ABSTRACT
The neighborhood activation model (NAM) (Goldinger, Luce, and Pisoni, 1989; Luce, 1986; Luce, Pisoni, and Goldinger, 1990) hypothesizes that spoken word recognition is characterized by two successive stages: (1) similarity neighborhood activation, and (2) a frequency-biased decision process. In the first stage, stimulus input (i.e. a spoken word) activates a set of similar sounding representations of words in memory. In the second stage, these multiply activated form-based representations –or neighbors– are decided among by a decision process that is influenced by the acoustic-phonetic similarity of the neighbors as well as their frequency of occurrence. NAM further posits that the decision process among the competitors in the activated set of representations is influenced by three characteristics of the similarity neighborhoods. These characteristics are: (1) target word frequency, (2) neighborhood density, and (3) neighborhood frequency.
