TB7 D&L Lecture 1; Connectionist (Triangle) models

?
  • Created by: mint75
  • Created on: 19-12-15 13:19

TB7 D&L Lecture 1; Connectionist (Triangle) models

Advantages

  • A 'BIOLOGICALLY PLAUSIBLE' structure. Each element in the model acts similar to a neuron with decisions made via intra-model patterns of activation
  • Rules are learnt via TRAINING algorithms, not pre-specified.
  • *This process involves creating MAPPINGS between orthography and phonology, and orthography and phonology via semantics
  • *'Words' represented as patterns of activation, with the relative importance of each pathway changing at diff levels of training
  • Accounts for similar phenomena as the DRC after sufficient training
  • Differing from the DRC in that a DEVELOPMENTAL perspectve is posed. Learning algorithms simulate a childs development into a skilled reader.

Disadvantages

  • OVERSIMPLIFICATION of a biological neural network. Learning 'rules' do not capture the messiness and complexity of 'real life' learning

Evaluation

Although creating a more biologically sound model which does have applications in real-world data, the DRC's simplicity does overshadow the learning algorithm explanation. 

Comments

No comments have yet been made

Similar Psychology resources:

See all Psychology resources »See all TB7 D&L resources »