next up previous contents
Next: 6 Discussion and Future Up: 5 Aftereffect Experiments Previous: 5.5 Time course of

5.6 Conclusion

The RF-LISSOM model was shown to exhibit tilt aftereffects that are very similar to those measured in humans. This result links the detailed low-level behavior of the orientation map from chapter 4 with measurements of human performance at the level of conscious perception. It also demonstrates that the same self-organizing principles driving the development of the map can account for behavior in the adult. The model suggests a new explanation for the indirect effect, namely that limited synaptic resources require balancing of synaptic strengths. Having a large-scale computational model also permits quite detailed analysis that shows exactly how the TAE occurs. For instance, weight changes and the corresponding changes in activity patterns can be visualized directly in the model, and data for the time course of the TAE can be obtained at a level of detail impractical for human subjects. This analysis represents a comprehensive set of predictions of what may be going on in human perception, and thus provides opportunities for later experimental work. Chapter 6 discusses how the RF-LISSOM results relate to biological evidence, and describes specific predictions for human and animal experiments as well as suggesting areas for future research with the model.


next up previous contents
Next: 6 Discussion and Future Up: 5 Aftereffect Experiments Previous: 5.5 Time course of
James A. Bednar
9/19/1997