Is earlier always better for learning a second language in terms of cognitive development?

Material Information

Is earlier always better for learning a second language in terms of cognitive development?
Rietbergen, Marpessa
Publication Date:


Subjects / Keywords:
Adjectivals ( jstor )
Adult education ( jstor )
Adults ( jstor )
Child psychology ( jstor )
Cognitive linguistics ( jstor )
Foreign language learning ( jstor )
Grammatical constructions ( jstor )
Language ( jstor )
Learning ( jstor )
Second language learning ( jstor )
Child development
English language
Second language acquisition
Spanish language
Undergraduate Honors Thesis


The purpose of this paper is to reflect on cognitive development and its effects on child second language acquisition. In the works done by Rice (2001), she proposes an empirical question concerning whether or not older age of onset is better or worse for acquiring an L2. Studies conducted by Paradis (2010) similarly suggest that delayed onset produces comparatively sharper progression in acquisition than early onset. Paradis’ study contends that children of 6-7 years are not fully developed cognitively; therefore, she proposes unlike matured children, they do not have the metalinguistic advantages of acquiring a second language. The current study tests two age groups exposed to sequential bilingualism in a classroom environment on L2 passive constructions in Spanish and English. The objective is to collect data regarding the implications of age and cognitive maturation within second language learners. The results could aide in disputed questions relating to cognitive developmental sequences in child L2 acquisition, and ultimately gauging when best to introduce an L2. Is earlier always better for learning a second language? ( en )
General Note:
Awarded Bachelor of Arts; Graduated December 20, 2011 summa sum laude. Major: Linguistics
General Note:
Advisor: Jason Rothman
General Note:
College of Liberal Arts and Sciences

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Marpessa Rietbergen. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.


This item is only available as the following downloads:

Full Text


! 1 Is earlier always better for learning a second language in terms of cognitive development? Marpessa Rietbergen University of Florida : Department of Linguistics 12.5.11 Advisor : Dr. Jason Rothman Abstract: The purpose of this paper is to reflect on cognitive development and its effects on child second language acquisition. In the works done by Rice (2001), she proposes an empirical question concerning whether or not older age of onset is better or worse for acquiring an L2. Studies conducted by Paradis (2010) similarly suggest that delayed onset produces comparatively sharper progression in acquisition than early onset. Paradis' study contends that children of 6 7 years are not fully developed cognitively; th erefore, she proposes unlike matured children, they do not have the metalinguistic advantages of acquiring a second language. The current study tests two age groups exposed to sequential bilingualism in a classroom environment on L2 passive constructions i n Spanish and English. The objective is to collect data regarding the implications of age and cognitive maturation within second language learners. The results could aide in disputed questions relating to cognitive developmental sequences in child L2 ac quisition, and ultimately gauging when best to introduce an L2. Is earlier always better for learning a second language ? SLA and the Importance of Child L2 Acquisition There has been an influx of new and developing theoretical linguistic research in second language acquisition (SLA). The newly ignited interest in part can be argued to be a result of the educational framework many nations have adopted when introducing a sec ond language (L2) within a classroom environment. For the field of linguistics this has proved to be very stimulating because it has opened the way for numerous empirical questions concerning bilingualism and second language acquisition. Earlier investigat ions were built upon predicated notions and founded on monolingual acquisition, producing conclusions on second language acquisition from a formulaic and narrowed approach. In recent years many researchers have


! 2 begun broadening their approach by focusing o n new variables therefore, as a result their findings have aroused what was once a stagnant field in linguistics This paper will be approached from a generative standpoint, which presupposes that from birth individuals acquire their first language thr ough a specific language faculty called universal grammar (UG; Chomsky 1981; Montrul 2004). The UG provides language principles and parameters that are influenced by input allowing able minded individuals to create a grammar and ultimately the means to com municate to one another (Chomsky 1981; White 2004). This has led many empirical studies to question whether an individual can employ UG similarly when learning a second language, and if so to what extent can she or he access her or his UG? In previous year s many studies compared native (L1) performance with L2 learning individuals. The methodological implications that arise from constructing such a paradigm for researching SLA provide us with surfacing differences; however, we are left to solve for the u nderlying cognitive similarities; if any, between the L1 speaker and the L2 learner. The polarity in metalinguistic knowledge between L1 acquirers and L2 learners has resulted in an effort to bridge the cognitive gap between the two. According to Schwartz (2003) and Unsworth (2008), child L2 learners in effect could be the bridge that could potentially compare cognitive development between L1 acquirers and adult L2 learners. Schwartz (2003) proposes that assuming adult L2 learners use metacognition to help advance themselves in their second language, there is still the issue of the poverty of stimulus Adults are able to acquire L2 features absent in their L1 without exposure to prior input. When con sidering L2 children, out of economy they subconsciously transfer learned properties from their L1 to their L2. However, if their L1 is insufficient in terms of acquiring certain elements in the L2, they use UG to compensate for their L1 to provide for the L2 grammar (Paradis 2010b; 2010c; Schwartz 2003).


