"HOMO RATIONALIS" AND HUMANIANITY

We humans do good things, sometimes really wonderful things, which bring us happiness and joy. But we also do bad things, sometimes really terrible things, which bring us unhappiness and suffering.

What would life be like if we did more of the good things and stopped doing the bad and really terrible things? Would not such a change promote the good life for all of us? And what would be required for us to make this change? What we are talking about is changing our behavior. What is required to change our behavior?

We must understand what the DETERMINANTS of our behavior are, and how to CHANGE those determinants into ones that help us do the good things and help us avoid doing the bad things. That is what this book is about.

But I wish first to encourage the reader to persist in reading, and to study in detail, what I predict will be the most difficult chapter to read, difficult because it attempts to deal so specifically and basically with concepts that are generally taken for granted in ordinary conversation. The effort will be rewarded with a much more thorough understanding of the rest of the book, and a much greater sense of the importance of the ideas presented in this book. We will be developing a highly useful model of the determinants of our behavior, useful especially by virtue of the development of an agreed-upon set of words that will be unusually precise in their definitions. Such precision of definition, though a difficult undertaking, will enable a much greater capacity for communication and therefore agreement.

Just prior to this undertaking, let us note and understand that this book is about trying to achieve, as much as possible, optimal living on the part of our species.

I will use the term optimal to refer to the hypothetical best. It is a goal to aim for. In some cases, it will indeed be possible to say that a particular entity (act, belief, outcome, etc.) is, has been, or will be optimal. Usually, of course, there is some degree of uncertainty. Sometimes an entity will obviously be non-optimal, or less than optimal, even though what instead would be optimal is not clear. "More optimal" will mean closer to optimal. Aiming for the optimal is an effort to improve, insofar as is possible. To optimize is to improve as much as possible.

By living, I am referring to all decision-making.

By optimal living, I am referring to the hypothetical set of all decisions most likely to lead to the survival of and best quality of life for everyone, now and in the future.

This book, then, as an effort to help our species to achieve optimal living, is a recommendation to the reader that he or she put forth some effort to develop some skills that will not only benefit the reader personally, but will benefit those around him or her and will be a contribution to bringing about a change in our species that will promote the survival of and the good life for our species in general, meaning for all of us, now and in the future.

In order to develop these skills, we will have to understand (believe accurately) how to do so. In order to understand how to do so, we will have to have a set of concepts (models) to guide us. These concepts, to be of use, must be as simple, clear, and consistent as possible, because the alternative is ambiguity, uncertainty, lack of agreement, indecision, and inefficiency, and probably even failure. The purpose of this chapter, therefore, is to present a basic set of concepts that will serve as our tools in this effort.

By basic, I do not mean that the concepts cannot be further analyzed, nor do I mean that the reader should not do so, if he or she wishes. What I do mean is that, for our purposes, I believe (predict) the reader will not find it necessary to do so, beyond what is carried out in this chapter.

The reader is reminded that any word, like "behavior," may have different meanings to different people, and that the method of this book includes establishing how words will be used in this book, for the purposes of this book. Thus, the reader is asked to agree to these usages while reading this book, in order to grasp the ideas presented in the book. However, I also believe that the model I am proposing is approximately the usual one according to which all of us operate and communicate most of the time, even though we do not explicitly say so, and even though we do so with much inaccuracy, ambiguity, and variability of meaning.

So now our first basic term will be "behavior" (and related words, such as "behave," "behaving," etc.).

Let us, for our purposes, agree that behavior shall mean that manifestation of animal life in which an animal does something when, as far as we can tell, there would have been other things that it could have done. For instance, the animal could have stayed where it was (rather than moving), it could have gone in the opposite direction than it did, it could have taken a bite (rather than refraining from doing so), it could have remained silent (rather than barking), etc.

(There are many processes that occur in animals that we will not be referring to. For instance, we will not be talking about metabolism or digestion. We will also not be talking about reflexes, such as the knee-jerk reflex. The reason for excluding these processes from the definition is that they are too "automatic" and invariant. It should also be noted that there will of course be things that the animal does that are on the border, such that we might have difficulty deciding whether to consider them behavior or not. This difficulty would simply be an example of how no definition can be perfectly precise, since the world is not divided up into discrete entities like our words are. Our tool, the symbol, "behavior," does not have to be perfectly defined in order to be useful. The reader, I believe, will come to conclude that this lack of precision does not represent a problem, for the purposes of this book.)

Notice that I am not just referring to human behavior, though that is what this book is primarily about. Much of our behavior is similar to that of many of the other higher animals on this planet. In this book, we will be looking at behavior by our species both that is a manifestation of our basic animal nature (i.e., also engaged in by many other higher animals) and also that is different from the behavior of any other species on this planet. Our model is to include ALL behavior, of ALL animals, including humans.

I will also be using the term, "act." I wish to clarify the relationship between "behavior" and "act." There will be no definite dividing line between these two concepts, nor any significant difference in their definitions. Any statement containing one of the words could probably always be changed to use the other word instead. We would be more inclined to say that our behavior consisted of a series of acts, but we could also consider each act to be an example of a behavior, and the overall behavior could be called an act. In other words, we could call an act a series of behaviors, but it is a little more typical to call a behavior a series of acts. Whichever way the words are used will make no difference in this book. However, I generally will use the term "act" to label a more limited, specific, or concrete example of behavior.

(Before we go on, I wish to call to the attention of the reader a kind of behavior that we humans engage in that we are not generally accustomed to regard as behavior, namely, "thinking." We are more accustomed to contrast thinking with behavior in our usual conversation, by emphasizing the fact that thinking is "hidden" from others, unless revealed through communication, whereas behavior is presumed to be "observable," at least in principle. However, using our definition of behavior, we can see that thinking might well qualify. In the first place, there is nothing in our definition that excludes hidden behavior from behavior in general. Also, we note that we have some "control" over our thinking. We can think about something or put it out of our minds. We can pray inwardly. We can practice reciting a poem in our heads. We can go over our lines in a play that we are participating in. We have at least some ability to control what words or sentences we think, and even whether we think those words or sentences at all. Also, we know that when we think words, there is some slight activity that can be detected in the musculature of the voice apparatus. When we imagine carrying out a motor act, there is similar overflow into the musculature, though slight enough as not to be observable without instruments. We can imagine that thinking probably first came about through the inhibition of overt speech, which in turn had developed from the vocalization that most animals engage in and that evolved into communicative behavior. It will not be important until later to consider thinking a form of behavior, however, and there are indeed important differences between thinking and other forms of behavior that are more overt and have a more obvious effect on the world. So I wish the reader just to make a note of this paragraph for future reference.)

Our next basic term will be "decision."

This term, "decision," is a "psychological" one rather than a "physical" one, in that, at least with our current level of understanding, this concept has little or no connection with the concepts used in the "natural sciences" (physics, chemistry, biology, anatomy, physiology, etc.). ("Psychological concepts" refers to concepts such as "feeling," "thought," "fantasy," "wish," "idea," etc. "Physical concepts" refers to concepts such as "molecule," "photon," "volt," "acid," "action potential," "cell," etc.) We do assume that behavior can also be explained to some extent by concepts from the "natural sciences," but in our daily living, attempting to do so would be very limiting and much less helpful, compared to the psychological models that we can construct. (The reader may recall the analogy of cutting the orange in different directions to find out what it is like inside. Some ways of cutting it will be more valuable than others for certain purposes.) But despite my use of psychological terms and concepts, the reader should remember that I am really referring to whatever is going on in the nervous system that is the process referred to by the psychological term. And so, for instance, the question as to whether a phenomenon is "conscious" or not will usually not be relevant.

There is an added benefit in refraining from requiring that the phenomenon, "decision," be locatable or observable in a nervous system of an animal. We use the same term to refer to a phenomenon that is manifested by a group. "The group decided that it should take the following course of action." "Our country decided to go to war." "The decision was made by group vote." We will find, I believe, that the concept of decision, and related concepts, will be helpful in understanding group behavior as well as individual behavior. In doing this, we will be using a metaphor in which the group is regarded as an individual. I predict that the concepts that we will be using to understand individual behavior will be found useful in understanding group behavior, also. Remember that we are simply constructing a model that will help us to predict and influence what will happen. If it works (allows us to predict accurately and optimize our behavior), our goal will be accomplished.

It would probably also be possible to construct a psychological model of behavior that did not involve the concept of "decision," but, again, I believe the reader will come to agree that, for the purposes of this book, no such model would be as helpful as the one I am proposing. Also, the model I am proposing is close to and most consistent with the ones we usually use in our daily lives.

What I shall mean by "deciding" is "choosing" one particular act or behavior out of a set of possible acts or behaviors that the animal could engage in, considering its biological (primarily neurological) makeup (i.e., its capabilities and limitations) and the SITUATION that it currently is in.

"Situation" means, for our purposes, everything that is going on, at a point in time or during a period of time, that could possibly have a direct or indirect effect on the animal(s) about which we are talking.

Consider a monkey in a tree. It leaps from one branch to another. For the purposes of this book, we will consider the monkey to have made the decision to do so. Considering its makeup and the situation that it was in, it could have "chosen" not to leap, it could have leaped to another branch, it could have leaped off to one side such that it would have missed the branch and fallen, etc. It is this possibility of choosing among a set of options that allows us to consider behavior to be taking place, and the choosing of one of those options will be what we shall mean by the making of a "decision."

The term "decision" could be defined differently, such as to apply only to human behavior, and perhaps only to certain behavior of humans that involves verbal activity or conscious thought, but such a definition would undermine the purposes of this book, as I believe the reader will come to agree.

Notice that I am not at all saying that the monkey engages in "thinking," or that the monkey is even "aware of" having made a decision. In fact, using this definition of "decision," I would maintain that most decision-making, even by our own species, is unaccompanied by "thinking" or even the experiencing of some mental phenomenon that could be called "the awareness of having made a decision." For instance, when the reader is driving, he or she is turning the steering wheel from side to side, depending on the moment-by-moment conditions of the road and traffic (i.e., the situation), but this "decision-making" could be called "automatic," "intuitive," "unconscious," etc. Similarly, when walking, each step the reader takes "could have been" taken differently, and with different results, and therefore each step, according to this definition, represents the result of having made the decision to take that step in that manner. However, only in something like ballroom dancing would there be an awareness of deciding to step in a particular way.

But notice that, on the other hand, the idea of "decision" also does indeed apply sometimes to conscious processes the reader has engaged in numerous times in his or her life, involving much thinking, inquiring of others, even research, etc.

