Circuit Switching Versus Packet Switching

Nurhazimah Binti Mohd Za’ba
Nursyafikah Binti Farakkasi 
Nursyahirah Binti Mohd Sanusi 
Nur Hidayu Binti Salleh
Nur Syafiqah Binti Zulkiflee
Nurul Ain Binti Mohd Nassir Adabi

The purpose of this study is to describe difference between packet switching and circuit switching. Circuit switching is dedicated communication between two stations that connected sequence of links between network nodes. It consist 3 phase communication which are establish, transfer and disconnect. Packet switching is data transmitted in small packets that each packet contains a portion of user data plus some control information. There are a few difference between packet switching and circuit switching include the bandwidth, dedicated path, call setup delay and so on.
In telecommunication networks they carried the information signals among an entity involved in the process of information transfer which may be in the form of a telephone conversation. In order to connect multiple devices, it must be point to point connection between pair of devices which we called as switches. Unfortunately, it will increase the number of connection. There are two types of switching techniques including circuit switching and packet switching. Switching is a collection of switching elements arranged and controlled to setup communication path between any distant points.

Get Help With Your Essay
If you need assistance with writing your essay, our professional essay writing service is here to help!
Essay Writing Service

Circuit switching is a network which allows the associated voice which it will followed between the two respective users refer in the Figure 1 is the circuit switching network.. The end to end communication was established during the duration of the call. It is a dedicated link or path which established between the sender and the receiver. It is maintained for the entire duration of the conversation.

Figure 1: Circuit SwitchingNetwork
Packet switching is a network which it does not requires establishing the connection in initially. Next, the connection or channel available usually used by many users. But, if the capacity of the users increases, it will lead to congestion in the network. This network mainly used for data and voice applications requiring non-real time scenarios. It also can handle in two ways which are datagram and virtual circuit refer Figure 2 is the Packet Switching network.

Figure 2: Packet Switching Network.
Data communications have been achieved by using a variety network such as PSTN, leased-lines and more recently ISDN and ATM/ Frame relay. These networks are partly or totally analogue or digital using technology such as circuit-switching, packet-switching etc. Main difference between Packet Switching and Circuit switching is that communication lines are not dedicated to passing massage from the sources to the destination. In Packet Switching, different message (and even different packet) can passed through the routes, and when there is a ‘dead time” in the communication between the source and the destination, the lines can be used by other sources. Oguntala George Adeyinka(2013) said that packet switching is an attempt to make a better utilization of the existing network by splitting the message to be sent to packets. Each packet contains information about the sender, the receiver, the position of the packet in the message as well as part of the actual massage. There are many protocols defining the ways packet can be sent from the sender to the receiver.
Then, Pablo Molinero-Fernandez & Nick McKeown (2004) said that Circuit Switching can be decomposed into a fast path without per-packet processing, and a slower path for establishing/releasing circuit, which is similar in complexity to forwarding a single packet in Packet Switching. However the slow path needs to be taken much less often (for example, the average TCP connection lasts more than 10 packets, which mean that connection are established at an average rate at least 1/10 that of packet processing in router ). For these, we argue that circuit switches can operate much faster than packet switches.
The concept of packet switching had two independent beginnings, with Paul Baran and Donald Davies [4]. Leonard Kleinrock [5] conducted early research and authored a book in 1961 in the related field of digital massage switching (without explicity using the concept of the packet), and also later a played a leading roles in building and management of the world’s first packet switched network, namely the ARPANET.
After that, K.Giridhar stated Packet switching is also called connectionless networking because no connections are established. The advantage of the connectionless packet model in that packets are forwarded independent of other packets. Packets are forwarded on the fly by routers based on the most current best path to a destination. If a link or routers fails, packets are quickly diverted along another path.
Dr.FaridFarahmand&DR.Qiong (Jo) Zhang stated the most common example of a circuit-switched network can be found in public telephone network (PTN) supporting services such as POTS (plain old telephone system) and long distances calls. Other examples of a circuit switched services are integrated service digital network (ISDN) and switched 56, 64, 384 (Kbps) services. The majority of wireless application protocols (WAP) enables phone also operate on top of circuit-switched networks. Furthermore, many public networks dedicated to data transport also use circuit-switching techniques; an example of a network in Europe is circuit-switched public data network (CSPDN), which transport data on circuit-switching networks using the X.21 protocol. Circuit-switching also has wide applications in optical networks including wavelength division multiplexed (WDM) systems and WDM SONET networks.
Topic for our assignment is packet switching. We make a research to get the information about this topic in library and also internet. There’s a lot information we get from internet that give information about definition, benefit and function. The definition of packet switching give us simple introduction to us to understand the concept of packet switching. After we understand the concept, we find out the article or journal from internet that can give us a more information about packet switching. The article must be from year 2012 to make sure that information are relevant with our topic and current study. This article can be found from OPAC system that can used in library website. Every article has their own writer and bibliographic databases that make us easier to find the reference if there’s a question refer to packet switching if we are not understand to topic. In internet, we find extra information by inserting the term of packet switching in key words in Google to get more explanation about function of packet switching. The bibliographies are used for us for further study and this way give us more information packet switching. The journal is created by other university and this journal help us to understand the topic. The history about packet switching also found in internet and the technology in packet switching also change and it make data can be transfer in good ways even though the data is large. Based from the information get from internet will make us more understand about the topic and help us to success give simple explanation about the topic of packet switching in report.
After make research in internet, we go to library to find information regarding our topic which is packet switching. There’s some book that have some explanation about the benefit of packet switching. We also know about the main concept of packet switching from the book in library. The main concepts give more information about how the packet switching work in real situation. After that, we know the weakness and benefit of packet switching in transfer data in computer or cellphone. Even this packet switching have a disadvantage when transferring the data and it still can function more efficient after they develop a new technology this make sure the data can send to location. We also can refer from our senior journal in library after they done their final year project which related with topic packet switching. This information can help us to know the function packet switching with a technology that still use the concept during transferring data in destination address in packet. We also make a small interview with a friend who already work in IT department but the information that we get from him are not related to our topic. So, we are not use that information to be write in this report.
Lastly, we combine all the information about packet switching that we found from internet, journal and book to be write in this report. Before we write this report, we already discuss and make some explanations in this report to make sure this report are occupy our objective to understand the topic of packet switching. We try to help each other understand the topic based from information we get from internet , journal , article and book. Everyone have knowledge and information after done this project that can be used in career life. We take about more than one month to finish this research and this report can be use as reference for our junior if they want to continue this research in their present time in future.
At the end of this article, we present the two switching techniques used in networks: circuit switching and packet switching; whereas datagram packet switching and virtual circuit packet switching. Then, we also are able to compare a difference between circuit switching and packet switching.

