In the previous section I followed Neisser's characterization of different aspects of our knowledge of the self that builds from the foundational level of an organism as an autonomous adaptive agent to a self-concept that guides action. I have emphasized the multiplicity of mental mechanisms that have evolved and serve to control our behavior. Nonetheless, the self-concept we develop characterizes ourselves as unities. But what insures that we are unified? In fact, nothing does since we are often not unified. The mechanisms that comprise us operate on their own and often without any awareness on our part. Wherever we look at a scene, our eyes saccade to different parts of it, typically without any awareness on our part. When we walk, we walk with a distinctive gait and follow a particular path. When we are required to choose items from a display, our choice may be influenced by factors such as relative location of which we have no awareness. These are all actions we perform, but they do not reflect choices made on the basis of our unified self-concept.
Cases such as these are relatively unproblematic unless we are under the mistaken impression that everything we do is determined by a unified central self that is aware of what it is doing. Sometimes we are aware of a divided self. Weakness of will is a ubiquitous phenomenon in which we behave in ways opposed to what we claim to want. We go to a lecture desirous of attending closely to the whole presentation, but find ourselves struggling to stay awake. We set out to eat right and exercise regularly, but find ourselves enjoying ice cream and missing workouts. As disturbing and frustrating as these cases are, we are generally not helpless in the face of them. We can take action to reduce the chance that we will succumb to the choice we, in our reflective moments, don't want to make. We can, for example, always take notes or draw sketches during lectures so as to maintain ourselves in an active state where we are less likely to feel sleepy. We can keep ice cream out of the house and arrange to meet others for workouts. In these and other ways we can shape ourselves to be more the sorts of persons we desire to be.
Creating one's is, in part, an activity in which we are constantly engaged. But it is not something we undertake from a position outside ourselves and our engagement with the world. Rather, we are autonomous adaptive agents situated in the world and behaving in the world. As a result of a host of mental mechanisms, we have an ecological and social point of view of the world, engage in mental time travel, have personal experiences, and construct personal narratives. As humans, these mechanisms enable us to create a concept of one's self, a concept that may be both descriptive and prescriptive insofar as we use it to regulate our lives. Like other mental mechanisms, regulating our lives in accord with a self-concept often involves suppressing the operation of other more basic mental mechanisms. In some cases, the operation of those more basic mechanisms can be recognized even as we work to counter them.
Tricky reasoning problems provide a good example of this. Psychologists such as Peter Wason, Amos Tversky, and Daniel Kahneman have identified numerous ways in which human judgments in reasoning problems are mistaken (see Piattelli-Palmarini, 1994, for a very instructive overview). Most of these problems are ones in which a response that seems highly plausible turns out to be incorrect when proper methods are applied. Yet the temptation to make the response remains compelling. Consider the following problem:
You visit your doctor with an unusual symptom. Your doctor tells you that it might well be due to a rare condition that affects one percent of your age group and requires very painful treatment. There is a test she can do which has been shown to identify seventy-nine percent of those with the condition, but in ten percent of cases produces a false positive. You decide to have the test and it comes back positive. Should you check in for the painful therapy?
Most people think in this situation that they are likely to have the condition. Yet, this is a probabilistic reasoning problem for which a theorem developed by Thomas Bayes provides the normatively correct result, and the result is that we are very unlikely to have the condition. Let p(A/B) be the probability of having the condition given a positive test, p(B/A) be the probability of a positive test given the condition, p(A) be the probability of the condition itself, etc., then:
If you supplied the information stated in the problem above, the probability that you have the disease given the positive test is .0739, a very low probability.
A more intuitive way to arrive at the same result is to construct the following table using the same information:
Has condition Lacks condition
Test is positive Test is negative Total
790 210 1,000
9,900 89,100 99,000
10,690 89,310 100,000
From the table one can recognize the overriding importance of the fact that only one percent of the total population has the condition. A test which correctly picks out seventy-nine percent of the positive cases among those with the condition and misdiagnoses just ten percent of those without the condition yields a situation in which, of those who test positive, very few actually have the condition. You can also see that if you assume even a much lower rate of false positives and a much higher rate of correct detection, the test still offers little guidance as to whether you have the condition.
The proper inference to draw from this example is that it is necessary to pay close attention to the base rate, and if the base rate is extremely low, no test will provide useful information unless it is extraordinarily accurate. Nonetheless, it is extremely hard for us humans when faced with such a problem not to assume that the test will give highly reliable information about our condition, and to be greatly alarmed when it comes back positive. We readily ignore the crucial information about base rates. A theoretical understanding of the importance of base rates, moreover, typically does not generate a better intuitive feel for such problems. We still feel as if the test must be highly informative. The remedy is to recognize problems that rely in such a way on base rate information, put our intuitive judgments aside, and perform the normatively correct calculations. We must then suppress our intuitive judgments and abide by what we have established as the correct response.
This strategy for dealing with tricky reasoning problems is simply an example of how, by developing our explicit knowledge, including knowledge about ourselves and our tendencies to make certain kinds of errors, we are in a position to exert control over our actions. Developing the capacity to reason in normatively correct ways on problems such as these is not easy. Many people resist efforts to be taught such modes of reasoning, preferring to rely on their own instincts. In gambling situations they can easily be taken advantage of in what is referred to as a Dutch Book, a gambling situation in which the odds are such so as to insure that the bettor will lose money. One form of Dutch Book arises as a result of inconsistent preferences—for example, different preferences for the same circumstances described differently. For example, people tend to prefer an action when it is described as possibly saving three hundred out of four hundred who would otherwise die than when it is described as resulting in one hundred out of four hundred dying, but in fact the outcomes are the same. We have the ability to develop ourselves in ways that protect ourselves from such situations, but only through effort.
The strategies for developing into reasoners who maintain consistent preferences, attend to base rates, etc., are just examples of a broader class of strategies for deploying mental mechanisms in ourselves so as to better promote ourselves as autonomous adaptive agents. A major factor in developing such mechanisms to the level that they can best serve our functioning as such agents is our representation of the self. By developing a concept of one's self as striving to be a cohesive rational and emotional agent and employing that to regulate our various mental mechanisms, we are able to act as responsible agents. This capacity, importantly, arises because of our being constituted of various mechanisms which function in our composite organized being to maintain ourselves as autonomous adaptive agents. As autonomous adaptive agents, we are engaged in acting in the world, but as a result of our mental mechanisms, especially those involved in creating a self-concept, we are able to regulate those activities to subserve focused ends.
Was this article helpful?