Nov 8 ===== Recasens:2018 ------------- What factors lead to coarticulatory resistance and coarticulatory aggressiveness being often positively correlated? Which coarticulation type (carryover or anticipatory) is more difficult for hearing impaired or deaf individuals? What role does auditory feedback play with regard to coarticulation? "The general hypothesis being that languages with more vowels should exhibit less transconsonantal coarticulation than those with fewer vowels, since contrast preservation restricts variability; in other words, the former tend to avoid coarticulation because vowels are more likely to be misperceived as a different vowel" -> Would the same be true of consonants? Would languages like Hawaiian with very few consonant phonemes show relatively high levels of consonantal coarticulation? "Thus, coronal palatalization in English /C#j/ sequences is more liable to apply when /j/ belongs to a highly frequent than a less frequent word (you versus union, unit; Shi et al., 2005)" -> I am curious about the underlying psychological(?) mechanism. If a word is very frequent for me but not for most people (i.e., y'all for speakers of certain dialects), will I take that into account when "deciding" how much to coarticulate if speaking to someone with a different dialect? "Whether bearing lexical or sentence stress, stressed vowels are encroached on less by coarticulatory effects from transconsonantal vowels than are unstressed, more reduced vowels (Fowler, 1981)." -> In the end, this was basically the only information we got about stress. I wonder if there are any other known/relevant effects of stress (for example on consonants). Recasens writes that "According to electroglottographic data for Catalan, where the voiced consonant trigger may also be a sonorant, regressive voicing is less when the syllable-final target consonant is a fricative versus a stop." I thought this was curious because as a speaker of Catalan, I've learned the phonotactic "rule" that a word-final consonant is voiced when followed by a vowel if it is a fricative (e.g., golf estret -> golvestret) whereas it is voiceless if it is a stop (e.g., àrab escrit -> arapescrit). This pattern seems to go against that described in the quote from the paper, which is admittedly about consonant- and not vowel-triggered voicing. I'm wondering if there's a phonetic explanation. The paper notes that coarticulatory mechanisms involved in speech production and speech perception might vary across individual speakers. To what extent can these differences complicate communication? How might the coarticulatory patterns of a native language influence the way second-language learners acquire phonemes in a new language? That is, could these native coarticulatory influences contribute to a lasting "accent" in the second language? The assigned research summary outlines how certain phonemes exhibit more coarticulatory resistance than others, depending on phonetic context. Can focusing on coarticulatory resistance help improve models of speech perception in noisy environments, where subtle articulatory cues might be less detectable? In section 4 of Recasens(2018), it is stated that certain coarticulation increases together with the increase of the speech rate. Would this also entail a decrease in coarticulation if someone would speak slower than the average speech rate? Or are there also coarticulatory effects that specifically increase/appear if language is spoken much slower than usual?