COGS 102

COGS 102 - COGS 102A DISCUSSION Sec$on: Week 1 ...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: COGS 102A DISCUSSION Sec$on: Week 1 Monday, September 26th Who is this Luke kid? •  I’m a 2nd year graduate student. •  I work with Ayse Saygin on ac$on percep$on and the brain. •  Like most of you I’m new to DCog! •  Office hours: Wednesday noon to 2 in CSB 233 •  Email: [email protected] Important Links •  Class website: hLp://thiscourse.com/ucsd/cogs102a/fa11/ •  NetLogo: hLp://ccl.northwestern.edu/netlogo/ •  Calibrated Peer Review (CPR): hLp://cpr.molsci.ucla.edu/ –  Remember to establish your user ID before class! •  Plagarism Quiz & Tutorial: hLp://libraries.ucsd.edu/loca$ons/sshl/guides/ course- subject- guides/index.html –  Don’t forget to take it before class tomorrow!!! Course Overview Grading Plagiarism Tutorial 5% 4 wriLen assignments 18% x 4 = 72% Midterm exam (Oct. 27?) 8% Final Exam 15% Tips for class success •  There are readings, do them! •  Think criDcally about what you learn/read. •  Come to every sec$on (ideally with ques$ons/ comments!). Sec$on Overview •  Sec$on is every Monday in WLH 2207. •  Sec$on is op#onal, however it is to your benefit to come! •  Things to do before sec$on: –  Read the reading(s) for Tuesday. –  Look over the “Reading Prep” •  Don’t be afraid to ask ques$ons! •  This sec$on will be a group discussion (this is a DCOG class aker all!) Hutchins (2001). Cogni$on, Distributed •  What dis$nguishes distributed cogni$on (DCOG) from classical approaches to cogni$on?! •  Two theore$cal principles 1.  2.  Unit of analysis is not just in the head, but can encompass the individual and the environment. The range of cogni$ve mechanisms/events is not limited to the “skin or skull of an individual” but can extend to the material environment. •  Three kinds of distribu$on of cogni$on. 1.  Cogni$ve processes may be distributed across social groups. 2.  The opera$on of the cogni$ve system may involve coordina$on between both internal and external structure. 3.  Products of earlier events can transform the nature of later events. Mind in Society “In the same way that epidemiology addresses the distribu$on of pathogens in a popula$on, anthropology should treat ques$ons about the distribu$on of representa$ons in a community.” •  Processes that a distributed among social groups include… –  Memory, beliefs, representa$ons •  Science is a great example of a distributed process! What are some examples that demonstrate this? The Society of Mind “to explain intelligence we need to consider a large system of experts or agencies that can be assembled together in various configura$ons to get things done.” “Both what is in the mind and what the mind is in are socie$es” “Every high- level cogni$ve func$on appears twice: first as an interpsychological process and only later as an intrapsychological process.” •  Interac$ons among component parts can give rise to new levels of organiza$on that can only exist in aggregate. –  Linguis$c structure as arising from the interac$on of other speakers The material environment “Cogni$ve ar$facts are involved in a process of organizing func$onal skills into cogni$ve func$onal systems.” •  Why don’t cogni$ve ar$facts (e.g. wri$ng) merely amplify cogni$ve process (e.g. memory)? •  Examples from boat naviga$on –  “3- minute rule” and “nomograms” –  The cogni$ve work being done by the individual alone is insufficient for the necessary computa$on. •  Manipula$on of the environment can also lead to differences in later cogni$ve performance. Any Ques$ons/par$ng thoughts? Resnick (1994) Learning about Life •  Centralized vs. Decentralized views of life. –  Example: Aggrega$on of slime- mold cells –  Centralized view: the clustering of slime- mold cells are due to “founder” cells (a leader). –  Decentralized view: clustering occurs due to simple rules. –  Other examples: gene$cs, origin of fossils, ex$nc$ons •  What are some examples related to cogni$ve science? Key Concepts •  Self organizaDon: Order arises from group interac$on. •  Emergence: complex structures (and behavior) arises from simple parts. –  E.g. line- following LEGO turtle, termite program •  Decentralized local interacDons: The simple pieces act with those around them, not the whole. •  Mutual causaDon: Causa$on at different levels goes both ways. •  Levels of organizaDon: Mul$ple levels of interac$on. Five guidelines •  PosiDve feedback isn’t always negaDve –  Exponen$ally posi$ve inputs can some$mes lead to self- organiza$on. •  Randomness can help create order –  Random fluctua$ons act as seeds from which paLerns and structures grow. •  A flock isn’t a big bird –  The behavior of one level (e.g. the birds) can be completely different than the behavior of another level (e.g. the flock). •  A traffic jam isn’t just a collecDon of cars –  Large collec$ons of individuals can be viewed as “emergent objects” (irrespec$ve of the individuals that compose it) •  The hills are alive –  The environment is something to be interacted with, not just acted on! •  What about some NetLogo examples? *Hint: This will help you conceptualize your first CPR assignment! Slime- mold Cells NetLogo Link Rules: 1.  Follow the chemical gradient of other turtles (slime- mold cells). What about the guidelines? Posi$ve Feedback? Randomness? Levels? Changing composi$on? Environmental interac$on? Termites NetLogo Link Rules 1.  2. If the termite bumps into a woodchip, it picks it up. When it bumps into another woodchip, it puts it down in an empty space. What about the guidelines? Posi$ve Feedback? Randomness? Levels? Changing composi$on? Environmental interac$on? Flocks of Birds NetLogo Link Rules: 1.  2.  3.  Align with the heading of other close birds. Separate from birds that become too close Move towards nearby birds (unless they are too close). What about the guidelines? Posi$ve Feedback? Randomness? Levels? Changing composi$on? Environmental interac$on? Traffic Jams NetLogo Link Rules 1.  2.  A car accelerates when it doesn’t see a car ahead. A car decelerates when it does see a car ahead. What about the guidelines? Posi$ve Feedback? Randomness? Levels? Changing composi$on? Environmental interac$on? Any ques$ons?!?!!? ...
View Full Document

This document was uploaded on 10/16/2011.

Ask a homework question - tutors are online