Probabilistic multidimensional scaling using a city-block metric

Using a probabilistic model, exact and approximate probability density func tions (PDFs) for city-block distances and distance ratios are developed. Ti le model assumes that stimuli can be represented by random vectors having m ultivariate normal distributions. Comparisons with the more common Euclidea n PDFs are presented. The potential ability of the proposed model to correc tly detect Euclidean and city-block metrics is briefly investigated. These results are then contrasted to those obtained using a deterministic, nonmet ric model. (C) 2001 Academic Press.