! 3 Child L2 learners are not fully cognitively mature because they have yet to reach puberty. Although children as young as three have had substantial L1 input, theoretically full utility or transfer of th eir L1 will not occur until linguistic development is complete (Schwartz 2003; Paradis 2010b; 2010c). Both child L2 learners and adult L2 learners will lack L2 features unavailable in the L1, and therefore there is a question of poverty of stimulus A r easonable comparison could be made between child L2 learners, who use L1 properties and UG to reach their final state, and adult L2 learners of the same L1. Adults transfer L1 metalinguistic knowledge and may possibly access UG similarly to children to rea ch a plausible L2 final state Ultimately, these findings could aide in bridging the cognitive gap between L1 acquirers and child and adult L2 learners. We will examine an idea proposed by Paradis' in which she highlights the uncertainties regarding ch ild L2 acquisition during cognitive development, and how children's development may influence their underlying L1 metalinguistic knowledge. Paradis' research on second language impairments (SLI) shows that there is variation when eliciting L2 grammar based on the extent of cognitive maturity of the child L2 learner. Although the exact constraints of when and how the critical period occurs are still heatedly debated, many can agree that there is a deficit in implicit language learning mechanisms as we mature post puberty (Montrul 2004). Paradis asserts that before the age of seven, children are not likely to acquire a second language in its entirety because of their limitations in working memory and cognitive processes. Paradis (2010b) questions whether l earning the L1 may delay the development of an L2 in early childhood. She proposes that once basic infrastructure of an L1 grammar system has been established, on an abstract level, it could advance or "bootstrap" the acquisition of the L2. Certain grammat ical features would not have to be relearned, providing an advantage for older children to advance


! 4 more quickly than younger children (Paradis 2010b; 2010c). This study aims to assess what Paradis has proposed by testing the degree of variability that the age of L2 onset, child cognitive maturity, as well as input can generate when learning an L2 (White 2004; Unsworth 2008; Paradis 2010b; 2010c; Rothman & Fuentes 2010). Utilizing Passive constructions: In order to measure these factors we wanted to test children's knowledge of a specific linguistic property, passives and the formation of argument chains (A chains). On the bases of Paradis (2010b; 2010c; 2011), it is hypothesized that passives should b e affected by cognitive development because of their dependence on linguistic external cognition. It can be argued in light of Borer and Wexler's (1987) view that A chain passive constructions are innate but mature biologically. Fox & Grodsinksy (1998) pro pose that children are adept to comprehend passives; however, initially children's interpretation of passives is limited to an adjectival or actional passive construction (using get in place of be'). Chocarro's (2009) study working with L1 children of Cat alan similarly surveyed children's trouble with passives constructions and concluded that there is a noticeable delay, specifically until five years of age, in the use of truncated and non truncated passives. Pierce's (1992) work with Spanish L1 children e licited similar results whereby increased production parallels with increased age, potentially a result of emerging working memory. There is empirical evidence that can refute Borer & Wexler's Maturation Hypothesis (MH; 1987); however, most studies that h ave focused on English or a variety of romance languages, have concluded that there seems to be an upward trend in usage of passive constructions with age. In the work done with L1 children of Sesotho (Demuth et al. 2009) they observed children as young as three fully producing passive constructions, refuting the MH


! 5 because in reference to limited passive usage it is not copacetic across all languages. This can be questioned by input, which verifiably the use of passive constructions are exceedingly greater in adult Sesotho production than compared to adult English and Spanish production. Aside from their high input and production, this Bantu language uses morphological inflection to assign passivity, avoiding further constructive ambiguity by assigning diff erential inflection for adjectival passives, a chain passives, etc. (Demuth et al. 2009). Irrespective of their conclusions in terms of English and Spanish, there is a verifiable trend that passives are produced in greater frequency and variation (A chai n, non actional, non adjectival, etc.) as children mature In light of this, for our study we can use this construction as one particular factor that could suggest cognitive maturation variability across both groups of children. If children mature cog nitively overtime, specifically linguistic mechanisms and working memory, when comparing a younger to an older group of children theoretically the older group with the concrete L1 will outperform the younger group in terms of passive constructions, consid ering the latent manifestation of passives. So if we can theorize that this is true, then we can hypothesize older children will make less interpretable errors when assessing the varying passive constructions in their L2. However, if younger and older grou ps of children perform equally in their L2, then the younger group caught up cognitively with the older group of children and little support can be granted to cognitive maturation under these circumstances. If this were the case, then one could make the a ssertion that the L1 has little "bootstrapping" effects on child L2 acquirement, and that it is plausible that until adolescence children may only depend on UG to formulate constructions


! 6 Research Questions: We will be testing the following research questions in an effort to assess what Paradis has proposed using passive constructions as a means of measuring cognitive maturity: 1. Is it fruitful to expose children to an L2, especially in non naturalistic envir onments, as early as possible? 2. Is there a minimum in terms of the child's age during L2 onset, and does this have a cognitive explanation? Our findings could provide future investigations more insight on cognitive maturity and its influence on the facilitation of attainment of the L2 final state. From a more pedagogical perspective we must also consider the adopted general assumption that ear lier is considered better for learning a second language. Expose the child to their L2, as early as possible to get them familiarized and as a result they will learn faster. This is practiced generally across European school systems, and programs have foun d this method successful. This study may provide further support for this method, or may suggest otherwise. Methodology : Spanish Speaking Study The study reported in this paper is in relation to a larger study that has been ongoing since 2010, overseen by Rothman et al (2011) The following will explain how the reported study is a mirror representation of the work that was completed in Cali, Colombia. Rothman et al. worked with a dual immersion program that specialized in a 50/50 immersion program i n English and Spanish instruction. The researchers selected students from this program who came


