Hampshire College

SPRING 2008 CS WEDNESDAY TALKS

February 6th 
Neil Stillings, dean of the School of Cognitive Science and professor of psychology
The Cognitive Modularity Hypothesis: What It Is and Is Not
Jerry Fodor's 1983 book The Modularity of Mind popularized the use of the term module to refer to hypothesized functional, computational subsystems of the mind. Since then, the theoretical claims being made about cognitive modules have changed a great deal, and debates about the existence and character of cognitive modules have shifted accordingly. In this talk I take no position on whether some version of the cognitive modularity hypothesis is correct. Instead, I try to characterize what the central claims are in several versions of the hypothesis, focusing mainly on the version being pursued by evolutionary psychologists and anthropologists (e.g. Cosmides, Tooby, Pinker, Sperber, Hirschfeld, Boyer, etc.), and trying to clarify what they are and are not claiming.

February 13th
Jay Trudeau, visiting assistant professor of psychology
Meaning and Made-Up Words: Demonstrating Semantic Effects in Reading With a Trained Pseudo-Word Lexicon
Does what a word means really make a difference in how easily we are able to read it? Some words are naturally easier to read than others, but what determines that ease is much debated. Some proponents of a classical Dual-Route model of the reading process claim that how much meaning a word has should make no difference. Advocates of a single-route Interactive Triangle model claim that semantics should play a significant role in decoding the written word. Experiments in this arena are frequently complicated by confounds and overlaps in spelling and sound in the English language (bare/bear, read/read). Fabricating new pseudo-words constructed to follow the rules of English but avoid common confounds may offer some insight into the value of meaning in reading and the structure of the underlying cognitive system.

February 20th
Lee Spector, professor of computer science
Evolution, the Programmer and Mathematician
It is now possible to breed computer programs instead of writing them in the traditional way. Genetic programming systems allow one to combine ideas from evolutionary biology with ideas from computer science to solve difficult and important problems in disciplines ranging from jazz composition to quantum computation. In this brief and mostly-non-technical presentation I will show how genetic programming works, and I will present some of the results that it has produced. I will focus on new results, achieved by a team that includes Hampshire students Ian Lindsay and Brad Barr, mathematician David M. Clark of SUNY New Paltz, and myself, in which genetic programming has solved long-standing mathematical problems (in the study of finite algebras) in ways that improve on all prior methods by several orders of magnitude.

March 5th
Lisa Scott, assistant professor of psychology at the University of Massachusetts
Perceptual Specialization in Infancy: Building Biases in the Brain
Infants refine and narrow their ability to discriminate between things they see and hear during the first year of life, revealing what appears to be a decline in ability at a time when most other skills and functions are dramatically increasing. For example, in one study, 6-month-old infants were able to differentiate two human faces as easily as two monkey faces, but 9-month-olds could only differentiate between two human faces. In this presentation I will describe these data in more detail and report the results of several studies, utilizing both behavioral and electrophysiological techniques, to investigate the specific behavioral and neural mechanisms mediating the development of visual perceptual narrowing.

March 12th 
Kathryn Flack, visiting assistant professor of linguistics
Avoiding Ambiguity in Japanese and ASL
Ambiguity arises fairly often in most languages. In English, for example, we frequently say and hear structurally ambiguous sentences like Ali showed Polly her room or Marc and Ed’s mother went running. Most sentences are unambiguous, though; in some languages, in fact, grammatical processes seem to occur specifically in order to prevent sentences from being ambiguous. This talk will discuss two very different grammatical phenomena – one from the syntax of Japanese and the other from the morphology of American Sign Language – which both occur only when sentences would otherwise be ambiguous.

April 2nd 
Mark Feinstein, professor of linguistics, and Steve Johnson, the University of Massachusetts
Syntax of Robin Songs
Despite the fact that almost everyone can identify an American Robin, even those who study bird song know very little about the song behavior of this species. Steve's dissertation work at the University of Massachusetts initially focused on song learning and development in robins, but soon expanded as surprisingly complex patterns were discovered in the "syntactic" combinations of song elements sung by wild robins. The investigation of song element sequence patterns and cultural transmission in robins has continued here at Hampshire College (in collaboration with Mark Feinstein and a group of students) since 2006. We've found several intriguing and surprising aspects to robin vocal behavior, and have learned a great deal about how they learn and use song.

April 9th 
Salman Hameed, assistant professor of integrated science and humanities
Darwin vs. Gods from Outer Space
The science and religion debate takes a fascinating turn when we look at UFO religions – religions explicitly based on the notion that alien civilizations have been responsible for the creation of human beings and other creatures on Earth. While the idea of extraterrestrials has been around for centuries, modern aliens really took shape in the 20th century. Speculations of astronomers about life on other worlds, the discovery of a vast universe, and developments in modern biology, paved the way for imagining worlds with beings more intelligent and technologically more advanced than humans. While many UFO religions are explicitly pro-science, their origin stories also clash with ideas of evolution via natural selection. For example, Raelians believe that humans were created in a test tube by an alien race on another planet, a few millions years ago. Thus, most UFO religions face a dissonance of having a faith based on advanced science and at the same time having to reject some of science’s fundamental ideas.I will explore this tension with evolution over UFO origin stories and how it shapes their attitude towards modern science.

April 16th
Chris Perry, assistant professor of media arts and sciences
An Argument for Bug Art
For as long as artists have worked with computer-based tools, artists have had to contend with bugs. Typically these bugs are considered problems: when a piece of software doesn’t work as expected, there is a general rush to fix it. In this talk I will introduce Bug Art, a movement suggesting that these so-called broken pieces of code should be explored as novel and valuable art-making technologies before being hastily fixed and lost forever.

April 23rd 
Jane Couperus, assistant professor of developmental cognitive neuroscience
Ignoring Distraction: Is There Active Suppression in the Brain?
As we travel through our daily lives we are constantly attending to some things while ignoring others. For example, we pick out our friend in a crowd, not noticing much about the people around us. How do we select out just our friend? How do we select just the relevant information, do we facilitate processing of the important information or do we suppress processing of distractors? Do we do both? In this talk I will discuss research using electrophysiological methods that examines brain activity to distracting information to determine if distractor suppression is a passive or active process. The results of this research may help inform how we think about selective attention and potentially how we can help those who have trouble ignoring irrelevant distractors.

April 30th
M.J. Wraga, associate professor of psychology at Smith College
Effects of Video Game Playing on Low- and High-Level Cognitive Tasks
In the past few decades, video game play has become a common diversion for many people of different ages and backgrounds. As a consequence, researchers have begun to examine the potential negative and positive effects of video game playing on gamers. In this study, we focused on the potential benefits of video game playing by testing non-video game players (NVGPs) and video game players (VGPs) on three tasks that utilize both low and high-level cognition. For our low-level cognitive task, we chose a flanker task that measures the flanker compatibility effect (FCE), a paradigm used in attention studies to measure attention capacity. Our two high-level tasks were a mental self- rotation task, which measured spatial reasoning, and the Tower of Hanoi (TOH) task, a problem-solving task. The results revealed trends that supported previous research for some tasks, but not others. Although we found a trend that VGPs performed faster than NVGPs on the flanker task, VGPs did not have a significantly larger FCE on high-load trials than NVGPs either for RTs or errors. VGPs also did not outperform NVGPs on the viewer rotation task. On the TOH, however, VGPs made an average of 22 fewer excess moves than NVGPs, showing a trend in favor of the VGPs. Our findings on the TOH task are promising in that they suggest a possible influence of video game playing on high-level cognition. Other cognitive advantages for VGPs, however, may be weaker than previously has been thought.