Current Projects

L1 vs. L2 attrition

Investigator: Jennifer Cabrelli Amaro

The Phonological Permeability Hypothesis accepts the notion of a critical period for optimal phonological acquisition and posits that adult L2 phonological systems, even when they appear native-like, are representationally different than native systems. However, unlike previous investigations of a critical period that center around ultimate attainment in the L2, the PPH defines the critical period in terms of differential stability of the phonological system, predicting that a system acquired in adulthood should be more vulnerable to influence from a third language than a system acquired in childhood. Within an Optimality Theoretic framework, we examine the stability of constraint rankings in grammars acquired in adulthood. Currently, we are testing the descriptive and explanatory adequacies of strong and weak versions of the hypothesis across the domains of phonology and syntax. We have initiated a longitudinal study of multiple phonological phenomena that present differently in Spanish and Brazilian Portuguese to determine whether the constitution of these two types of systems is fundamentally different, or if domain general mechanisms such inhibitory control might explain potential permeability. Longitudinal observation crucially allows for the control of individual variation, such that participants act as their own encapsulated control.

L1 attrition of perception and production of illusory vowels
Investigators: Jennifer Cabrelli Amaro, Elizabeth Goodin-Mayeda (University of Houston)

This study investigates L1 modification in the perception of illusory vowels by L1 Brazilian Portuguese (BP)/L2 English speakers. (C)VC syllables with a coda stop consonant violate syllabic structure constraints in BP, but are licit in English. As a result, BP speakers perceive an illusory /i/ between illicit consonant clusters and produce an epenthesized /i/ after coda stop consonants in word-medial and final position. Existing perceptual attrition studies indicate that perception of segmental phenomena is vulnerable to attrition while perception of suprasegmental phenomena may not be. However, effects on L1 perception when the L2 phonology allows a syllabic structure that is illicit in the L1 are unknown. To determine the degree of the illusory effect due to phonotactics, we implement vowel identification and ABX perception tasks as well as a concatenation production task.

Regressive transfer in L3 syntactic development
Investigator: Jennifer Cabrelli Amaro

This study investigates the extent to which early-acquired versus late-acquired (i.e., adult) syntactic systems resist influence from a third language (L3). Based on the tenets of the Phonological Permeability Hypothesis, we test the hypothesis that adult-acquired syntactic systems, even when evidencing native-like representations, are different from early-acquired systems with regards to the stability of the grammar. We investigate the effects of Brazilian Portuguese (BP) on Spanish feature configurations in two groups of English/Spanish bilinguals that are advanced L3 BP speakers. The first group is made up of sequential L1 English/adult L2 Spanish speakers, while the second group consists of L1 Spanish/L2 adult English speakers. To test our hypothesis, we examine the phenomenon of raising across an experiencer (RExp). Spanish and BP differ with regard to RExp; while BP allows RExp due to defective T, Spanish does not. Data from a grammaticality judgment task will allow us to determine 1) whether both experimental groups accept RExp in BP, i.e., they have converged on the L3 target and the BP configuration is available for transfer, and 2) whether the late Spanish learners are more accepting of RExp than the early Spanish learners and the Spanish controls, while the early Spanish learners do not differ from the control group.

The roles of the L1 and L2 in L3 morphosyntactic development
Investigator: Jennifer Cabrelli Amaro

Much recent L3 initial stages research points to the transfer of a single system that is determined by typological proximity of the L1 vs. L2 to the L3. Under such a scenario, non-facilitative transfer is expected to occur to the extent that the language selected for transfer—the L1 or L2— differs from the L3. It is conceivable that non-facilitative transfer from the L2 creates an L3 learning task of feature reconfiguration that is qualitatively different than the learning task that transfer from the L1 creates. With this in mind, the aim of this study is to identify potential differences in the reconfiguration process (developmental sequence) between learners that have transferred their L1 versus their L2.  Specifically, we examine whether reconfiguration will be slower for learners that have transferred their L1. We follow the hypothesis that L1 dominance impedes the occurrence of parsing failures, which stimulate reconfiguration. To test the hypothesis, we examine reconfiguration of the featural specification of embedded T in English/Spanish bilingual learners of L3 Brazilian Portuguese (BP). for English/Spanish bilinguals, only the English configuration will facilitate L3 BP acquisition. We compare initial stages data from a grammaticality judgment task with data from advanced L3 BP learners to determine whether the rate of reconfiguration is slower for L1 Spanish speakers, i.e., whether transfer from an L1 takes longer to undo.

L3 phonology: Initial stages transfer and early development
Investigator: Jennifer Cabrelli Amaro

A growing body of research indicates that structural similarity drives transfer of one system over another at the L3 initial stages (see Rothman, 2015, for an overview), with a particularly robust effect for English/Spanish bilinguals acquiring L3 Portuguese. However, there is a dearth of evidence from the phonological domain. This study aims to fill that gap via examination of several phonological phenomena that differ between Portuguese and Spanish (nasal vowels and nasalization, mid vowel contrasts, postvocalic stop lenition, word-final vowel reduction, and coda repair strategies). We follow L3 Portuguese learners throughout a semester of Portuguese for Spanish speakers. Learners complete a delayed repetition task four weeks into the semester to determine initial transfer of English or Spanish, and again in week 16 to track development in the case of initial non-facilitative transfer. 

The Acquisition of the Spanish palatal nasal
Investigators: Sara Stefanich, Jennifer Cabrelli Amaro

Description: The purpose of this project is to determine whether formation of a new category for the Spanish palatal nasal /ɲ/ is blocked by equivalence classification in L2 Spanish development.  Both the English and Spanish production of beginner and advanced L1 English/L2 Spanish learners is examined, with respect to the Spanish palatal nasal, in order to determine if the learners approximate a target-like Spanish /ɲ/ or if they are instead employing a more English-like [n+j] sequence.

Word Internal Switches: An Acoustic Analysis
Investigators: Sara Stefanich, Kay E. González-Vilbazo, Jennifer Cabrelli Amaro

Description: This project investigates the claims made by MacSwan (2000,2005), Rao & DenDikken (2014)  and MacSwan & Colina (2014) that there can be no phonological switches within a word. An experimental methodology is employed in which Spanish/English bilinguals are encouraged to produce code-switched nonce words composed of an English root and Spanish affixes (verbal inflection). An acoustic analysis will be run in order to determine if these bilinguals produce the four target phonemes (list here) as part of the English root in the code-switched word, or if the bilinguals produce the entire code-switched word with Spanish phonology, thus invalidating or lending support to previous theoretical claims.

CLLD in L2 Spanish: Findings from Learners’ Production Data
Investigator: José Sequeros Valle,  Bradley Hoot (DePaul University), Jennifer Cabrelli Amaro

Description: This project examines the acquisition of the Spanish information structure by second language learners whose first language is English. The goal is to test the Interface Hypothesis at the syntax-discourse interface by looking at the left dislocation and the clitic-doubling required to produce a grammatical CLLD, and at the discourse context for which CLLD is appropriate. Unlike previous works on the same topic, this project analyzes utterances produced by learners instead of acceptability judgment tasks. In future stages, this project will use psycholinguistic methodologies since processing seems to be the problematic cognitive domain at the interfaces.