So it is the "choosing" of one act among a set of possible acts that we will be calling the "decision," not the awareness of having done so, even though we humans are indeed aware of making some of those decisions.

There is a specific reason why this concept, "decision," will be useful to us, as distinguished from the concept of "behavior." There is generally the passage of a certain amount of time between the making of the decision to act and the actual act itself. It has been determined, I understand, that even when the decision to act and the act itself seem to take place at the same time, there is a momentary delay that can be observed to take place in the nervous system. But even if this were not true, we do have clear examples of the lapse of time, sometimes a very large amount of time, between the decision and the act. A simple example would be my deciding now to go to the store tomorrow.

We might hypothesize that a certain state of the brain and/or mind exists between the making of the decision and the carrying out of the act, but I wish to point out that we are not able at this time to identify, observe, or describe such a state, utilizing the "natural sciences." And, for our purposes, it is unnecessary to do so.

What is important is that we are distinguishing between the decision to act and the act itself.

Notice that this concept allows us to consider that a person or other animal might make a decision, and before acting as decided upon, change that decision, resulting in the occurrence of a different act. Thus the decision to act might not be followed by the act, but if the act does occur, the decision did also, according to our definition. Any act or behavior, then, implies that the decision to engage in that act or behavior has occurred, even though some decisions may actually be changed and not result in an act or behavior.

There is no clear dividing line between "decision" and "act." In the nervous system, something must take place that ultimately results in the state of affairs that we would refer to as "the animal having acted." There is just as much vagueness in when the process of deciding ends and the process of acting begins as there is in where our atmosphere ends and where space begins. What we are referring to as "deciding" extends over some period of time, from close to instantaneous to perhaps months or years, depending upon what we are referring to by the term. Where to step when walking is a decision that is close to instantaneous. Whether to buy a house is a decision that might take months or years. Of course, while that decision (whether to buy a house) is being made, the individual(s) probably would be deciding to do and therefore would be doing a fair number of things in order to make the decision easier. And if, metaphorically, we talk about a group making a decision, we might see the decision-making process as also occurring over months or years, and as including much behavior decided upon to help make that decision. So a decision may actually consist of a set of "smaller" decisions. We must remember that our words may be discrete from each other, but reality comes with no such well defined lines drawn in it. The most important aspect of the concept of "decision" is that it is a process in the nervous system that results, at least sometimes, in behavior, and that it is a process that can be influenced. And influencing decision-making in order to achieve more optimal behavior is what this book is about.

So we are saying that we are going to look at everything that an animal does that appears to be optional, that is, appears to be among a set of behaviors that we could imagine might take place instead, and to consider that the behavior is the result of a "decision" to behave that way, whether conscious or unconscious, whether deliberate or automatic, whether immediately preceding the behavior or much earlier, whether described with words or not, and whether engaged in by a human or by some other animal.

• "I have decided to go to the store."
• "They made the wrong decision, because they didn't give it any thought."
• "The rat decided to turn left in the maze."
• "I have no idea why I decided to do that."
• "I decided it was better to do nothing."
• "Who knows why people decide to do such things?"
• "I keep on making the wrong decisions."
• "I decided to do it, but someone stopped me at the last minute."
• "I am deciding what to do."
• "I am having a hard time deciding what to do."

So our first three basic terms are "behavior," "situation," and "decision."

Our next basic term will be "outcome."

We know that when an animal does something (engages in an act), there will be a resulting situation that is likely to be different than if the animal did something else. Furthermore, to some extent we know what result to expect. But we also know that this expectation is quite limited. We may be able to figure out that the act will cause a particular situation to arise, and that this situation in turn will make another situation more likely, and so on. But the further away from the act that we go in time, the less likely we are to know what to expect. So when we speak of the outcomes of a particular act, we have a poorly defined set of subsequent situations, which, as we move ahead in time, can less and less be said to be outcomes of the act. The further into the future we try to predict, the lower becomes the probability that we will be correct. And all of the above can be said without even referring to the current concepts in quantum science about uncertainty.

So if we take a look at the "events," or situations, that occur following a particular act, we must ask ourselves which of those we will consider to be outcomes of the act and which ones not. Right away we can see that there will be no way of drawing a dividing line between "outcomes" and simply "subsequent events." Yet, there will definitely be events that everyone can agree could be considered outcomes. If I turn my steering wheel left and the car turns left, the car turning left surely can be considered an outcome. If I immediately run into something, this may well be considered also to be an outcome. If someone else suddenly makes a turn at the same time, such that I run into that car, it is harder to say that the accident was an outcome of my turning my steering wheel left. If the accident occurs a half block later, after I have made my turn, we would probably say that the accident was not an outcome of my having turned my steering wheel left, even though, if I had not done so, the accident would not have occurred.

Yet, for our purposes, we must have some way of defining "outcome" that will allow us to proceed with a relatively consistent and simple set of concepts. I will propose that we consider "outcome" to refer to that set of events, or situations, that we could reasonably be able to predict if we had as much knowledge of the "laws" of the world, or the "regularities" of the world, as we could ever come to have, and that we recognize that the word "outcome" should always be considered to be shorthand for the phrase "probable outcome." This will be our way of acknowledging the fact that all knowledge is basically probabilistic, in that we realize that we will never be able to understand the way the world works sufficiently to predict everything that is going to happen in it. (This idea is separate from and in addition to the current idea, in quantum physics, that the basic nature of the world itself may be probabilistic.)

Our basic terms now consist of "behavior," "situation," "decision," and "outcome."

The next basic term is "mistake."

We will be making the following basic assumption: It is possible to make a mistake.

I believe that the reader will automatically agree to the above statement. It is a part of our normal way of thinking about behavior. However, we need to have a precise definition in order to think clearly about this area.

For our purposes, let us assume that a mistake is a decision that, in the opinion of the person or persons using the label, is a less optimal decision than some other decision that could have been made in its place. We might then say that mistakes are that subset of decisions that we consider to be "faulty," or even "bad." In this book, mistake and suboptimal behavior will be used synonymously.

I believe the reader will agree with me that there are many, many decisions made by members of our species, both by individuals (including ourselves) and by groups, that are less than optimal. In fact, there is seldom a day that goes by that we do not become aware of some decision (by virtue of observing or learning about some act), made by some individual or group, that we wish had not been made, some of these decisions even being very heartbreaking and tragic. This book, then, is an effort to promote a basic way of life that reduces such decisions ("mistakes") to a minimum.

Now, as is so often true, the word "mistake" is used in many different ways at different times by different persons. Also there are a number of different words and phrases that persons may use for the same concept. Therefore, because of our effort to be simple, clear, and consistent, it is important for us to clarify how the concept is going to be used in this book.

First, let us clarify that, for the time being, we are talking about only one kind of mistake, namely, a faulty decision, as opposed to a faulty perception, a faulty judgment, or a faulty belief about something in the world. For instance, we are not currently talking about a mirage or an optical illusion. We are not currently talking about the incorrect estimate of a distance. We are not currently talking about thinking that the day of the week is Tuesday rather than Wednesday. We are at this time talking only about the decision to act in a particular way.

We also should remember that we are talking about the decision, not the act, though it will almost always be true that the observation of, or contemplation of, the act with its actual or probable outcomes will be the cause of regarding the decision as a mistake. A shorthand way of speaking might indeed be to say that the person's action was a mistake, in which case most would readily admit that they also meant that the decision to engage in that act was a mistake. But a person could make the decision to do something, and prior to doing it, could come to the conclusion that the decision was a mistake, and therefore change his or her mind and decide to act differently, before ever having carried out the original decision. In such a case, a mistaken decision, or mistake, was made, even though no corresponding act took place.

Now notice that we are essentially talking about what might be called a "value judgment," that depends upon whoever is making that judgment. For instance, one person might consider a particular decision to be a mistake, while another might consider the decision to have been the best decision that could have been made (i.e., optimal). In fact, we could probably take any one of a large percentage of specific decisions that have been made, and find at least someone who thought the decision was a mistake or a bad decision. On the other hand, there are certainly many decisions about which the vast majority of us would agree.

Let us take a look at some decisions that almost everyone would acknowledge were "mistakes," using the word in the manner proposed in this book. The monkey leaps for the other branch, which happens to be too small to carry the monkey's weight, so the monkey falls. The rat turns in the wrong direction in the maze and fails to get the food it is seeking. A person believes he is walking on level ground and steps off the edge of something and falls. A person turns the steering wheel too far and runs into the side of a car traveling beside him or her. A person believes he or she can make it into the intersection before the light turns red, but, failing to do so, crashes into another car. A person incorrectly thinks someone at his door is dangerous and therefore shoots an innocent individual. A person eats some food that is spoiled and dies of food poisoning. A group decides to spend its money on a scam. A nation decides to adopt a fiscal policy that later is recognized to be a disaster. In these examples, there would be different reasons for the various mistakes, but in each case, the decision turned out not to be optimal, and therefore, according to our definition, was a mistake.

Notice that, with this definition, one of the causes of making a mistake may be insufficient knowledge, as in the mistake of unknowingly eating spoiled food.

Let us avoid a quite possible misunderstanding. Some would use the word "mistake" to mean "unintended," and they would mean by this that a person would not be considered to have made a mistake if what he or she had done was done "deliberately," or "intentionally," even if the outcome was indeed bad. Using this definition, a person would say, "That was no mistake; he meant to do that." In speaking this way, he or she would be using, as the criterion to judge the decision, the "state of mind" of the person making the decision at the moment the decision was being made. Although this is certainly a perfectly acceptable way of defining "mistake" for certain purposes, it is not the definition being used here, for the purposes of this book. In this book, considering the outcome to have been bad, when an alternative decision would have produced behavior with a better outcome (better set of probable outcomes), is, by definition, considering the decision to have been a mistake. Similarly, a decision would be considered a mistake, even if no act occurred because of a subsequent change in decision prior to acting, if the outcome(s) of that original decision would be considered likely to be less optimal than would be the outcome(s) of an alternative act.

Of course, we recognize that some mistakes are an expectable part of life and have no detectable deleterious consequences (actually helping us to learn from our mistakes), whereas some mistakes are very unfortunate, and even disastrous and tragic. Target practice, with anything less than a bull's eye being regarded as a mistake, would be an example of mistakes not having detectable deleterious consequences. So some mistakes are much more important to avoid than others.

