1. Introduction

With the development of globalization, English is becoming a lingua franca and is learned by people all around the world. More and more people can speak more than one language, and bilingualism as well as multilingualism, has become a very common phenomenon. In this context, many parents give priority to developing their children's bilingual abilities at a young age, aiming to provide them with a competitive advantage in the future, especially in Asia [1]. Asian countries have placed particular emphasis on English language education, influenced by globalization and the growing importance of English as a key language in education and the economy. The popularization of bilingual education leads to the natural occurrence of the phenomenon of code-mixing in people's daily communication, which has become a common pattern of linguistic behavior. Code-mixing is defined as the practice of interchanging two or more languages in a single conversation or sentence [2-6]. It is important for educators to understand what factors influence code mixing in bilingual children to fully understand the broader meaning of bilingualism in a globalized world. There are various factors that influence the occurrence of code-mixing, but bilingual children tend to be deeply shaped by family factors such as their family language practices which frequently occur in informal daily communications. Bilingual children are sensitive to differences between languages at an early age [7-8]. Parental attitudes toward language use and the linguistic input provided by family members influence the frequency and pattern of children's code-mixing, subtly shaping the way they manage their bilingual identities. Such linguistic interactions not only determine how children switch between bilingualisms but also influence their bilingual identity and their propensity to make language choices [7-12]. For instance, in some immigrant families, parents continue to use their first language to communicate with their children; some cross-cultural families use a one-parent-one-language policy to develop their children's bilingual abilities.

Although bilingualism and code-mixing have been extensively researched, there are still gaps in understanding the manifestation of these phenomena in Asian contexts, particularly concerning the impact of bilingual Mandarin-English families. The purpose of this paper is to fill these gaps and provide unique insights through a comprehensive review of past literature, with a special focus on family language policies, and linguistic choices about code-mixing in parent-child interactions. This paper will begin with a brief review of the language acquisition of bilingual children, highlighting the importance of input and interaction for bilingual education. Subsequently, the introduction of code-mixing and the Markedness Model will be presented. After that, the focus will then shift to bilingual children in Mandarin-English families in Asia, reviewing some research to explore how family factors influence the occurrence of code-mixing of bilingual children in daily communications. Understanding these dynamics is important for educators and parents to grasp the complex interactions between language, culture, and identity in bilingual families.

2. Language acquisition of bilingual children

According to Kessler [6], the nature of language acquisition is a social behavior. In other words, successful language acquisition needs to be developed and achieved through social interaction with others. The developmental track of bilingual children follows the same principles as other children, requiring rich language input and social interactions to improve language skills [13]. Krashen [14] proposes the input hypothesis, which emphasizes the significant influence of the quantity and quality of language input on language acquisition. He believes that comprehensible input is the key to language acquisition. Although input has always played an important role, it is not possible to fully realize language acquisition by relying on input alone. In the 1980s, Long further proposed the interaction hypothesis, which highlights social practices and language use. Long [15] believes that language learners not only receive language input through interactions but also improve their language abilities through feedback and correction. While the input hypothesis focuses on the quality and comprehension of language input, the interaction hypothesis emphasizes the importance of input and language practice obtained through interaction. Both theories, though different in perspective, highlight the key role of input and interaction in language acquisition.

For children, the primary source of language input is typically their parents and the family serves as the primary environment for social interaction. Nevertheless, research [8] showed that although bilingual children are exposed to two different language inputs from an early age, they will still go through a stage dominated by their mother tongue. It is not until around the age of three that they truly become bilingual. However, their bilingualism develops as their social interactions expand and their second language input gradually increases. This process is closely related to the quantity and quality of linguistic input they experience and the frequency of interaction.

3. Code-mixing

Code-mixing refers to a phenomenon of blending elements of two or more languages within a discourse or conversation and reflects the complex linguistic dynamics of multilingual societies, frequently occurring in bilingual individuals' speech [1-5]. This phenomenon frequently occurs among bilingual speakers. Code-switching is another term commonly used alongside code-mixing, but they are different in concepts and applications. Some scholars [16-18] have made distinctions between them. Code-switching refers to a transition from one language to another, typically occurring between sentences, whereas code-mixing is a more regulated linguistic strategy that combines words or phrases from different languages within a single sentence to realize direct language exchange.

In 1993, Myers-Scotton [19] introduced the Markedness Model, which believed that speakers could select the language to use based on the social meanings they want to convey. This model also explains the rules of language choices and structures of code-mixing between bilingual speakers. Code-mixing, according to this model, occurs when speakers try to manage complex social identities. In this process, language is viewed as a tool to enable individuals to align themselves with a particular culture or social group. Thus, bilingual parents can appropriately use code-mixing strategies when communicating with their children to guide them in adapting to social norms and shaping their self-identity. By encouraging children to take marker assessments in different environments and make language choices accordingly, parents can help them better navigate and identify with their bilingual identity.