Circuit Switching

Datagram Packet Switching

Virtual Circuit Packet Switching

Fixed bandwidth

Dynamic bandwidth

Dynamic bandwidth

Overload can block packet delay

Overload increases packet delay

Both; can block and increases packet delay

A dedicated path

Not a dedicated path

Not a dedicated path

Path is established before data transmission begins

Route is established before any packets sent

Route is established before any packets sent

Call setup delay

Packet transmission delay

Both; call setup and packet transmission delay

No speed or node conversion

Speed and node conversion

Speed and node conversion

[1] Oguntala George Adeyinka, “Network Solution, Applications and challenges of Mobile Computing In Africa,” International Journal of Scientific & Engineering Research, volume 4, issues 10, pp. 884-889, October 2013.
[3] Pablo Molinero-Fernandez & Nick McKeown, The Performances of Circuit Switching in the Internet, Computer System Laboratory, Stanford University (2004).
[4] Roberts L.G (1978). The Evolution of Packet Switching, Proceedings of the IEEE, vol.66, no11, pp.1307-1313. 1978.
[5] Kleinrock L. (1961), Information Flow in Large Communication Nets, RLE Quarterly Progress Report, July 1961.
[6] K.Giridhar, “Packet Switched Data Network and Its Evolution”, Information Technology and Communications Resources for development, Telecommunications and Computer Networks (Tenet) Group, Department Of Electrical Engineering, Indian Institute Of Technology Madras, Chennai-600036, India (2007).
[7] Dr.Farid Farahmand & DR.Qiong (Jo) Zhang, Circuit Switching, Central Connecticut State University & Arizona State University at West Campus (2007).
[9] Telecommunication System Engineering, Technical University of Malaysia Malacca, chapter 5.

Delay Aware Multipath Switching Zone Traversal (MSZT)

Delay Aware Multipath Switching Zone Traversal (MSZT) Approach for MANET
Route discovery and data transmission in Mobile Ad-Hoc networks (MANETs) are the key procedures which influence the performance of the network. The data transmission through the discovered route with in minimum delay is considered as a major problem in the MANET. Hence this paper proposed an approach called Multipath Switching Zone Traversal (MSZT) approach to minimize average delay. In this approach, the broadcast has been converged to two zones after the initial broadcasting in order to minimize the broadcast delay. Based on the data size and the Time to Live (TTL), the data transmission is initiated via multipath or single path. Game theory approach (GTA) is a powerful mathematical tool for analyzing the strategic communications among several decision makers. Hence this paper utilizes the GTA for switching the path between inter-zonal and intra-zonal nodes for different source data to provide a successful data transmission. The simulation results show that the approach is efficient in terms of delay, packet delivery ratio, and localization error when compared to the existing approaches.
Keywords: Routing, MANET, MSZT, game theory approach, delay
1. Introduction
A multi-hop wireless Ad hoc network (MANET) is composed of mobile nodes, which can communicate without any aid of centralized Infrastructure (T.Durga , 2015) The demand of different multimedia applications such as surveillance system and video on demand service over MANET has been increasing rapidly in recent years. However, it is not easy to support the data transmission according to the end-to-end delay requirements over MANETs. A number of routing protocols have been proposed for Ad hoc mobile networks to improve the QoS ((K.S.Dinesh , 2014).