I ask the reader to acknowledge that there are many decisions that are made that turn out to be undesirable and can be called (for our purposes in this book) "mistakes," and that the quality of our lives becomes worse the more of these mistakes we make.

As noted above, we also have other words that mean approximately the same thing. Examples would be "blooper," "misdeed," "sin," "atrocity," "blunder," "oversight," "slip-up," "fault," "misdemeanor," "crime," etc. These words all have the same primary meaning, that the decision was not optimal, but they generally have additional connotations. Since we are trying to acquire a basic understanding of our topic, it will be helpful if we do not use alternative words that might have connotations that would distract us from our task.

It will be important to acknowledge that there can be said to be two types of mistakes, with significant differences between them. Let us call them "type one mistakes" and "type two mistakes." (There will be no clear dividing line between these two concepts, however.)

Type one mistakes are non-optimal decisions that the individual made when he or she had, at the time of making the decision, the knowledge that allowed for the prediction that the probability of a non-optimal outcome was increased. Examples would be failure to wear a life jacket or seatbelt, speeding, stealing, etc.

Type two mistakes are non-optimal decisions that the individual made when he or she did not have, at the time of making the decision, the knowledge that allowed for the prediction that the probability of a non-optimal outcome was increased. Examples might be investing in certain stock, taking action based upon someone's lie, eating contaminated food in a restaurant, etc.

This distinction between the two types of mistakes is not relevant to the current discussion, but I am including it here only to make clear that we are at this point talking about all decisions. Later, we may, for instance, discuss how guilt about type one mistakes may be considered rational (appropriate), whereas guilt about type two mistakes may be considered irrational (inappropriate). (Irrational guilt is often found in depression, and even in daily living, given certain irrational aspects of our cultures, including certain frequent phenomena found in current, natural child rearing. Again, though, such ideas are not relevant for our current discussion.)

Now we have noted that inherent in our definitions of "behavior," "decision," and "mistake" is the idea that since we are talking about the choosing of one option to act among a set of options to act, we can assume that each of those possible acts will have a somewhat different outcome. Each step the reader takes produces a new situation, and if the reader took a different step, that new situation would be at least a little different. But even more noteworthy is the apparent, rather obvious fact that there is a strong tendency for the act that is chosen to be one more likely to result in a desired or favorable outcome. There are various ways to point to this phenomenon. We may say that animals are more frequently successful than not (otherwise they would not survive). We may say that behavior becomes adapted to the environment. We may say that animals learn to survive by doing the "right" (adaptive) things. We may say that somehow we "know what to do," and we are usually more right than wrong. We may say that we try to avoid making mistakes. In other words, the acts that are chosen are not random, but somehow seem to be chosen because of "the outcome that is likely to result."

Thus, there seems to be "something" in the nervous system of the animal that corresponds to the connection between its potential acts and the situations that would result from those acts, making it possible to choose an act that is more likely to be successful, that is, to have a particular outcome. For example, the rat develops something in its nervous system that results in it choosing to turn left in the maze, such that it obtains food. There is something in the reader's nervous system that guides his or her feet according to the terrain he or she is walking on and the destination that he or she is seeking. There is something in the person's nervous system that enables him or her to say the right words to other people (words that have the desired effect). It is as if we, and other animals, "somehow know" that choosing to act in certain ways is likely to accomplish certain goals (desired outcomes), and that choosing to act in other ways is likely to result in bad outcomes, or outcomes that are not desired. For this reason, certain acts are more likely to be chosen and others more likely to be avoided. Because of "something" in the nervous system, we somehow "know what will happen" (at least to a certain extent) if we do certain things, and for this reason we decide to do them or not do them.

And this "something" in the animal's nervous system is made possible by the fact that regularity exists in the universe. Regularity essentially means predictability. We can, at least to some extent, "count on" certain things happening "under certain conditions," that is, in certain situations. This "something in the nervous system" helps the organism to make use of this "regularity in the world."

To model this "something" in the nervous system (the "something" itself being a model), that makes use of the regularity that exists in the world, we are going to introduce the next two basic terms, "belief" and "prediction" (to be added to "behavior," "situation," "decision," "outcome," and "mistake").

Let us let "belief" refer to "whatever it is in the nervous system that corresponds to, or models, something in or about the world, such that this something in or about the world could conceivably have some effect on the animal's decision-making (and therefore behavior) in certain situations."

We currently have very little idea what physical condition is actually present in the nervous system that constitutes "belief," but this knowledge is not necessary. (We will speculate about this a little later.)

We see evidence all the time for an animal having some pattern of behavior that implies that its nervous system has acquired an altered state that produces a change in behavior that represents an adaptation to something in or about the world. We often refer to this as "learning." So when the rat "learns" that there is food likely to be found in the maze only if the rat turns left, the likelihood of turning left increases. The nervous system now is altered in a way that corresponds to the fact that food has been found only to the left. There is what has happened in the world (food being only on the left) and there is what has happened in the nervous system (manifested by an increased tendency to go to the left in certain situations). All learning implies some alteration of the nervous system, some new "state of affairs" in the nervous system, that somehow corresponds to something in or about the world (that the animal has "experienced," or has been influenced by, in some manner).

The above description of "belief" has referred to a "state of affairs" in the nervous system brought about by learning. It is conceivable (and even perhaps probable) that there are similar "states of affairs" (beliefs) in the nervous system that are brought about by the genetic make-up of the animal. This might be said to be true, for instance, in the case of "inborn fears," repugnance for certain smells that are associated with danger, or even certain reflexes that "work" for the animal because of gravity and other aspects of the world. It certainly can be imagined that the process of natural selection could result in the nervous system having built into it certain beliefs that have survival value. But whether this is true or not will be irrelevant for our purposes. Insofar as there is a state of affairs in the nervous system that "corresponds to," or models, something in or about the world, we may label that state of affairs "belief," no matter how the state of affairs came to be.

Now notice that our concept of belief does not imply that there will always be an observable manifestation of it at any given point in time. For instance, the reader might believe that his or her car is in his or her garage, but that belief may not be "doing anything" at the moment, because the reader has no use at this time for the car, or for the knowledge as to where the car is. Nevertheless, there is in the reader's nervous system something such that if the reader suddenly wanted the car, he or she would be able to go right to it. So we say that the belief exists in the nervous system at all times, including times when it is having no effect on decision-making (and thus no effect on behavior), and in fact is having no effect on anything that we can imagine.

On the other hand, there are times when the belief does indeed have an effect on decision-making. So what term should we use to indicate this effect? I propose that the best term will be "prediction."

Notice that whereas "belief" refers to a "state of affairs" in the nervous system, "prediction" refers to something that the nervous system is "doing" at a particular point in time.

(But also notice that this statement is partly a reflection of our language, because the "state of affairs" in the nervous system may turn out to be something that the nervous system is indeed actually "doing" all the time, even though we do not know what it is that it is doing. Nevertheless, for our purposes, I believe we can make this distinction.)

Again, to try to clarify these concepts, let us take the simple example of our laboratory rat. We place it in the maze over and over, always putting food to the left, and the rat learns to turn to the left (especially if hungry). Now we put the rat back into its cage. The nervous system of that rat has been changed. But one would never know it to look at the rat in the cage. We don't see the rat continuously running in circles to the left, for instance. We therefore conclude that there is a state of affairs in its nervous system that is not having any effect at this moment on its behavior (or, for that matter, on anything else, such as emotion). But if we put the rat back into the situation of the maze, especially when it is hungry, we will see evidence that we are correct about the nervous system having been altered, because now the rat will indeed go to the left more frequently (whereas before we began the experiment it did not necessarily do so). So when we observe the rat go to the left, and we conclude that it is doing so because of the altered state of affairs in its nervous system, we are assuming that "something is happening" at that moment in time, and it is that "something that is happening" that we will be calling "prediction." The "altered state of affairs" of the nervous system, that is present even when the rat is in its cage, we will designate as "belief," and that BELIEF is the reason for the PREDICTION occurring in CERTAIN SITUATIONS.

(Notice that when the rat consistently turns left, we consider that the rat has some "memory" of which way to turn, and we can say of that "memory" that it is a model of a part of the world. The prediction is therefore a model, based upon or related to memory, of a part of the world. So all of these processes, memory, imagination, belief, prediction, etc., are examples of models occurring in the nervous system, produced by stimuli coming from a world that has some degree of regularity, or predictability, to it. Memory and imagination will be discussed later.)

Notice also that, so far, the ONLY manifestation of a belief is the set of predictions that may occur because of the existence of the belief. Of course, many of these predictions may actually never occur. For instance, if we never put the rat back into the maze, we will never see behavior that reflects the operation of the prediction that food is to be found only to the left when in the maze. On the other hand, we can indeed imagine situations in which the belief would be made manifest by certain predictions occurring and thereby producing certain behavior. So we can say that the belief existing in the nervous system of the animal is indistinguishable from the complete list of all of the predictions that could ever result from the having of the belief. We can therefore simplify our model, I propose, by saying that a belief is that list, or set, of all the predictions that would possibly result from the having of that belief. We are, then, referring to a set of potential predictions, ones that may never actually occur but would have a tendency to occur in certain situations, should those situations occur. Belief is whatever it is in the nervous system that makes for the tendency for these specific predictions to occur in certain situations.

In our model, then, there is included in "belief" nothing different from "prediction" other than whether something is manifesting itself or not. When prediction is occurring, belief is manifesting itself, so to speak. So, prediction is indeed belief, belief that is currently influential in the nervous system. It is easy to see that one can say that a prediction is a belief. When I predict that X will happen, I believe that X will happen. I may for a long time have believed that X will happen, but this belief may not have manifested itself because it was not relevant to my current situation. But should the situation change, the belief may become relevant, such that it will manifest itself as a prediction that may affect my decision-making.

So we are saying that a belief is a set of potential predictions. (By "potential" I mean nothing more than that they could occur, that is, would occur in certain situations.)

However, a particular belief could not be the set of all potential predictions, or an arbitrary or random set of potential predictions, but only the set of all potential predictions that would be brought about by that belief, and therefore would be consistent with that belief and therefore also consistent with each other. And since we are defining "belief" in terms of these predictions, we want to remove the concept of "belief" from its own definition. So we would define "belief" as a set of consistent, potential predictions. I will acknowledge with the reader that the word, "consistent," is vague and undefined at this point. But it is not meaningless. For instance, we certainly could say that if our rat made the prediction that food would be to the right (and therefore perhaps went to the right), then that prediction would be inconsistent with the belief that we have been attributing to it in our example (the belief that food in the maze is only to the left). In fact, we could even say that it is only by virtue of the fact that certain predictions were consistent with one another that we could conclude that a belief existed. (If the rat seemed to predict that food was on the right as often as it seemed to predict that food was on the left, we would question whether the rat had any belief at all about the location of the food.) We will focus on the meaning of "consistent" later.

