Controlling the Intelligibility of Referring Expressions in Dialogue

Listeners recover /t/s that speakers reduce: Evidence from /t/-lenition in Dutch

“…This variable did not have a significant contribution (p ¼ 0:29) when used as predictor in a logistic regression with /t/-deletion as dependent variable. Secondly, the null-effect might be due to the fact that we did not count realizations of the target within the conversation context by other speakers, which was suggested to be an important variable for reduction of articulatory effort by Bard et al (2000). Third, the role of repetition may be smaller than suggested in some studies because of the confound of repetition with prosodic factors, such as accent (Hawkins & Warren, 1994).…”

Why is conversation so easy?

“…They even repeat syntax between languages, when one interlocutor speaks English and the other speaks Spanish [33]. There is evidence for alignment of clarity of articulation [34], and of accent and speech rate [35]. Finally, alignment at one level increases alignment at other levels, with people being more likely to use an unusual form like 'the sheep that is red ' (rather than the red sheep) after they have just heard 'the goat that is red ' than after they heard 'the door that is red ' [31].…”

Givenness and Locality

“…Bard et al (2000)) report that in addition to deaccentuation, reducing token length and decreasing articulatory detail is also a gradient reflex of givenness. Constituents that are accented and thus not marked as given might still be subject to such gradient reduction effects when they are contextually salient and 'predictable' (Jurafsky et al 2000).…”