Friday, April 06, 2007

Mahzarin Banaji: Bugs of the Mind

Today, we start Session 7 of Beyond Belief 2006. Our leadoff speaker is Mahzarin Banaji, Richard Clarke Cabot Professor of Social Ethics at Harvard and Carol K. Pforzheimer Professor at the Radcliffe Institute for Advanced Study.

She is here to talk about ways in which the brain misperceives reality.

Some of these examples are visual, and I invite any reader to watch her presentation. The conclusion she listen to her draws from these examples is clear enough – it is extremely easy, and in fact even natural, to draw mistaken conclusions from a given set of evidence.

Banaji shows a shape in two different context, side by side. Our brains interpret these as two different shapes. Yet, a simple demonstration shows that they are in fact the same shape. Our brains have misinterpreted the context in which the two shapes occur to give us a misperception of their relative sameness, perceiving a difference where none exists in fact.

How does this relate to anything important?

Banaji mentioned that studies routinely show that when people are asked if the gender of their boss is important, responders will report that it is not. However, when presented with a series of questions that are designed to root out biases, people are willing to give up $3000 in salary to have a male boss. (Or, in other words, a company can lower its payroll costs by having male managers.)

Another set of studies asks people to choose a partner in an intelligence-based game. They are provided with a set of statistics about various partners they can choose from – years of college, profession, and the like. They are also given a picture of their partner. The research shows that people will give up 9 points of IQ to have a slender partner. This 9 IQ points represents 41% of the population.

She mentions other examples in which “standardized patients” are sent to 100 doctors. These standardized patients are trained to give doctors an identical description of symptoms. By an 8 to 1 margin, males are more likely to be sent on to specialists than females, and by a 24 to 1 margin are more likely to be sent to surgery.

She provides a demonstration in which she proves a gender bias in the audience of participants at the Beyond Belief 2006 conference among people who substantially tend to be atheist and liberal.

When I discussed Patricia Churchland’s presentation, she gave a similar demonstration, showing that two objects that are, in fact, the same color are perceived as being of different color, because the brain assumes that it is looking at a three-dimensional world with a source of light, when it is in fact looking at a two dimensional drawing. Churchland uses this in a discussion of our ability to draw inferences of ‘ought’ from the perceptual world, without stopping to consider the possibility that her example demonstrates, that these perceptions of ‘ought’ could easily be misperceptions of ‘ought’.

In a great many cases, I would argue, people often see an illusion of ‘ought’ where none exists and are making poor judgments about who shall remain free or go to prison, who shall live and who shall die, based on these illusions. Also, just as atheists are not immune to the illusion of the shapes appearing different when they are they same, or the gender bias that Banaji demonstrated in her audience, they are not immune from perceiving illusionary oughts (or values) and using them to ‘justify’ real-world harms.

Having tools for the mind sciences to show that there are these bugs in our mind that make us do and feel and think things that are not true, the onus is on us to design and develop techniques of shoving their faces in front of it so that they can see, and that is in a sense the simple hope of this research program

It is extremely important to know how the mind generates illusions so that we do not act on the false impressions they generate, prudentially in ways where we harm ourselves, and morally in ways where we harm others.

I have spoken somewhat tentatively in the past, and will likely speak more forcefully in the future, of the need for rationalists to set up sort of curriculum for children (aged 12 and above) to teach them the many ways in which it is possible to make mistakes and to derive truth.

During “Wish Week” last year, my first wish was for “Logic Circles”. These would be informal clubs where members would meet to improve their knowledge and understanding of areas of reasoning. At the time, I suggested:

(1) The Informal Fallacies

(2) The Scientific Method in Everyday Life

(3) Techniques of Neutralization (Rationalization)

(4) Formal Propositional Logic.

In light of Banaji’s presentation, I would like to include:

(5) Bugs in the Mind (unconscious bias).

Because the more people know and understand these bugs, the less likely they are to give them authority in their decision making. Once a person sees the demonstration whereby the two shapes are shown to be identical, they still look different, but the perceiver is now far less likely to conclude that they are, in fact different.

It is unreasonable to expect that the public school system in a vast majority of the country would ever adopt a ‘rationalist’ curriculum. Parents simply will not allow it.

To show this, all one needs to realize is that “The Scientific Method” module would have to include a less on in why ‘intelligent design’ is not science. Anybody who thinks that intelligent design is science does not understand science. Yet, I assume (though I could be wrong) that only a very small fraction of the school districts would allow a teacher to teach and test their student’s understanding of why intelligent design is not science.

Schools may well adopt some of these ideas. Yet, the few examples that I have run across – examples that I mentioned in have been diluted to such an extent that they seem useless. Logic classes do not use actual real-world arguments. Rather, they use examples such as:

(1) All dolphins eat peanut butter

(2) Sam is a dolphin

(3) Therefore, Sam eats peanut butter.

These are examples that pretty much ‘teach’ students that logic is concerned with entirely stupid things that no intelligent person would ever want to bother with. However, as soon as a teacher starts using real-world, relevant examples, some kid is going to go home and accuse their parent of using an “argument from ignorance” or “affirming the consequent,” that that teacher’s career will be over.

So, if one wants children to learn a rationalist curriculum, this curriculum will have to be developed and taught outside of the public school system. It will have to be an independent class – an independent project – with meetings in the evening or on the weekend, where children are provided with this type of knowledge without the schools getting involved.

Then, maybe, in some future generation, there will be enough rationalists to get the curriculum adopted in more and more schools.

I do not know what resources are already available for teaching these things to 7th through 12th grade children in a way that shows real-world relevance. So, I would like to ask you, the reader, to comment and identify any resources that you are aware of.

I would like to take a look at them.

I consider it important that we do so.