In summary, our key words at this point are: "behavior," "situation," "decision," "outcome," "mistake," "belief," and "prediction."

And we say that an animal has beliefs (in its nervous system) that manifest themselves in certain situations as predictions of outcomes of behavior, those predictions having an effect on the animal's decision-making and therefore quite likely its behavior, and that under certain circumstances the animal may make a decision that we can label as a mistake, because the outcomes of that decision will or would be less than optimal. What we would like to do is to find ways to reduce mistakes to a minimum, since the more mistakes we make, the worse is the quality of our lives. Reducing mistakes to a minimum, in order to promote the good life for everyone, or, in other words, OPTIMAL LIVING, is what this book is about.

It will be helpful at this point to speculate, for the next 20 paragraphs, about how the brain works. The validity or adequacy of the speculations will not be essential for understanding and using the concepts in this chapter, but they will again serve as a best-model-so-far to help us think more clearly and easily about the other material in this chapter. We will be trying to tie the above "psychological" terms tentatively to a "physical" model of brain functioning, not only to help us conceptualize more clearly, but also to produce a greater sense of confidence that our model will not produce beliefs that are contradictory to beliefs acquired by means of the physical sciences. We must remember that these speculations will necessarily be oversimplified and inaccurate, but I believe they will be of help to the reader.

An animal's current situation is causing input into the brain through a number of different sensory organs and other sources of nerve impulses. We know that these various inputs are actually going to several or many different parts of the brain. (For instance, visual impulses go to the occipital cortex, whereas auditory impulses go to the temporal cortex.) We could refer to each of these areas of the brain as responding at the level of "sensation." These various parts of the brain then model that input (classifying and quantifying it), and send the results of that process to a "higher level" area of the brain, that in turn "combines," or models, that input from some or all of these lower level models. We might be inclined to refer to these higher level models as examples of "perception." The modeling that goes on in these various areas of the brain may be influenced by other models from elsewhere in the brain, especially from even "higher" levels. (This is why we refer to perception as being influenced by various factors, including past experience.) As the modeling of stimulation from the current situation takes place at higher and higher levels in the brain, the resulting models become more and more general. Even the lowest levels of modeling involve the stripping away of, or loss of, some of the specific details of the incoming stimuli.

It is often true that the more times that one nerve cell (neuron) stimulates another, the easier it gets for that stimulation to take place. Therefore, when a particular pattern of activity in the brain occurs, there is a tendency for that pattern of activity to occur more easily in the future. Thus, the more certain "pathways" in the brain get used, the more easily they become activated in the future. So when a particular area of the brain creates a model of stimuli coming from a situation, or creates a model of input from several areas of the brain that have responded to a situation, that model will become active a little more easily in the future in response to some stimulation. Such models, especially perhaps the lower level ones, could be considered to be "memories," in that if they were somehow activated again, the animal would experience something like the original experience. However, it is apparent that the original experience could not be reproduced exactly, in the same detail, with all of the sensations exactly the same, since much of the input has been stripped away. Memory does not work like a video recorder. What is reproduced is actually the activity of a somewhat generalized model of the experience, with specific details tending to have been lost. (The brain can add back in details from other memory fragments or other models in the brain, this being why memory can be so inaccurate and "creative," and thus at times "false.")

Now, as noted, the "activation" of a persisting model of a previous experience could be called "remembering," and we could call that model a "memory" of the previous experience, which, however, might lie dormant in the nervous system until possibly "activated" by incoming stimuli from a new experience (that serves as a "reminder"), or by some other process in the brain, such as the output from a higher level model. The model activated might itself be of a "lower level" specific memory and/or a "higher level" generalized "kind" of situation.

When new stimulation, by a new situation, is being modeled, the new model may be very similar to a model that already exists from previous experience. The new model will most likely have a tendency to activate the already existing model. Thus, a new experience may become "recognized" as an example of a kind of situation, or as being very similar to another specific memory. There does appear to be some degree of comparison of current experience with previous experience. This tendency toward comparison is obviously brought about by the fact that there is regularity in the world, and that recognition of such regularity promotes adaptation and survival.

Of course, the animal is also engaging in behavior. This behavior is the result of nerve impulses leaving the brain and traveling to various muscles (and other components of the body, including glands). This output is the result of patterns of activity in various parts of the brain that we could say constituted a model of the behavior that would result. These behavioral models also probably exist somewhat hierarchically, such that lower level models would produce simple behaviors, while higher level models would "orchestrate" various combinations of and sequences of the simpler behavioral components. These behavioral models not only produce actual behavior, but also probably send input into other developing models, such that the behavior that is occurring in response to a situation becomes a part of the situation that is being modeled as a memory of that situation. Therefore, the animal may have a "memory" of behaving in a particular way in a particular situation, or perhaps even of engaging in a kind of behavior in a kind of situation.

Animals appearing later in evolution and capable of more complex functioning probably accomplish such functioning through having more (and therefore "higher") levels of modeling. (Most of this higher level development seems to be in the frontal lobes of the brain.) Their patterns of perception, memory, and behavior become more complex and generalized, and these more complex and generalized models are modifiable, of course, by incoming information (current stimulation) and by memory (persistent models of earlier stimulation).

Note that "remembering" is really a kind of primitive imagination, the activation of an "image" (the memory) in a manner different from the simple ongoing modeling of currently incoming stimuli ("perception"). (Of course by "image" I do not mean just visual experiences, but any kind of possible input, even from the brain itself, as in the remembering of having had a thought or feeling.) The use of the word "imagination," however, is usually reserved for processes more complex than just memory. If, for instance, various of these (memory) models were activated at the same time (by a higher level model, for instance), a new combination of them would be an "image" that had never before been experienced in its totality, but was instead a new creation. Thus something new could be "imagined," the parts of which, however, would indeed have been actually experienced, probably at various times in the past.

An important part of the modeling that the brain does has to do with time. Somehow, the brain sometimes models the chronological sequence of the states of the various parts of the brain as the output of those parts passes on to higher levels, such that the memory of at least some situations includes the chronological sequence of events (the order in which they occurred) in those situations. This means that, to a certain extent, the reactivation of this memory model produces a sequence of states over time. (Under certain neurosurgical circumstances, stimulating part of the brain results in a person reporting experiencing a "memory" of an unfolding event that continues while the stimulation continues, and begins back at the beginning if the stimulation is interrupted and then resumed.) Under normal conditions, the speed with which "remembering" occurs may be substantially faster than the speed of the original development of the model by virtue of the original stimuli. (In other words, it does not take as much time to remember a sequence of events that has occurred as it did for the sequence originally to have occurred.) In fact it may even be that the brain has a way of modeling some experience such that the model of a sequence of events may be active at essentially one time, just as a graph of some process may display instantly the variation of some quantity over the course of a period of time.

Now as we have said, when a new experience is occurring, incoming stimuli are bringing about the ongoing production of a new model. Then, if the new model begins to resemble a previous model (a memory), the previous model somehow begins to be activated also (the process of being reminded), and that model includes components that are ahead of the incoming stimuli in time. Thus, as the new model is being developed of the incoming stimuli, there is the possibility of matching the currently developing model with the model from the past. In this way, the animal has the capability, while receiving the current input, of responding to whether the current input is consistent with the memory of the previous input. It can tell whether what is happening now is turning out to be similar to or different from what happened last time. It can, then, tell whether something is happening "as expected" or not. (We could say that the incoming pattern of stimuli is being found to be "familiar" or "unfamiliar.") And what is expected is what is predicted, using our terminology. In other words, in a particular situation, the animal may predict that certain things are going to happen, because they have happened that way in the past, and this prediction may be confirmed or disconfirmed as the experience proceeds.

So we may say that as a model of a previous situation is being activated by a current situation, that part of the previous model that contains parts of the situation that have not yet occurred in the current situation are what we mean by prediction.

For example, in Pavlov's famous experiment in "classical conditioning" of a dog, every time he rang a bell, he would place meat powder on the tongue, causing salivation to occur. Thus, a model occurred in the dog's nervous system that included, sequentially, the hearing of a bell and the tasting of meat powder (which in turn activated salivation). Then Pavlov would ring the bell without putting meat powder on the tongue, and the hearing of the bell would activate the whole model, including the memory of the taste of meat powder (in response to which salivation would occur). This "imagination" of the taste of meat powder, as an expected experience, would thus be equivalent to our psychological term, "prediction." (Remember, we are making no assumption that the animal is consciously experiencing anything, despite our use of these "psychological" terms.)

Now on the other hand, if the model of the past that is being activated is one that contains "memory" of the animal's own behavior (that is, the internal and external stimuli produced by the animal's own acts, including input from the activated behavioral models that produced the behavior), then the later events within the (memory) model from the past would be the predicted outcome of the behavior. Let us remember that there is a period of time, perhaps at times quite brief, between the decision to act and the actual act. So we have the model from the past that contains memory of the animal's behavior in a situation and the outcomes of that behavior, and stimulating it we have the developing model of the present that at first contains the situation that the animal is in but then begins also to contain the behavior that is about to be engaged in, and at that point the rest of the activated model from the past is a prediction of what the outcome of that behavior will be. Notice also that if the predicted outcome is aversive, that is, "undesired," the effect could be to suppress that model and to activate one that contained a model of different behavior to be engaged in, the outcome of which was less undesirable or more desirable. Something like this process is probably what is involved in the process of decision-making within the nervous system, and it is an explanation, in "physical" terms, of the previously mentioned fact that animals "somehow know" what the outcomes of their behavior are likely to be. It also describes "instrumental conditioning," the changing of behavioral tendencies by the rewarding or aversive consequences, or outcomes, of those behaviors.

