Second Language Processing

An Introduction

Nan Jiang

• Publisher: Routledge
• Year: 2018
• ISBN:
  9781134608324
• Edition: 1
• Topic: Sprachen & Linguistik
• Subtopic: Sprachwissenschaft

Second Language Processing: An Introduction is the first textbook to offer a thorough introduction to the field of second language processing (SLP). The study of SLP seeks to illuminate the cognitive processes underlying the processing of a non-native language. While current literature tends to focus on one topic or area of research, this textbook aims to bring these different research strands together in a single volume, elucidating their particularities while also demonstrating the relationships between them. The book begins by outlining what is entailed in the study of SLP, how it relates to other fields of study, and some of the main issues shared across its subareas. It then moves into an exploration of the three major areas of current research in the field—phonological processing, lexical processing, and sentence processing. Each chapter provides a broad overview of the topic and covers the major research methods, models, and studies germane to that area of study. Ideal for students and researchers working in this growing field, Second Language Processing will serve as the go-to guide for a complete examination of the major topics of study in SLP.

Chapter 1
Introducing Second Language Processing

1.1 Introduction

The use of a language other than one’s native language is common. This can be seen in the number of people who can use a language other than their native tongue. According to a survey by the European Commission published in February 2006, 56% of the 450 million people in the European Union “are able to hold a conversation in a language other than their mother tongue” (p. 8). In the USA, the number of people who speak a language other than English at home has increased significantly over the past four decades, from 23.1 million in 1980 to 60.6 million in 2011. The percentage of this population has also increased steadily from 11% in 1980 to 21% in 2011, according to the US census data. A majority of these people supposedly speak their native language at home and speak English as a second language (ESL) outside of home. These numbers do not include English native speakers in the USA who are able to speak another language such as Spanish, French, Japanese, and Chinese. Additionally, there is even a bigger number of ESL speakers in other parts of the world. In China alone, over 100 million school and university students are studying English, according to the Chinese Ministry of Education.

The use of a non-native language or a second language (L2) has become a phenomenon for scientific enquiry. Multiple approaches have been taken in this endeavor. One of them is to examine this phenomenon from a cognitive and psycholinguistic perspective. Second language processing (SLP) represents such an approach. In this approach, L2 use is viewed primarily as a cognitive event rather than, for example, a cultural, sociolinguistic, or pedagogical one. The primary goal of this research is to understand the mental processes and mechanisms involved in L2 use and what such knowledge can tell us about L2 acquisition. This focus on cognition is reflected in its reference to language use as language processing, which represents an information processing approach to the study of human cognition, including language. In this approach, the human mind is considered as an information processor (or language processor) that has a limited processing capacity and storage and that relies on a set of algorithms or grammatical rules in handling linguistic input and output. In this context, the term language processing refers to both receptive and productive use of a language. Thus, SLP is concerned with the mental processes involved in the receptive and productive use of a second language.¹

The study of the mental processes involved in language use by L2 or bilingual speakers can be traced back at least to James McKeen Cattell’s (1887) research. As part of his effort to study the temporal issues related to mental processes, he examined the amount of time bilinguals needed to name pictures in their native language (NL) and non-native language (NNL) and to translate in both directions. He found that individuals needed 149–172 milliseconds (ms) more to name pictures in a NNL in comparison to naming them in their native or first language (L1), and that translation from L2 to L1 took less time than the reverse, both of which findings were replicated in more recent research.

The 1980s witnessed a significant increase in SLP studies. Some of the earliest SLP research included studies by Flege and Davidian (1984) and Flege and Hillenbrand (1984) on phonological processing, by Muchisky (1983), Koda (1988, 1989) on word recognition, and by Harrington (1987) and Kilborn and Cooreman (1987) on sentence processing. Substantial progress has been made in the past three decades in the development of theoretical frameworks (e.g., the proposal of L2 speech processing models), in the expansion of the scope of investigation (e.g., the wide range of structures involved in sentence processing research), in the number of discovered L2 processing phenomena (e.g., the interaction of L1 and L2 phonological representations, a larger frequency effect in L2 than in L1 in visual word recognition, a stronger presence of form-based lexical representation in the L2 lexicon), and in methodological advances (e.g., the application of the masked priming paradigm, the use of eye trackers, the electroencephalogram, and magnetic resonance imaging).

