Conceptualisation and planning
المؤلف:
Paul Warren
المصدر:
Introducing Psycholinguistics
الجزء والصفحة:
P17
2025-10-29
40
Conceptualisation and planning
Conceptualisation is pre-linguistic. That is, it does not involve forms of language, but is all done in the head’ in abstract terms. The speaker needs to make some very general decisions about what to say, taking into account facts about the situation that they are in, and relying on their general knowledge as well as on their understanding of how communication including conversation works. They will also need to think about whether specific aspects of their message need to be highlighted. The result of the process of conceptualisation is a pre- erbal message, i.e. it is still not language. It consists of a set of ideas in what has sometimes been called mentalese, or the language of thought. These ideas form part of the mental model Johnson-Laird, 1983 of what the speaker wants to say. They are not yet organised into an ordered string of phrases and words, which will be the task of the formulation component.
Let’s imagine the speaker wants to communicate the idea that a cat hunted a mouse Figure 2.3. For this example sentence, the pre-verbal message will reflect their intention to communicate at least three concepts the concept of hunting of one animal by another, the concept of a cat as the entity carrying out this action the AGENT, and the concept of a mouse as the entity on which the action is carried out the THEME, sometimes referred to as the PATIENT. It will also include information about whether the speaker wants to highlight any of these concepts, e.g. by focusing on the AGENT or on the THEME.
A key source of information about the processes of conceptualisation and planning in producing spoken output is the pattern of pausing that speakers produce. The argument is that planning involves mental activity that competes for our attention resources with the actual process of speaking. The more planning we need to do, the less easy it is for us to continue speaking and the more likely it is that we will hesitate.
When is a pause a pause?
Before we explore more closely this idea that pauses can tell us about conceptualisation and planning, it is important first to consider what counts as a pause. A key issue concerns how long a silence has to be before we can safely say it is a pause. This is because in the normal articulation of speech we will find silences that we would not want to include in our catalogue of pauses. Take for instance the pronunciation of /p/ in the word space. This /p/ requires the speaker to close the lips tightly, preventing any air or sound from coming out, and then to release that closure rather suddenly. During the closure part of this sound, there will be a measurable though very brief silence, as shown in Figure 2.4 . This is an articulatory pause. Because of such silences – and also in early research because of the degree of accuracy of the equipment used – researchers tend to use a minimum duration for what counts as a pause, such as 200 milliseconds. At the same time, however, listeners often hear pauses where there are none. This is because speakers sometimes slow down their rate of speaking without actually producing a silence, and this change in rhythm can be enough to lead to the perception of a pause. Not surprisingly, some studies also look at such changes in the durations of syllables or other parts of words as part of the study of pausing.
Note that pauses also have functions other than planning. eliminate pauses can occur at places where a written text might have punctuation, breaking utterances into constituent parts, possibly to help the listener. Intonation and other aspects of prosody can also contribute to this function. Physiological pauses are those that help speakers to regulate their breathing while speaking. In practice, speakers mostly breathe at points where they might have to pause for some other reason. Accordingly, the use of pausing for breathing is largely disregarded in psycholinguistic studies.
Some researchers have looked in more detail at filled pauses, and in particular at whether different types of filled pause might have different functions. So for example, while Clark and Fox Tree 2002 found a variety of reasons why speakers might use filled pauses – to gain time to search for a word, to indicate that they have not yet finished their turn, etc. – they found that speakers were more likely to use uh’ to signal a short delay and um’ to signal a longer delay in speaking. Chapter 5 looks at how such behaviour indicates that speakers are continually monitoring their output and effecting repairs when things go wrong.
Pausing and the amount of planning
Everyone has probably planned an utterance in some detail – a speech, an answer prepared for an interview, a marriage proposal, an apology for that broken vase that has been preying on your mind, and so on. Most speech events are less well prepared but still differ from one another in the amount and types of planning involved. These differences are reflected in the number and distribution of pauses in speech. When the task is reading aloud, most of the planning has already been done in preparing the text in the first place. As a consequence, when fluent readers speak aloud from a prepared manuscript, they do not need to pause for planning purposes but instead pause almost exclusively at points marked by punctuation, with longer pauses at paragraph breaks than at full stops.
This close coordination of pausing and sentence structure during reading can be contrasted with pause patterns in spontaneous speech. Here, it is much more likely that sentence constituents will be interrupted by pausing and other hesitation types see the transcribed monologue given earlier. Pauses are more frequent and often much longer in spontaneous speech than in reading, but there are also many short pauses. In one of my own unpublished studies of speakers’ performance in different tasks, one speaker had average pause durations of 870 msec in spontaneous speech and 479 msec in reading. She spent a total of 31 of her time pausing in spontaneous speech but only 10 during reading aloud.
These rather obvious differences between read speech and unprepared speech reflect the equally obvious differences in the planning involved in the two tasks. When we read aloud a prepared text, we need to plan when to pause in order to mark the structure of the text for our listeners and to grab an opportunity to breathe. We also need to organise how we are going to articulate the speech sounds that correspond to the words. This is not a trivial task, since it involves recognising the written form of the word, finding a pronunciation for that word, coordinating the movements of various muscles to produce the sounds, and so on. In spontaneous speech, we have to carry out some of the tasks just listed and a whole lot more besides. This is because we also need to decide what we want to say and what sentences and words we want to use in order to say it. Compared with reading aloud, spontaneous speaking involves consider ably more planning, and planning of quite different kinds. It is not surprising then that it includes many more pauses, and longer pauses. There are also many more disfluencies and other types of hesitation in spontaneous speech. Since we continuously monitor what we are saying and frequently decide that there is a better or more appropriate way of getting our message across, spontaneous speech also contains more self-interruptions, false starts, and so on.
Between these extremes of reading and spontaneous speech we find other speech tasks that vary in the amount and nature of the planning involved. They also vary in the amount and distribution of pausing. For instance, in a pioneering psycholinguistic study of pausing, Goldman Eisler 1961 asked speakers to carry out two different tasks using a cartoon story made up of a series of pictures, usually a story with some kind of moral behind it. One task was simply to describe the action in the pictures, and the other was to interpret the story. In the description, speakers do not need to generate the basic ideas which are given by the pictures, but they do need to find the right words and sentence structures to convey those ideas. In the interpretation, speakers need to explain the moral behind the story. This involves a greater amount of original and abstract thought, and the conversion of this to speech involves more planning than the description task. This difference is reflected in more pause time in interpretations than in descriptions. Goldman-Eisler found that interpretations of cartoon stories had an average ratio of pause time to speech time of 3.31 1, while the corresponding figure for cartoon descriptions was just 1.24 1.
الاكثر قراءة في Linguistics fields
اخر الاخبار
اخبار العتبة العباسية المقدسة