And we humans (and probably some other animals) can spontaneously, in the absence of immediately relevant stimuli, because of higher level models that activate lower level models, create in imagination a model of a situation, that includes our own behavior, leading to a prediction of the outcome of that behavior. (I can imagine doing something when I am at a certain place tomorrow, and imagine what will happen if I do.) And even more, we humans can try out in imagination several different behaviors in response to an imagined situation, to see what the different predicted outcomes will be, and then finally decide upon the behavior that is most strongly predicted to result in the best (most desired) outcome. We may also even imagine different situations and different decisions within each of those situations, and then compare the results so as to be prepared with decisions based upon whatever situations arise, this process being referred to as "contingency planning." Thus, to some extent we can go beyond "learning from experience," to "figuring things out," simply because we are more able than many animals to activate models of past experience, and even models of new experience (imagination), independently of currently incoming stimuli. It is probably primarily "even higher level" models that are doing this "activation."

We have said that the repeated use of a particular set of "pathways" in the brain leads to a greater tendency for those pathways to be used. It should be apparent, then, that the more times a situation has been similar, activating a similar set of pathways in the brain, the stronger will be the tendency for the model of that situation to be activated by newly occurring, similar situations. Thus, the more times one event follows another, the stronger will be the prediction that the second event will occur if the first event has just occurred again. The "confidence" in the prediction is increased by each repetition.

So the more times the dog experiences the taste of meat powder following hearing the bell, the stronger will the prediction of the taste of meat powder become, and the greater will be the tendency of the dog to salivate in response to the bell. (And conversely, we can imagine that repetition of the experience of hearing the bell without subsequently experiencing the taste of meat powder would result in the gradual weakening of the confidence in the prediction of the taste of meat powder, with a diminishing tendency to salivate.)

And if the first event, or situation, includes the behavior of the animal itself, then the strength of the predicted outcomes of the animal's behavior would be increased by the repetition. For instance, the more times food is found when the rat, in the maze, turns to the left, the more strongly the rat will come to predict, in the maze, that finding food will be the outcome of turning to the left. (And it will then take a longer time, or more repetitions, of finding that food is no longer to be found by turning left for the prediction to become "unlearned," the process of "extinction.")

(Since it seems that connections in the brain grow stronger with repetition, just as muscles grow stronger with repetition of contraction, exercising accurate predictions in the brain would seem to be as important as exercising the muscles, if the goal is maintenance or enhancement of function, as understood by "use it or lose it." But there will be more about this under "belief management.")

So we are saying that a situation brings about a prediction, because the situation activates a model in the nervous system that contains within it the potential prediction. This model then must be at least close to what we are referring to as a BELIEF. In our example of the dog "conditioned" to salivate to the sound of the bell, the model (belief) lies dormant in the nervous system until activated by hearing the bell (situation). The model is something in the nervous system that corresponds to a regularity in the world, and we would therefore say that it represents a belief about the world, that nevertheless is not doing anything unless activated by a situation, or some other process, in which case it then becomes manifest as a prediction.

And the "strength" of a belief, that is, the "confidence" in those (potential) predictions that would be brought about by certain situations, would correspond to the "strength" of the "connections" in the model, produced by its repeated activation.

Let us note that we have described even a simple "memory" as a model residing in the nervous system, which causes predictions to occur when stimuli that are coming in construct a model that "reminds" the animal of the already existing memory model, such that the animal predicts that the same thing is happening again (the experience of finding the current events "familiar"). That being so, even the simple memory model qualifies as a belief. And this conclusion is certainly acceptable, in that we might readily say that a memory is a belief about something that happened in the past. Basically, if it is a model in the nervous system that causes a prediction, even a "weak" one, to occur in some situation, or could cause one, then it is acceptable to call it a belief, whatever else we might also call it (such as "memory").

The above speculation about how the brain works, which remains to be verified, elaborated on, and improved by scientific observation and experimentation, has been partly for the purpose of giving the reader an idea of the underlying similarities between very primitive behaviors and brain processes of lower animals and the more complex behaviors and brain processes of higher animals, especially humans. This is important because we are attempting to understand the determinants of all behavior (simple and complex) of all animals (lower and higher). And the reason that doing so is important to us is that our own behavior ranges from the simplest to the most complex, the more complex having been added to, rather than having replaced, the simple. And by understanding as much as possible the determinants of all of our behavior, we can optimize our ability to change it and thereby achieve a much better quality of life.

Now we have been talking about beliefs and predictions in general. But we will also want to talk about specific beliefs and predictions, so we will need to have a way of referring to them, designating them, labeling them, differentiating between them, or, in other words, modeling them. So for the next 33 paragraphs, we will be developing this method of modeling specific beliefs and predictions.

It so happens, of course, that we humans have already done this. And in fact our way of doing this has had a profound effect on our beliefs. As we shall see, our method of modeling beliefs in the nervous system has led us to be able to have a much larger number of beliefs and also to drastically improve the precision, consistency, and accuracy of those beliefs (the tendency for those beliefs to produce predictions that turn out to be what actually happens). However, this method of modeling our beliefs, which has developed naturally, has only gradually come to produce these improvements, and only to a very limited extent, compared to what is ultimately possible. We have not yet reaped anywhere near the full benefit of this new capability.

So what I suggest to the reader is that we "start from scratch" and develop our own method, designed to be as precise as possible, but keeping in mind that this method will be the same as that which we already use naturally, but use of course with great clumsiness.

We will model beliefs and predictions using PROPOSITIONS, these amazing tools that our species has come to be able to use.

But first, in order for us to proceed successfully, a somewhat subtle but important distinction must be made when we talk about a "belief" or a "prediction." Our language can confuse us here because of the fact that, as is so often true, the same words may be used to mean different things. This clarification will be best carried out by looking first at "prediction."

Suppose I go around to people and ask them who they believe is likely to win the next election, and I write down what they tell me in a list, using the format "X will win the election." Then I go to someone and I point to the sentences on the piece of paper and say, "Look, here is a list of predictions." Notice that these "predictions" could be said to be "disembodied." I am pointing to some propositions that anyone might indeed call "predictions," in that they have the form, "X (or Y or Z) will win the election." But these "predictions" are NOT processes that are occurring in a nervous system at that point in time. In fact, by the time I show the list to someone, something may have happened such that none of the contributors would make the same statement again. (So I could even say that these sentences, or propositions, are a set of predictions that no one is making anymore.) The reason that these propositions are being called "predictions" is essentially because of their reference to the future. This meaning of "prediction" is perfectly acceptable and frequently used. However, we are not trying to model (with a proposition) a "prediction" in a "disembodied" sense, as, for instance, something we might find in a book, but are instead trying to model a process actually going on in the nervous system of an animal (human or other). In the same way, "belief" could mean a "disembodied" belief, such as a belief that no one has anymore, or it could mean a state of affairs actually existing in a specific animal's nervous system. We are currently using only this second meaning of "belief."

We would be tempted to make this distinction clearer by saying that our concept of "prediction" refers to the "act" of predicting, whereas the other kind described above does not, except that this use of words would be contradictory to our previous use of "act," as being similar to "behavior." In our model, "prediction" may lead to, or influence, decision-making and thus may be a "precursor" of an act, so it would be confusing to call prediction itself also an "act." But we do consider it to be a process or event-of-some-sort occurring in the nervous system of an animal. (Remember, we excluded certain activities, such as digestion, from our set of animals' activities that we define as "behavior," because they were too automatic. We might consider prediction to be another example. But, of course, if a human were to decide to "predict" something, then doing so would indeed be considered an "act," or "behavior.")

Now let us start trying to develop our method of modeling beliefs and predictions, existing or occurring in a nervous system, with propositions.

Let us take the simplest sort of prediction, one that does not involve creative imagination and one that does not involve decision-making. The stimulus input from the current situation is activating a model in the brain that includes what is likely to happen next. Our proposition would be "Since X is happening, Z will happen." We could also say, "Because X is happening, Z will happen." Note that Z may (or may not) be something that induces fear, anger, sadness, hunger, sexual arousal, etc. In the case of Pavlov's dog, the proposition modeling the prediction would be, "Since (or because) the bell is ringing, the taste of meat powder will appear." The taste of meat powder produces certain automatic responses that include activation of the autonomic nervous system to produce salivation. This is classical conditioning described by the use of propositions.

Now let us add decision-making to the situation. In this case, our proposition would be, "Because X is happening and I am about to do Y, Z will happen." In the place of "about to do Y" we could use "deciding to do Y" or "planning to do Y," or "am in the process of doing Y." In other words, this part of the proposition is modeling whatever is occurring in the nervous system that we would refer to as "deciding," or "decision-making." In this case, "Z will happen" represents the predicted outcome of the decision. "Z will happen" is the model of the remainder of the general model in the nervous system that has been established by one or more repetitions of a similar set of events that have included a behavior or kind of behavior. Again, we may note that Z may produce a reaction within the nervous system that either inhibits or promotes this model (for instance, produces fear or hunger), and therefore decreases or increases the tendency for the model to unfold to completion (producing the behavior). The predicted outcome thus influences decision-making. This is instrumental conditioning described by the use of propositions.

Now we have noted that "imagination," as described in this book, is a capability that has been developed by some species, in which a model of a situation or sequence of events can be activated, not by incoming stimuli, but by a "higher level" model that calls forth this model. The animal "imagines" X happening and therefore imagines Z happening. We would model this with the proposition, "If X happens, then Z will happen." Note that when the animal makes this prediction, even though it is not produced by incoming stimuli, but instead by an imagined situation, the prediction may nevertheless also induce a motivational state. For example, the animal could imagine a situation that would produce a prediction that induces fear, anger, etc. If Pavlov's dog were to have the capability of imagination, it might imagine the bell ringing and thus begin to salivate, even though no bell had rung. For us humans, obviously we are able to decide to imagine, for instance, a favorite meal, and thereby induce an increase in hunger. The same would be true for imagining a sexual situation and experiencing sexual arousal, or, for some, imagining a storm and experiencing fear.

And now we can add decision-making to the imagined situation. Our proposition would become, "If X happens and I do Y, then Z will happen." Once again, Z may produce a motivational state, either inhibiting or promoting the tendency to imagine doing Y in response to X, and therefore increasing or decreasing the likelihood of actually doing Y if X eventually happens.

Remember that beliefs manifest themselves as predictions in response to situations. As we can see, if the animal is not capable of imagination, then the belief will be manifested only in certain actually occurring situations that activate the model that we are referring to as the belief. If the animal is capable of imagination, however, then the belief may be manifested as the animal imagines various situations. We, capable of imagination, looking at the animal and attempting to model with our propositions the beliefs of that animal, can imagine various situations that the animal might be in and the predictions that the animal would make in those situations. Therefore, we would say that the animal's belief would consist of all of the predictions that we might imagine that the animal would make in response to all of the situations that the animal could conceivably find itself in. Whatever predictions that belief would lead to, in all possible relevant situations, would be what we would MEAN by that belief. We humans can use our imagination and our ability to use propositions to construct a list of propositions that model many of the predictions that would occur in various situations, and this list of propositions would be our approximate meaning of the proposition that we used to specify the belief. The greater the number of such propositions for imagined predictions we listed, the more completely described would be the meaning of the proposition modeling the belief. Of course we could never list all of the predictions that we could imagine could possibly occur, partly because we could not list all of the possible situations, but we are saying that the list is what the proposition modeling the belief actually means, or stands for.

