How to Model a Neuron in Our Visual Cortex, A Layman’s Understanding (Part I)

Close your eyes and listen to the brain.

What does it sound like?

Scientists did just that — they used amplifiers to listen to how neurons in our brain sounded, so they could discover how our eyes form an image of the outside world.

Where would you start if you wanted to model this process?

One thing that is generally helpful is to first identify the different components of a system. We have:
(1) the image in front of us
(2) some representation of the neuron
(3) an output of that image we see

How should we put those pieces together?

But what is wrong with this? That it’s probably more like:

But the brain is made up of many neurons, how would we model just one?

Now what if we do this for every neuron? Then we would have a whole picture.

The next question that arises is: How does each neuron know which slice of the picture to look at?

Well, what are the different components of a picture? A picture is fundamentally made up of:
(1) colors
(2) edges

Edges seem like they would give us more information about a picture, so let’s go down that route. What is an edge and how can we break that concept down? A major component of edges is the orientation of an edge.

So maybe each neuron has a preferred orientation it favors. And all of these neurons work together to create one picture.

So we can take lots of bars with different orientations, run them through a gabor filter (a fancy term that refers to a function that is similar to those in our visual system), and see when the gabor filter produces the most excited response.

If we run these through a gabor filter, we end up with a curve looking like this.

Thus, this neuron’s preferred orientation is ~40 degrees. If we do this for every neuron, we’ll have a starting model of how our brain sees edges.

You just learned a few fundamentals about how your retina sends signals to your brain, creating a picture for the world in front of you.

To come: Code as to how you would model this.

The Cost of Making Your Work Public

The other day my co-worker Bret Victor was speaking of the requests he gets to release the source code for his prototypes on the future of creation tools.

While he’s been getting requests for years, Bret said that releasing his code to the public could potentially stifle innovation.

Why?

The release creates a standard for what these new products should look like.

When an inventor creates a tool and makes it available to the world, the public has a tendency to simply accept the tool as is, use it, and stop thinking of new ways to do things, thus halting development of the product).

Example #1: the Pie Chart

Pie charts are pervasive…

image

image

… And a poor way to represent information. (See Edward Tufte, The Worst Chart in the World, and Oracle’s Reasons Not to Use a Pie Chart)

A simple example illustrates that pie charts make it difficult to make comparisons between two quantities. See:

image

What if we represented the same information like this, instead? This illustration enables us to make direct comparisons between quantities.

image

Because pie charts cannot spatially fit all information, they also are pleasantly accompanied by a key (right-hand side), which attempts to illustrate what all the components of the pie chart designate.

image

Could we design another way to represent this information — one that doesn’t require you darting your eyes back and forth to understand the information?

What about this?

image

Okay. Maybe not as aesthetically pleasing. Perhaps a reason why people use pie charts is because all the circles look pretty. But I would argue that we should not sell ourselves short — can we not have a representation that is both intuitively functional, and aesthetically pleasing?

I re-designed a representation of the same information; illustrated below.

image

image

So what?

These are just pie charts. Okay, it’ll take a millisecond longer to process the percentages. Why does it matter?

I think we vastly underestimate how good the quality of an experience (in this case, understanding and exploring the world) could be, because we already have models in our heads of what the medium is currently like.


One step further: Our tools for everyday statistics

Does the same tragedy that happened with pie charts also extend to other branches?

Let us look at the field of statistics.

The traditional probability problem presents you with several numbers on the percentage of women who get a positive or negative result when they get a mammogram.

image

You are typically asked:

Assuming you have a positive test, what is the chance you have cancer?

When calculated out, the number seems much lower than expected.

Some of us simply accept that our brains do not intuitively grasp probabilities, but I’d argue that statistics is only unintuitive because we don’t have proper representations.

How would you design another representation of Bayes’ Theorem?

More to come.

Analogizing Information to Nutrition

paleo

Few would disagree that the foods they eat affect how they feel. I feel much better eating vegetables and rice for dinner than I do a cheese pizza.

I take this analogy one step further by being selective about the information I expose myself to.

Media geared for average consumption : food geared for average assumption.

Consuming popular television, movies, magazines is akin to consuming McDonalds, potato chips, and deep-fried Twinkies.

Delicious in the short run (reading about gossip and exploits, even love songs on the radio are scary) — deadly in the long run (how do they impact your values and what you think about?).

Think of it as a mental paleo diet. Tasty.