! 7 solely from Spanish speaking mid to upper class backgrounds, and had three years of exposure to English instruction within the program prior to testing. The pa rticipants were placed into two test groups based on age. The first group of students began their 50/50 instruction at the age of four, and was tested approximately at seven years of age. The second group began English instructions at the age of seven, and was tested at 10 years of age. Their study involved two stages of testing. The first stage of testing used a standardized picture story task, to elicit production, completed with a question answer comprehension task based on the picture story task followe d by an oral interview. After examining each individual's MLU, stylistics, error rates, and overall fluency: there were 25 students within each group that had accurately completed this first experiment. The 50 Spanish speaking students were then asked to complete a second experiment testing their knowledge of passive constructions, in which they were asked to complete a picture matching task in English. Researchers used puppets from Sesame Street to encou rage a relaxed and familiar environment, so that the participants indirectly answered the researchers by sharing their answers amongst the puppets. The pictures that were used included cartoon ch aracters from a popular TV show shown in Spanish and Engli sh SpongeBob Squarepants. The second experiment was broken down into two parts; the first examined a total of 16 constructions comprised of passive A chains intermixed with active sentences. The second part focused on a total of 12 A chains and adjecti val passive constructions collectively, preceded by three to four brief sentences introducing the scenario. Students were presented with a phrase, spoken by a native English speaker, and asked to select the corresponding picture. Pictures were placed side by side depicting identical situations where characters would switch roles in each picture by


! 8 doing so the participants had to distinguish between agent and patient roles presented in the passive construction. The following pictures are examples of th e experiment broken down by task: Task 1 Active: 1a. SpongeB ob kisses Gary. 1b. SpongeB ob besa a Gary. Passive (a chain): 2a. SpongeB ob was found by Gary. 2b. SpongeB ob fue encontrado por Patrick.


! 9 Task 2 A chain: Mr. Krabs owns a restaurant. He wanted to change the brown color of his walls. So, Mr. Krabs decides to paint the walls red. Which picture best describes this sentence? The walls are painted red by Mr. Krabs. Mr. Krabs es el due–o de un restaurante. Est‡ cansado del color marr—n de las paredes del restaurante por lo que decide pintarlas rojas. Cu‡l de estos dibujos describe mejor esta oraci—n? Las paredes est‡n pintadas rojas por Mr. Krabs. No A chain/adjectival passi ves: Mr. Krabs owns a restaurant. He doesn't like the color of the walls, so he plans to paint them a different color, probably red. Which picture best describes this sentence? The walls were painted brown. Mr. Krabs es el due–o de un restaurante. Est‡ cansado del color de las paredes del restaurante por lo que decide pintarlas de un color diferente, probablemente rojas. Cu‡l de estos dibujos describe mejor esta oraci—n? Las paredes estaban pintadas de color marr—n. The full Spanish transcription of both tasks can be found in the Appendix.


! 10 English Speaking Study : The results from the second experiment completed in Colombia sparked much interest, so the aim of this study was to conduct an analysis involving the second experiment amongst English speaking children learning Spanish as their L2, in an effort to provide more validity for the original study. Both of the aforementioned tasks conducted in the experiment were completed in an identical manner, with the exception of the L2. The experiment was conducted entirely in Spanish and the tasks were directly translated from English to Spanish to maintain the continuity of the experiment. All participants were from English households, receiving no additional language contact, with the exception of Spanish in school. All children had approximately three years of exposure to Spanish, and in congruency to the study completed in Colombia, were split into two groups. The first began instruction at approximately four years of age, and the second group was exposed to Spanish at the age of approximately seven years old. Each par ticipant was then tested three years after initial instruction, at the ages of seven and 10. The test solely consisted of five participants, three in the younger group and two in the older group. All participants were students within a 50/50 dual immersion program and had been exposed to similar instruction across the years. The experiment had a total of 28 slides, and for every slide where participants chose the corresponding picture with the phrase provided, one point would be given. No points were added for inaccurate responses. The points were then calculated individually for each participant, all points were then collectively calculated to determine the combined average for each group across all four constructions presented in the two tasks. No statisti cal analysis was conducted due to the low number of participants. The results are represented in the tables below


! 11 Results: The data displayed in both Tables 1 and 2 compared the varying age groups from both the English and Spanish backgrounds. We controlled for deviations or inconsistencies that might have arisen in the tasks presented in the experiment. We had 11 participant s that classified as L1 Spanish speaking adults that use English as their L2, as well as a total of 12 pa rticipants that were L1 English speaking adults learning Spanish (L2). Both groups were tested in English, and as the data shows, the L1 English speaking adults performed at near ceiling level, scoring 95% overall, and the L1 Spanish speaking adults scored an overall 85%. The experiment for L1 Spanish speakers was administered in English, therefore the lower score was expected due to the question of age of initial onset and ultimately proficiency for a few participants. Regardless the adult speakers for both cases make for an adequate comparison to the fourth grade children. At first glance we can surmise that all adult and child groups consistently score above 75% for all active constructions, but for the remainder of this paper we will solely focus on the passive construction s. If we average the scores from Table 1 by combining the categories of passives, a chain, and no chain (adjectival) arguments, then we can look at the overall scores per group relevant to their overall performance on passive constructions. The Spanish fourth graders interpreted the constructions with the matching picture accurately 71%, whereas the Spanish first graders responded accurately 54% of the time. If we compare the Spanish speaking children to the English speaking first and fourth graders, the English fourth graders scored 93%, whereas the English firs t graders scored 55% on average. Overall, the first graders perform ed comparatively