As it stands today, three SLP areas can be identified that have received the most attention: phonological processing, lexical processing, and sentence processing. The first area deals with speech perception and production among L2 speakers primarily at the segmental and suprasegmental levels, the second is about L2 word recognition and lexicosemantic representation and development, and the third focuses on syntactic structure building and the representation and processing of morphosyntactic and lexicosemantic knowledge in understanding sentences.

These three areas of SLP focus on three different aspects of language, and they sometimes deal with issues that are unique to that area. For example, the relationship between perception and production and the impact of degraded input on processing are topics unique to phonological processing research, the decomposition issue, i.e., whether certain linguistic units are represented and processed holistically or in an analytical manner, is primarily examined in lexical processing, and the relative importance of syntactic and semantic information in language processing concerns sentence processing only.

However, these three areas also share a great deal of similarities. They study the same population of L2 speakers and the same general phenomenon of language use by L2 speakers, and they share the same goal of understanding how L2 processing is similar to or different from L1 processing, they share the general methodological approach of comparing L2 speakers to L1 speakers, and they also share the same experimental approach and often draw from the same repertoire of experimental techniques and paradigms. Additionally, they also deal with very similar research questions. For example, the role of an individual’s L1 linguistic knowledge in L2 processing is an important research topic across all three areas. Individuals’ L2 learning profile, e.g., the age of onset, length of residence in the L2, and their proficiency, are factors frequently considered in the study of many topics across these areas. Consequently, research findings across these areas are likely to be related in one way or another.²

In the remainder of the chapter, I hope to highlight a few characteristics of SLP research shared by its subareas, and discuss some recurrent themes across these areas. In so doing, I hope to enhance the awareness of the similarities in research goals, research topics, research findings, and research methods across the different areas of SLP and promote SLP as a unified and coherent field of inquiry.

1.2 Characterizing SLP Research

SLP research enjoys several shared characteristics across its three areas. These characteristics help make SLP a coherent and unique field of inquiry. At the same time, these characteristics may help define the relationship between SLP and other fields, particularly psycholinguistics and second language acquisition (SLA). Four such characteristics can be identified.

1.2.1 A Cognitive Focus

First, the primary goal of SLP research is to understand the mental processes involved in L2 use, even though L2 use can be approached from a number of different perspectives. The study of foreign accent may serve as an example to illustrate this focus. Most adult L2 speakers speak with an accent. From a sociolinguistic perspective, one may ask whether the degree of accentedness has to do with their attitude toward the new language or culture, or whether it is affected by the person an individual is speaking to. These issues are legitimate and interesting issues to explore but they are of less interest to SLP researchers. Instead, their focus is more on the cognitive underpinnings of accented speech. They explore what acoustic features are similar or different in the mental representations of phonological categories between NNS and NS and the causes of any representational deviation that contributed to accented speech. Similarly, in studying lexical processing, SLP researchers are less concerned with how word recognition or literacy development is affected by individual differences in motivation, attitude, or socioeconomic background. Instead, they attempt to understand how lexical knowledge is represented and accessed in L2 word recognition in general and pay attention to linguistic and individual factors that may potentially affect lexical representation and access.

