Replication of Finn & Hudson Kam (2008) The curse of knowledge: First language knowledge impairs adult learners’ use of novel statistics for word segmentation, Exp. 1

Abstract

We replicated Exp. 1 of Finn & Hudson Kam (2008) The curse of knowledge: First language knowledge impairs adult learners’ use of novel statistics for word segmentation, Cognition, 108, 477-499, as part of a multi-year effort to replicate every adult statistical word segmentation study. Unlike the original study, we do not find clear evidence that English-speaking adults fail to successfully segment words that violate the phonotactic constraints of English.

Second Replication of Finn & Hudson Kam (2008) Experiment 1

“…A previous replication attempt failed to replicate this finding (Garcia, van Horne, & Hartshorne, 2017). Our goal here is to attempt to replicate with a larger sample size and hence greater statistical power.…”

Robustness of the adult statistical word segmentation literature: Part 1

“…Either way, the differences in results would be of significant theoretical importance. We have exhaustively documented all such methodological decisiones in individual preprints for each experiment, allowing the reader to judge for themselves and design follow-up experiments if needed (Garcia, Iozzo, et al, 2017;Garcia, van Horne, & Hartshorne, 2017;Hartshorne & Skorb, 2018;Iozzo et al, 2017;Mu et al, 2017;Murdock et al, 2017a,b). Note that our materials, data, and code are available as well (see preprints for details).…”

“…Exhaustive descriptions of each experiment are available in individual preprints posted on PsiArXiv (Garcia, Iozzo, et al, 2017;Garcia, van Horne, & Hartshorne, 2017;Hartshorne & Skorb, 2018;Iozzo et al, 2017;Mu et al, 2017;Murdock et al, 2017a,b). These write-ups include additional methodological and statistical details and provide links to Open Science Framework repositories that contain all data, materials, and code.…”

“…Similarly, the original enforced a five minute break between training segments, whereas ours was self-paced. An exhaustive comparison of the methods in the three replications and the original can be found in the preprints (Garcia, Iozzo, et al, 2017;Garcia, van Horne, & Hartshorne, 2017;Hartshorne & Skorb, 2018;Iozzo et al, 2017;Mu et al, 2017;Murdock et al, 2017a,b).…”

What happens to large changes? Saltation produces well-liked outputs that are hard to generate

“…However, it also leaves the observed biases susceptible to explanations based on first language phonological experience rather than differences in change magnitudes (see also Skoruppa et al, 2011, for similar concerns regarding their alternations; though cf. Garcia, van Horne, & Hartshorne, 2017;Mitrović, 2012;Wang & Saffran, 2014; for evidence against first-language transfer in miniature artificial language learning). In particular, English has alveolar palatalization patterns that are productive in specific contexts: before glides in frequent phrases like would you and bet you and in words like creature (cf.…”