! 12 less accurate than the fourth graders. Although the English speaking children population used for the study does not reflect an accurate statistic in comparison to the study conducted in Colombia out of the few English speaking participants that were tested they reflect similar findings Table 1 Passives Actives Chain No Chains ____________________ _____________________ Groups N M SD M SD M SD M SD L1 Eng. Ctrl 12 7.8 0.4 7.7 0.5 5.7 0.5 5.3 0.8 L1 Spn. Ctrl 11 6.9 2 7.3 1.2 5.4 0.7 4.2 1.4 English 4 th 2 8 0 8 0 6 0 4.5 0.7 Spanish 4 th 25 4.96 1.9 7.3 1.1 5.5 0.8 3.7 1.2 English 1 st 3 2 1 7 1 4.7 1.2 4.3 1.5 Spanish 1 st 27 2.9 2.4 6.1 1.8 4.9 1.1 2.9 1.3 Let us survey solely the figures for the second tas k. The second task assessed each groups' performance of argument chains and adjectival passive constructions. Comparatively none of the groups performed better when interpreting adjectival phrases with the coinciding picture to that of argument chain phrases. The Spanish first grader s underperformed exceedingly, 49%, compared to the Spanish fourth graders, 62%, but the English first graders, 72%, did not vary much with the English fourth graders, 75%. Overall it seemed that all groups had difficulty choosing the corresponding picture with the adjectival phrase. The exclusion of the nominative role or agent within the passive construction, where the passive construction is actually acting as a modifier, resulted in more inaccurate interpretations than when the agent is realized in the A


! 13 chain argument Table 2 Discussion : Age effects: The hypothesis proposed that exposing children to a second language, specifically in non naturalistic classroom environments, early in childhood might not be as fruitful as later on in childhood. We went into a n American 50/50 English & Spanish immersion program, examining children who's L2 was Spanish. We assessed two groups of prepubescent children, a seven year old group and a 10 year old group, in order to test for age and discern a minimum for wh en cognitive maturation could aide in L2 attainment more effectively. If we consider the findings "! #! $! %! &! '! (! )! *! +! ,#!-./0!1234! ,#!56.0!1234!! -./4789!&! 56:.789!&! -./4789!#! 56:.789!#! !"#$%&'(#$)*$+%,-#' .$*/01' ;<27=>! ?:887=>! 19:7.8! @A!19:7.8!


! 14 from the study completed in Colombia, it showed that the older group of children displayed with more frequency accurate associations between passive construct ions and corresponding pictures. The younger group tested in Colombia indisputably displayed a weaker performance in associating passive constructions with correlating pictures. The results from the reported study showed that comparatively the English spea king children tested in Spanish (L2) displayed a relatively similar disparity between the two age groups, with the older group outperforming the younger group of children. If we examine the results focusing on English L1 speaking children, the average d ifference between the first and the fourth graders were comparatively akin with those of the Colombian children. The fourth graders outperformed the first graders on all tasks in both passive constructions as well as active constructions Both the Spanish and En glish first graders responded accurately to under half, 40%, of the passive constructions that were presented in the first task. However, they responded over 75% accurately to active constructions when presented in the L2 during the first task Discernably, on the first task, the first graders performed at a far lower level when interpreting the passive phrase with the appropriate picture. Both the fourth grade groups accurately selected over 60% of the correlating pictures with the representat ive passive constructions. The results grossly exhibit a notable gap in performance level between that of the fourth and first graders across both L2s. The data reflects a consistent gap between the first grader performance and that of the fourth graders However if we consider the control groups that performed at near ceiling levels, there is some discrepancy between the fourth grade groups. The English L1 fourth graders perform the first task at ceiling level. Although the Spanish L1 fourth graders ou tperformed the Spanish L1 first graders, the data indicates that the Spanish L1 fourth grade performance was


! 15 less accurate and deviated more from the English L1 fourth grade performance. Ultimately the variation between both the fourth grade groups is a di rect result of having only two participants involved in the study conducted with the English L1 fourth grade speakers. Both fourth graders happened to have scored perfectly on all passive constructions in the first task. Comparatively, out of the Colombian fourth graders, none of the students scored perfectly on the passive constructions, but four of the students did score a 90%. We can project that with more English L1 fourth graders, the population would have been more representative of the profile, an d would have displayed more variation. If we look more closely at the second task, we see similar trends between the comparative L2 groups. The fourth graders outperformed on both A chain and adjectival passives to the first graders. There was very littl e deviation between both groups of fourth graders when assessing A chain passives. Both groups were fairly accurate when matching the phrase with the corresponding picture after hearing the insertion of the agent, and overall the first graders performed th e least accurate amongst all the groups. Nevertheless, the overall average for adjectival passive constructions was significantly lower, even for the adult English L1 speaker controls. This could have been the result of a faulty depiction of a particular p hrase, or underlying semantic ambiguities that may have manifested in the phrase itself. The English L2 children differed from each other significantly in terms of the adjectival passives, whereas the English L1 fourth graders outperformed the English L1 f irst graders by only a mere three percent Many researchers agree that children have the cognitive skills and syntactic values to produce passives (Martin & Snyder 2009) going along with the innate theory (IT), but empirical evidence counters with the IT showing that production may be unrealized until external cognitive mechanisms mature. Martin & Snyder (2009) argue that rather than producing A chain passives


