Delivering Aid in Kenya

On Thursday, December 1, Mike Veatch, professor of mathematics and chair of the mathematics and computer science department, will be giving a talk on campus (KOSC 125 at 4:45 p.m.) about the “Spatial Distribution of  Aid Recipients in Kenya.”

In it he’ll explore how GiveDirect, a non-profit organization, recruits and delivers aid to residents of Kenya. The aid delivered is cash funds that are directly transferred to the recipients. The organization is able to operate with low overhead costs and has small transaction costs after recipients are identified. But recruiting recipients produces additional costs. The project studies where GiveDirect should deliver aid, and a model that was developed to minimize the recruitment costs while achieving the desired diversity and total number of locations visited will be discussed. (Former students Matt Forsstrom and Hang Yang worked with Veatch on the project.)

Voting by the Numbers

Yes, the presidential election season is upon us again, and that’s good news for assistant mathematics professor Karl-Dieter Crisman as he continues exploring several theories related to voting. In early January, he’ll be presenting a talk entitled, “Symmetry in Voting Theory: The Borda-Kemeny Spectrum and Beyond” in Boston at a special session of the American Mathematical Society joint meeting with the Mathematical Association of America, the largest annual mathematics meeting in the world. Below is an introduction of his talk (intended for nonspecialists):

“Why Math and Voting? One of the myriad applications of math is in aggregation of preferences. Economists, psychologists, and political scientists all need our tools to talk about this. In my city, we just elected new councilors-at-large, and of course, the primaries and caucuses for the next presidential race are almost upon us. So it’s always a hot topic for the public as well as undergrads. Plus, relatively elementary math yields big results.

Most American elections use one rule to aggregate our preferences – the plurality vote. If you get more first-place votes than anyone else, you win! But there are many other ways to aggregate voter preferences. This talk will discuss a surprising relation between two relatively prominent ones in voting theory circles.

(Arrow’s Theorem) There is no voting rule that satisfies all typical ideas of ‘fairness.’ However, much of the (non-mathematical) debate in voting theory revolves around whether certain axioms of fairness are really red herrings. So my talk focuses on two methods. In both, each voter lists all n candidates in her preferred order: 1.) The Borda Count: Points are assigned based on each ranking – say 0 for last, 1 for second-last, up to n − 1 for first. The candidate with the most total points wins. 2.) The Kemeny Rule: For each possible (full ranking) outcome, check for how many pairwise votes it disagrees with each voter’s ranking, and add up these disagreements. The (highest ranked candidate in the) ranking with the fewest points wins.”

“The Last of the Rowanberrys” or The Distinctions of Gordon College

When Provost Mark Sargent started preparing his fall presentation about Gordon’s distinctives, he wondered if he would be able to get the marketing jargon and agenda from the past decade out of his head.  He told the faculty and staff who gathered for the special faculty forum, that what “saved” him was reading Wendell Berry. By exploring five key areas of distinctions, Dr. Sargent described Gordon’s unique qualities, using his recent encounter with Berry’s story, “Are You All Right?” to help frame his perspective. The text of his speech is printed in full below:

The Last of the Rowanberrys, or The Distinctions of Gordon College  

By Mark Sargent

October 10, 2011

Not long ago the members of the Cabinet were in the President’s Office preparing for a conference call with the Board of Trustees, and we were scurrying around to insure that we had all the right papers and notes ready for the event, when we suddenly ventured onto one of the thornier philosophical topics we had faced in some time. Was it possible to taste the difference between Coke Zero, Diet Coke, and Diet Pepsi? While we are at it, why not add Pepsi One and some generic store brand to the mix. President Lindsay, like most of the others, seemed certain that he could identify the difference by taste.

So I asked Jerry Logan, our academic programs coordinator, if he would conduct a taste test this past week among the visitors to our office. We found what many studies have confirmed: That virtually everyone had a very strong preference among the five choices. Before they took a sip, they generally knew what drink they would buy if they saw these choices in the store. And yet only 13 percent were successful in picking that flavor simply by taste. You would have better odds if you rolled dice.

What this implies, of course, is a triumph of marketing. The actual distinctions between colas are relatively slight: this one has a few more milligrams of corn syrup or artificial sweetener than that one. But most all of us who have spent any time watching television have absorbed certain images associated with each drink. We may not fully realize how effectively Coca Cola has targeted Coke Zero at men and Diet Coke at women, feeding the absurd notion that a sweetener that comprises less than 1 percent of the volume defines the boundary line between genders and gender roles.