Get Help With Your Essay
If you need assistance with writing your essay, our professional essay writing service is here to help!
Essay Writing Service

Most of the routing protocols use the minimum hop numbers as a metric of route cost while taking routing decisions. However they ignore the some important link capacity properties because of simplicity and ease of implementation. Each node in the network may have different traffic load and hence, nodes which have more active neighbors may experience more collisions (Rakesh Kumar, et, al., 2010). Uncertainly some of the over loaded nodes may fall on the minimum hop route, it may considerably introduce a longer delay, even though the number of nodes in the chosen route is minimum. In addition, if some of the over loaded nodes are congested; it may result in huge packet drop rates and consequent retransmission. This may increase the end – to – end delay between two end points.
Thus in this paper Multipath Switching Zone Traversal (MSZT) routing approach has been proposed in order to reduce average delay. The approach reduces the number of broadcast after the initial broadcast in the network. Multipath data transmission is enabled based on the TTL and the data size. The game theory approach is used to switch the path between inter-zonal and intra-zonal nodes for different source data, which is a powerful mathematical tool. The path switching while during the data transmission provides a successful data transmission by checking the path availability. Thus this approach enhance the QoS demanded MANET applications
The rest of the paper is organized as follows: Section 2 presents the recent related work on the delay aware MANET. Section 3 describes the system model of the proposed Multipath Switching Zone Traversal (MSZT) routing approach. The explanation of the Multipath Switching Zone Traversal (MSZT) routing approach is presented in the section 4. Section 5 describes the simulation results. Finally, section 6 renders the conclusion
2. Related Work
(Saad M. Adam, et, al., 2013) presents an reactive routing protocols overview in terms of QoS requirements in MANET. Due to the requirements has been fulfilled for the commercial, real-time, and multimedia applications in MANET. The delay has been considered as one of the important QoS metric to satisfy the application requirements.
The delay optimization approach has been presented by the authors (Syed Jalal Ahmad, et, al., 2015) for multimedia application in MANET. The Knapsack algorithm is used for buffer management to maximize and minimize the in order and out order packets simultaneously. The buffer internals are exploited and the adjusting the buffer usage dynamically makes the node to transmit the packet in the preferred order to its successive nodes. The simulation results show that the approach transmits multimedia data packet without loss and in minimum delay.
(K. Sasikala,et, al., 2014) proposed the finite state machine for queue and transmission management mechanism to minimize the packet delay time in MANET. Information about the delay for each data flow in the queue is maintained in a node. A timer is fixed to each flow in a node and it has been updated dynamically. Based on the nodes characteristics the queue is controlled to minimize the packet delay.
(Obaidat, M, et, al., 2011) proposes a multipath routing protocol for MANETs. The protocol establishes the route, which having the lowest delays relied upon the communication of various layers. The performance of the protocol is compared with the AODV and AOMDV.
(V. R. Budyal and S. S. Manvi, 2013) presents a clever agent based on-demand QoS routing methodology in MANET. The intelligent agent utilizes the neuro-fuzzy logic aided by Q-learning. The DSR protocol has been enhanced to discover all the multiple paths and the condition of the path from source to destination. The fuzzy membership function has been optimized by the software agents and if then rules are made to take decision in the system.
A cross layer design is made for delay concerned Node Disjoint Multipath AODV in Ad Hoc Network by the author (GawasMahadev A, et,al., 2014). The channel and link information has been obtained by applying the cross layer design between the MAC and routing layer. The path status has been updated subsequently by concerning the lowest delay attained at each intermediate node. The protocol is compared with the AODV and NMDR and it perform better than the other two in terms of routing overhead and packet delivery ratio.
3. Multipath Switching Zone Traversal (MSZT)
The Multipath Switching Zone traversal routing approach is proposed in this paper to minimize the average delay in the network. The approach is implemented after the initial broadcast from the source. The MSZT is composed of two parts such as route discovery and path switching.
3.1 Route Discovery
During the initial broadcast the source node obtains the information about the location of the destination node by using the GPS service. After the initial broadcast, the source node divides its communication range into four equal zones for further broadcasting. The node broadcast the RREQ message through one or at most 2 zones, where the broadcast should be destination oriented zone based on the location information of the destination node. Only one node must be selected in each zone for forwarding the broadcast. Alike the source node, the forwarding node also divides its communication range into four equal zones and forward the broadcast in the same manner. The process will take place until the destination node is reached.
Before the data transmission via the shortest path, the Expected Transmission Time (ETT) and the TTL value has to be estimated to enable multipath through the different zones in order of increasing the packet delivery factor (PDF).
Expected Transmission Time (ETT) is the time taken for transmitting a data packet successfully to the destination. The ETT is estimated based on the data size of a packet.
Here ETX is the expected transmission count i.e. expected number of transmissions that a node needs to transmit a packet successfully to a neighbor. The ETX can be estimated as follows
Where dforward is the received probes from a A dreverse is the received probes from B
Where t is the average time a single data packet requires to be delivered and the t can be estimated as follows
The maximum Time to Live (TTL) is a timer data part in the IP header which specifies RREQ packet life time before it is discarded (Cheng-Yuan Ho,et, al., 2007). All the RREQ fields in its route cache have been maintained in order to find the route minimum Time to Live (TTL) value after TTL time expires and the information about the TTL is available in the RREP packet.
If the ETT is greater than the TTL, then the packets are fragmented and then the packet has been transmitted over the multiple paths.
3.2 Path Switching Algorithm based on the Game theory Approach
Game theory is a mathematical tool for analyzing and estimating how a person behaves in strategic situations. The game is composed of three fundamental components: a set of players, a set of strategies, and a set of payoffs. Nodes (players) are the decision takers in the game. The strategies are the various options available to nodes such as (TTL, node availability) (Badr Benmammar, Francine Krief, 2014). At last the utility function (payoffs) make a choice of all possible outcomes for each player. The components of the proposed game theory approach are given in the table 1.

