From Computational Memory Lab
Revision as of 18:17, 22 October 2012 by Anonymous (Talk) (Pcrutchl moved page Cmlwiki/InternalWiki/Magellan to Magellan2)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Chronological Lab Notes: Magellan model of spatial navigation

Reverse Chronological Order From May 20, 2011

May 20, 2011, JRM

NL is developing a geometric solution to SeeClearly. He will present at the first Magellan meeting today and add notes to the Wiki.

At our meeting yesterday, NL raised two important issues:

  1. SeeClearly assumes that colors are represented as RGB values ranging from 1 - 255. However, Matlab actually represents RGB values on a scale from 0 - 1. This may reflect a change in Matlab over the past few years. We should fix the code accordingly.
  2. Memorizing routes. NL observed that there are (5*4/2) = 10 unique pairs of stores in each town, and that this means that 5 delivery paths per environment could simply reflect memorized routes between each pair of stores. We talked about potentially testing the hypothesis that participants memorize routes by comparing the shapes of the delivery paths between repeated pairs of stores.

TL and JRM are adding a discussion of egocentric vs. allocentric spatial encoding to the Magellan manuscript. We agreed to read the following papers:

  1. Burgess, 2006 (TICS)
  2. Kelly et al, 2007 (JEP)
  3. Avraamides & Kelly, 2010 (JEP)
  4. Chen et al, 2011 (Neuropsychologia)
  5. Mou et al, 2006 (JEP)

We were also going to check out another paper that talks about a similar navigation experiment:

  1. Etchamendy & Bohbot, 2007 (Hippocampus)

Our plan for the summer is to:

  • Submit updated manuscript to Psych Review by the end of May
  • Add new modes to experiment (implemented by NL):
    • Replay mode (read from log file)
    • SeeClearly mode (read from log file, replace textures with solid colors, count pixels of each color)
  • Potentially add a distance estimation task in addition to the existing pointing task (NL or TL)
  • Run many new participants (80?) -- this will primarily be facilitated by a new research assistant, yet to join the Magellan team
  • Make good progress on TL's senior thesis project