Vocabulary Networks: Program-6

  Use these boxes to set the basic parameters for your model

  NTWK sets up the network for this simulation. Choose a number between 0000 and 9999 for this parameter.
  

  ORWds determines the number of OR words in the network when it is set up. Choose a value between 0 and 1000.
  

  INIT randomises the current state of the words in this simulation. Choose a number between 0000 and 9999 for this parameter.
  

  nEv sets how many events will occur in your simulation. Choose a number between 0 and 800 for this parameter.
  

  rEv determines which words are affected by your instructions. Choose a number between 0 and 9999 for this parameter.
  

  WdsON adds some random noise to the background. Choose a number between 0 and 200 for this parameter.
  

   




  This simulation works in the same way as Program 5
  but it injects some random noise into your network
  at the start of each update cycle. This means that
  your network is NOT in its attractor state when an
  upgrade event takes place. You can control the
  amount of noise using the WdsON parameter.

  Parameter nEv controls the number of events in the
  simulation. Each EVENT turns ON a single word and
  changes its TYPE to OR.
 
  THINGS TO DO.
  Set the value of nEv to 0. Set the value of ORWds
  to 400, and find a network with a low level
  attractor state. Check how this network responds
  when you vary the value of WdsON.
  Now raise the value of ORWds in small increments. How
  does the behaviour of the network change as a result?
 
  QUESTIONS TO ASK.
  Which parameter affects your network most?
  Can you identify a threshold where the value of ORWds
  always results in a network where most words are ON?
 
  What implications does this set of behaviours have
  for theories of L2 vocabulary acquisition?