Components of a game

Elements of a wireless network




Multiple path


TTL, node availability

Table 1 Components of the proposed Game Theory Approach
After each transmission, the next hop zonal node must update the path availability (zone availability) to its predecessor node. The competition is between the multiple paths available between the source nodes to the destination node. After each successful data transmission the payoff is considered such as the availability and the TTL in order to stay on the same path otherwise path is switched to the next highest probability value of the available path.
For each available path a probability value is estimated and it is formulated as follows
Here the total hops is the number of hops in the path and the not available hops is the number of hop which is not available for next data transmission due to the reason “the node is available for other source node in the network”. This information is updated to the source after each successful transmission. The proposed path switching algorithm based on the game theory approach is shown in the figure 1.

For each source node after each successful data transmission


For each available multiple path


Next hop zonal node must update the path availability (zone availability) to its predecessor node.

Calculate the not available node based on the collected zone availability information from all the predecessor node

Calculate the probability value using equation 4


Switch over the path based on the probability value of each path


Figure 1 Path switching algorithm based on the game theory approach
5. Simulation Results
The proposed Multipath Switching Zone Traversal (MSZT) routing approach make an effort to minimize average end to end delay to enhance the network performance. The NS2 simulation has been adopted to evaluate the performance of the MSZT approach and the simulation setup is shown in the table 2. The performance of the MSZT is compared with other existing approach such as Zone Tree Routing (ZTR), Path State Routing (PSR). The metrics used to measure the performance of the proposed approach are broadcast delay, localization error and packet delivery factor.

Simulation Parameter




Topology size


Number of nodes

50,100, 150, 200

Transmission range

250 m

Traffic type


Packet size

512 bytes

Pause time


Min speed

1 m/s

Max speed

5 m/s

MAC protocol

IEEE 802.11

Simulation time


Table 2 Simulation Setup
A. Performance metrics
Broadcast Delay:
Broadcasting is the fundamental process where the probe transmission from the source to all other forwarding nodes to arrive at the destination. The broadcast delay is the difference between the packet receiving time (prt) by destination and the packet transmitted time (ptt) from the source node.
Localization Error:
Localization error (LE) is accounted as a one of the performance metric in the scattered network and it yields the LE of a node.
Where EMax is the maximum LE
Where a is the radius based on the communication range of node i in a network model and it can be determined as follows
Where N is the number of nodes A is the area of the network and is the connectivity order ie the number of neighbor nodes linked to the node i
The average value of LE for a network can be determined as follows
Packet Delivery Factor
Packet delivery factor is defined as the ratio of the number of packet arrived at the destinations to those transmitted data packets by the source.
B. Discussion
In the proposed approach the broadcast delay has been minimized by converging the broadcasting to one or at most two zones. The nodes distance increases then the broadcast delay increases simultaneously. Figure 2 shows the broadcast delay concerning the distance. The broadcast delay attained by the proposed MSZT for 10 m distance is 15 ms, while the PSR and the ZTR incurred 16ms, 19ms.

Figure 2 Broadcast delay
The localization is directed in our approach by broadcasting towards the destination oriented zones after the initial broadcast. The increase in distance between the nodes affects the localization error. Figure 3 shows the localization error concerned with the distance. The localization error attained by the proposed MSZT for 10 m distance is 2%, while PSR and ZTR attained 16%, 18%.

