One of the best-known and interesting findings in speech perception research is the “phonemic restoration phenomenon”. It is a beneficial and amazingly utilized human ability by which, “under certain conditions, sounds actually missing from a speech signal can be synthesized by the brain and clearly heard”(Kashino, 2006. P.318). This shows the brains sophisticated ability in comprehending speech in the everyday life noisy settings.
This paper explores - with illustrative demonstrations - four queries concerning different aspects of phonemic restoration:
1. Is phonemic restoration a conscious act? Are there any conditions for its activation?
2. What exactly takes place during the process of phonemic restoration? How is it best modeled?
3. Would visual/ auditory integration of a stimulus enhance phonemic restoration?
4. Does hearing impairment affect phonemic restoration?
The phonemic restoration phenomenon was first demonstrated by Warren (1970). He conducted an experiment on a number of listeners where he got them to listen to the sentence, “The state governors met with their respective legislatures convening in the capital city”. However, in the word “legislatures” a cough completely masked the medial “s”. Listeners stated hearing the masked phoneme. This showed that when phonemic restoration occurred, listeners were not able to identify the masking sound position in the sentence. What they received was unconsciously interpreted. In other words, the efficiency of phonemic restoration reached the extent that the listeners did not notice it. (Kashino, 2006). Amazingly, the phonemic restoration effect was not observed when “s” was masked with silence!
Bregman (1990) explains the mechanism of phonemic restoration as a process that consists of two stages. The first stage is a bottom-up innate mechanism built with the use of certain information about the sound, such as, its spatial location and pitch. The second stage is of a top-down nature. It is a schema-based stage where the knowledge of a sound input that is mentally stored is used to supply the primitive bottom-up stage.
Bregman (1981) demonstrated this idea in a visual depiction of phonetic restoration as shown in Figure 1. Figure 1/a demonstrates scattered fragments of a number of images of the letter “B”. Here “B” patterns cannot be perceived from these fragments. Figure 1/b demonstrates the same fragments but in the presence of an occluding pattern. With this occluding pattern it is possible now to recognize the fragments as parts of the letter “B”.
Figure 1. a visual depiction of phonetic restoration Bregman (1981)
This visual phenomenon can be explained in a standard way in terms of “amodal completion, which is the process of perceptually completing occluded visual surfaces” (Nakayama et al., 1995, P.64)....