About

Joseph Casillas is an Associate Professor at Rutgers University, holding a Ph.D. from the University of Arizona. He is part of the Department of Spanish and Portuguese and is based in the Academic Building-CAC at Rutgers' New Brunswick campus. His contact information includes an office at AB-5181, available for office hours on Monday from 1:00 to 2:00 pm and Wednesday from 11:30 am to 12:30 pm, with additional availability via Zoom or by appointment. He can be reached by phone at 848-932-6941 or through his website at https://www.jvcasillas.com/. The page does not specify his research focus, key contributions, or detailed background beyond his academic affiliation and educational qualification.

Research topics

• Computer Science
• Psychology
• Political Science
• Linguistics
• Sociology
• Data science
• Mathematics
• Medicine
• Artificial Intelligence
• Statistics
• Econometrics
• Cognitive psychology
• Engineering ethics
• Public relations
• Engineering
• Epistemology

Selected publications

• Opening open science to all: Demystifying reproducibility and transparency practices in linguistic research

  2025-04-17

  preprintOpen access1st authorCorresponding

  In recent years, numerous fields of research have seen a push for increased reproducibility andtransparency. As a result, specific transparency practices have emerged, such as open accesspublishing, preregistration, sharing data, analyses, and code, performing study replications, anddeclaring positionality and conflicts of interest. While many agree that open science practicesrepresent a positive step forward in improving scientific rigor, these practices, by and large, havenot been adopted in the field of linguistics (Bochynska et al., 2023). Few, if any, researchershave had explicit instruction on the practices of open science as part of their professionaltraining. Nonetheless, today’s speech researcher is expected to be up to date on the currentprotocols of open science in order incorporate the methodological practices aimed at improvingreproducibility/replicability. The present work intends to help make open science practicesunderstandable and accessible to researchers in linguistics from all backgrounds and at everystage, from students/early career researchers to senior researchers and advisors. We outline eightspecific open science practices that linguists can adopt to make their research more open,transparent, inclusive, and accessible to a wider audience.

  PublisherOA PDFDOI

• Bayesian Statistics

  The Encyclopedia of Applied Linguistics · 2025-12-02

  other1st authorCorresponding

  Abstract Over the past two decades, advances in computational power have facilitated the use of complex statistical modeling in academic research. As a result, sophisticated analytic strategies, such as multilevel modeling and structural equation modeling, are becoming commonplace, and computationally costly procedures, like Bayesian data analysis, have moved to the forefront. In particular, the use of Bayesian methods is gaining ground in the field of applied linguistics. The present overview provides a brief introduction to Bayesian data analysis, covering both why and how one might decide to use Bayesian estimation for statistical modeling, along with some of the commonly referenced advantages and disadvantages of this approach. This primer serves as a starting point for any linguist interested in beginning the journey of understanding and implementing Bayesian methods.