Figure 3 Localization error
The multipath switching algorithm in the proposed approach improves the packet delivery factor by considering the TTL and path availability. When the number of group (paths) involved in the data packet transmission increases then the packet delivery factor (PDF) will decrease. While in the proposed a minimum number of zones have been utilized than the existing methods in the network. Figure 4 shows the packet delivery considering the number of groups. For 2 groups the proposed approach MSZT acquired 0.94, while the PSR and ZTR incurred 0.91, 0.83.

Figure 4 Packet Delivery Factor
6. Conclusion
This paper proposes a Multipath Switching Zone Traversal (MSZT) routing approach for achieving a minimum average delay in a network. The number of broadcast has been converged to two zones after initial broadcast in order to minimize the broadcast delay. Multiple paths have been selected through different nodes presents in different zone. While during data transmission the TTL and the data size has been checked to provide a successful data transmission with minimum transmission delay. Furthermore a game theory approach based path switching algorithm has been proposed after each successful data transmission to improve the delay metric for each data transmission. The simulation results show that the proposed routing approach performs better than the existing approaches in terms of Localization error, broadcast delay and Packet delivery factor
1.Syed Jalal Ahmada , V.S.K. Reddyb, A. Damodaramc and P. RadhaKrishnad, “ Delay optimization using Knapsack algorithm for multimedia traffic over MANETs”, Expert Systems with Applications, Elsevier, vol 42, issue 20, pages 6819-6827.
2. K. Sasikala, Dr. R. S. D. Wahidabanu, “ Adaptive packet scheduling technique to minimize the packet delay time in MANET by maintaining a Queue for each flow through FSM Mechanism”, Journal of Convergence Information Technology(JCIT), Vol 9, No 3, 2014.
3. Saad M. Adam , Rosilah Hassan, “Delay aware Reactive Routing Protocols for QoS in MANETs: a Review”, Journal of Applied Research and Technology, Elsevier, vol 11, issue 6, pages 844-850, 2013.
4. Obaidat M, Ali M.A, Obaidat M.S, Obeidat S, Shahwan I, “A Novel Multipath Routing Protocol for MANETs”, International Conference on Wireless Communications, Networking and Mobile Computing (WiCOM), pages 1-6, 2011.
5. Cheng-Yuan Ho, Yaw-Chung Chen, and Cheng-Yun Ho, “Improving Performance of Delay-Based TCPs with Rerouting”, International conference on Communication letters, IEEE, vol 11, no 1, 2007.
6. Badr Benmammar, Francine Krief, “Game theory applications in wireless networks: A survey” International Conference on Software Engineering, Applications of Information Systems in Engineering and Bioscience, 2014.
7. V. R. Budyal and S. S. Manvi, “Intelligent Agent Based Delay Aware QoS Unicast Routing in Mobile Ad hoc Networks”, International Journal of Multimedia and Ubiquitous Engineering, vol 8, no 1, 2013.
8. GawasMahadev A, Gudino, Lucy J, Anupama K. R, Rodrigues, Joseph, “A Cross-Layer Delay-Aware Node Disjoint Multipath Routing Algorithm For Mobile Ad Hoc Networks”, International Journal of Wireless & Mobile Networks, vol 6, issue 3, page 39, 2014.
11. K.S.Dinesh , “Routing Overhead Reduction and Selection of Stable Paths in MANET”, International Journal of Inventions in Computer Science and Engineering (IJICSE), vol 1, issue 9, pages 2348-3539, 2014.
12. T.Durga , “Maximizing the Network Lifetime by Using a Mobile Aware Topology Control Algorithm in MANET”, Journal of Recent Research in Engineering and Technology, pages 2349-2252, vol 2, issue 3, 2015.
13. Rakesh Kumar, Manoj Misra and Anil K. Sarje, “A Simplified Analytical Model for End-To-End Delay Analysis in MANET”, International Journal Computer Application (IJCA), no 4, article 6, 2010.

Code Switching, Emotions and Trauma Narratives

Bilingualism or multilingualism being described merely as a phenomenon where an individual is able to speak and write in two or more languages is not exactly an accurate definition; the term also applies to those who think and feel in two or more languages and this can take place in multilingual communities or not. Early research in bilingualism shows that language 1 or L1 is often preferred to express emotions and the second language or L2 to easily expresses emotional detachment. According to Dewaele[1], this explains that people who learn a L2 later in life might find it easier to swear or talk about sex taboos in L2, as this language has not been socialized into appropriateness in a particular context. On the contrary, if the person has acquired two or more languages through primary socialization, the expression and decoding of emotions would be provided with a broader variety that would enable a more complex expression and decoding of emotion; this variety would line up with language-specific sociocultural norms.

