Testing the Limits of Long‐Distance Learning: Learning Beyond a Three‐Segment Window

Abstract

Traditional flat-structured bigram and trigram models of phonotactics are useful because they capture a large number of facts about phonological processes. Additionally, these models predict that local interactions should be easier to learn than long-distance ones since long-distance dependencies are difficult to capture with these models. Long-distance phonotactic patterns have been observed by linguists in many languages, who have proposed different kinds of models, including feature-based bigram and trigram models, as well as precedence models. Contrary to flat-structured bigram and trigram models, these alternatives capture unbounded dependencies because at an abstract level of representation, the relevant elements are locally dependent, even if they are not adjacent at the observable level. Using an artificial grammar learning paradigm, we provide additional support for these alternative models of phonotactics. Participants in two experiments were exposed to a long-distance consonant harmony pattern in which the first consonant of a five-syllable word was [s] or [∫] ('sh') and triggered a suffix that was either [−su] or [−∫u] depending on the sibilant quality of this first consonant. Participants learned this pattern, despite the large distance between the trigger and the target, suggesting that when participants learn long-distance phonological patterns, that pattern is learned without specific reference to distance.

The privileged status of locality in consonant harmony

“…In the present experiment, participants generalized from a non-local pattern to a local pattern, but it is unclear whether participants learned a general non-local pattern that would extend to distances greater than a syllable. Recent research has shown that it is possible to learn consonant harmony patterns with two intervening syllables, suggesting that the consonant harmony pattern used in the present study is not limited to short words (Finley, submitted). Future research will continue to investigate the role of distance in non-local dependencies.…”

Learnable vs. Unlearnable Harmony Patterns

“…A body of research has shown that both children and adults can learn phonotactic patterns by extracting the regularities exhibited in an artificial language (e.g., Infant studies: Fisher 2003, Seidl et al 2009; Adult studies: Dell et al 2000, Onishi, Chambers, and Fisher 2002, Wilson 2003, Goldrick 2004, Peperkamp, Skoruppa, and Dupoux 2006, Finley and Badecker 2009a,b, Finley 2011, 2012, Koo and Callahan 2012. Each pattern examined in the previous studies falls into one of two categories: adjacent dependencies or nonadjacent dependencies.…”

“…Nonadjacent dependencies are also readily learned (Pycha et al 2003, Wilson 2003, Newport and Aslin 2004, Onnis et al 2005, Finley and Badecker 2009a,b, Finley 2011, 2012. The patterns tested in these studies require some segment x to agree with some segment y, and the two agreeing segments can be separated by a number of nonparticipating segments.…”

An onset is an onset: Evidence from abstraction of newly-learned phonotactic constraints

“…For instance, English-speaking adults can learn a consonant harmony pattern based on non-adjacent consonants, and generalize it to different word positions (e.g., based on whether a word started with either the sound s or sh , participants predicted the suffix to be either su or shu ; Finley, 2012), suggesting that participants were able to encode information about patterns occurring in structures larger than the syllable.…”

“…On the one hand, there is evidence that the syllable is a unit that is readily used when learning about speech (e.g., Fowler et al, 1993; Mattys & Melhorn, 2005; Sevald et al, 1995; Treiman, Fowler, Gross, Berch, & Weatherston, 1995). On the other hand, there are suggestions that syllable-based units are not necessary (e.g., Endress & Mehler, 2010) or sufficient for phonotactic learning (e.g., Cutler et al, 1986; Finley, 2012; Fowler et al, 1993; Hay et al, 2004; Kabak & Idsardi, 2007; Mattys et al, 1999; Richtsmeier et al, 2009; Seidl & Buckley, 2005; Steriade, 1999; Treiman & Danis, 1988; Treiman & Zukowski, 1990). The availability of multiple levels of representation (e.g., syllable and word) might help explain why natural phonotactic constraints can be substantially, though not fully, described by reference to syllable structure.…”