Vibrotactile learning and the locus dimension

Forty subjects, each of whom was randomly allocated to one of four experimental groups, learned a 26-element vibrotactile code. Each group used either 3, 4, 5 or 6 vibrators. As predicted, the performance of each group improved in the order 3-, 4-, 5- and 6-vibrators. Although there were no significant differences between the groups there was a highly significant learning effect over ten trials. Error analyses showed that duration errors were the most frequent type of error and that intensity errors were the least frequent. It is suggested that both the types of errors and the optimum number of loci which can be separately stimulated in this type of experiment will depend upon the spacing and arrangement of signal elements.