If the proposition modeling the dog's belief is, "Ringing bells are followed by the taste of meat powder," then one of the propositions that this proposition would mean would be the prediction, "Because I hear a bell ringing right now, I am about to taste meat powder." If the proposition modeling the rat's belief is, "Food of some sort is found to the left in mazes," then one of the propositions that would be in the list representing the meaning of that proposition would be the prediction, "If I turn left in this maze, I will find food." If I, myself, believe that "My car is in my garage," then one of the propositions constituting the meaning of this proposition would be the prediction, "If right now I ask Mary to look to see if my car is in my garage, she will report back that it is."

Notice that the proposition modeling a belief (an example being, "It is raining") does not necessarily have to have all of the ingredients that a proposition modeling a prediction would have (including X, Z, and perhaps Y, as described above), but a proposition modeling a prediction (with all of its components) would certainly also be a proposition modeling a belief, just as we have previously acknowledged, when we have said that any prediction, itself, could also be regarded as a belief.

Remember, however, that although a proposition may model a belief, it might instead be a lie, or a joke, or a metaphor, or something that is not believed or no longer believed, or something that is to be tested or examined further but not currently believed. There must therefore be some designation, overt or implied, that the proposition is indeed modeling a belief. We actually do this all the time. The context usually is what designates the proposition as a belief. For instance, all of my propositions in this book are representing my beliefs, and the reader is assuming that. A proposition could be preceded by an indicator, such as "I believe that...," as in "I believe that it is raining," (or "He believes that it is raining" or "It believed that the food is only to the left"), the proposition (it is raining, or, the food is only to the left) being thus contained in another proposition that designates what the original proposition is. An example of the proposition not being the model of a belief would be it is raining in, "He doesn't believe that it is raining," although the whole proposition might indeed represent a belief (of the author of the statement).

In summary, we may state simply that any BELIEF may be modeled by a PROPOSITION. So this proposition is our model of the model in an animal's nervous system of SOMETHING IN OR ABOUT THE WORLD. This belief consists of a set of consistent potential predictions (activated by possible situations), each prediction of which can be modeled by a proposition, and it is the TOTAL SET of these propositions, modeling those predictions, that is the MEANING of the proposition modeling the belief.

At this point, I wish to discuss the previously-mentioned extremely powerful effects that our having developed this ability to use propositions to model beliefs and predictions has had on our beliefs, compared to all other species. (But then I will also attempt to point out how little of this power we actually use, compared to what would be possible.) The four amazing effects on our beliefs are on their number, their precision, their consistency, and their accuracy.

The number of beliefs that we can attain is enormously larger than the number that other species can attain. We are able to bring about the development of new beliefs in each other simply by using these propositions, that is, by conveying "information" to each other, about things that we have no direct experience of. Our symbols work by means of a kind of imagination, such that when a symbol is used, it produces in the nervous system an activation of a model of something previously experienced (memory). With our propositions, these individual memories or memory fragments can be rearranged and combined in an almost infinite number of ways. We can therefore create stories (models) of events that have not yet happened, or happened long ago. We can create models of things which have never existed, have existed long ago, or will exist in the far future. And we can create models of (beliefs about) the way the world is, was or will be, consisting of things, events, and processes that we can never directly experience. For instance, someone can tell me about atoms and energy fields that I have never seen, and never will, and thereby leave me with beliefs about those things. A rat can have no beliefs about atoms and energy fields. A rat's set of beliefs have been acquired through its own, personal experience, and not through induction in the rat's brain (mind) of symbols by other rats. Only we humans can come to have beliefs about science, history, politics, religion, etc., as well as beliefs about what a friend has just done, or is planning to do, in another part of the world. Only we humans can come to believe the sun will one day engulf the earth. And increasingly higher level models in the brain may be modeled by increasingly abstract propositions.

The precision of our beliefs is enormously greater than that of other species. We have noted that the phenomena in the nervous system (belief and prediction), at least as far as we can describe them, are quite amorphous, vague, and confluent. But our use of propositions can be quite precise, primarily by virtue of precise, agreed-upon definitions of the symbols (words) in those propositions, along with use of agreed-upon rules of syntax. For instance, I can tell you that I will reward you if and only if you refrain from opening the door unless you hear three knocks, a pause, and then two knocks. I can say it once, and you will have a very precise belief that you can act according to, if you wish. The reader may imagine what would have to be done to train a member of another species to carry out this procedure, even if the animal was eager to obtain a reward by obeying correctly.

The consistency of our beliefs is enormously greater than that of other species. Whereas we are dealing with amorphous, vague, and confluent phenomena in the nervous system, we are using a system of modeling of those phenomena that has built into it a rather marked capacity for consistency, by virtue of the ability to use the rules of logic. Building upon our acquired ability to model beliefs and predictions with precision, the rules of logic allow us to conclude whether, if proposition A is true, then proposition B must also be true, by virtue of whether a contradiction would otherwise occur. Thus, with the rules of logic, we can test our beliefs to see if they are contradictory to one another. And we can test a specific proposition, modeling a potential prediction, to see if it follows logically from another proposition that models a belief (to see if the prediction is a part of the meaning of the belief). Of course, these judgments are determined also by any number of other propositions, many of which are implied. For example, believing my car to be in my garage of course leads to predictions as to what will happen if I look in the garage, but such predictions are also dependent upon beliefs as to whether I have a car or not, whether I have a garage or not, whom I can trust, etc., and such beliefs, in turn, have their own sets of potential predictions. So we are able to develop "belief systems" or sets of beliefs, with their potential predictions, that are internally consistent with each other, in that their combination does not result in contradiction.

The accuracy of our beliefs is enormously greater than that of other species. Because we can produce such precise and consistent predictions, we are then able to determine if our predictions indeed turn out to be correct. Failure to predict accurately what will happen implies that something is wrong with our beliefs. The accuracy of our beliefs has allowed us to do with confidence a large number of things that we would otherwise not even think about doing. Despite the fact that many of the things we do would be quite dangerous if not done correctly, we feel very confident most of the time when doing those things. Driving would be an example. Undergoing surgery or going to the moon would be others.

But, as stated above, we must realize that we fall far, far short of what we really could be able to do with regard to precision, consistency, and accuracy of belief, and this is because we have not yet learned to do well what is involved in attaining these goals.

Please note that these goals (precision, consistency, and accuracy) are only appropriate for certain kinds of propositional behaviors, and they would actually be detrimental to other kinds of propositional behaviors. They are usually appropriate for making good decisions, sharing helpful information, giving directions or instructions, studying the way the world really is, was, or will be, coordinating our behavior, entering into contracts, carrying out legal proceedings, etc. They are usually not appropriate for fostering creative new ideas (including new scientific theories), "brainstorming," sharing one's feelings, entertaining, stimulating emotion, etc., or for fiction, poetry, vocal music, games, etc. The thesis of this book has to do with changing our behavior such that we stop making so many mistakes, and so precision, consistency, and accuracy of belief are the important focus of this book. Ideally, we should be able to aim for and to a great extent achieve those goals when appropriate, but also be able to do other things that add to our quality of life, including making use of some of the natural lack of precision, consistency, and accuracy that exists in the functioning of all of our brains. And we should have accurate beliefs as to which goals are appropriate to which kinds of situations.

But how do we fall far, far short of the precision, consistency, and accuracy that we really could attain?

Regarding precision, even in our most important efforts to solve our most important problems, we are unable to engage in efficient, effective comparison of ideas in order to choose those that are optimal. We have many different words for approximately the same thing, and we use the same word for many different things. Thus, when we debate, there is continuing apparent disagreement, and there is very little ability to focus on one idea to the extent that it is fully explored. Apparent disagreement because of different word usage leads to wandering discourse, and almost never to the changing of anyone's original beliefs.

Regarding consistency, the above lack of precision fosters another problem, namely, failure to use the rules of logic. Logical inconsistency becomes most apparent only when the same words are used for the same things in sequences of propositions. In a debate, we often make logic impossible through "creative" use of words to promote confusion. On the one hand, we have seen that we are able to develop "belief systems," sets of beliefs that are indeed internally consistent, but we also quite readily look the other way when apparent contradiction in our own belief systems is occurring. We do not have a strong need or commitment to apply logic to our most cherished beliefs.

Regarding accuracy, we do not readily follow, or even learn, the rules of evidence (generally acquired currently only through formal education), and therefore easily draw conclusions from our own personal experiences (which have been shown to be highly unreliable), and from listening to others who present what is to be believed in such a manner as to produce very good feeling that rewards accepting the beliefs offered. There are many, many belief systems, to some extent internally consistent (at least on the surface), but many of which are different from and contradictory to one another, and therefore cannot all be correct, and some of which have caused our greatest tragedies. Yet the adherents of those belief systems have strong beliefs that produce very confident predictions, and thus have strong tendencies to make mistakes, sometimes tragic ones. Consistency of belief is not enough. If we are consistently wrong, our quality of life becomes terrible, because of the mistakes we make. It takes our growing use of the rules of evidence to help us develop beliefs that accurately model the way the world really is and thus cause us to make decisions based upon predictions of the outcomes of our behavior that turn out to be the same as what actually happens. The rules of logic are necessary, but not sufficient. The rules of evidence are also essential. (And something even more is needed, something that is the primary focus of this book, and will be specifically discussed in the next chapter.)

So we humans have acquired, by virtue of our ability to use propositions, an almost unlimited number of beliefs, but we have a very long way to go before we globally, as a species, reward, teach, and model for identification the effort to attain, when appropriate, precision, consistency, and accuracy in our use of our propositions to model and manage our beliefs.

I wish now to call to our attention how, because of our acquired ability to use propositions, our behavior has become so drastically different from all other species.

