Skip to main content
Social Sci LibreTexts

13.4: Evidence for ‘top-down’ effects of word knowledge on perception

  • Page ID
    192749
    • Catherine Anderson, Bronwyn Bjorkman, Derek Denis, Julianne Doner, Margaret Grant, Nathan Sanders, and Ai Taniguchi
    • eCampusOntario

    \( \newcommand{\vecs}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \)

    \( \newcommand{\vecd}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash {#1}}} \)

    \( \newcommand{\id}{\mathrm{id}}\) \( \newcommand{\Span}{\mathrm{span}}\)

    ( \newcommand{\kernel}{\mathrm{null}\,}\) \( \newcommand{\range}{\mathrm{range}\,}\)

    \( \newcommand{\RealPart}{\mathrm{Re}}\) \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\)

    \( \newcommand{\Argument}{\mathrm{Arg}}\) \( \newcommand{\norm}[1]{\| #1 \|}\)

    \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\)

    \( \newcommand{\Span}{\mathrm{span}}\)

    \( \newcommand{\id}{\mathrm{id}}\)

    \( \newcommand{\Span}{\mathrm{span}}\)

    \( \newcommand{\kernel}{\mathrm{null}\,}\)

    \( \newcommand{\range}{\mathrm{range}\,}\)

    \( \newcommand{\RealPart}{\mathrm{Re}}\)

    \( \newcommand{\ImaginaryPart}{\mathrm{Im}}\)

    \( \newcommand{\Argument}{\mathrm{Arg}}\)

    \( \newcommand{\norm}[1]{\| #1 \|}\)

    \( \newcommand{\inner}[2]{\langle #1, #2 \rangle}\)

    \( \newcommand{\Span}{\mathrm{span}}\) \( \newcommand{\AA}{\unicode[.8,0]{x212B}}\)

    \( \newcommand{\vectorA}[1]{\vec{#1}}      % arrow\)

    \( \newcommand{\vectorAt}[1]{\vec{\text{#1}}}      % arrow\)

    \( \newcommand{\vectorB}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \)

    \( \newcommand{\vectorC}[1]{\textbf{#1}} \)

    \( \newcommand{\vectorD}[1]{\overrightarrow{#1}} \)

    \( \newcommand{\vectorDt}[1]{\overrightarrow{\text{#1}}} \)

    \( \newcommand{\vectE}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash{\mathbf {#1}}}} \)

    \( \newcommand{\vecs}[1]{\overset { \scriptstyle \rightharpoonup} {\mathbf{#1}} } \)

    \( \newcommand{\vecd}[1]{\overset{-\!-\!\rightharpoonup}{\vphantom{a}\smash {#1}}} \)

