The String-to-String Correction Problem
Journal of the ACM (JACM)
Modeling melody recognition using a sequence recognition neural network with meta-level processes
IJCNN'09 Proceedings of the 2009 international joint conference on Neural Networks
| Hi-index | 0.00 |
This research models human performance in Dalla Bella, Peretz, and Aronoff (2003) melody recognition study. They compared performance between musicians and nonmusicians in the perception and recognition of isolated melodies. Inspired by cohort theory, they used a gating task to identify three cognitive events in the melody perception/recognition process. These events were the familiarity emergence point (FEP), the isolation point (IP), and the recognition point (RP). Dalla Bella et al. explain their results using cohort theory. In the present research, we develop a simulation based on a sequence recognition neural network (SRN) that is compatible with cohort theory. We use it to model the hypothesized cognitive processes underlying the occurrence of these events. Two separate networks with different stored memory sizes are used to model musicians and nonmusicians. Our simulations capture the qualitative pattern of results in Dalla Bella et al. and show how stored memory size may affect the melody recognition process. The IP is modeled using the core SRN that contains a winner-take-all (WTA) mechanism. The FEP and RP are modeled by introducing meta-level readout units that monitor the dynamically evolving state of the WTA network. The unit that models the FEP plays a causal role in determining the RP. Therefore, the FEP is not an epiphenomenon in our simulation. It was not previously known whether the core SRN contained enough information to allow simple meta-level readout of dynamic cognitive states. Our simulations show that such readout is in fact possible.