About Speech-errors and Disfluencies

What are Speech errors and Disfluencies?

These two terms refer to different phenomena.

Speech errors include incorrectly uttered phrases, words, syllables or phonemes as well as incorrect/unintended intonation stress and timing patterns.

Disfluencies are defined as such, because they constitute interruptions to the normal smooth forward flow of speech. Typical examples of disfluencies are (abnormally long) silent pauses, phrase, word or sound repetitions, and prolongations of continuants. Speech errors, in themselves, are not disfluencies and do not necessarily always result in disfluencies. However, many disfluencies do result from our tendencies to interrupt and attempt to repair the errors we make. Another important source of disfluencies, especially in older speakers, is difficulty retrieving the words we want to say. Under such circumstances, speakers often use “fillers” such as “um” or “er”, perhaps to signal to the listener that they are still trying to speak.

The study of speech errors and disfluencies.

Although the scientific study of speech errors and disfluencies has a long history, until relatively recently most of the research into these phenomena has tended to be carried out by clinicians and researchers with an interest primarily in pathological conditions, such as stuttering and aphasia, in which they manifest as symptoms. However, “normal” speech is rarely completely free of errors and it too is rarely completely fluent. The study of normal patterns of errors and disfluencies has the potential to reveal additional valuable information about the normal workings of our language and speech production and comprehension systems. Nowadays our ability to carry out detailed studies of the mechanisms that lie behind both normal and pathologically produced speech-errors and disfluencies owes much to the availability of specialised recording and computing resources, which together have enabled researchers to make more detailed analyses of error and disfluency frequencies, distributions, and spectral and temporal characteristics; both in spontaneous speech as well as under experimental conditions.

In recent years the study of (both normal and pathological) speech errors and disfluencies has played a particularly important role in enabling researchers to develop and test models of language production, and of the work of the Edinburgh Disfluency Group is has contributed significantly to this development.

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License