Melinda Fricke's research while affiliated with University of Pittsburgh and other places

Publications (10)

Preprint
Full-text available
Natural language processing (NLP) models trained on people-generated data can be unreliable because, without any constraints, they can learn from spurious correlations that are not relevant to the task. We hypothesize that enriching models with speaker information in a controlled, educated way can guide them to pick up on relevant inductive biases....
Article
Full-text available
Previous research has shown that as the level of background noise increases, auditory word recognition performance drops off more rapidly for bilinguals than monolinguals. This disproportionate bilingual deficit has often been attributed to a presumed increase in cross-language activation in noise, although no studies have specifically tested for s...
Article
Full-text available
Although variation in the ways individuals process language has long been a topic of interest and discussion in the psycholinguistic literature, only recently have studies of bilingualism and its cognitive consequences begun to reveal the fundamental dynamics between language and cognition. We argue that the active use of two languages provides a l...
Article
Full-text available
This paper presents evidence that 'spirantization', a cross-linguistically common lenition process, affects English listeners’ ease of segmenting novel “words” in an artificial language. The cross-linguistically common spirantization pattern of initial stops and medial continuants (e.g. [gußa]) results in improved word segmentation compared to the...
Article
Structural priming has played an important role in research on both monolingual and bilingual language production. However, studies of bilingual priming have mainly used priming as an experimental tool, focusing on cross-language priming between single-language sentences, which is a relatively infrequent form of communication in real life. We inves...
Article
During language production planning, multiple candidate representations are implicitly activated prior to articulation. Lexical representations that are phonologically related to the target (phonological neighbours) are known to influence phonetic properties of the target word. However, the question of which dimensions of phonological similarity co...
Article
We exploit the unique phonetic properties of bilingual speech to ask how processes occurring during planning affect speech articulation, and whether listeners can use the phonetic modulations that occur in anticipation of a codeswitch to help restrict their lexical search to the appropriate language. An analysis of spontaneous bilingual codeswitchi...
Article
Three groups of native English speakers named words aloud in Spanish, their second language (L2). Intermediate proficiency learners in a classroom setting (Experiment 1) and in a domestic immersion program (Experiment 2) were compared to a group of highly proficient English–Spanish speakers. All three groups named cognate words more quickly and acc...
Article
In the last two decades, there has been an upsurge of research on bilingualism recognizing that bilinguals may be more representative language users than their monolingual peers (e.g., Kroll, Dussias, Bogulski, & Valdes Kroff, 2012). The excitement about bilinguals is related not only to their neglected status in the past literature but also to a s...

Citations

... While certain aspects of language use are nearly universal -such as basic grammatical acceptability (Warstadt et al., 2019;Linzen and Baroni, 2021) or simple lexical judgements (Wang et al., 2019) -these often seem to be the exception that proves the rule. Contextual variation is ubiquitous in language, where predictions may differ as a function of speaker identity (Blodgett et al., 2016;Yang and Eisenstein, 2017;Ostapenko et al., 2022), time (Lazaridou et al., 2021;Sawhney et al., 2020;Schlechtweg et al., 2019), or usage domain (Dai et al., 2020;Nguyen et al., 2020;Lee et al., 2020). Although such variation has long been recognized in psycholinguistics (Clark, 1998) and sociolinguistics (Eckert, 2012), the dominant approach in NLP has been to train monolithic models (Flek, 2020;Hovy, 2015) and fine-tune for individual domains if necessary (e.g. ...
... (Davies, 2008) An awake patient is more complicated that an asleep patient. (Davies, 2008) Such variation may not be random but may instead reflect divergent and systematic tendencies of how speakers engage linguistic and cognitive processes (Fricke et al., 2019;Green et al., 2006). However, little is known about the underlying processes that regulate such choices in production. ...
... In Experiment 3, we model these lenition patterns using a series of deep neural networks and find that, even with limited training data, we can achieve reasonably high accuracy in the automatic categorization of lenition patterns. The results of this research both complement recent work on the phonetics of lenition in the world's languages (Katz and Fricke, 2018;White et al., 2020) and provide computational tools for modeling and predicting patterns of extreme lenition. ...
... Previous studies on the code-switching patterns between typologically distant languages (languages that demonstrate systematic discrepancies in their syntactic, morphological, phonological, and phonetic rules or inventories) have used various methodologies to explore topics beyond sociolinguistics-a common area of focus of code-switching studies-which 2 syntactic structures trigger code-switching (Fricke & Kootstra 2016), how the morphological structures of the two languages are handled near code-switches (Myer-Scotton 2002; Stefanich et al. 2019), how the phonetic details of one language shifts toward the other language near code-switches (Balukas & Koops 2014;Goldrick et al. 2014;Olson 2013;Olson 2016b;Piccinini & Arvaniti 2015;Simonet 2014), and how bilingual speakers perceive code-switched words (Li 1996;Grosjean 1988;Olson 2016a;Piccinini & Garellek 2014). However, it remains unclear whether the findings of the studies on the typologically distinct language pairs would extend to typologically similar languages that share common syntactic structures, morphological constructions, phonological inventories, and even phonetic properties. ...
... CLEARPOND for English (Marian et al., 2012) was used to obtain English lexical frequency rates of the target words, initial plosive and initial bigram frequency rates. In addition, neighborhood density (i.e., number of phonological neighbors of the English target words differing by one phoneme) was calculated, as this variable plays an important role in lexical processing of first and foreign languages (Fricke, Baese-Berk, & Goldrick, 2016). ...
... Consistently, in spontaneous speech, switch costs were also typically paid before switch words were produced. For example, Fricke et al. (2016) divided the Bangor Miami Corpus into a series of "utterances" that consist of one main clause in each and measured the average syllable production duration of each utterance. In the corpus, bilinguals may switch in both directions, and the results showed significantly longer syllable production duration in utterances before language switching occurred in general. ...
... Cognates are also named faster and with fewer errors in naming tasks (Jacobs et al., 2016;Schwartz & Kroll, 2006). ...
... Indeed, it was Grosjean (1989) with his denial of the definition of bilingual as two monolinguals in one person, who triggered the series of actions that lead to understanding language attrition as a sole phenomenon that occurs as a result of the joint existence and the continuous interaction of two or more languages. As a result, many scholars (Cook, 2003;Grosjean, 2016;Kroll & Fricke, 2014;Yılmaz & Schmid, 2018, among others) have argued that L1 attrition is an expected and reasonable phenomenon in multilinguals. Indeed, individuals that are able to communicate in more than one language have a linguistic supersystem (Cook, 2003) that shows the reciprocal influence between languages. ...