Nd. Cook et al., THE CEREBRAL HEMISPHERES AND NEURAL-NETWORK SIMULATIONS - DESIGN CONSIDERATIONS, Journal of experimental psychology. Human perception and performance, 21(2), 1995, pp. 410-422
The conclusions concerning hemispheric specializations based on neural
network simulations, which were previously reported by Kosslyn, Chabr
is, Marsolek, and Koenig (1992), are shown not to be valid. Difference
s in network performance on tasks said to be ''categorical'' and ''coo
rdinate spatial'' in nature were due to imbalances in the input stimul
i and cannot, in principle, be related to differences in performance o
n such tasks in human subjects. The use of truth tables and correlatio
n coefficients in the design of neural networks is discussed.