Get Help With Your Essay
If you need assistance with writing your essay, our professional essay writing service is here to help!
Essay Writing Service

For Panayiotou[2], bilingual and multilingual individuals are often described as multilayered selves that experience the world through them, the expression and interpretation of emotions would be included in these layers and would play an important role in determining the expression chosen. Therefore, bilingual speakers behavior present quite often the use of code switching (CS from now on). Hoffman says when talking about CS “each language makes the other relative” as it can be used to represent the best the specific emotional state.

Some researchers have questioned that certain languages excuse expressions of negative emotions. A bilingual speaker who finds him/herself in a situation would prefer to switch from L1 (Japanese) to L2 (English) when having to express anger. (Dewaele) Therefore, bilinguals can resort to dual sources to express emotions, being the reason to change the lack of an unequivocal concept in the other or others languages. CS can be used in two or more codes as a way to deal with difficult emotional situations. When we learn a L2 in a formal classroom context, the emotional part is not as involved as in L1 and strong feelings and repressed emotions are likely to be best expressed in L1 and CS would be used to get closer or far from an emotional conflict. Research in CS has mainly concentrated on perception and recollection of experiences whereas language and emotion focus on linguistic behavior. CS has been particularly drawn on while narrating difficult experiences.

Then, what is the logic behind the use of CS? Can we relate CS to a specific narrative discourse? According to Burr[3] (2015), social constructionism can help us understanding how narratives are conceptualized; she argues that narrative situates whereas dynamic discourses construct. Social constructionism proposes that when a person tells a story they also present and negotiate inevitably their social identity. In a social constructionist approach, storytelling would be context and the people would represent the actors (Augoustinos, Walker, & Donaghue 2014). For default narratives, there is one active storyteller but also a dynamic co-constructed narrative with multiple tellers. Traumatic narratives have proved to present most stories with multiple storytellers. Narrative therapy assumption is that helping people to reinterpret their life stories may help them to be able to change their present lives.

Why CS and storytelling is so much connected? Storytelling permits individuals to responds to at least five principal purposes: contributes to coherence through synthesizing personal experiences, helps creating distance by creating a story, supplies a communicative function by the narrator to the audience, promotes the reinterpretation and reevaluation of past events by giving the narrator the possibility to narrate past events and encourages explorative and therapeutic functions.

Dewaele[4] research informs that multilingual individuals feel “significantly less logical, less serious, less emotional and increasingly fake when using their L2, L3 and L4 compared to L1”, endorsing the idea that L1 is a more authentic and a closer language. Nonetheless, different languages help people to enact in a range of facets and certain languages line up better with specific emotional expressions.

As for CS, sociolinguistic have disagree that bilingual may use it to show an affective side. Bond and Lai[5] (1986) concluded in an experimental research with Cantonese (L1) and English (L2) that CS to L2 took place in order to permit to speakers to give speech to subjects that were too disturbing emotionally in L1. Martinovic & Altarriba[6] (2013) explain that CS in L2 would respond in a detachment effect, but in fact, this detachment will allow the individual to be able to express difficult experience without being overwhelmed with upsetting feelings.

Some scholars like Pavlenko[7] respond to CS in bilinguals as a response to non-perfect equivalents concepts between the different languages, not always being L1 a more emotional language. Other scholars like Panayioutou insist on bilingualism as being found and founded in two languages, depending the importance of the emotion on the context and the needs of the individual. Martinovic & Altarriba provide a double-side vision of CS phenomenon. On the one hand, CS constitutes a resource to provide a better understanding of what the individual wants to communicate. On the other hand, CS is considered a tool that provides distance into far too emotional subjects in one of the bilingual speaker languages. Numerous questions about the relation of emotions and discourse context remain unanswered though.

Dewaele insists on the fact that L1 is no longer the dominant language for emotional expression. More recent research has shown a more complex picture of this situation enabling to categorize L1 as a law of nature. Even if multi-linguals dominant language will often be L1, we can also say that the more frequently LX is used, the more likely that this language will also respond to both emotional and non-emotional objectives. Dewaele also points out that CS is often used to avoid disturb emotionally the interlocutor. He also claims that the use of CS responds to avoid upsetting interlocutors, speakers will make use of their linguistic repertoire and therefore adjust it to more or less emotional in order to respond to their needs. He also clarifies that there is no clear justification for L1 to be a more emotional language than LX, but that there is a connection between language and emotion and that the use of CS by bilinguals allows them to express more clearly certain feelings and also permits them to evade subjects or terms with negative connotations in one of the speaker’s languages.

Dewaele also points out that due to the complex nature of the CS research carried out by humanities and social studies, the conclusion should be interpreted in a more holistic way.

