1.3.2 The development of the musical memory

"Every musical activity, whether it be perception, performance, improvisation or composition, involves memory" (Sloboda, 2001a). In a fairly broad way, the acquisition of a memory for sounds begins when the auditory system is fully developed. The first external sound to be recognised reliably is the mother's voice. (Lecanuet, 22). The fact that the foetus recognises its mother's voice shows that there is some form of memory retention at work, so the assumption can be made that a human's memory for sound begins before birth. It seems that prenatal infants can also acquire music, as studies have shown that neonatal infants will be soothed by music that was played to them in the last stages of prenatal development. (Lecanuet, 21) This is a very basic form of musical memory that develops continuously throughout childhood. Studies have shown that Piaget's theory of conservation also applies to music. This demonstrates the developing nature of the musical memory in children (Hargreaves, 1986, 43-48).

Memory for sound takes the form of seven major parts. These are episodic, semantic, echoic, explicit, implicit, recognition, and recall (Carlson and Buskist, 1997; Kilgour et al, 2000; Peretz et al, 1998; Peynircioglu et al, 1998; Schulkind et al, 1999; Sloboda, 2001a). These forms can and do overlap, and some have more than one name. The other aspects of sonic memory that will be discussed are autonoetic and noetic awareness and intraopus and extraopus expectation.

Episodic memory, (also known as veridical memory), is a type of long-term memory, which serves as a record of life events. It is a memory for specific songs, and the nostalgic/emotional experience of music most probably relies on this form of memory.

Semantic memory (also known as schematic) is also a type of long-term memory but serves as an information repository. It is a memory for conventions and models (scales, harmonic theory), which govern the way we hear music.

The next form of memory for sound identified is echoic memory. Carlson and Buskist identify two types of sensory memories, echoic and iconic. Iconic is concerned with sight. Echoic is "a form of sensory memory for sounds that have just been perceived" (Carlson and Buskist, 1997, 235). This would perhaps be the equivalent to Fraisse's (1978, 1982) perceptual present. This may also explain why De Nora (2000) found that musical fragments were more important than the whole musical work for an individual to transform it into a resource for identity construction.

Butler (1992) says "we compare ongoing musical events with our memories of what we have already heard in this piece, making connections with earlier listening experience and making tacit predictions about what we will hear." (Butler, 146) This relates to the intraopus and extraopus processing explained by Narmour (1999) as "within a given work (intraopus), we recognize when we are experiencing something we heard earlier, and between pieces (extraopus), we recognize when we are hearing something we learned elsewhere." Levinson (1997) paraphrases Gurney, "…once a melody is known, the opening bars seem to convey in prospect all the beauty of the rest… one seems to construct the melody… by the very act of listening…"(Levinson, 5) Intraopus and extraopus expectation reside under the areas of echoic, episodic and even semantic memory. Echoic memory allows a human to hold a short phrase in the memory in order to compare and anticipate. This anticipation/comparison may be because we know the piece; therefore the memory accessed would be episodic. The anticipation/comparison may also occur in an unfamiliar piece. In this situation, the person must use semantic memory (memory for conventions and models) to make educated guesses about what will come next in the piece.

Explicit memory refers to knowledge that one knows one has. This could be a musician actively knowing a blues scale, or knowing exactly why a particular piece is so well liked. Explicit memory allows a person to describe what is heard using specific terms, for example, "I can hear a diminished seventh."

Implicit memory refers to a more unconscious type of memory, for example a non-musician has learnt by exposure that a cadential sequence needs to resolve to the tonic, but it is very unlikely that this person would be able to verbalise this need. This person knows that it "sounds right", but cannot describe what it is that is heard.

As mentioned above, the forms of memory can overlap. Semantic memory for music can be explicit or implicit. In the case of a trained musician, semantic memory is mostly explicit. In the case of non-musicians, the semantic structure of music is largely implicit. Logically however, a musician's semantic memory would have begun as implicit memory (exposure throughout childhood) but once the learning process takes place the memory becomes explicit (learning musical theory at school or university).

Episodic memory can also be implicit or explicit. Explicit episodic memory is probably more common, and would equate to life events that one can describe (e.g. rock concert). An implicit form of episodic memory would be hearing a piece of music but not remembering having heard it, and perhaps reproducing it as an original melody (see Appendix 5, for an example).

Recognition is defined by Sloboda (2001a) as the experience of similarity, (known elsewhere as a "feeling of knowing"), or in a more advanced state, as identification or naming. Recall is defined as "the reproduction, either in imagination or behaviour, of a previously experienced sequence [of notes, rhythms, i.e. a tune]." The article goes on to say:

As early as 1936, Van de Wall discussed the faculty of recalling to mind music which humans have heard, and labelled it as an intrapsychic experience. He states:

Given the definitions above, it is apparent that this intrapsychic experience is audiation and that audiation fits into the section of musical memory known as recall. This may be because hearing music is an episodic event, and as such people associate certain events, emotions, and people with certain types of music. Examples of such triggers include listening to the radio, repetitive manual labour, physical exertion, an intense mood, a phrase, or a stray thought.

Cutietta and Booth (1996, 222-236) showed that cognitive categorisation takes place within melodic recall. Their study showed that the predominant musical elements used by subjects for categorisation appeared to change over the trials, from meter and mode in the early trials which were supplanted by interval-type and melodic contour in the later trials. A reason that they suggested for this was the greater amount of melodic information one could obtain from these elements which would assist recall, as opposed to knowing what key or time signature the melody was in.

Another two aspects of memory that should be mentioned are Gardiner and Radomski's (1999) autonoetic and noetic awareness. These are primarily alternate terms for episodic and semantic memory, but the authors' emphasis is on the awareness of consciousness. Autonoetic awareness is a subjective awareness of the self in relation to past, present and future events. This is the episodic version. Noetic awareness is a more objective awareness of past, present and future, things that we "know" but haven't actually experienced. General knowledge (for example, trivia, knowing the capitals of each country, in a musical sense an example would be a knowledge of the lives of the great composers) fits into this semantic version. The authors state that "remembering" and "knowing" are independent states of awareness. These are alternate ways of naming episodic and semantic memory.

Humans begin acquiring a musical memory from before birth, and as with generic memory (i.e. general knowledge, life events), this acquisition usually stops only at death. It should be noted that humans acquire a catalogue of sounds, beginning prenatally with the mother's voice and increasing, through everyday exposure to sound, throughout life. Sounds can be recognised, recalled and reproduced. The inner soundscape of the mind consists of audiation, thoughts, and random sounds that people hear in their minds. People often "talk" to themselves in their minds, they do not necessarily specifically carry out a conversation with themselves but rather they sound out theories, read with an internal voice, make comments on their actions or the actions of others, or say what they can't say out loud for fear of the consequences. Random sounds are just that, sounds from our life that just occur in our mind for as many and varied reasons as audiational triggers. For example, when a person is anticipating an important phone call, the brain may try to alleviate his/her anxiety by making him/her hear a phone ringing. However, the phone is not ringing, nor is the neighbour's phone. It is a random sound occurring within the inner soundscape. Once a sound or a piece of music is established in our memory it can be recalled, voluntarily, involuntarily, inwardly (in the mind) or outwardly (through vocalisation, noise or musical expression). Therefore audiation is a constructed human reality, and as such has been used in film music, either consciously or unconsciously.