Behavioral%20Graph%20Coloring%20Tue%20Jan%2024%20

1
Behavioral Graph ColoringTue Jan 24 Wed Jan
256-8 PM207 Moore

• Networked Life
• CSE 112
• Spring 2006
• Prof. Michael Kearns

2
Important notes The experiments will take
place in the workstation lab of 207 Moore.
Please find out how to get there well
in advance. We will begin seating for the
experiments at 6 PM SHARP. At 605, we will
stop giving credit for participating, so please
don't be late. YOU MUST BE PRESENT FOR THE
ENTIRE SESSION TO RECEIVE CREDIT OR
COMPENSATION. The 8 PM end time is
only approximate. If you have commitments
afterwards, you should not participate. YOU
MUST BE PRESENT IN TODAY'S LECTURE in order to
participate in either night's experiments.
3
Below are the lists, according to my records, of
who will NOT be able to be present either tonight
(Tue) or tomorrow. These people will make
alternate arrangements with me later in the term
to make up the assignment. If your name does NOT
appear on a list, it means I am expecting you to
appear for that night's experiments. Tuesday Jan
24 CARGO, B. GREENBERG, D. GUILLAUME,
A. JOHNSON, K. LI, D. NA, D. PHILLIPS,
N. Wednesday Jan 25 CARGO, B. GOLDSTEIN,
I. GREEN, A. PELHAN, S. SHETH, A. SOMANI,
A. STERENSIS, L. O'HARA, R.
4
Overview

• Experiments in human collective problem-solving
  on networks
• Directly related to almost all class themes
• Well discuss the results in due course
• Our immediate agenda
• Review of Background and Information for Subjects
• Please ask questions!
• Compensation Details
• Additional Logistics

5
Compensation Details

• All participants will be compensated in
  accordance with performance
• Performance will be measured differently on the
  two nights
• TUESDAY JANUARY 24 COLLECTIVE COMPENSATION
• For every graph that is solved (properly
  colored), ALL participants will receive 5
• No compensation for anything less than full
  solution
• One conflicting edge ? no compensation for entire
  group
• Minimum compensation 0 Maximum compensation
  195 95
• WEDNESDAY JANUARY 25 INDIVIDUAL COMPENSATION
• Compensation will depend on the individual
  instead of the group
• If at the conclusion of an experiment, you have
  no color conflicts with your neighbors, you
  receive 5
• If you have even one color conflict with a
  neighbor, you receive 0
• Minimum compensation 0 Maximum compensation
  195 95
• No compensation for experiments with system
  problems or protocol violations, etc.
• Compensation will be distributed later, details
  TBD

6
Additional Logistics

• System Issues
• PLEASE FOLLOW ALL INSTRUCTIONS!
• do NOT randomly click around in between
  experiments
• when in doubt, do NOTHING and raise your hand
• Layout problems
• remember that bold edges indicate your conflicts
• remember you can toggle colors to see which
  generate fewer conflicts
• Mapping People to Workstations
• After everyone is seated, we will come around and
  note at which workstation you are sitting
• This record is for compensation purposes, and
  will be destroyed later
• Overbooking Protocol
• An orderly line of arriving participants will be
  formed
• Up to a limit, parties will be seated first come,
  first serve
• We will then take parties from the line who can
  only come that night
• SOME OF YOU MAY BE TURNED AWAY
• ALL who are in line by 605 will receive credit!