The behavior that I am referring to is VERBAL (communication and thought). As we carry on conversations, we are, according to our model, continuously deciding (not necessarily consciously) what to say, and then usually saying it. Some of this verbal behavior is the saying of one proposition after another, many of these propositions consisting of a model of a belief that we are reporting we have. And as we rehearse (imagine) what we are going to say, what we are doing again is saying (internally) propositions, this being one part of what we usually mean by "thinking." So for us humans, as opposed to all other species, much of our behavior consists of the production of propositions internally in the form of thinking and externally in the form of speech. We may refer to this internal and external behavior as "propositional behavior."

Furthermore, our attention tends to be focused on the relationships between these propositions, rather than the relationships between the propositions and the things that those propositions are modeling. It is as if we humans have constructed a whole new world, consisting of propositions, that ultimately is supposed to model the world that all animals, including ourselves, are faced with. Much of our preoccupation is with this new world, which is essentially unavailable to all other species on this planet. But as we have dialogue with each other, it is somewhat infrequent that we imagine in our minds those parts of the world that our propositions are referring to or modeling. That is, when we are talking, or thinking, it is seldom true that each word in the sentence evokes an image of what the word is supposed to stand for. The meaning of the word has more to do with the other words being said or thought, such as whether such words are usually used with each other in this manner. Much of what is said consists of phrases, or collections of words, that have become frequently used together in this manner and are recognized as accepted sets of words, implying that those collections of words have already been agreed upon as either "true" or "meaningful." The propositions themselves are what primarily occupy our minds.

This "propositional behavior" (communication and thought) is subject to "instrumental conditioning," that is, can be promoted or inhibited by consequences. We humans engage in complex rewarding and punishing behavior in response to our communication with each other, and secondarily then in response to our own thinking. What we say and what we think is subject to powerful sanctions by others. Our beliefs, modeled with propositions, come to be molded by our use of those propositions, and by the consequences of doing so. Systems of belief come to be promoted and defended vigorously by individuals and groups. Whether beliefs are accurate or not is not necessarily the strongest determinant of many of our beliefs. Belief, as manifested by the recitation of propositions, is frequently required as an act of obedience, deviation from which may result in ostracism, punishment, and even murder.

So we, and all other animals, live in a world in which decisions are required. But we humans have built another world, constructed of propositions, that is a model (or set of models) of our beliefs about the first world. We have discovered ways to make this second world much more precise, consistent, and accurate, and this process of doing so has had much beneficial effect on our beliefs about the first world. It has been a part of what, in this book, we mean by "rationality." It is this world of propositions, or this continuous production of propositions (externally as a form of communication and internally as thought) that make us "human." It is superimposed upon our basic animal nature (all those aspects of ourselves that we share with at least some other animals), and has the capability of producing good behavior that is drastically different from what our basic animal nature would have us do. But it also is to a great extent in the service of our basic animal nature, and since our basic animal nature has developed by natural selection only to promote survival (and has nothing to do with "the good life"), that which is our most valuable tool has also become our most awful weapon and most dangerous source of mistake. Optimizing our use of propositional behavior (thought and communication), in order to promote the good life, is what this book is about.

The reader should recall at this point what we are attempting to do.

We are beginning to develop, as a tool, a model that includes a list of simple, basic concepts, with which we will work in trying to change our behavior to promote the survival of and the good life for ourselves and our species. So far, we have said that behavior is considered to be the result of decisions to act or behave that way, based upon beliefs that manifest themselves in certain situations as predictions having to do with the outcomes of these acts. Some decisions are good (desirable) ones and some are bad (undesirable). The bad decisions are, for the purposes of this book, being called mistakes. Whether a decision is a mistake or not will be a matter of opinion, but will depend upon whether the resulting outcomes of that act are or would be considered good or bad (the meaning of "good" and "bad" to be covered later). Although this will always be a matter of opinion, there will be widespread agreement regarding some of these decisions, as to whether they are mistakes or not.

We have digressed from building this basic model by discussing the development, nature, and limitations of our new tools (symbols, syntax, propositions, rules of logic, and rules of evidence), so that we can use these tools in the construction of our model.

We now will continue to construct that model.

We now need one further concept to add to our conceptual model, to help us understand, in a very basic way, how decisions are made and therefore how mistakes are made, so that we can develop ways of influencing these processes for our benefit.

It is easy to see that just having beliefs in the nervous system, or even having them active in the form of predictions, is insufficient to account for behavior, that is, for the animal actually doing something. I may believe that I can get a particular item at the store, but this belief, in itself, is insufficient to cause me actually to do so.

I will now propose for our model the idea that there are two main sets of factors, which, operating together, are the determinants of decisions. (This model will correspond closely with how we talk in our normal language about decision-making.)

Any decision may be considered to be the result of one or more MOTIVATIONAL STATES and one or more BELIEFS relevant to the motivational state(s) and the situation.

The new term is "motivational state."

We know that animals have a tendency to behave, to do things. This tendency may vary from time to time anywhere from an apparent lack of such tendency to a very strong tendency. For instance, we can place a harness around a rat and attach it to a string that is also attached to a scale, such that we can measure the amount of force with which the rat is pulling to try to get somewhere. In some experiments it can be shown that the closer the rat is to a goal (e.g., food), the harder it will pull to get there. We do not know at this point exactly what is going on in the nervous system and other parts of the animal that results in this variable intensity of motivational states, but it is quite an accepted fact that it happens. So, motivational states have variable amounts of strength.

These motivational states are quite varied, and go by different names. The most common general names for these states are "drives," "emotions," "feelings," "wishes," "wants," etc. All of these states tend not only to produce behavior, but to produce specific kinds of behavior that are related to different kinds of outcomes, depending on the motivational state. For instance, hunger motivates different behaviors than does thirst, sexual drive, pain, itching, anger, fear, envy, etc. What motivational states have in common is the tendency to produce behavior or the tendency to change behavior from what it might otherwise have been if the motivational state had not been present.

Just as we have developed a way of modeling specific beliefs and predictions with propositions, we have also come to model specific motivational states with propositions. The "motivational state," then, is a hypothetical state existing in a nervous system or "mind" (or metaphorically in a group) that can be modeled with a particular kind of proposition, namely, a sentence that follows the format, "(subject) wants (object)." The subject is the animal that has the motivational state. The object is a sought-for outcome. Let's look at examples:

The sought-for outcome or desired situation, may be very specific ("I want to see X movie tomorrow afternoon") to very general ("I want this bad feeling to go away" or "I want something, but I don't know what").

It is important to note here that there is no implication that the human or other animal has an "awareness" of the presence of these motivational states. We humans are often aware of some of them, and that is the main reason why we have symbols, or words, for them. However, we know that sometimes the strength of these motivational states is so low that we may not notice them, and we know also that sometimes the awareness of them is inhibited within us for "psychological" or "defensive" reasons. In fact, it is not at all unusual for us to conclude that a motivational state has indeed been present just by looking at the behavior of the person (self or other), even though the person does not feel or perceive its presence. (Examples are, "He denies feeling that way, but just look at what he did!" or "I must have been awfully scared, without realizing it.")

What leads to the appearance of these motivational states? Of course, we do not know exactly, and perhaps we never will know the whole story. But we know enough to make some statements that are most likely accurate enough.

We know that some motivational states somehow arise because of certain processes that increase over time, that occur in and between cells in the body. For instance, we are aware of the build-up of motivational states such as hunger, thirst, sexual drive, and drive for sleep over a period of time.

But we also know that some motivational states, such as pain, anger, fear, sexual arousal, etc., sometimes occur more or less immediately in response to specific events, independent of passage of time.

Furthermore, it is important to note that motivational states may become "attached" to certain other states in the nervous system, such that when one of these other states occurs, a specific motivational state will also occur. These other states include memories, fantasies, perceptions, thoughts, and even other motivational states. The memory of having been treated badly is likely to produce anger. If one imagines (fantasizes) falling off of a high building, one is likely to experience at least a little fear. Seeing one's favorite food is likely to produce hunger. Certain thoughts may produce sadness or embarrassment. The experience of anger may produce fear. The experience of sexual arousal may produce guilt. Some, but probably not all, of these connections appear to be acquired through learning or "conditioning."

Closely related to the above is the fact that certain predictions can come to produce motivational states. If I have the belief that whenever situation X arises I will experience pain, then if I begin to predict that situation X is about to arise, I will experience fear. Therefore, certain beliefs will lead to the appearance of motivational states under certain circumstances (situations) that "activate" those beliefs into actual predictions. I have, since childhood, had the belief that tigers are prone to eat people. But it is only when I see the tiger that has escaped coming toward me that I begin to experience fear based upon the prediction that this tiger may attack me. I may not have started to experience hunger until I saw them bring me my favorite food. Or, my hunger increased when I saw this food. I have beliefs about what the experience will be like to eat this food, based upon old memories, and my prediction that I am about to have this experience again increases the motivational state of wanting to eat this food (similar to the rat pulling harder, the nearer it is to the food).

So, in general, situations produce motivational states. These situations may include intracellular and intercellular processes, perceptions of environmental objects and events, memories, fantasies (imagined situations), thoughts, predictions, etc.

When a situation produces a motivational state, it is often labeled as a "pleasant" or an "unpleasant" situation. "Pleasant" situations are those in which the behavioral tendency is to do that which will prolong or intensify the experience, whereas "unpleasant" situations are those in which the behavioral tendency is to do that which will diminish or avoid the experience.

When a situation produces a motivational state, the situation then includes that motivational state. This situation activates relevant belief(s) into specific prediction(s), which promote specific decisions, and therefore specific behavior.

Modeling with propositions, let us look at simple examples of the relationship between motivational states, beliefs, and decisions.

Notice that BOTH the motivational state and the belief are necessary factors, and that one without the other is insufficient to produce or explain the decision (behavior). For instance, if I am not hungry, I am not as likely to eat this substance. Also, if I don't believe this substance will remove the hunger, I am not as likely to eat it. If I don't care about passing the test, I may not study, and if I don't believe that my studying will help, I will be less likely to study.

Metaphorically, it is as if there is a "push" (motivational state) to do something, and then a decision (based upon belief) as to what to do.

Now let us return to our central task.

I am proposing that we look at ALL BEHAVIOR according to the following basic model. Because of a situation, an animal develops a motivational state and, because of one or more beliefs that it has that are relevant to that situation and motivational state, it makes predictions that lead to a decision to act in a particular way. This means that a change in either the motivational state(s) or the belief(s) may result in a change of decision, and therefore probably of behavior.

The satisfactory understanding of the reasons for a decision requires the knowledge of both the motivational state(s) and the belief(s).