Given its focus on mental processes involved in L2 processing, SLP is deeply rooted in psycholinguistics and cognitive psychology. Indeed, many early L2 processing studies arose in the context of exploring general language processing issues such as the study of L2 sentence processing within the competition model (Gass, 1987; Harrington, 1987; Kilborn & Cooreman, 1987) and the study of the processing of non-native speech segments associated with the perceptual assimilation model (e.g., Best, McRoberts, & Goodell, 2001). Today’s SLP research continues to rely heavily on psycholinguistics for conceptual frameworks, empirical findings, and research methods. Psycholinguistic findings obtained from L1 speakers often serve as a point of departure and baseline for comparison in many L2 processing research, such as the analysis of acoustic cues in speech perception (Section 3.2.1), morphological priming (Section 5.2.2), and relative clause attachment (Section 6.2.2). In light of this relationship, SLP may be considered as an extension of psycholinguistics from native language processing to L2 processing. It is not surprising in this sense that many leading scholars in SLP teach in a psychology department and many SLP empirical studies were published in psycholinguistics journals.

1.2.2 A Broader Scope Than L1 Psycholinguistics

Second, in extending language processing research from L1 to L2 speakers, SLP has to consider issues that are not usually considered in L1 psycholinguistic research, as L2 processing is more complicated and involves more factors. For example, for most L2 speakers, the L2 is learned after the establishment of an L1 and a related conceptual system. Thus, L1 influence and the nature of the L2-concept connections become unique issues to consider in studying L2 processing. Individuals also learn an L2 at variable ages, which raises the issue of how maturational factors may affect L2 acquisition and processing. The L1 and age factors combine to raise the acquirability issue, i.e., to what extent NNS are able to develop nativelike competence in a new language. For many individuals, an L2 is learned with less than optimal input. For example, classroom learners often do not have sufficient contextualized input or ample opportunities for meaningful interaction. This is likely to affect L2 learning and processing, too. Thus, L1, age, and input become three factors that are uniquely relevant to L2 processing research but less so in L1 processing research.

Other acquisition differences may also have processing ramifications. For example, unlike children learning their L1 where the development of speech and literacy is sequential, i.e., speech before literacy, L2 learners often learn to speak, listen, read, and write at the same time. This creates a unique situation with regard to the role of phonology and orthography in visual word recognition. Psycholinguists have found that phonology is actively involved in visual word recognition and attribute this finding to the development of speech before literacy (Bosman & De Groot, 1996; Davis, Castles, & Iakovidis, 1998). Now we have to wonder if phonology also plays a significant role in visual word recognition in an L2, as well. Additionally, the tendency for adults to rely on explicit knowledge or explicit learning may also result in two opposite patterns of behavior between L1 and L2 speakers, i.e., the presence of processing competence without explicit knowledge among L1 speakers and the presence of explicit knowledge without processing competence (i.e., automatic access of knowledge) among L2 speakers.

Thus, SLP research explores both issues that concern language processing in general and issues that are unique to L2 processing. The exploration of these L2-specific issues represents the unique contribution SLP makes to the broader field of cognitive science. It also provides important insights for understanding issues related to second language acquisition (SLA).

1.2.3 From Processing Data to Representation and Acquisition Issues

Representation, acquisition, and processing are closely related in SLP research. Mental representation refers to the information and knowledge that is stored in our mind in the form of symbols, images, or propositions. It reflects what we know. We behave in a certain way in language use because of the linguistic representation we have developed. I point to a tree and tell my daughter that it is a willow because I have a mental representation for this type of tree in my mind, i.e., a concept and/or an image, which includes probably its color and shape among other things, and a word that refers to this concept. A participant considers two acoustic signals, e.g., /i/ and /ɪ/ as in seat and sit, different sounds because there are two different phonetic categories represented in his or her mind, each corresponding to a different signal. From a cognitive perspective, the mental representation of our linguistic knowledge determines our linguistic behavior. The essence of language learning lies in the changes in an individual’s mental representation of the language being learned. These changes may include the differentiation of two phonetic categories in phonological development, the creation of a new lexical entry in our lexicon in vocabulary acquisition, and the formation of a subject-verb agreement rule in learning syntax.

However, mental representations and their changes are often difficult to study directly, even with today’s technology. Instead, we have to infer what linguistic knowledge is represented or how it has chan...