From a cognitive perspective, the occurrence of code-mixing illustrates the enhancement of bilinguals’ cognitive flexibility and meta-linguistic awareness. Bialystok [20] argues that bilinguals who code-mix regularly show better language system-switching abilities, which facilitate cognitive processes such as problem-solving and multitasking. Nicoladis and Genesee [7] point out that children have a strong sensitivity to language at a very young age, enabling them to distinguish differences between languages. Therefore, the occurrence of code-mixing does not indicate linguistic confusion. Instead, it reflects children’s strategy for efficient communication in a complex linguistic environment. Bilingual children who engage in such practices exhibit stronger executive functions, including better task-switching abilities and attention control [10].

4. Family factors influencing code-mixing in Mandarin-English children in Asia

In Asia, due to cultural diversity and globalization, bilingualism is widespread, and Mandarin-English code-mixing is frequently observed in children's daily communication. For children, parents are the main source of language input, with the family representing the social context for interaction. In bilingual families, parents' language choices directly influence children's language patterns. Similarly, parents' attitudes toward different languages and their support for bilingualism also impact children's code-mixing behaviors.

4.1. Family language policy

Family language policy refers to the explicit or implicit rules governing language use within the household [12]. As the primary source of language input for children, the type of home language policy implemented by parents has a significant impact on children's bilingual development. Bilingual parents always employ a Bilingual Support Policy where code-mixing is seen as a very normal phenomenon and usually represents a natural component of language proficiency. In this linguistic environment, children have the flexibility to switch freely between the different languages to meet different communication needs. This policy not only gives children greater freedom in their language choices but also potentially enhances their linguistic creativity and expressive abilities. A study by Qiu and Winsler [12] supports this view. They studied the language use, proficiency, and code-mixing behaviors of the Mandarin-English bilingual child growing up under a “one parent, one language” (OPOL) policy. Data were collected on a three-year-old child whose mother used to speak Mandarin while the father spoke English in the family through observations, interviews, and assessments. Major findings of the study illustrate that all family members showed some degree of code-mixing behaviors in communication despite the parents' OPOL policy. Another point of this study is that the child used more verbs in Mandarin (a verb-dominant language) and more nouns in English (a noun-dominant language), reflecting the structural differences between the two languages. Meanwhile, the prior research of Min [9] also explored how bilingual parents' discourse strategies influence their two-year-old child's Mandarin and English code-mixing behaviors. In this family, the parents had different attitudes towards code-mixing: the father was more tolerant of code-mixing, while the mother insisted on using English only. The final results showed that the child’s code-mixing behaviors differed: code-mixing was more frequent in conversations with the father and less frequent when communicating with the mother.

4.2. Parental attitudes and practices

Parental attitudes and practices toward bilingualism and language use also shape the occurrence of children's code-mixing behaviors. Parents who view bilingualism positively and actively support their children's use of both languages tend to create an environment that normalizes code-mixing [7-13].

Goodz's [10] research demonstrated that children will use the linguistic data available to them to form hypotheses about acceptable language use. This research highlighted the significant role of parental language behavior in shaping children's bilingual development, emphasizing that children's code-mixing might naturally result from imitating the mixed language input from their parents. If parents actively support and encourage the use of multiple languages, children are more likely to engage in code-mixing, viewing it as a normal and effective mode of communication [7-13]. Similarly, Kwan-Terry [21] conducted a similar study. Data were collected from a bilingual child from Singapore from the age of 3.5 years to 5 years by recording and observing scenes of his daily life. These data included conversations between bilingual children and their family members. The final results show that the language use patterns of parents and family members have a significant impact on the child's language choices. For example, although the mother primarily decides to talk with her child in Mandarin, she also uses English in some special situations. This flexibility is also reflected in the child's language choices. Meanwhile, this study also paid particular attention to how the child chooses and switches languages according to different conversation partners and contexts, and how he uses code-mixing to fill in gaps in his vocabulary knowledge.

These researches all highlight the importance of understanding the relationship between family language influence and children's bilingual development. Parents' positive attitudes towards support and flexibility in language use not only increase the frequency of code-mixing but also enhance children's linguistic creativity and adaptability. Children's code-mixing often reflects the language dynamics within the family, highlighting the need for well-considered family language policies to effectively support bilingual development.

5. Conclusion

To sum up, family factors are a critical source of Mandarin-English bilingual children's code-mixing behaviors. As Asia becomes a multilingual region, it opens new possibilities for further research and support to be done in this direction, which shows the potential benefits of raising bilinguals in terms of linguistic flexibility and cultural identity in the younger generation. Language policies and attitudes employed by parents in family education have a significant impact on children's bilingual development. These dynamics need to be understood because educators, parents, and policymakers work together through them to build the environments for nurturing and supporting bilingual talent in an increasingly globalizing world.