  PublisherDOI

• Editorial: Experimental approaches to the acquisition of information structure

  Frontiers in Psychology · 2025-06-17

  editorialOpen access

  Despite the surge, investigating how information structure is acquired remains in its early stages, with uneven coverage across populations and languages. As shown in Figure 2, the representation of individual languages in our survey follows a power-law distribution, with 52% of the articles focusing on just four languages. Only 25% of articles in our sample include at least one term related to acquisition or bilingualism.foot_0 The contributions to this research topic address these lacunae by expanding the cross-linguistic scope, incorporating data from child L1 acquirers, L2 and heritage bilinguals, and contexts of societal multilingualism, and utilizing both traditional and innovative methods.Lozano and Quesada use CEDEL2 corpus texts to examine anaphora resolution in Spanish native speakers and English-speaking Spanish learners. Their findings challenge the Position of Antecedent Strategy (Carminati, 2002) as the default strategy, showing anaphora resolution is more complex than experimental data suggests, with overt pronouns rarely used and often substituted by repeated noun phrases.Uth, Gutiérrez-Bravo, and Fleissbach demonstrate, using an oral production task and a corpus study, that focus in Yucatec Maya is incompatible with progressive aspect marking. Appealing to a semantic account, they argue that progressive aspect blocks focus fronting because the marker itself functions as a type of focalization.Seraye Alseraye examines how incomplete speech representations affect processing of garden path sentences in L2 Arabic, finding faster reading times in unambiguous contexts and when disambiguating segmental information is present. Overall comprehension remained unaffected, even in the presence of incorrect disambiguating information. The study supports the 'good-enough' model of language processing (Ferreira, et al., 2009) among L2 learners of an understudied language.Slioussar and Harchevnik explore how L1 Russian speakers and Mandarin Chinese L2 Russian learners process SVO and OVS word orders. Using online (reading times) and offline (sentence ratings) tasks, they show that both groups benefit from given-before-new structures, although L2 learners struggle more with processing non-canonical word orders and are less sensitive to discourse constraints.Lorenzen, Roessig, and Baumann employ a novel paradigm-an interactive reading task-to increase the ecological validity of spoken data. They examine how information status affects prosodic prominence in German, finding that paradigmatic effects appear mainly in F0, while syntagmatic effects vary across speakers and depend on the specific acoustic parameter.Destruel, Lalande, and Chen investigate the acquisition of French prosody using a virtual robot-mediated picture-matching task. Unlike younger children, 7-to 8-year-olds and adults use prosody to distinguish focus from non-focus. Furthermore, this study finds subject-object asymmetries, attributed to the dominant use of syntactic strategies for subject focus in French.Yang, Cho, Kim, and Chen examine how young children acquire prosodic phrasing to mark focus in Korean. Using a picture-matching task, they find that children (ages 4-5) pattern like adults in distinguishing narrow from broad focus and prefocal material, but not from postfocal material or contrastive focus. By age 11, patterns are adult-like, with acquisition speed linked to form-meaning transparency.Smeets uses two tasks to test clitic-doubled left dislocation in Romanian, which has received less attention than other Romance languages. The finding that L1 Romanian speakers who learned L2 Italian show attrition-unlike those who learned L2 English-highlights the role of L1-L2 similarity in reshaping L1 information structure via feature reassembly.Luchkina, Ionin, and Goldshtein used two aural identification tasks (with and without contexts) to investigate how English-Russian heritage bilinguals process Russian non-contrastive focus, examining constituent order and prosodic cues. While higher-proficiency heritage speakers patterned with native speakers, the group overall tended to assign focus to nouns with nuclear stress in SVO orders-unlike native speakers-which highlights the challenges external interface structures pose (Sorace, 2011).Neocleous and Sitaridou examine information-structural reflexes of contact between VO and OV languages. Romeyka, an Asia-Minor Greek variety (VO), has coexisted alongside Turkish (OV) for centuries. As a result, left peripheral configurations like focus movement occur in a wider range of contexts than in other Greek varieties.Each article fills the literature gaps we identified, offering directions for future research to build on. At the methodological level, a key desideratum in information structure research is to improve the ecological validity of data, minimizing lab speech artefacts. Several contributions address this by proposing novel experimental designs (e.g., Lorenzen, Roessig, and Baumann) or combining experimental and observational research (e.g., Uth, Gutiérrez-Bravo, and Fleissbach). We envision future studies in which these avenues will be further pursued.Another major challenge in studying information structure is disentangling the roles of different linguistic layers involved in its expression. The interplay between syntax and prosody in particular is central to several contributions. Destruel, Lalande, and Chen examine the syntax-prosody complementarity in French focus expression, while Luchkina, Ionin, and Goldshtein investigate how prosodic and syntactic cues contribute to focus processing in Russian. We see a continued need for such nuanced, multi-layered approaches to the cross-linguistic inventory encoding these distinctions.Finally, studying different populations beyond literate adult monolinguals-such as naturalistic and instructed bilinguals, L1 acquirers at different stages, and speakers of vernacular varieties-is imperative to understanding how grammars vary within and across languages. Some contributions show effects on attrition (Smeets) or adaptation under language contact (Neocleous and Sitaridou), while others reveal particular processing challenges in L2 learners and other bilinguals (Slioussar & Harchevnik). These findings enrich broader discussions on how dynamic processes like acquisition and language contact shape the representation and processing of information structure across diverse linguistic systems.

  PublisherOA PDFDOI

• Opening open science to all: demystifying reproducibility and transparency practices in linguistic research

  Linguistics · 2025-07-01

  articleOpen access1st authorCorresponding

  Abstract In recent years, numerous fields of research have seen a push for increased reproducibility and transparency. As a result, specific transparency practices have emerged, such as open access publishing, preregistration, sharing data, analyses, and code, performing study replications, and declaring positionality and conflicts of interest. While many agree that open science practices represent a positive step forward in improving scientific rigor, these practices, by and large, have not been adopted in the field of linguistics (Bochynska et al. 2023. Reproducible research practices and transparency across linguistics. Glossa Psycholinguistics 2(1). 1–36). Few, if any, researchers have had explicit instruction on the practices of open science as part of their professional training. Nonetheless, today’s speech researcher is expected to be up to date on the current protocols of open science in order to incorporate the methodological practices aimed at improving reproducibility/replicability. The present work intends to help make open science practices understandable and accessible to researchers in linguistics from all backgrounds and at every stage, from students/early career researchers to senior researchers and advisors. We outline eight specific open science practices that linguists can adopt to make their research more open, transparent, inclusive, and accessible to a wider audience.

  PublisherOA PDFDOI

• Sound communities: A quantitative proposal for studying bilingual speech

  2025-03-03 · 1 citations

  preprintOpen access1st authorCorresponding

  Bilingualism researchers have intensively studied how learning and using multiple languagesaffects all levels of linguistic structure. In this strand, examining diversity in the bilingualexperience and the extent to which variables like language dominance regulate crosslinguisticinteraction has been of special interest. However, most studies sample small groups of bilingualsfrom a single research site, creating a twofold generalizability problem. First, with small samplesit is unlikely that researchers will be able to fully capture and quantify the range of variablesknown to affect findings. Second, when bilinguals are recruited from a single site, it isimpossible to determine if findings are site-specific or apply to bilinguals more broadly. Toaddress these issues, we propose a large(r)-scale, multisite approach to bilingualism research. Webelieve that such an approach, when informed by open science practices, has the potential tosignificantly advance the state of the art.

