Spoken Word Recognition: Historical Roots, Current Theoretical Issues, and Some New Directions

The most distinctive hallmark of human spoken word recognition (SWR) is its perceptual robustness to the presence of acoustic variability in the transmission and reception of the talker’s linguistic message. Normal-hearing listeners adapt rapidly with little apparent effort to many different sources of variability in the speech signal and their immediate listening environment. Sensory processing and early encoding of speech into lexical representations are critical for robust SWR. However, audibility and sensory processing are not sufficient to account for the robust nature of SWR, especially under degraded listening conditions. In this chapter, we describe the historical roots of the field, present a selective review of the principle theoretical issues, and then consider several contemporary models of SWR. We conclude by identifying promising new directions and future challenges, including the perception of foreign accented speech, SWR by deaf children with cochlear implants, bilinguals, and older adults.