Listening involves physiological and cognitive processes at different levels (Field, 2002; Lynch, 2002; Rost, 2011). Several theories have been advanced to aount for listening processes, with two being particularly influential on research.
Anderson (1983 & 1995) proposed a cognitive framework presenting listening as a three-stage process of Perceptual Processing, Parsing, and Utilization. In the perceptual processing phase, attention is focused entirely on the text, and phonemes are segmented from the speech stream (1995, ). Therefore, such listening strategies as “selective attention”and “directed attention”are crucial in this stage (Vandergrift, 2003a). In the parsing stage, meaning representations are formed from words and phrases by h linguistic inforener's long-term meal representations. “Grouping”and “inferencing”strategies are dominant in the parsing stage. And finally in the utilization phase, inforted from the previous two stages is linked with the schema—the previous knowledge of the listener. As a factor related to the present study, schema is further reviewed later in this chapter. Listeners use their prior knowledge to aid prehension and recall. At this stage, “elaboration”strategy is a crucial strategy (Vandergrift, 2003a).
This it provides recognizable stages in the process of listening, and thus facilitates research into each of the stages (as in O'Malley, Chaheless, this model presents listening merely as a linear process with utilization as the final product of the process. As argued by Graham & Macaro (2008, ), “it is perfectly possible for listeners to start by utilizing fragments of parsed text and then draw incorrect inferences. ”
In light of the parallel processing capacity offered by working lelland & Rucing, recursive model was suggested in which listeners operate within he listening stages—an interactive top-down and bottog model of listening. Listeners use “bottom-up”processes when they use linguistic knowledge of sounds and word forms and build up to plex lexical items and grammatical relationships to interpret the input. By using bottom-up processes, listeners construct retion, gradually bining increasingly larger units of meaning from the phoneme-level up to discourse-level features. Listeners also use “top-down”processes when they employ familiarity with the listening context and prior knowledge (topic, genre, culture, and other schema knowledge in long-terual fraprehension. Listeners use content words and contextual clues to form hypotheses in an exploratory manner.
Listening prehension is not just top-down or just bottoive and interpretive process in which listeners use both linguistic knowledge and prior knowledge to understand messages. In other words, a listener es to a listening task with two sets of resources: his/her own linguistic and scheowledge (Rumelhart, 1980) and the inforual listening text. Within an interactive model, a listener might begin by activating his/her schemata as a result of knowing the topic of the text, or of understanding a few words of the text, and thus perceive, parse and ing speech stream with the elaborations that he/she previously activated (Graham & Macaro, 2008).
Also, while these processes interact in some form of parallel distributed processing, the degree to which listeners may use one process her will depend on their knowledge of the language, familiarity with the topic or the purpose for listening. Research (e. g., lelland & Rumelhart, 1986; O'Malley, Chamot, & Kupper, 1989) on these cognitive processes suggests that L2 listeners need to learn how to use both processes to their advantage, depending on their purpose for listening. For example, listening for gist involves prig, whereas listening for specific inforher broadcast, involves primarily bottoprehend all the details. The above research has also shown that suessful and less suessful listeners process input quite differently. Peterson (2001) states that less suessful listeners tend to rely priher top-down or bottog and spend a great aious effort on perceptual activity (e. g., identifying boundaries, recognizing meaningful sound units) so little is left over for high-level operations (e. g., relating new information to information stored in long-tertrast, listeners with higher-proficiency use both top-down and bottoand oral input, which is also known as the use of ening strategies, and is to be further discussed in the last section of this chapter.