! 16 most children produce get passives that denote a predicated action, which potentially mirror a djectival passives (Crain et al. 2009). This means that although children may not produce passive constructions, they can still exemplify passive utility through means of alternative constructions. In relation to the current findings the prior epistemic no tions in regard to children's tendencies to produce non truncated passives (adjectival) more heavily than truncated (Chocarro 2009), our findings may suggest otherwise: lower averages for adjectival passives ranging between the different groups of children L2 learners were found. However, this experiment did not test for production, additionally it is reasonable to question the clarity of the second task that was designed. Both the adults and children may have been predisposed or primed to select the pictur e with a character acting as the agent because the characters were introduced in the foreground prior to hearing the actual adjectival construction. Results may have varied if we had left out the foreground so as not to possibly habituate the responses eli cited from the participants. The empirical issues in the methodology of the study must be taken into consideration, but the surfacing findings from this study and the study conducted by Rothman et al. (2011) provide a strong argument that there may be age effects during cognitive maturity in terms of linguistic mechanisms. The fourth graders outperformed the first graders across all variables. Age may strongly play a large role in terms of aiding and facilitating grammar, in the L1, and we can reasonably s uggest this for L2 grammars as well. This study suggests that based on the principles of economy, introducing young children to an L2 before all principles in their L1 are concrete may not be the most effective or successful approach rather than introducin g the L2 when children's linguistic capacity has become more cognitively concrete. This study


! 17 demonstrates older children, performing near or at ceiling levels in their L2 compared to their younger counterparts. Factors of Success: Metalinguistic knowled ge, Input, & Motivation It is imperative that we discuss the varying degree of success between that of the L2 Spanish and L2 English learning students. Here we want to discuss why age may not be the only factor pushing the fourth graders to success. So wha t can we determine in terms of metalinguistic knowledge ? Paradis propose d that older children could be utilizing a "bootstrapping" effect when it comes to acquiring an L2. The data that was collected strongly points in that direction, and by testing the ir level of understanding the experiment does not deny young children the capacity to produce passives; the experiment was designed to test their facilitation and associ ations when the passive presented itself. It would seem that the first graders could st ill be developing syntactic and semantic information to stabilize the abstract constraints of passives in their L1 and L2 grammars. The older children displayed a better handle on deciphering between passive and active scenarios, as well as more intrica te f orms of the passive in their L2. W e can deduce that these children may have a more concrete understanding of what the passive construction involves semantically and are able to adjust parameters syntactically more readily We want to discuss if indee d the children in fourth grade had more success due to existing metalinguistic knowledge. Fourth graders have the knowledge to produce and decipher the semi syntactic properties of the passives by ten years old (Paradis 2010b). With this property in their grammar already in place, if a fully functioning system is already in place there would be no purpose of re conceptualizing the passive in their L2, and all a child must do is alter their


! 18 parameters to adjust to the L2 A child with acquired knowledge regarding passives, in three years time must only learn how to construct the syntax in the L2 to employ the passives (Rothman & Fuentes 2010). The faculty of language is efficient, common to all cognitive processes, and will not permit a child to reconstruct existing principles already established in the UG. By having access to their UG and possessing existing metalinguistic knowledge, as represented in our data, older children are more likely to learn the L2 more quickly than their younger peers. If the variability between the seven year olds and 10 year olds is evident, then there must be a reason why there is variability between the two groups of fourth graders It is difficult to deduce that the English speaking fo urth graders outperformed the Spanish speaking fourth graders, because of the lack of Anglophone participants that marginally reflect the population. However, let us take into consideration that both groups of fourth graders had the same time of exposure t o the L2, within a 50/50 immersion program, L2 contact remaining entirely in school, and without L3 influence. Yet out of the two Spanish L2 fourth grade participants, both students scored a perfect score on passive constructions during the first task, whe reas the English L2 fourth grade participants performed drastically lower, scoring only 62% on the same task. We must evaluate what the L2 quality input and quantity input is comprised of. Although both groups of fourth graders received L2 input in a c lassroom environment, we cannot dismiss that there is minimal opportunity to receive input or put forth any output when as little as five or less English speakers attend the class (including the instructor in most cases). We do not doubt that the quality o f i nput may be exceptional in Colombia, but the quantity can be in question. In Miami, Florida, children learning Spanish in the classroom have ample amounts of input, as well as opportunity for output. This type of classroom environment provides