    \(\newcommand{\avec}{\mathbf a}\) \(\newcommand{\bvec}{\mathbf b}\) \(\newcommand{\cvec}{\mathbf c}\) \(\newcommand{\dvec}{\mathbf d}\) \(\newcommand{\dtil}{\widetilde{\mathbf d}}\) \(\newcommand{\evec}{\mathbf e}\) \(\newcommand{\fvec}{\mathbf f}\) \(\newcommand{\nvec}{\mathbf n}\) \(\newcommand{\pvec}{\mathbf p}\) \(\newcommand{\qvec}{\mathbf q}\) \(\newcommand{\svec}{\mathbf s}\) \(\newcommand{\tvec}{\mathbf t}\) \(\newcommand{\uvec}{\mathbf u}\) \(\newcommand{\vvec}{\mathbf v}\) \(\newcommand{\wvec}{\mathbf w}\) \(\newcommand{\xvec}{\mathbf x}\) \(\newcommand{\yvec}{\mathbf y}\) \(\newcommand{\zvec}{\mathbf z}\) \(\newcommand{\rvec}{\mathbf r}\) \(\newcommand{\mvec}{\mathbf m}\) \(\newcommand{\zerovec}{\mathbf 0}\) \(\newcommand{\onevec}{\mathbf 1}\) \(\newcommand{\real}{\mathbb R}\) \(\newcommand{\twovec}[2]{\left[\begin{array}{r}#1 \\ #2 \end{array}\right]}\) \(\newcommand{\ctwovec}[2]{\left[\begin{array}{c}#1 \\ #2 \end{array}\right]}\) \(\newcommand{\threevec}[3]{\left[\begin{array}{r}#1 \\ #2 \\ #3 \end{array}\right]}\) \(\newcommand{\cthreevec}[3]{\left[\begin{array}{c}#1 \\ #2 \\ #3 \end{array}\right]}\) \(\newcommand{\fourvec}[4]{\left[\begin{array}{r}#1 \\ #2 \\ #3 \\ #4 \end{array}\right]}\) \(\newcommand{\cfourvec}[4]{\left[\begin{array}{c}#1 \\ #2 \\ #3 \\ #4 \end{array}\right]}\) \(\newcommand{\fivevec}[5]{\left[\begin{array}{r}#1 \\ #2 \\ #3 \\ #4 \\ #5 \\ \end{array}\right]}\) \(\newcommand{\cfivevec}[5]{\left[\begin{array}{c}#1 \\ #2 \\ #3 \\ #4 \\ #5 \\ \end{array}\right]}\) \(\newcommand{\mattwo}[4]{\left[\begin{array}{rr}#1 \amp #2 \\ #3 \amp #4 \\ \end{array}\right]}\) \(\newcommand{\laspan}[1]{\text{Span}\{#1\}}\) \(\newcommand{\bcal}{\cal B}\) \(\newcommand{\ccal}{\cal C}\) \(\newcommand{\scal}{\cal S}\) \(\newcommand{\wcal}{\cal W}\) \(\newcommand{\ecal}{\cal E}\) \(\newcommand{\coords}[2]{\left\{#1\right\}_{#2}}\) \(\newcommand{\gray}[1]{\color{gray}{#1}}\) \(\newcommand{\lgray}[1]{\color{lightgray}{#1}}\) \(\newcommand{\rank}{\operatorname{rank}}\) \(\newcommand{\row}{\text{Row}}\) \(\newcommand{\col}{\text{Col}}\) \(\renewcommand{\row}{\text{Row}}\) \(\newcommand{\nul}{\text{Nul}}\) \(\newcommand{\var}{\text{Var}}\) \(\newcommand{\corr}{\text{corr}}\) \(\newcommand{\len}[1]{\left|#1\right|}\) \(\newcommand{\bbar}{\overline{\bvec}}\) \(\newcommand{\bhat}{\widehat{\bvec}}\) \(\newcommand{\bperp}{\bvec^\perp}\) \(\newcommand{\xhat}{\widehat{\xvec}}\) \(\newcommand{\vhat}{\widehat{\vvec}}\) \(\newcommand{\uhat}{\widehat{\uvec}}\) \(\newcommand{\what}{\widehat{\wvec}}\) \(\newcommand{\Sighat}{\widehat{\Sigma}}\) \(\newcommand{\lt}{<}\) \(\newcommand{\gt}{>}\) \(\newcommand{\amp}{&}\) \(\definecolor{fillinmathshade}{gray}{0.9}\)

    The preceding sections provided evidence for how a speaker’s knowledge of the phonology of their language can influence the way they perceive speech. In this section, we will see that a speaker’s knowledge of the words of their language also has an influence on the way language is perceived. In particular, we will focus here on ways in which a speaker’s word knowledge has an influence on the speech sounds that they hear.

