The role of consolidation in learning context-dependent phonotactic patterns in speech and digital sequence production

Abstract

Speakers implicitly learn novel phonotactic patterns by producing strings of syllables. The learning is revealed in their speech errors. First-order patterns, such as "/f/ must be a syllable onset," can be distinguished from contingent, or second-order, patterns, such as "/f/ must be an onset if the vowel is /a/, but a coda if the vowel is /o/." A metaanalysis of 19 experiments clearly demonstrated that first-order patterns affect speech errors to a very great extent in a single experimental session, but second-order vowel-contingent patterns only affect errors on the second day of testing, suggesting the need for a consolidation period. Two experiments tested an analogue to these studies involving sequences of button pushes, with fingers as "consonants" and thumbs as "vowels." The button-push errors revealed two of the key speech-error findings: first-order patterns are learned quickly, but second-order thumb-contingent patterns are only strongly revealed in the errors on the second day of testing. The influence of computational complexity on the implicit learning of phonotactic patterns in speech production may be a general feature of sequence production.

Rapid phonotactic constraint learning in aging: Evidence from speech errors

“…As such, explicit learning capacities seem to follow an inverse U-shaped pattern, whereas implicit learning skills remain relatively stable across age. Furthermore, some recent studies that used a non-linguistic variant of the paradigm also showed a relatively weak syllable-position effect on errors involving arbitrary finger movements in young adults (Anderson & Dell, 2018;Rebei, Anderson, & Dell, 2019). They discuss the possibility that these weak effects are caused by limited experience with the event schema.…”

“…For this, we calculated the proportion of same-position errors for each constraint type and day (depicted in Figure 1), on which we carried out non-parametric Wilcoxon signed-rank tests (cf. Anderson & Dell, 2018). To compare between groups, we used Mann-Whitney U-tests.…”

“…Phonotactic learning has been widely demonstrated with young adults and children (see Anderson & Dell, 2018;Smalle, Muylle, Szmalec, & Duyck, 2017), but, as far as we know, not with older adults. Yet, with advancing age, such skills remain essential to foreign language learning and educational training or rehabilitation programs (e.g., after stroke).…”

“…This is referred to as the syllable-position effect (Dell et al, 2000;Fromkin, 1971;Warker, Dell, Whalen, & Gereg, 2008). Anderson and Dell (2018) argued that this effect might reflect a short-term version of trial-specific first-order learning (e.g., /f/ is an onset right now, in the literature also referred to as local-positional constraints, see Anderson & Dell, 2018) and provides a baseline against which long-term, incremental learning of the second-order EWC consonants can be LANGUAGE LEARNING AND AGEING 6 assessed. In other words, if EWC consonants slip more often to the same position than UR consonants, this indicates long-term learning of the EWC above what can be explained by a syllable-position effect.…”

“…Importantly, whereas EWC consonants may initially slip to the same position as much as the UR consonants, Dell and colleagues observed that this probability changed and exceeded that of the UR condition with only limited exposure (i.e., from a first day of training for first-order EWC, from a second day of training for second-order EWC: Warker & Dell, 2006, Anderson & Dell, 2018. This indicates that the participants rapidly adapt to novel sound-combination rules within their native phonological system.…”