! 19 Anglophon e children exceedingly more opportunities to retrieve L2 input. Culturally, in light of current political ease, Colombia has recently taken a stronger approach to the tourist industry employing English for economic reasons In contrast since 1960, Miami co ntinues to serve as one of the major summits for Hispanic influence and integration into the United States. The availability of Spanish input in Miami is overwhelmingly more than that of English in Colombia. The other prevailing factor that aids in exp laining English speaking fourth graders outperforming their Spanish speaking fourth grade peers, is the factor of motivation within the classroom. These predominantly English speaking 10 year olds are in constant interaction and contact with heritage Spani sh speakers and/or bilingual speakers in the classroom English speaking children's motivation to want to communicate within a more balanced bilingual environment and the amount of exposure to Spanish input is more than the Spanish speaking students lea rning English in Colombia. Take into further consideration the fact that children have little qualms about making errors in speech, their constant interaction with friends and teachers motivates them to want to communicate effectively in a bilingual enviro nment. Most young students want to establish an equal ground between peers; therefore, quickly learning the language is motivation enough to quickly excel in their L2. When coming into contact with Spanish at the age of seven for 50% of one's entire school day, the advancements in one's L2 grammar over the next three years would understandably be considerably rapid Limitations : There were several important limitations in the study. The time allotted to complete the study was a major constraint. The methodology and the experiment had already been designed; although this provided a reasonable window of opportunity to allocate more time to find participants, this proved to be a larger task than expected. We were under the notion that if we


! 20 could fin d a private 50/50 immersion program in Colombia, then finding such a private program in Florida where the largest majority happens to be Hispanic, our findings would prove to be fruitful. This was not the case. We were able to locate a few elementary schoo ls in the state of Florida and Georgia that provided 50/50 settings, predominantly public institutions. The issue with working with most public institutions is obtaining the necessary approval to work with the students. Aside from working with the scho ol's administration, many schools must follow county district regulations, which require schools to retrieve district approval before collaborating in any research from outside the school. Aside from the arduous work and time involved in obtaining district approval to work with the children, the main issue was that there was not enough time for this extensive approval process. This eliminated many prospective participants that could have provided for more potential data as well as providing a more accurate representation of the population and profile being tested. We had to assert our efforts elsewhere and work with smaller numbers and smaller specialized schools who could work on an immediate schedule, so that we could pursue direct means of communicating with the final administrative decision maker and ultimately the children's parents. This was much more time efficient but resulted in resigning to work with a smaller number of participants that didn't reflect an appropriate populat ion such as the mirror study in Colombia There are a surprisingly small number of immersion programs throughout the U.S. that would consider themselves a 50/50 immersion program, let alone provide second language learning (excluding ESOL), with a focus on Spanish. In some states it is illegal to have public programs created for an enhanced, fully immersed bilingual education program in Spanish & English. The political issues that revolve around this latent view of dual language immersion programs are hin dered by an extreme and negative view in regard to assimilating to US culture.


! 21 Although progress is slow, there is a gradual but notably small shift to SLA in most public school systems. However as presented earlier in the paper, foreign language learning programs are growing in number globally. In more modern institutions, these programs are a mandatory part of basic education, and are considered to be advantageous for societies in contact, as well as the individual. Aside from finding a school that was w illing and able to collaborate with us for the study, finding the participants that would match the profile and provide for the most congruency with the Colombia study was challenging. The younger group wasn't as difficult to find, considering most childre n start in kindergarten; however, the older group, whereby participants started learning Spanish at seven followed by three years of classroom exposure was difficult to come by. This resulted an insufficient amount of time to test for proficiency. The stud y completed in Colombia had a preliminary experiment that tested the children's MLU and comprehension by using a picture story experiment using the same profile. Although this picture matching experiment had little to do with the level of proficiency, and rather focused on children's intelligibility of the passive in the L2, a certain level of skill in the L2 was required in order to complete the task. Unfortunately, no proficiency exam was administered for the children involved in this study. A written pr oficiency test could have been examined, but the results of which would have been rather irrelevant for the purpose of our study. We also had to consider the amount of time we were permitted to work with the child outside of class. What we did to assess th e students and their level of comprehension was to begin the experiment with a few preliminary comprehensive questions in Spanish, regarding the instructions of the task. Although it may not have been a thorough approach, from observation it was closely no ted how both groups of children


! 2 2 comfortably switched to using Spanish with such ease and with little to no sign of delay. After the experiment was completed, we completed a brief final interview regarding their household and language contact. A ll of the pa rticipants appeared to be attentively listening. Future Research: The research completed and the information collected from this study will benefit in the ongoing work involved in child second language acquisition. The goal here is to ultimately revise t he methodology t hat was used here and apply it in a more suitable timeframe. By coordinating with a larger institution, and seeking out the necessary approval early on this would provide for a more concise comparison with the research that is being done i n Colombia. Administering a more thorough proficiency test to assess the students in detail would aide in the value of the experiment. Aside from employing a longer timeframe and having a higher number of participants, reassessing the written experiment an d extinguishing any ambiguities would provide for more impactful data. Essentially we are working towards bridging the gap between language acquirers and language learners, to explain what the cognitive linguistic differences are between children and adu lts. By assessing how children differ when learning a second language during their critical period can help us better understand how adults may approach a second language. Overall this type of data could help us teach language more efficiently and effectiv ely across the varying periods of maturation, as well as understand how we conceptualize language as a means of communication and an innate form of expression. Furthermore, f rom a micro perspective we aim to work through the limi tations presented above, po tentially work ing towards a more meticulous study in terms of measuring cognitive development and metalinguistic knowledge and how they influence the learning of another language.


