Research
Psycholinguistically motivated patterns in phonology
Just as physiological factors in acoustics and articulation have been linked to patterns in phonological systems, I am interested in using psycholinguistic findings as guides in seeking out cross-linguistic phonological patterns. One such finding is the neighborhood density effect, which suggests that distinguishing a morpheme from phonologically similar lexical items is a limiting factor in lexical access (cf. Goldinger, Luce and Pisoni 1989, Cluff and Luce 1990). If a large number of very similar neighbors makes it more difficult to process a lexical item, cross-linguistic lexical patterns might be expected to show evidence of evolution to avoid particularly high neighborhood densities. Two potential examples of lexical access effects on phonological patterning are described in the two links below:
- Neighborhood Density and
the Root-Affix Distinction (2002). With: Adam Ussishkin.
Proceedings of NELS 32 (pdf)
- Neighborhood density and effective contrast may play a role in deriving a crosslinguistically attested distinction in roots and affixes.
- Phonological Alternation,
Lexical Neighborhood Density and Markedness
in Processing (2002).
Poster presented at LabPhon 8, Yale, New
Haven, CT (pdf).
- Neighborhood density effects may account for a crosslinguistic avoidance of phonological alternation.
Self-organization as a source of higher-order patterns in language
Systems can self-organize in interesting ways when:
- System elements are repeatedly subject to conflicting forces.
- Local interactions between system elements are distinct from non-local interactions.
- The system does not reach equilibrium on the time-scale of local change.
Using computer simulations, I am currently exploring models of the lexicon meeting these criteria, with the goal of identifying cross-linguistically attested patterns that can be accounted for through self-organizational processes. An abstract of a recent paper delivered at the meeting of the Berkeley Linguistics Society 29 (2003).
A few news stories have recently appeared on more recent work of mine on contrast maintenance in language, framing this work in terms of the innate/emergent debate on language patterning. To clarify what I think this work contributes to this ongoing debate, I provide here my own short summary of this work in these terms.