Han L. Ladegaard[8] conducted a research on life stories of Filipina and Indonesian domestic migrant workers (from now on DMWs) and their trauma narratives while working overseas, in particular in Saudi Arabia and Malaysia. His research did not have the intention of analyzing the use of CS and emotions, but after conversations with DMWs for four years, data results brought some light for researchers use of trauma narratives of repressed groups.

The results showed that DMWs would switch from L1 (Bahasa) to L2 (English) when they had to narrate a high difficult experience. Testimonies containing a high emotionally content will be accompanied by intensive weep, while recollecting and explaining. Being able to switch to a less emotionally charged code will help DMWs making them capable of their experience to be told. Ladegaard’s research showed that language and emotions were intensively related as was shown in a consistent switch of language that would take place at specific moments of difficulty that would be exteriorized by crying or followed by a painful memory.

Encoding process for bi-multi-linguals can be a complex process. Emotional experiences in L1 show normally a higher level of emotional aspects and intricacy than events recollected in L2. Despite this, Tehrani &Vaughan[9] sustain that for balanced bilinguals the quality and intensity of the emotional content would be determined by the language and context in which that experience was encoded. Tehrani &Vaughan provide the example of a French-English bilingual teacher that had been bullied in France. The use of English could not help this person when trying to remember the emotional impact suffered, as the experience happened in French and her traumatic experience was lost in the process of translation. The use of English in therapy, on the contrary, weakened the negative emotional respond to the incident helping the victim to recuperate without having to recollect or provoke negative feelings; being able to work in English restored her French persona.

As for Ladegaard’s project, he says in his article that he is unaware of the language in which the traumatic events took place, but that they are most probably to be in English; as DMWs did not normally speak Arabic and in Hong Kong, where the interviews took place, the shared language is English. Taking Tehrani & Vaughan’s explanation, their experiences would have its most powerful expression of emotional response when narrating them in English. That being said, the use of English could also contribute to experience feelings of detachment, which will help them to share their stories without feeling overwhelmed by negative emotions. And, even if DMWs are considered late bilinguals, their fluency were not exactly the same in both language, which would tell us that the use of English turned into an instrumental language and therefore a more detached language emotionally speaking.

To sum up, there is no evidence that can prove L1 to be a more emotional language and L2 a less emotional language in Ladegaard project, but there is evidence from his investigation showing the CS is used by DMWs in order to align to their emotions. Proof of this is that a change in the linguistic code will occur when anxiety and emotions are at its peak.

The conclusions about the change to L2 accompanied with its peak intensity, as this is the language in which DMWs encoded their experiences, are not conclusive. Yet, the intense weeping is a factor that could imply this, but also a sign that could have helped the process. Weeping is an expression of the healing process that helps victims to be able to talk about trauma. Victims, through intense crying, can attenuate the anxiety caused by remembering. There is also the fact that DMWs expressing in L2 connects the victims to the experience, enabling them to ask for compassion.

There is no agreement about language and emotions findings and some scholars conclude that they are imprecise. However, what we can certainly say is that the use of language is emotionally loaded and that when a speaker uses CS there is a change in an emotional sense.

Another aspect we need to be aware of is the disturbing character of the traumatic testimonies and that the stories will be incoherent with missing components in the narratives and rhythmically inconsistent. Trauma survivors try to accommodate their experiences as a coherent story, but in fact (Langer, 1991) explained that life-threatening experiences cannot be taken on in a temporal continuity as they reside in out of time. In addition to that, people do not want to hear to incoherent traumatic stories and favor linearity and stories allowing recovery.

According to Harvey[10] and her colleagues when talking about sexual trauma, “if they cannot avoid listening, then they prefer coherent stories, ones that make sense by following a culturally-preferred plot from a state of suffering and pain to one of wholeness and recovery” (2000, 294). We do not want to hear traumatic events, but if we have do it, we prefer to listen to a coherent story when we have proof of the almost impossibility of the survivors to do so.

Example of this incapacity is the following testimony of Sari:

(1) Sari, thirty-four years old, four years in Saudi Arabia, four years in UAE. Four

more DMWs were in this sharing session. A female interpreter (Int) and a male

fieldworker (FW) were in all of the sharing sessions. Bahasa is in bold.1

1 FW: has anybody helped you talk about this after you came back?

2 Int: oke waktu//

‘okay when//’

3 Sari: //iya (sobs) pernah saya ngomong, tapi kan saya juga

4 susah kadang kalau ngeluarin itu, susah gitu

‘//yes (sobs) I’ve talked but I also sometimes have difficulties

expressing [myself], difficult like that’

5 Int: okay, it’s not easy for her to share her story, she’s tried but

6 it’s kind of not easy (quietly to FW)

7 Sari: the trauma is so deep (sobs)

8 FW: yeah, how many years now?

9 Sari: three years

10 FW: three years ago okay (1.0) yeah okay, how do you feel now?

11 Sari I think everything is the same, what, what I feel is, until

12 now is the same (sobs)

13 FW: just take your time, right? (6.0) have you thought about this

14 many times or is it the first time? have you talked to Mr (name)

15 or is it the first time?