    The first example comes from an influential experiment by Ganong in 1980. Remember from Section 13.2 that the difference between a voiced stop like [d] and a voiceless stop like [t] is the stop’s Voice Onset Time, or in other words the time between the release of the stop closure and the beginning of the following vowel. Many years of research have shown that although Voice Onset Time is a continuous variable – it can take on any millisecond value – listeners perceive a sharp boundary between speech sounds like [d] and [t]. So in an experiment that presents participants with different values of Voice Onset Time and asks them to indicate whether they heard [d] or [t], you would get nearly all [d] responses below a threshold and nearly all [t] responses above that threshold, with only a small area of uncertainty in the middle. This figure shows a typical pattern found by experiments that vary Voice Onset Time in increments. Below the boundary between [d] and [t], nearly all participants would identify the sound as [da]. Above the boundary, the responses would be all or nearly all [ta]. There is only a short window around the boundary where responses are not all one way or the other.

    Graph of theoretical results of a phoneme identification task where Voice Onset Time is varied on a continuum. At short Voice Onset Times, nearly all responses identify a sound as [d]. At high Voice Onset Times, nearly all responses identify the sound as [t]. There is a short window of mixed responses around the [d]-[t] boundary, creating a sinusoidal (s-shaped) curve.
    Graph showing a typical pattern found in phoneme identification tasks. This particular graph does not show real data, but rather an idealized pattern.

    This pattern is called categorical perception and holds for a variety of phonetic continua other than Voice Onset Time as well. Categorical perception was initially thought to reflect phonological knowledge, but subsequent studies provided evidence against that interpretation. For example, categorical perception is found for contrasts that are not a part of the phonology of a listener’s language, for babies who have not had sufficient time to tune their mental grammar to the phonology of their language, and even for non-human animals, who we presume do not have human-like phonology (see Kuhl and Miller, 1975, for a study on chinchillas). This is why, in this section, we use the square bracket notation for [d] and [t].

    Building on the robust findings of categorical perception, Ganong examined whether presenting sounds on a continuum from [d] to [t] as a part of words would influence participants’ perception of the sounds. In particular, words were chosen such that at one end of the continuum, the word was a real word of English (e.g., dash), while at the other end, the word was pronounceable but not a real word of English (e.g., tash). The critical hypothesis was that listeners might be more likely to categorize a sound that was intermediate between [d] and [t] as the sound that would create a real word. This is what Ganong found. When a sound near the threshold was presented as part of Xash, where X represents the critical stop consonant, it was more likely to be perceived as dash; (the real word of English) rather than tash (which is pronounceable but not an English word). When the same sound is presented with Xask, it would be more likely to be perceived as task (a real word of English) rather than dask. So people’s knowledge of words of English influenced which speech sound they heard, even when the acoustic signal is the same. This finding, which now gets called a Ganong effect, is an example of a top-down influence on perceptual processing. We call information that flows to the brain from the outside world through sensory surfaces, like the vibration of the inner ear or an image on the retina, bottom-up information. It is no surprise that bottom-up information has an influence on processing, as it couldn’t really be any other way. What we call top-down influences are cases where our knowledge, for example our knowledge of words, has an effect on perceptual processing. In this chapter we have seen several ways in which our knowledge of language has an influence on the way we process linguistic input from quite early stages of perception.

    References

    Ganong, W. F. (1980). Phonetic categorization in auditory word perception. Journal of Experimental Psychology: Human Perception and Performance, 6(1), 110–125. https://doi.org/10.1037/0096-1523.6.1.110

    Kuhl, P. K., & Miller, J. D. (1975). Speech Perception by the Chinchilla: Voiced-Voiceless Distinction in Alveolar Plosive Consonants. Science, 190(4209), 69–72. www.jstor.org/stable/1740887


    This page titled 13.4: Evidence for ‘top-down’ effects of word knowledge on perception is shared under a CC BY-NC-SA 4.0 license and was authored, remixed, and/or curated by Catherine Anderson, Bronwyn Bjorkman, Derek Denis, Julianne Doner, Margaret Grant, Nathan Sanders, and Ai Taniguchi (eCampusOntario) via source content that was edited to the style and standards of the LibreTexts platform; a detailed edit history is available upon request.