The authors analyzed the effects of second language interference on first language abilities in 50 Spanish-English bilinguals living in the US. They examined syntactic comprehension, verbal memory, and calculation abilities in both Spanish and English. Performance was generally better in Spanish than English, and difficulties arose when Spanish syntax differed from English. The study suggests bilinguals may be at a disadvantage in either language due to interference effects, and special testing norms should be established.

1. Syntactic Comprehension, verbal memory, and calculation Abilities in Spanish- English Bilinguals Alfredo Ardila Monica Rosselli FeggyOstrosky-Solis Jose Marcos Gladys Granda Miguel Soto

2. The research The authors analyzed the interfering effects of second language on the first Language in Native speakers living in the United States. They examined 3 linguistics aspects: Syntactic comprehension Verbal memory Calculation abilities

3. The context The Hispanic population represents about 10/30% of the general population Spanish/English is not socially, academically, economically and politically equivalent to Spanish/English and frequently it is maintained as a marginal language. Children attend English/Spanish speaking schools In the particular context that the research was conducted, Spanish-English Bilingualism represents and active Bilingualism.

4. The Background Interference from the first language ( L1) to L2 is a well-established observation in bilinguals ( e.g., Bahrick, Hall, Goggin, & Bahrick, 1994) Inference from L2 to L1 is rarely/ frequently mentioned in bilingualism literature Everyday observations illustrates that Spanish speakers living for some time in the United States frequently present difficulties in using Spanish/ Englishortography, minor phonetic abnormalities may be observed in some phoneme productions ( most frequent ,/r/,/p/, /l/, and /t/. ), and Spanish verbal fluency decreases/increases.

5. The Background…cont Marcos and Ostrosky (1995) observed that, in the interpretation of sentences, normal participants followed different strategies. Verbal memory in bilinguals has been the focus of several research studies ( Peynircioglu & Durgunoglu, 1993) however, research have been mixed. Sometimes performance in memory task is similar/different in both languages, sometimes bilinguals seems to be in disadvantage/ advantage, and sometimes bilinguals appear to be at an advantage/disadvantage ( Magiste, 1980; Ransdell & Fischer 1987; Snodgrass, 1984).

6. Paivio and Lambert (1981)found that words coded bilingually were retained better/worse than words coded unilingually. Cullum and Puentes ( 1995) observed that balanced/ nonbalanced bilinguals assessed in English learned fewer words overall and demonstrated lower retention scores compared to monolinguals.

7. The hypothesis Bilinguals, and particularly balanced bilinguals, are required to simultaneously rely on two different morphosyntactic systems. A certain degree of interference may be anticipated. Verbal memory and calculation tasks would be better performed in L1/L2 than L2/L1. When similar syntactic structures between L1 and L2 were used, syntactic structures between L1 and L2 were used, syntactic understanding would be higher/ lower.

8. Study 1

9. Method Participants Fifty middle socioeconomic status Spanish- English bilinguals ( 18 men and 32 woman; M= 28.7). Participants with a minimum of 12 years of schooling were selected. L1 was Spanish and L2 was English All learned English Early in life and had attended to English schools. 28 participants had been exposed to English before the age of 5, and 22 had been exposed between the ages of 5 and 12.

10. Method Participants All participants were living in a monolingual/ bilingual community in Miami, FL. Spanish/ English was the home language of all participants.

11. Instrument The Spanish Syntactic Comprehension Test ( Marcos & Ostrosky, 1995) was used. The Spanish Syntactic Comprehension Test is a normalized instrument used to assess the phonological /syntactic comprehension of Spanish. A forced election task is used in which participants listens to 190 different reversible sentences and is asked to select one of 4 options on a plate. The four options correspond to : Same action but performed by the wrong animal Same action and correct animal Same animal but performing a different action Different pair of animals and same action

12. Procedure Participants were instructed to point at the picture corresponding to the sentence read by the examiner. 190 sentences were read with normal intonation and contour by the examiner No feedback regarding the accuracy of the answers was provided.

13. Results The performance on the Spanish Syntactic Comprehension Test was higher in the bilinguals /monolinguals sample than the bilinguals/monolinguals sample Differences were most evident in sentences with a pseudocleft A, followed by active reversible without preposition. Pseudocleft – “ fue el tigre lo que a un leongolpeo” Active reversible without preposition –“ El leongolpeo a un tigre”.

