely sure.

 

[Dymanic]

Originally posted by number9


Emotion happens outside of thought. "Emotional thinking" is like "basketball thinking". You can think about basketball, and basketball can provide an input to your thinking, but you can't think with a basketball, because it's not inside the part of your brain that's doing the thinking.

Among the hidden assumptions I objected to above, and which I see repeated here, is a sort of tacit homunculism inherent in your treatment of the thought/emotion distinction.

Emotion is semantic content inserted into logical equations as symbols from outside.

How do you see this as different from the treatment of other 'symbols'? Is source alone the main basis for the distinction, or do you propose a qualitative difference as well which continues to distinguish them from other symbols once the symbols are 'inside'?

Lots of things become symbols to be operated on by logic. Emotion is just one of them.

Are you proposing that we 'adopt the intentional stance' with regard to this logic itself, or do you see it as merely a tool, employed by something else to which we should attribute intent?

When the logic is impaired, thinking stops, and reactions to emotion become atavistic and reflexive.

I tentatively agree, at least with regard to atavism. Reflexivity I'm less sure about. How do you see reactions to logical thinking being less so?

I see essentially two main directions from which these questions may be approached. The first is more intuitive, and posits that the best way to frame the questions is in terms of low-level activity being accessed to greater or lesser degree by high-level processes; the second, less intuitive, posits that high-level processes are accessed to greater or lesser degree by low-level processes (i.e., they compete for high-level access and control).

So far, you seem to favor the first approach. I am more partial to the second.

 

[number9]

"Question with boldness even the existence of a God; because, if there be one, he must more approve of the homage of reason, than that of blind-folded fear."
-Thomas Jefferson

That's apropos of nothing here, it just happened to be in my cut-buffer when I opened this post.

Anyway, you're expanding the argument when I'm trying to reduce it.

Simply put, emotion is a computational paradigm outside of logic. But emotion has its own logic, which comes from evolution, instinct, learned hormonal response, whatever. It's separate from logical thought.

To mimic a human in terms of logical thought and emotion, you need to set the two subsystems up separately, and have them cross-coupled.

Think of it as interprocess communication rather than integration of the two paradigms. Emotion acts as afferent input to logic the same way that sight or smell do. To logic, emotion is a sense. We don't control the conversion of situation to emotion, though we can create imagination that feeds back as situation and creates emotion.

I can't make myself feel sad, but I can feel sad when I think of something that I know will make me feel sad.

Subtle difference, but key.

You can test various emotional afferent/efferent models, and you can test various logical-impairment models, separately.

And then you can add things like autonomic hormonal changes, which care about systemic physical function more than emotion or logic.

 

[Dymanic]

Originally posted by number9

Anyway, you're expanding the argument when I'm trying to reduce it.

I agree.

-----------
Simply put, emotion is a computational paradigm outside of logic.
-----------
But I don't think I agree with that.

-----------
But emotion has its own logic...
-----------
And it doesn't look like you do either.

-----------
It's separate from logical thought.
-----------
Although you seem not to have entirely made up your mind.

-----------
I can't make myself feel sad, but I can feel sad when I think of something that I know will make me feel sad.
Subtle difference, but key.
-----------
Yes, I think that may be important.

-----------
You can test various emotional afferent/efferent models, and you can test various logical-impairment models, separately.
-----------
What would I be testing for again?

 

[number9]

What would I be testing for again? [/B]

Behavior.

Fit of your model to a "model of real emotional behavior in human beings".

Reaction of prospective grantors to your proposal.