! 23 Conclusion: The findings that were made currently in Miami and comparative ly to the ongoing research conducted by Rothman et al. (2011) in Colombia, aims to shed some light on the empirical questions that surround child and adult L2 acquisition. We hope that the data that was collected will help shape future ideas and possibly a ide in the further investigation of cognitive development The stages of cognitive development may coincide with linguistic development before, during, and after the critical period. Our findings may point in the direction of our hypothesis, but the assess ment of the methodology used must be taken into consideration. Whether or not it is more effective to introduce the L2 at a later period in childhood, after the L1 grammar is considered established, must be examined in much further depth in order to make a ny solidified conclusions. Acknowledgements: I first would like to thank all the students who participated in the study with the support of their parents. Their genuine interest and support was the sole reason I could complete the study in such a limited time. Thank you to my mentor and advisor, Dr. Jason Rothman, for all of his guidance and enthusiasm throughout the past year in my undergraduate career and research. Furthermore, I would like to thank Hans Duque for taking time from his family, and assisting me with my research down in Miami. To concl ude, I would like to express my gratitude to all of my fellow undergraduate peers, alongside the esteemed graduate students and staff from both the Spanish and Linguistic Departments. It has been my greatest pleasure working alongside such a motivating and encouraging group of individuals.


! 24 References : Borer, Hagit & Kenneth Wexler. (1987). The maturation of syntax. In T. Roeper & E. Williams (eds.), Parameter setting. Dordrecht: D. Reidel Publishing Company. Chocarro, P. (2009). The acquisition of act ional passives in Catalan (pp. 1 69). Retrieved from Chomsky, N. (1993). Lectures on government and binding: The Pisa lectures Germany: Walter de Gruyter. Craik, F., & Bialystok, E. (2006). Cognition through the lifespan: Mechanis ms of change. Trends in Cognitive Sciences 10 131 138. Crain, S., & Lillo Martin, D. C. (1999). An introduction to linguistic theory and language acquisition Hobokon, NJ: Wiley Blackwell. Crain, S., Thornton, R., & Murasugi, K. (2009). Capturing t he evasive passive. Language Acquisition 16 123 133. Demuth, K., Moloi, F., & Machobane, M. (2010). 3 Year olds' comprehension, production, and gerneralization of Sesotho passives. Cognition, 115 238 251. Fox, D., & Grodzinsky, Y. (1998). Children's pas sive: A view from the by phrase. Linguistic Inquiry, 29 311 332. Johnson, J., & Newport, E. (1989). Critical period effects in second language learning: The influence of maturational state on the acquisition of English as a second language. Cognitive Psyc hology 21 60 99.


! 25 Levin, B., & Rappaport, M. (1986). The formation of adjectival passives. Linguistic Inquiry, 17 623 661. Lillo Martin, D., & Snyder, W. (2009). Commentary on "capturing the evasive passive". Language Acquisition, 16 118 122. Montrul, S. (2004). The acquisition of Spanish, morphosyntactic development in monolingual and bilingual l1 acquisition and adult l2 acquisition The Netherlands: John Benjamins Pub Co. Paradis, J. (2010a). Bilingual children's acquisition of English verb morpho logy: Effects of language exposure, structure complexity, and task type. Language Learning, 60 651 680. Paradis, J. (2010b). The interface between bilingual development and specific language impairment. Applied Psycholinguistics, 31 227 252. Paradis, J. (2010c). Response to commentaries on the interface between bilingual development and specific language impairment. Applied Psycholinguistics, 31, 345 362. Paradis, J., Nicoladis, E., Crago, M., & Genesee, F. (2010). Bilingual children's acquisition of the past tense: A usage based approach. J. Child Lang., 38 554 578. Paradis, J. (2011). The impact of input factors on bilingual development: Quantity versus quality. Linguistic Approaches to Bilingualism, 1 67 70. Pierce, A. (1992). The acquisition of pas sives in Spanish and the question of a chain maturation. Language Acquisition, 2 55 81.


! 26 Rice, M. L., & Wexler, K. (2001). Test of early grammatical impairment New York, NY: Psychological Corporation. Rothman, J., & Guijarro Fuentes, P. (2010). Input quality matters: Some comment on input type and age effects in adult SLA. Applied Linguistics 31 301 306. Rothman, J., Long, D., Lingwall, A., Halloran, T., Chakravarty, T. Firestine, J, & Rietbergen, M. J. (2011). Child Secon d Language (L2) Acquisition and Cognitive Development: Is earlier always better? Paper presented at EuroSLA (European Second Language Association). Stockholm: Sweden. Schwartz, B.D. (2003). "Child L2 acquisition: Paving the way." In B. Beachley, A. Brown & F. Conlin (eds.) Proceedings of the 27th Annual Boston University Conference on Language Development, Vol. 1 Somerville, MA: Cascadilla Press. Unsworth, S. (2008). Comparing child L2 development with adult L2 development: How to measure L2 proficiency. C urrent Trends in Child Second Language Acquisition, 6 301 333. White, L. (2004). Second language acquisition and universal grammar United Kingdom: Cambridge University Press.


