Magazine Archive

Home -> Magazines -> Issues -> Articles in this issue -> View

The Psychology of Music (Part 3)

How is it some musicians can create music spontaneously, seemingly at whim, whilst others cannot? New pieces seem to just flow from their fingertips, but it's often true when those who are able to improvise are asked how they manage it, they are invariably at a loss as to how to answer this taxing question - "I don't think about it, it just happens", "I hear a tune in my head...". Perhaps the difficulty we have in trying to fathom this stems from the unconscious and automatic nature of the processes involved.

An analogy of the improvisation phenomenon is spontaneous speech, in which we talk first without thinking consciously about what are about to say. This, however, does not further elucidate how it happens, but rather it points our line of inquiry to the development patterns which exist in both learning to speak and learning to play an instrument. It is likely that in the early stages of learning to do either, the processes were very much under conscious control. As our competency and technique increased in complexity and automation the 'soft'-control systems moved from conscious to unconscious levels. At some stage the systems containing the motor programs and action plans (which we introduced in Parts 1 and 2) would cease to be directly accessed by our consciousness.


In improvisation, a piece of music is spontaneously created. What triggers this is difficult to say, but the process must be governed to a certain extent by a set of rules that limit the final form of the improvisation. These rules involve the overall structure of pieces such as the initial statement of the melody, development, chordal and harmonic progressions, second melody, bridge, conclusion, and so on. Interestingly, it seems that these rules apply to all musical improvisation regardless of its cultural setting: what holds true for Western jazz holds true for Indian sitar music.

Models of Improvisation

There are several perspectives from which the phenomenon of improvisation can be studied but perhaps one of the more interesting is to consider it as a sort of problem, rather like chess, having a starting state and a goal state. A problem exists when there is a difference between the two. Let's take the chess analogy a little further. The starting state is when all the pieces are arranged in their respective places on the chessboard. The goal state is the checkmate which both players aim for. Each time a player moves a piece, a new state is formed. In this way there are an endless number of intermediary states between the starting and goal states. The range of possible moves at each state can be set out in what is called a problem space which looks on paper something like an elaborate tree diagram. For a complex problem like chess, and also improvisation, the number of branches or possible future moves becomes infinite after the first few moves.

If we consider the improvisation process as having a starting state in the particular note or chord that is played first, and a goal state in the note or chord that finishes the piece, then a problem exists, since the aim is to achieve the goal state using as many (or as few) 'sensible' moves as the performer wishes. Each new chord or note played after the first represents a new state, and at each of these there exists an almost infinite number of possible 'moves' which the performer can make based on experience and technique.

Before, let's look at another but far simpler problem. By analysing the sorts of processing we would use to solve this simple problem, we may gain some insight into the processing involved in improvisation. A tall story? Then read on.

Missionaries and Cannibals

For the benefit of those who have not come across this little puzzle, here are the essential details: at the start there are three missionaries (M) and three cannibals (C) on the left bank of a river. There is a boat which is used to transport them to the opposite bank. The only rule of the puzzle is that the cannibals must never outnumber the missionaries on either bank. The aim is to transport everyone to the right bank without losing any of the missionaries to the bloodthirsty cannibals!

In Fig. 1, the starting state is shown with the numerical representation that will be used from now on. From this state there are five possible moves, which, in the interests of parsimony, we will denote M, MM, C, CC, and MC. These represent the limited ways in which the M's and C's can be moved across the river, and will be called. By applying each of these operators to the starting state a series of new states is formed (level 2). Each of these should bring us closer to the goal state i.e. the answer. The number of 'branches' at any level of the problem space tree is directly proportional to (i) the number of operators (in this instance five) and (ii) the level of that state. For instance, after say four moves, the number of possible branches would be 54 (625 moves). Of course, some are 'illegal' as they result in the C's outnumbering the M's on one or other bank, and some result in infinite loops which are the bane of computer programmers. These are terminated. An example of this is seen in Fig. 2, where a cannibal is transferred to the right bank only to be returned to the left. The application of this operator does not bring about a state which is any closer to the goal.

A New Explanation of Improvisation

It has been necessary to look closely at this strange type of problem analysis in order to try to explain how musicians improvise pieces of music. Improvisation can be represented in a similar fashion to the previous problems by a problem space, in which the operators are defined as the motor programs and action plans required to perform specific notes or chords. The application of an operator follows a rapid 'mental calculation' which is based on the previous note or chord. This is outlined schematically in Fig.3. The number of possible operators is almost infinite, and depends almost entirely on each individual performer's musical knowledge, experience, and technique. It is conceivable that an improvisation could take each separate state as it comes, calculate and execute a new operator in a relatively arbitrary manner, and produce an entirely satisfactory sound. But unfortunately, it is in this last respect that improvisation separates the men from the boys.

Expert improvisors generally play in a certain style which tends to reflect the way their improvisatory repetoire is chosen. Their choice of each operator is far from arbitrary. Meanwhile the mediocre improvisator has a much more restricted set of operators, has less physical familiarity with the instrument, and consequently a poor style. He or she solves real-time problems by choosing operators from a restricted and unadventurous set, creating a piece that is repetitive and uninspiring. In fact, the improvisation my take the form of a simple tonic to dominant, dominant to subdominant, and back to tonic progression. This is analogous to the short-term infinite loop we saw earlier.

Infinite loops which involve many more progressions, do have their uses in improvisation, serving to repeat and develop melodic ideas.


And so to putting all these ideas together. The starting state of any improvisation is the first note or chord. Depending on the performer's experience and ability, there are several ways to create a new state by the application of operators which reflect the result of a rapid mental calculation. This is the result of a rapid access to the performer's own library of musical knowledge, containing information about the pitch, timbre, duration, and rhythm of notes, as well as chord structure, harmony and more general information about the overall structure of tunes and their aesthetics.

In a jazz improvisation, for example, it may be necessary before choosing the appropriate closing sequence, to determine where the previous sequence will lead, and how much time is required to return to the starting state. Short-cuts can be taken if the performer knows already the result of a certain operator, and so it is obviously not enough that an improvisor knows how a performance is to be structured (such as, that a certain chord sequence must be negotiated in a given time); he or she must have rapid access to a massive compilation of knowledge, and musical experience.

It is understandable that the complexity of the improvisatory skill has severely inhibited experimental investigation and psychological theorising. At present, we cannot account for the ability which allows some people to spontaneously create music as opposed to others who just cannot. It is likely that we may never know, as the innate and learned components of most behaviours are inextricably linked.

Previous Article in this issue

The Kit & Accessories

Next article in this issue


Electronics & Music Maker - Copyright: Music Maker Publications (UK), Future Publishing.


Electronics & Music Maker - Jan 1984

Donated & scanned by: Stewart Lawler


Music Theory


The Psychology of Music

Part 1 | Part 2 | Part 3 (Viewing)

Feature by Andrew Morris

Previous article in this issue:

> The Kit & Accessories

Next article in this issue:

> Patchwork

Help Support The Things You Love

mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.

If you value this resource, you can support this project - it really helps!

Donations for January 2021
Issues donated this month: 0

New issues that have been donated or scanned for us this month.

Funds donated this month: £46.00

All donations and support are gratefully appreciated - thank you.

If you're enjoying the site, please consider supporting me to help build this archive...

...with a one time Donation, or a recurring Donation of just £2 a month. It really helps - thank you!

Small Print

Terms of usePrivacy