Friday, August 5, 2011

My First Experiment in Jerry Lettvin’s Lab


Dr. Jerome Y. Lettvin, my de facto Ph.D. advisor, passed away on April 23. Many impressive obituaries have been written, but here is a reminiscence…
 
Dr. Jerome Y. Lettvin (Feb. 23, 1920 – Apr. 23, 2011) was a professor of Electrical Engineering and of Biology in the Research Laboratory of Electronics at MIT. He is best known for the 1959 Proc. IRE article, “What the frog’s eye tells the frog’s brain,” which he wrote with H. Maturana and W. Pitts. He is also known for his televised debate with Timothy Leary in 1967, in which he used the uncensored word “bullshit” to describe Leary’s rationale for endorsing drug-induced euphorias. To color science he gave a Scientific American article, and perhaps more significantly, “The colors of colored things” [1], which was formative to all who studied color with him (see the only English title in [2]).   

Obituaries for Jerry abound (e.g., [3], [4]). His sons David and Jonathan have both created Web postings containing information and memorabilia (see [5], [6]). So rather than another obituary, I offer here a story from personal experience.

Having read “The colors of colored things” and heard Jerry’s intriguing (to me spellbinding) lecture, I obtained permission to write my PhD. dissertation under Jerry in absentia from Syracuse University, I reported to Jerry’s lab ready to create profound theories. Jerry had other ideas. He declared that I must first do a few experiments. We started talking about color effects, and he mentioned Abney’s effect. I was eager to show off, so I said the effect was that most monochromatic lights shift toward yellow when mixed with white light. Then the conversation went something like this:

Jerry: No. All lights get yellower when mixed with white light.
Me: Surely not all lights, Jerry. Surely the yellow lights near the spectrum locus don’t get yellower.
Jerry: Yes, they do. In fact, that is the first experiment I want you to do: Show that the yellows get yellower. You can use the materials around the lab.
Me: Surely there’s some sort of trick. Can you give me a hint?
Jerry: Just remember what I said in “The colors of colored things.” Pay attention to spatial boundaries.

Well, I found a 35-mm Wratten 15 filter, several lenses, and two projectors, and thought I would just project a spot of white light onto a diffuse yellow field created by the other projector. But I couldn’t get the effect. I had to make the white spot have a very sharp edge, and to do this I directed a lot of light through a diaphragm aperture at the end of a collimating lens. That made the white spot too bright.

So I had to dim the white spot. I couldn’t do it by decreasing the power to the projector, because that would make the light redder and it wouldn’t be the same white that referenced the other projector. I would simply be adding one yellow to the other, and that wouldn’t be fair. So I needed neutral-density filters---a lot of them.

Jerry suggested I enlist the help of John McCann (then at Polaroid, a short walking distance away). After some cajoling from Jerry, John offered his facilities to me. He had great projectors and as many neutral-density filters as I needed. John was very gracious. What harm could it do?

So I took my lens and diaphragm over to John’s lab, set up the experiment, and started putting neutral-density filters one after the other in the various slots in his projector until the white spot had dimmed a lot. Then I noticed the effect. What Jerry had said was true! On the edge of the white spot, on the white side, a band of yellow appeared, which was much more saturated than the yellow in the dim diffuse field. Evidently the jitter of my eye was causing the edge to induce yellow color into the white field. Even now I am not sure of the exact mechanism, but it does work.

With great excitement I rushed down the hall to summon John McCann to be a witness, so I could put a checkmark “done” in the box that would bring me closer to theory and my Ph.D. dissertation. John came quickly, but not quickly enough. By the time I re-entered the room with the apparatus, smoke was streaming out of the white-light projector. The neutral-density filters were burning!

Fortunately John and I are still friends. And fortunately Jerry took my word for having achieved the effect, perhaps fearful that my re-creating it in his lab would imperil the far more flammable Building 20. Jerry never told me how he himself had achieved the effect.

You can try the yellow-light Abney effect at home. But perhaps you’d better use a calibrated monitor and not my dangerous projectors! 

Michael H. Brill
Datacolor

[1] J. Y. Lettvin, “The colors of colored things,” MIT RLE Quarterly Progress Report No. 85, 15 Oct. 1967, pp. 193-229.