! 27 Appendix: Diapositiva 1 Presentaci—n de los personajes Squidward Patrick, Mr. Krabs, Spongebob, Gary (Otros nombres si los participantes no conocen el programa) Diapositiva 2 Activa: Spongebob besa a Gary. ( Elmo a la derecha ) Diapositiva 3 Pasiva: Spongebob fue encontrado por Patrick. ( Grover a la derecha ) Diapositiva 4 Activa: Squidward patea a Patrick. ( Elmo a la derecha ) Diapositiva 5 Pasiva: Spongebob fue besado por Gary ( Grover a la derecha ) Diapositiva 6 Pasiva: Mr. Krabs fue tocado por Squidward. ( Elmo a la derecha ) Diapositiva 7 Activa: Spongebob encuentra a Patrick. ( Elmo a la derecha ) Diapositiva 8 Pasiva: Patrick fue pateado por Squidward. ( Elmo a la derecha ) Diapositiva 9 Activa: Mr. Krabs toca a Squidwar d. ( Grover a la derecha )


! 28 Diapositiva 10 Pasiva: Patrick fue encontrado p or Spongebob. ( Elmo a la derecha ) Diapositiva 11 Activa: Gary besa a Spongebob. ( Grover a la derecha ) Diapositiva 12 Pasiva: Gary fue besado por Spongebob. ( Grover a la derecha ) Diapositiva 13 Activa: Patrick patea a Squidward. ( Elmo a la derecha ) Diapositiva 14 Activa: Patrick encuentra a Spongebob. ( Grover a la derecha ) Diapositiva 15 Pasiva: Squidward fue tocado por Mr. Krabs. ( Elmo a la derecha ) Diapositiva 16 Activa: Squidward toca a Mr. Krabs ( Grover a la derecha ) Diapositiva 17 Pasiva: Squidward fue pateado por Patrick. ( Grover a la derecha ) Diapositiva 18 ---------En blanco ----------Diapositiva 19


! 29 Spongebob se acaba de mudar a Gainesville y necesita cosas para su casa. Spongebob compra una televisi—n en __________( Walmart). Al llegar a casa la rompe de manera accidental practicando Karate. Cu‡l de estos dibujos describe mejor esta oraci—n? La televisi—n fue rota por Spongebob. ( Elmo a la derecha ) Diapositiva 20 Squidward trabaja en el restaurante KrustyKrab. De ca mino al trabajo, ve un carro aparcado en frente del restaurante. Cu‡l de estos dibujos describe mejor esta oraci—n? El carro fue aparcado. ( Elmo a la derecha ) Diapositiva 21 En un d’a maravilloso, Squidward quiere tomar el aire as’ que abre la puerta. Cu ‡l de estos dibujos describe mejor esta oraci—n? La puerta fue abierta por Squidward. ( Grover a la derecha ) Diapositiva 22 Mr. Krabs es el due–o de un restaurante. Est‡ cansado del color marr—n de las paredes del restaurante por lo que decide pintarlas roj as. Cu‡l de estos dibujos describe mejor esta oraci—n? Las paredes est‡n pintadas rojas por Mr. Krabs. ( Elmo a la derecha ) Diapositiva 23 Spongebob escucha un ruido en la planta de abajo. Al bajar las escaleras ve que la televisi—n est‡ encendida. Cu‡l d e estos dibujos describe mejor esta oraci—n? La televisi—n estaba encendida. ( Grover a la derecha ) Diapositiva 24 Spongebob cocina una hamburguesa porque tiene hambre. La hamburguesa est‡ muy caliente as’ que la deja encima de la mesa para que se enfr’e. D espuŽs de lavarse las manos, vuelve y se come la hamburguesa. Cu‡l de estos dibujos describe mejor esta oraci—n? La hamburguesa fue comida por Spongebob. ( Grover a la derecha ) Diapositiva 25 Spongebob se acaba de mudar a Cali y necesita cosas para su casa Spongebob compra una televisi—n en __________(una tienda en Cali). Al llegar a casa se da cuenta de que la televisi—n que acaba de comprar est‡ rota. Cu‡l de estos dibujos describe mejor esta oraci—n? La televisi—n que Spongebob compr— estaba rota. ( El mo a la derecha ) Diapositiva 26


! 30 Mr. Krabs es el due–o de un restaurante. Est‡ cansado del color de las paredes del restaurante por lo que decide pintarlas de un color diferente, probablemente rojas. Cu‡l de estos dibujos describe mejor esta oraci—n? Las paredes estaban pintadas de color marr—n. ( Elmo a la derecha ) Diapositiva 27 En un d’a maravilloso, Squidward quiere tomar el aire as’ que va a abrir la puerta pero ve que ya est‡ abierta. Cu‡l de estos dibujos describe mejor esta oraci—n? La puerta estab a abierta. ( Grover a la derecha ) Diapositiva 28 Spongebob piensa que su casa es muy silenciosa. As’ pues decide bajar las escaleras y encender la televisi—n. Cu‡l de estos dibujos describe mejor esta oraci—n? La televisi—n fue encendida por Spongebob. ( Grover a la derecha ) Diapositiva 29 Squidward trabaja en el restaurante KrustyKrab. Squidward maneja su carro al trabajo y lo aparca en frente del KrustyKrab. Cu‡l de estos dibujos describe mejor esta oraci—n? El carro fue aparcado por Squidward. ( Elmo a la derecha ) Diapositiva 30 Spongebob cocina una hamburguesa porque tiene hambre. La hamburguesa est‡ muy caliente as’ que la deja encima de la mesa para que se enfr’e. DespuŽs de lavarse las manos, vuelve y observa como su mascota Pet se ha comido toda la hamburguesa. Cu‡l de estos dibujos describe mejor esta oraci—n? La hamburguesa fue comida. ( Grover a la derecha )

xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EQB9T2CXL_E2EHO0 INGEST_TIME 2017-07-11T21:52:50Z PACKAGE AA00057244_00001