The reader may already have noticed that there is almost infinite complexity inherent in this model, as is shown in what follows.

More than one motivational state relevant to the decision may be present at the same time. For instance, I may not eat the substance because, simultaneously, I have a strong need to go to the bathroom, a need that must be taken care of first to prevent a bad outcome.

Also, more than one relevant belief may be present at the same time. For instance, I may not eat the substance because, even though I believe it would stop the hunger, I also believe it has poison in it, and eating it would lead to a bad outcome.

With a little imagination, one could make up increasingly complicated scenarios that would change the decision back and forth. For instance, I might indeed believe that the food had poison in it, but I might also believe that the poison was very slow acting and that a harmless antidote was available, and that if I did not eat the food, I would let on that I knew poison was in it and therefore expose myself to an alternative attack on my life. But then I might also believe that if I do not die, many more will, etc. So the decision is produced by all of the motivational states present (in varying strengths, of course, and therefore of varying amounts of influence) and by all of the relevant beliefs.

But although there is this almost infinite complexity inherent in the model, it is also true that in many situations in which an effort is being made to understand and/or influence behavior, it will be possible to identify the main, or most important, motivational states and beliefs, and to come to conclusions with a fair amount of certainty. We are neither perfectly able to understand nor completely unable to understand, just as is true regarding our ability to predict the weather.

Now let us return to the concept of "mistake."

A mistake has been defined as a decision that has led to, or would lead to, a bad outcome. This is one use of the word, "mistake." Another common use of the word refers to mistaken belief, which, in this book, will be called inaccurate belief. Inaccurate beliefs lead to inaccurate predictions, and therefore to mistaken decisions. If, when I attempt to eat this apple, I find that the apple is made of wax, my prediction is found to be inaccurate, and I conclude that I have made a mistake (the bad outcome being a mouthful of wax). The mistake occurs because of an inaccurate belief, namely, that the apple is food. So, an important cause of mistakes is inaccurate beliefs.

But some mistakes are made primarily because of the strength of motivational states rather than the existence of inaccurate beliefs. (It is well recognized that strong emotions often lead to mistakes. "I knew it wasn't right, but I felt so upset I couldn't control myself.") On the other hand, strong motivational states perhaps most often are based upon beliefs that have been activated into predictions in certain situations. It is the interpretation of the situation, namely, the beliefs about the nature of the situation, that often produce unusually strong motivational states, and we have often seen such motivational states, brought about by accurate or inaccurate beliefs, lead to mistakes.

In the case of the faulty or "mistaken" perception, such as the mirage or the optical illusion, the mistake (in the sense of the decision) that might be caused by such a perception is caused by the belief that the perception is indeed accurate. (I go off course in my journey across the desert because I believe there is water in that direction, based upon what I am seeing. But seeing is not believing, because another person, already knowing about mirages and also having a map, does not believe what he or she is seeing.) Thus, there is no need to add another set of determinants (beyond motivational states and beliefs) to account for mistakes caused by faulty perceptions.

So we are saying that, at this point, our model will explain ALL BEHAVIOR in terms of just TWO phenomena, namely, motivational states and beliefs, and that all MISTAKES are therefore caused ONLY by aspects of one or both of these two phenomena.

The above being true, it is apparent that we must learn as much as possible about these two phenomena, including their determinants and their interactions. What follows, then, is a set of important observations and conclusions regarding motivational states and beliefs.

First, we need to expand upon the previously mentioned fact that situations, or the perceptions of them, can come to produce motivational states. Upon seeing an appealing plate of food, I may notice the appearance of or an increase in hunger (modeled by "I want to eat"). Upon seeing the low price offered on an item, I may notice the appearance of or an increased wish for the item ("I want to have it"). In each case, the perception appears to produce or increase the motivational state. But remember, this effect, according to our model, is mediated by beliefs. If I believe the food is just a wax model, or if I believe it has poison in it, the motivational state will become different. If I believe that there is an even better deal on, or style of, the item, or a better way of spending my money, my wish to have the item may diminish. So beliefs essentially can cause and/or modify motivational states.

Second, there is an interaction between motivational states and beliefs that can be modeled, as has already been done, by syllogisms. As a further extension of this idea, the reader is asked to consider the following propositions (motivational states are italicized, beliefs are not):

• I want to satisfy this hunger.
• If I eat some food, my hunger will be satisfied.
• Therefore, I want to eat some food.
• If I obtain some food, I will be able to eat it.
• Therefore I want to obtain some food.
• If I go to the grocery store, I will be able to obtain some food.
• Therefore, I want to go to the grocery store.
• If I turn left at the stoplight, I will get to the grocery store.
• Therefore, I want to turn left at the stoplight.
• If I turn my steering wheel left, I will turn left at the stoplight.
• Therefore, I want to turn my steering wheel left.

(The reader should once again remember that we are not talking about thoughts that go through the mind of the animal, human or otherwise; we are just constructing a useable propositional model to correspond as closely as possible to what happens in the nervous system, despite our knowing very little about what is indeed happening in the nervous system. We might speculate vaguely that this channeling process is the progressive activation by motivational states of lower and lower level models, or beliefs, in the brain.)

Notice that any change in the beliefs modeled above could change the motivational states following them.

Notice also that the motivational state at the beginning of this set of propositions is quite general, and that one of the functions of the beliefs is that of leading to more specific motivational states, and therefore more specific decisions. So we may say that beliefs "channel" motivational states into specific behaviors or acts (or, more accurately, decisions).

Notice also that the beliefs listed above are part of a broader network of beliefs upon which they are dependent. They are dependent upon such beliefs as "There is a grocery store," "There is food in the grocery store," "The street is not one way to the right or blocked," "My steering wheel works," etc. And these beliefs, in turn, are dependent upon other beliefs, such as that I can believe what others have told me, that I am not dreaming, that gravity is such that my car will not float away, etc. Therefore there is probably an ultimate connection between all of a person's beliefs, such that a faulty or incorrect (inaccurate) belief may have widespread effects on decision-making and therefore behavior.

Now I am not saying that our beliefs are the only point of intervention in order to optimize behavior. Our motivational states are also dependent upon the normal functioning of the brain. We do know that certain substances and certain illnesses can modify motivational states. We know, for instance, that cocaine can induce strong motivational states. We know that certain mental illnesses can spontaneously produce anger, fear, sadness, etc., and that medication can often help. (We also know that certain kinds of psychotherapy, aimed at helping an individual change his or her beliefs, this being one example of "belief management," can sometimes similarly help alleviate unpleasant or painful motivational states, even at times without the use of medication.) And sometimes the mental illness itself can produce, in ways not fully understood, but possibly primarily through the production of chronically abnormal motivational states, highly unusual and inaccurate beliefs ("delusions"). Nevertheless, I believe that the reader can readily see that it is INACCURATE BELIEFS, through the motivational states they promote, that contribute the most directly, and the most often, to MISTAKES.

In summary, our motivational states push us to do something, but it is primarily our beliefs that lead to our specific decisions as to what to do. And our beliefs produce and modify many of our motivational states.

Remember that we humans have constructed a "new world" of propositions that we have become preoccupied with. In other words, in addition to all of those behaviors that we share with other higher animals, we have an entirely new kind of behavior, propositional behavior (communication and thought). This is indeed behavior, and therefore has the determinants of any behavior, namely, beliefs and motivational states. What we say and what we think are determined by our beliefs and motivational states. The motivational states may conflict with each other, promoting different speech and thought, and it is the resultant of these "forces" that produces what we actually say and think. What we say produces responses from others, which in turn have an effect on how we feel. What we think has an effect on how we feel. Our basic animal nature tends to produce speech and thought that has the immediate effect of inducing positive feeling. As we all really have always known, sometimes that which produces immediate good feeling in the self may nevertheless result in outcomes that cause bad feeling in the self and/or others, and are therefore "mistakes," as the term is used in this book. It is only by the building up of a set of beliefs that produce predictions that cause the appropriate feelings (motivational states) that we can avoid doing what, in the long run, may turn out to have been mistakes.

But note that our beliefs promote certain thoughts and speech, using certain propositions, and then those propositions, when "managed" by the rules of logic and the rules of evidence, tend to have some backward effects on those beliefs. When we observe contradiction occurring among our propositions, we tend to try to correct that situation, because increasingly our culture, that is, the people around us, express disapproval if we are blatantly self-contradictory. To some extent, our beliefs, and therefore our behavior, are changed by the necessity to maintain apparent logic in what we say. We do not want to be considered "irrational." Therefore, there is an ever-growing, though still quite minimal, change in our beliefs and therefore behavior, toward consistency. Unfortunately, of course, we still, instead, often remove the awareness of contradiction by "removing" those situations or individuals that bring it to our attention.

In the next chapter, I will be dealing with a particular subset of beliefs, namely, ethical beliefs, that produce, I will maintain, our most important motivational states. We must, above all, want to do the right thing, and therefore it will be important to have accurate belief about what the right thing is.

The bottom line, then, is that "belief" is the most crucial determinant of behavior. Beliefs can bring about motivational states, change their intensity, and channel them into decisions and therefore into behavior. How optimal one's behavior will be will have to do primarily with one's beliefs. If we see non-optimal behavior, implying a mistake, we are likely to be able to find inaccurate belief. The most important concern that we should have is that our beliefs be accurate, in order to avoid making mistakes, some of which can be devastating. Belief is modifiable by experience. Coming to new conclusions, developing more accurate beliefs on the basis of our experience, happens all the time. What we need is a set of methods to optimize this process. There can be no more important effort, since all of our behavior, good and bad, is primarily dependent upon our management of our beliefs, that is, optimizing their accuracy.

Now I wish to refer back to what I covered earlier in this book, namely, that there has developed, especially recently, the idea that what a person believes should be left to personal taste, since no particular belief can really be proven. "What is true for you may not be true for me, so let us just agree to disagree, unless your beliefs make my life bad, in which case let's settle it ultimately according to who is strongest or most powerful (emotionally, physically, politically, or militarily)." I am letting the reader know that I have the diametrically opposed opinion to this, and that I believe that our most important task as a species is to make our beliefs as accurate as possible. But before the reader comes to a conclusion about this belief of mine, I recommend reading on to find out what I mean more specifically, especially regarding how we should attempt to achieve such increasing accuracy. But I will say that the changes in our way of life that need to be made are enormous, and that we, as a species, are just at the very beginning of making those changes. Our species, in certain ways, is just a toddler, with the future possibility of a wonderful adulthood. And that is what this book is about.