Pink Noise in Language Production: A Nonlinear Approach to the Multilingual Lexicon

Most theories of (second) language processing are based on component-dominant models that assume linear relationships between linguistic factors affecting language production and language comprehension. In this contribution we argue that in the wake of current nonlinear approaches to cognition, language processing must be seen in terms of interaction-dominant dynamics manifested in 1/f scaling relations. Our study showed that increased experience in second language use leads to scaling relations that more and more tend toward pink noise. Moreover, our data tentatively show that intensive immersion in a second-language context leads to steeper spectral slopes in response time series of that same language. These findings demonstrate the significant contribution of nonlinear analyses to the study of real-time language processing in multilinguals.