14. Results Passive/Activesentences were the easiest for Bilinguals The participants exposed to English between ages 5 and 12 performed better/worsethan those individuals who were exposed before years of age. Participants preferring Spanish outperformed those participants preferring English. The number of correct responses was about twice as high in women/menthan in men/women- these differences were statistically significant.

15. Study 2

16. Participants Participants were 69 middle socio-economic status Spanish-English bilinguals. Graduate students recruited in Miami, FL. Age range = 18 to 49 ( M= 30.28 years, SD= 7.97) 34 men and 35 women, 27 were born in the United States, and 42 were born in Latin America. They used Spanish, English and “Spanglish” to communicate. On a scale from 1 to 5 participants rated themselves on spoken Spanish and English at the level 3 (well) or above. The age of acquisition of English ranged from 0 to 35 years.

17. Materials and procedures A demographic questionnaire was administered to evaluate the characteristics of the participants’ bilingualism Five subtests were taken from the Wechsler Memory scale ( WMS) English version and its equivalent translation to Spanish language to assess memory in both languages. The Serial Verbal Learning Test ( SVLT) was administered in both languages. The SVLS test a. number of words recalled in the first trial, b. number of trials required to recall the 10 word list, c. delayed recall of words. To assess the participant’s calculation abilities in Spanish and English, three basic arithmetical operations and one numerical problem were performed aloud in both languages.

18. Materials and procedures Time and errors were scored. However, errors were so low that no statistical analysis of errors was possible. Results, in consequence, refer only on time.

19. Results Performance was better in Spanish/Englishthan in Spanish/English. Performance in was better in Delayed logical memory and the 2 digits scores. Spanish/English On the calculation abilities performance was significantly faster in Spanish/English in 3 out of 4 scores. Age effect was observed in the multiplication, and interaction was significant in the multiplication and division. The effect of preferred language was evident only in the WMS Digits. In calculation abilities subtest, a language effect was observed in all scores, whereas the preferred language affected the numerical problem.

20. Discussion

21. Discussion Although four answers were available on the Spanish syntactic Comprehension Test, only two answers were likely. Consequently the chance level may be supposed to 50/ 90%. Passive sentences are more frequently used in English than Spanish, so it is not surprising that performance in understanding Spanish passive sentences was virtually perfect/imperfectin Bilinguals. When Spanish syntax moved away/ closer from English syntax, difficulties emerged.

22. Discussion Differences in bilinguals participants and monolinguals participants were slight and nonsignificant due to an appropriate understanding of the use of preposition a. Sentences with a pseudocleft A were extremely difficult for bilingual participants. The closer to English syntax the Spanish sentences were, the easier/ harder it was for participants to understand them. Participants exposed to English between ages 5 and 12 outperformedthe participants exposed to English before 5. This means that the interfering effects of English was stronger in the participants exposed to English before the age of 5. Participants preferring Spanish or either had a better/worse syntactic understanding than those participants preferring English.

23. Discussion The sex effect on linguistic abilities remains controversial. Women outperformed men. Regarding differences in memory and calculation abilities, performance was overall better in the participant’s L1 (Spanish) than in their L2 (English). Although Spanish can be considered the L1 for this group, a significant percentage of the participants learned Spanish early in life, later attended to school in English, and were currently living in a mostly English-speaking environment. Preferred spoken language turned out to be a more significant variable than language acquisition .

24. Discussion Overall, differences in the calculation subtests were larger than differences in verbal memory tasks. Digits Spam was larger in English/Spanish than in English/Spanish. English speakers accomplish a 7 digit spam, while Spanish speakers are usually 5.8. Phonological length may account for this difference. The significant difference between both languages on numerical problems ability suggests a better ability for reasoning when using a L1 than when using L2.

25. Clinical Significance and suggestions Spanish Bilinguals may be at a disadvantage when using either language. The use either Spanish or English testing materials or norms Awareness of the syntactic structures Adjusting Scores The results do not necessarily reflect the participant’s real abilities

26. Research significance and suggestions Are the syntactic structures used on standardized test placing bilinguals at a disadvantage? Is the patterns equivalent for bilinguals of other languages ( e.g: Chinese- English bilinguals)? How would speakers of languages with few equivalent syntactic structures perform on these tasks? Special norms should be established for Spanish- English bilinguals.