LINEAR CORRELATES IN THE SPEECH SIGNAL - THE ORDERLY OUTPUT CONSTRAINT

Neuroethological investigations of mammalian and avian auditory system s have documented species-specific specializations for processing comp lex acoustic signals that could, viewed in abstract terms, have an int riguing and striking relevance for human speech sound categorization a nd representation. Each species forms biologically relevant categories based on combinatorial analysis of information-bearing parameters wit hin the complex input signal. This target article uses known neural mo dels from the mustached bat and barn owl to develop, by analogy, a con ceptualization of human processing of consonant plus vowel sequences t hat offers a partial solution to the noninvariance dilemma - the nontr ansparent relationship between the acoustic waveform and the phonetic segment. Critical input sound parameters used to establish species-spe cific categories in the mustached bat and barn owl exhibit high correl ation and linearity due to physical laws. A cue long known to be relev ant to the perception of stop place of articulation is the second form ant (F2) transition. This article describes an empirical phenomenon - the locus equations - that describes the relationship between the F2 o f a vowel and the F2 measured at the onset of a consonant-vowel (CV) t ransition. These variables, F2 onset and F2 vowel within a given place category, are consistently and robustly linearly correlated across di verse speakers and languages, and even under perturbation conditions a s imposed by bite blocks. A functional role for this category-level ex treme correlation and linearity (the ''orderly output constraint'') is hypothesized based on the notion of an evolutionarily conserved audit ory-processing strategy. High correlation and linearity between critic al parameters in the speech signal that help to cue place of articulat ion categories might have evolved to satisfy a preadaptation by mammal ian auditory systems for representing tightly correlated, linearly rel ated components of acoustic signals.