3.Children are more likely to acquire native grammatical competence. The critical period for grammar might be later than for pronunciation (around 15 years). Nonetheless, according to different studies, some adults might be able to acquire native levels of grammatical accuracy in speech and writing and in some cases even full linguistic competence. 

4.Independently of whether native-speaker proficiency is acquired, there are more possibilities for children to reach higher levels of attainment in both pronunciation and grammar than for adults. 

5.The process of acquisition in the case of an L2 grammar does not seem to be meaningfully affected by age but that of acquisition of pronunciation may be. 

The varied result of the studies just mentioned led researchers to believe that there must be, if not a critical period for L2 acquisition at least what could be called a sensitive period. The difference between the terms “critical” and “sensitive” period¨is based on the consideration that completely successful acquisition can only be possible within a given period of a learner’s life (that is, “critical”) or that acquisition is just easier within that period, but not impossible afterwards (“sensitive”). This is sustained by Seliger’s (1978) proposal that there might be multiple critical/sensitive periods for different aspects of language. According to him the span at which a native accent is easily achieved seems to end sooner than the period governing the acquisition of a native grammar. 

The truth is that researchers have not yet found conclusive evidence of any differences in the process of L2 acquisition by child and adult learners. In relation to phonology, they found some clear processing differences suggesting that children and adults depend on different mechanisms. However, with regard to grammar, they did find clear differences; this suggests that learners of all ages rely on the same learning mechanisms. 

In brief, we can conclude that it is not yet possible to have any conclusive results on the topic and we still need to find a justification for the fact that adults have continuous access to a language specific acquisition device. Nevertheless, we can conclude that, based on research, the acquisition of phonology, which does appear to be affected by age, differs from the acquisition of grammar, which seems to be much less sensitive. 

An interesting point is raised by Cenoz (2003) when she states that the implementation of foreign languages in pre-school education needs to be analysed with reference to the particular educational background where it takes place. For instance, in Spain, there are several bilingual communities where English has become the third language at school. The analysis of third languages and trilingualism is a fairly new area of research embedded in the field of Applied Linguistics and one that has received much attention in the last few years). 

According to Cenoz (2003), there are two key questions directly related to educational background: 

1.We need to know if the implementation of English in pre-school in bilingual programmes has some type of effect in the linguistic and cognitive development. 

2.It is necessary to find out the results of those programmes that have been already implemented.

1.2. The teaching of EFL from pre-school and its influence in linguistic and cognitive processes 

As Cenoz (2003) points out, a series of studies both in instructed and naturalistic contexts show that the acquisition of more than two languages is possible and it does not have a negative effect on cognitive development. Furthermore, bilingualism can promote metalinguistic awareness and creativity (Baker, 2001; Cenoz & Genesee, 1998b; Jessner, 1999, Muñoz, 2000). 

Studies both in instructional contexts (Cenoz, 1997) and naturalistic contexts (Quay, 2001) lead us to believe that instruction in English since pre-school and early trilingualism does not seem to cause cognitive problems, although further investigation is needed. However, research in this area (Cenoz & Genesee, 1998b; Muñoz, 2000) has highlighted the positive influence of bilingualism on the acquisition of a third language. This advantage could be due to different factors such as linguistic interdependence, the development of metalinguistic awareness and strategies for language learning. According to Muñoz (2004), research on monolingual and bilingual children has demonstrated that the latter obtain better results than the former at particular tasks as, for instance, numerical problem solving tasks (Bialystok, 2001). Other advantages brought about by bilingualism are the enhancing of the awareness of the relationship between words and objects (Rosenblum and Pinker, 1983) or the benefit of sharing mental representations of both languages in similar areas of the brain (Fabbro, 2001). 

Another aspect that deserves attention here is language transfer. It could be assumed that in the case of a third language there would be more confusion when English is introduced at an early age, but scientific results show that this is not the case and that errors due to negative transfer¨occur on fewer occasions than other types of errors (Cenoz, 2001; Bouvy, 2000). 

However, we need to highlight that all these benefits are found in the type of bilingualism that results from similar amounts of exposure to both languages. This is the type of bilingualism or trilingualism we find in certain areas where two or three languages aree spoken (bilingual o trilingual communities). This type of bilingualism/trilingualism cannot be achieved through formal instruction at school and it is not, by any means, the final goal of bilingual education. Therefore, we need to be very cautious when overgeneralizing regarding the advantages of bi- or-trillingualism and should take into account that natural bilingualism cannot be equated with bilingual or multilingual education. 

This does not mean that starting the teaching of a foreign language at an early age is inadequate or pointless: as Muñoz (2004) remarks, from an educational point of view, the advantages its implementation can bring include increasing linguistic awareness, becoming more conscious of linguistic and cultural diversity and encouraging a more open attitude towards these differences and a more positive one towards the learning process. 

Right now the predicted benefits of early instruction have not been yet found: we cannot see much difference between those students who started taking English when they were 11 and those who started when they were 6. Therefore, some changes in the educational system are necessary if we want make early learning of languages possible. This includes: 

·ØImproving teacher training, particularly in the case of those who are to teach in pre- school and especially when it comes to oral skills 

·ØIncreasing the amount of time students are going to be exposed to the foreign language: this can be achieved if some content subjects are taught in the foreign language (CLIL¨) 

·ØProviding students with the right type of instruction, that is, instruction that involves practicing of the language in real communicative contexts. 

The main objective, then, would be to advance the starting age of students not because of an unfounded belief that the younger a student is the better he/she will learn the language, but according to the actual characteristics and needs of our students.