  PublisherOA PDFDOI

• Sound communities: A quantitative proposal for studying bilingual speech

  2025-11-12

  articleOpen accessSenior author

  Bilingualism researchers have intensively studied how learning and using multiple languagesaffects all levels of linguistic structure. In this strand, examining diversity in the bilingualexperience and the extent to which variables like language dominance regulate crosslinguisticinteraction has been of special interest. However, most studies sample small groups of bilingualsfrom a single research site, creating a twofold generalizability problem. First, with small samplesit is unlikely that researchers will be able to fully capture and quantify the range of variablesknown to affect findings. Second, when bilinguals are recruited from a single site, it isimpossible to determine if findings are site-specific or apply to bilinguals more broadly. Toaddress these issues, we propose a large(r)-scale, multisite approach to bilingualism research. Webelieve that such an approach, when informed by open science practices, has the potential tosignificantly advance the state of the art.

  PublisherDOI

• Unraveling the Complexities of Second Language Lexical Stress Processing: The Impact of First Language Transfer, Second Language Proficiency, and Exposure

  Language Learning · 2024-01-16 · 5 citations

  articleOpen accessSenior author

  Abstract We investigated the role of cue weighting, second language (L2) proficiency, and L2 daily exposure in L2 learning of suprasegmentals different from the first language (L1), using eye‐tracking. Spanish monolinguals, English–Spanish learners, and Mandarin–Spanish learners saw a paroxytone and an oxytone verb (e.g., FIRma–firMÓ “s/he signs – signed”), listened to a sentence containing one of the verbs, and chose the one that they heard. The three languages have contrastive lexical stress, but suprasegmentals have a greater functional load in Mandarin than in English. Monolinguals predicted suffixes accurately with both stress conditions and favored oxytones, but learners predicted suffixes accurately only with oxytones, the condition activating fewer lexical competitors. Monolinguals predicted suffixes accurately sooner but at a slower rate than did learners. L2 proficiency, but not L1 or L2 exposure, facilitated L2 predictions. In conclusion, learners of a contrastive‐stress L1 rely on L2 suprasegmentals to the same extent as monolinguals, regardless of their L1. Lower L2 proficiency and higher cognitive load (more lexical competitors) reduce learners’ reliance on suprasegmentals.

  PublisherOA PDFDOI

• Sound communities: A quantitative proposal for studying bilingualism in context

  2024-07-08

  preprintOpen accessSenior author

  Bilingualism researchers have intensively studied how learning and using multiple languagesaffects all levels of linguistic structure. In this strand, examining diversity in the bilingualexperience and the extent to which variables like language dominance regulate crosslinguisticinteraction has been of special interest. However, most studies sample small groups of bilingualsfrom a single research site, creating a twofold generalizability problem. First, with small samplesit is unlikely that researchers will be able to fully capture and quantify the range of variablesknown to affect findings. Second, when bilinguals are recruited from a single site, it isimpossible to determine if findings are site-specific or apply to bilinguals more broadly. Toaddress these issues, we propose a large(r)-scale, multisite approach to bilingualism research. Webelieve that such an approach, when informed by open science practices, has the potential tosignificantly advance the state of the art.

  PublisherDOI

• A tutorial on generalised additive mixed effects models for bilingualism research

  Linguistic Approaches to Bilingualism · 2024-11-29 · 6 citations

  articleOpen accessSenior author

  Abstract While recent years have seen a shift towards random effects modelling, particularly in areas of linguistics in which nested structure is the norm (e.g., trial repetitions nested within participants), an over-reliance on standard linear modelling prevails, particularly in the cases of dynamic phenomena that may not constitute a linear relationship, e.g., vowel trajectories, pitch contours, acquisition processes, etc. Generalised Additive (Mixed) Models (GAMMs) are now commonly employed in phonetic research (given the naturally dynamic nature of speech data) and this is reflected by the availability of several tutorials which focus on phonetic data. This tutorial aims at making GAMMs accessible to researchers from other fields within linguistics. In particular, this tutorial is written for researchers in bilingualism and multilingualism who wish to be able to start using GAMMs for non-linear data, which is very common in developmental and learning phenomena. While only the basics will be covered here, we hope that researchers will get the necessary foundations to be able to learn GAMMs from existing resources.

  PublisherDOI

• The Speech Production of Bilingual Adults

  Cambridge University Press eBooks · 2024-11-14

  book-chapter1st authorCorresponding
  PublisherDOI

Frequent coauthors

• Timo B. Roettger

  53 shared

• Caitlin Halfacre

  Newcastle University

  52 shared

• Erin Michelle Buchanan

  52 shared

• Melanie Röthlisberger

  University of Zurich

  51 shared

• Kaidi Chen

  University of Oslo

  51 shared

• Agata Bochyńska

  University of Oslo

  51 shared

• Liam Keeble

  Harrisburg University of Science and Technology

  51 shared

• Nicole Rodríguez

  Universidad del Sinú

  17 shared

Education

• Ph.D.

  University of Arizona