In a previous post I committed to doing blog posts on failed conditions. One reason behind this was to be part of the solution to the hidden data problem that exists in psychology, and by extension, cognitive science. But not all hidden data are failures to find a relationship, some hidden data come from studies that have never been written up, and may never be due to time constraints. (That is a whole other problem and one I’m not going to tackle here.) In an effort to get some of those data out too, I’ve decided to post papers that for one reason or another are unlikely to ever be published.
This is the first such post. It’s a paper that resulted in part from an undergraduate honors thesis by Jessica Morrison at UC Berkeley. The paper was written up in late 2008 early 2009. It was submitted somewhere (I can’t remember where). It was rejected, but we got comments that could have been useful for reworking the paper a bit before sending it off somewhere else. Jessica left academia, and despite plans to send the paper elsewhere, it was not my highest priority. So the paper languished and now the citations are out of date. I have faced the fact that I will always have other things on my plate that are higher priority than this paper. But I think it’s a neat little project, and someone out there might find it interesting or useful. So here it is. I haven’t done anything to the manuscript other than add a statement about copyright, and put the tables and figures into the text. So it’s not the prettiest document, but it is now available, which is something it wasn’t when it was just sitting on my hard drive. And hey, it’s also open access! Just click on the highlighted names after the title to get a copy of the paper.
Title: Phonological form influences memory for form-meaning mappings in adult second-language learners. Morrison & Hudson Kam (2009)
Abstract: This study asks whether phonological form affects adult second language learners’ ability to learn the meanings of novel words. Specifically, we ask whether hard-to-pronounce words, defined as having phones/phone combinations not present in the learner’s native language, are more difficult to learn meanings for, and further, if learnability differences are due to interference from production problems or more general representational difficulties. We exposed participants to easy- and hard-to pronounce novel word-novel object pairings and tested their memory for the pairings. Participants who had either repeated words aloud, performed subvocal repetition, or heard another learner’s attempts to repeat the words during exposure performed worse on hard-to-pronounce words when tested immediately after exposure. When tested the following day, all participants, regardless of exposure condition, showed the effect. In a follow-up experiment, participants who engaged in an articulatory suppression task during learning did not have more difficulty with hard-to-pronounce words, suggesting that differences cannot simply be due to interference. Rather, we suggest that more difficult phonological forms lead to weaker representations which are then more difficult to link up with meanings in memory.