I tell you this to confess why I shuddered some when I was first asked to describe the distinctions of Gordon College. For the last ten years, when I have been engaged in earnest discussions about distinctions at Gordon, it has usually been part of a quest for that Holy Grail of sales: a market differential. In one respect, the term “distinction” should refer less to something precise than to something quite generic: we refer sometimes to a person of distinction as one with integrity, wisdom, perhaps even a gray-haired eminence.

To be fair, I am not fully discounting the need to find themes in our marketing that enable us to recruit students and to define our product. But I will confess to having listened to too many consultants and advisors on branding: Too many people who have pressed us to define our “brand” or our “brand promise” and then be sure that we conform our academic and co-curricular programs to match in order to secure our customer. Or too many advisors have told us to define reality by our marketing jargon. When I suggested to one marketer that MIT might quarrel with a tagline proclaiming Gordon now to be “New England’s best science college,” I was told I needed “aspiration.”

So I started thinking about this presentation about Gordon’s distinctions fearing that I would not get the marketing jargon and agenda out of my head.  What saved me, you might say, was reading Wendell Berry. Continue reading

The Color of Sound?

Are a person’s eyes the only way to see color? That’s one of many sensory related questions assistant professor of philosophy Brian Glenney has been exploring for the past several years in his scholarship and research, using color sonification sensory substitution devices (SSD) to test his theories. Next month at a conference in Scotland on sensory worlds, Dr. Glenney will present some of his findings. His abstract below describes the highlights of  his talk:

Hearing Real Colors with Adaptive Technologies

By Brian Glenney

“African Grey Parrots (Psittacus erithacus) have a tetrachromat visual system with a fourth pigment sensitive to ultraviolet wavelengths (Bennett and Cuthill 1994). So, trichromat humans cannot be expected to share the same color categories as their feathered friends. For one, parrots perceive more colors.  Two, this fourth band of wavelengths influences how all colors are perceived by parrots (Shepard 1997). For instance, it influences their categorization of colors at the opposite end of the spectrum, of red and orange colors, like the color of a ripe grapefruit (Pepperberg 1999). So, physiological differences suggest that color perception is better for parrots than humans and if not better, at least different. It is therefore prima facie unreasonable to think that the parrot and the human both see “real” colors, or colors that represent real color properties of objects.

The physiological details need not derail color realism across species as it may be argued that ecological differences—differences in how species respond to their environment with respect to different action potentials evolved out of their distinctive ecological niche—primarily influence a species’ color categories (Matthen 2005). For instance, the color of a ripe grapefruit for a parrot might be the color of an unripe grapefruit for a human. So, the real color of a ripe grapefruit might be a point of irresolvable contention in a debate with parrot, for both physiological and ecological reasons. That is, unless real color is just a label for what’s ecologically relevant to a species, a view advocated by Mohan Matthen’s (2005) called “Pluralistic Realism.”

The central question I would like to raise in this paper is whether the perception of real color might include stimuli from any sensory system whose input was within the visible and near-visible spectrum.  For instance, might color content be processed by an auditory system with color receptors? For instance, might subjects wearing a color sonification sensory substitution device (SSD) that processes color content using the auditory system hear real color much like a parrot or a human sees real color? To argue “yes” is to advocate “Radical Pluralistic Realism.”  What makes it radical is that, unlike Matthen’s view, color content can be processed by a visual system that did not evolve out of processes relevant to an ecological niche.”

Oil and Coral

Assistant professor of biology Walter Cho recently returned from St. Petersburg, FL, where he was a participant in the  Deepwater Horizon Oil Spill Principal Investigator One Year Update Workshop, sponsored by the National Science and Technology Council’s Sub-Committee on Ocean Science and Technology. Along with his colleague Timothy Shank from the Woods Hole Oceanographic Institute, Walter presented his research on the effects of the oil spill on coral communities in the Gulf of Mexico. The abstract of the presentation is below:

“Deep-water coral communities are thought to be vulnerable to disturbance due to their low rates of colonization, growth, and the high levels of host-specificity for associated invertebrates. A major concern resulting from the Deepwater Horizon oil spill is the vulnerability of these deep-water coral communities to the oil spill. Research cruises in 2008 and 2009 established a comparative baseline for changes in benthic community structure. Research cruises in 2010 and 2011 returned to some of the sites visited prior to the Deepwater Horizon oil spill and also explored new areas that may host deep-water coral communities. We assessed the potential impact of the Deepwater Horizon oil spill on coral-associated invertebrates (including ophiuroids, crabs, shrimp, barnacles) and the level of genetic connectivity between populations in the Gulf of Mexico, the Caribbean, and the North Atlantic.”