The Emergence of Speech Patterns

Buy this book »

There is little evidence of any direct relationship between early babbling and the language spoken around an infant prior to 9 months (Boysson- Bardies & Vihman, 1991). The seemingly independent development of perception and speech development may be related to the different areas of the brain devoted to the two functions (MacWhinney, 1998).

By 9 months of age, however, there is increasing evidence of a connection. Babbling changes first occur in intonational patterns. Called jargon, it consists of long strings of unintelligible sounds with adultlike prosodic and intonational patterns. Infants 7 to 10 months of age are sensitive to prosodic or rhythmic cues that help them segment speech into smaller perceptual units (Hirsh-Pasek et aI., 1987). Mothers' speech to infants includes pauses at sentence boundaries, while mothers' speech to other adults often does not. Thus, a child is given cues to a grammatical unit oflanguage (Nelson, Hirsh-Pasek, Jusczyk, & Cassidy, 1989).

A child's babbling gradually comes to resemble the prosodic pattern of the language to which he or she is exposed (Levitt, Utman, & Aydelott, 1993). Babbling patterns become shorter and phonetically more stable. The resultant jargon may sound like questions, commands, and statements. Many parents will swear at this point that their child is communicating in sentences, although the parents aren't exactly clear on what the child is saying. Apparently, the paralinguistic aspects of language are easier for the child to reproduce than the linguistic aspects.

Children's early intonation reflects the interaction of biological, affective, and linguistic influences (Snow, 2006). Although many modifications suggest the importance of linguistic input, the early expression of intonation in infants also points to the role of physiological changes and emotional experience.

• Speech recognition and production pose numerous problems for an infant (Lively, Pisoni, & Goldinger, 1994).
• Auditory processing is complicated by the variety of speakers and contexts.
• The relationship of spoken words to their meanings is essentially arbitrary. There is a lack of any systematic relationship between the sounds in a word and the word's meaning. Acoustic and speech production similarity is unrelated to semantic similarity.
• An infant must learn to produce comprehensible speech without any direct instruction.
• The processes of learning to comprehend and to produce speech must be closely coordinated by an infant.

While comprehension involves placing input onto meaning, production involves generating output from a phonological representation in order to convey meaning. To make it even more difficult, this must all be accomplished within ongoing communication.

Phonological representations consist of phonemes and syllable structures of the native language that are stored in the brain after repeated exposure. As such, phonological representations of words form a stable template or pattern against which both input and output can be compared. In addition, phonological representations or patterns play a critical role in facilitating acoustic input, articulatory output, and meaning interfacing (Plaut & Kello, 1999).