16 Sari: no this is not the first time, it’s always like this if I raise it,

17 also I cannot, I cannot, right? (1.0) the pain is the same,

18 I go home I’m sick (2.0) I cannot xx (sobs)

19 FW: it’s okay

According to Dewaele, psycholinguistics studies can help us understanding of the relation between CS and the attitudes towards this phenomenon. His studies show that there are certain patterns that can be described and categorized by personality, linguistic practices and learning history and socio-biographical variables.

Personality characteristics are related to the use of CS. It is proven that emotionally stable people will show more admissible attitudes towards CS. Dewaele’s data was recollected by online questionnaire and showed the revealing result that individual with high scores in TA (Tolerance of Ambiguity) and CE (Cognitive Empathy) were the individual with a more positive attitude towards CS.

His research also showed that individuals in the lowest or highest levels of the multilingualism spectrum held the most positive attitudes; leaving the middle groups spectrum with the lowest acceptance towards CS.

Some more logical evidence was also recollected: individuals that have lived and/or work in multilingual and diverse environments and/or abroad had more positive attitudes towards CS. As for some socio-graphical features, women participants presented a more positive attitude than men. Education level showed a similar spectrum as for the level of multilingualism, perceiving a more positive attitude in the lowest and highest dimension. On the contrary, age related findings showed the best attitude in the middle spectrum, being individual in their forties the one with the most acceptance towards CS and teenagers and older groups with the less positive attitude.

All this characteristics of age, gender and proficiency can help us elucidating CS phenomenon but, Dewaele denounces that the attitudes towards CS have rarely been taken into account and that this variables have significant effect on CS practices.

How can we explain the correlation of certain narratives and the use of CS? How are Ladegaard’s DMWs traumatic testimonies related to Montoya’s personal experience, the Outsiders or the case People v. Josephine Chavez?

The complex nature of CS and the fact that they have essentially been explained through recollection of testimonies, might also explain the need of more holistic explanation and that personal narratives would explain best the complexity and the uniqueness of the phenomenon investigating attitudes in depth.

In M. Montoya[11]’s article Máscaras, Trenzas, y Greñas: Un/masking the Self While Un/braiding Latina Stories and Legal Discourse, she talks about the madness of discrimination “la locura de la discriminación” in her speech given in Mexico City during the conference Encuentro Chicano México 1993: “Aqui estoy ocultada por mi máscara linguistica con sus aspectos subtextuales. Desde niña, he entendido el significado de los accentos, vocabulario, pronunciacion, sintaxis. En ingles estos elementos idiomaticos estan relacionados con my psique, con la persona quien soy. Por la primera vez entiendo que español tiene el mismo poder, a pesar de estar donde no soy parte de una minoria cultural or racial. Para mi, hablar español afuera de la casa me hace sentir vulnerable. Sobre todo, hablar español donde la mayoria lo habla mucho mejor que yo, tiene algun aspecto de como me sentia cuando era niña, cuando me sentia vulnerable antes de los gringos. Por eso es dificil quitarme la mascara que me presta el ingles y hablarles en español. Así es la locura de la discriminacíon.”

The complexity of Montoya’s expression cannot be just explained by the separation of L1, emotional language or “personal voice” as she names it; and L2, emotionally detached language or “academic” language. By “madness of discrimination” Montoya is defining her identity as being “ambiguous and contradictory” but also genuine in her own, as she truly is an equivocal and complex self.

Montoya’s numerous and complex masks are sartorial, ideological and cognitive; but also lexographic, rhetorical and linguistic. She gives the example of a Janus faced which shows two sides, her L2 would represent the adult side and the dominant culture and her L1 would represent her child or no artifice side. In her speech for the conference and her feeling towards her own L1, we can see that the delimitation of her L1 and L2 are intertwined. A Janus faced mask would be too limited for Montoya’s expression as she finds herself in between the two faces at times.

Multi-linguals would need to use multiple-face masks according to their needs, as their repertoires would allow a wider range and these could be presented in more intertwined zones.

Linguists have not come to a single definition for code switching and code-mixing. For some scholars both terms are synonym of the same linguistic phenomenon. For other scholars, they are two terms describing different concepts. According to Muysken[12] (2000), code-switching is the use of two language code in a speech event, while in contrast code-mixing refers to all situation where lexical and grammatical features of two different languages codes takes place in a single sentence. Muysken[13] comparative study identifies three different typologies of code mixing: insertion, alternation, and congruent lexicalization. The selection of the type in a bilingual setting would be determined by two factors: the grammatical characteristics of the languages used and a large quantity of sociolinguistic and psycholinguistic factors. Montoya’s speech shows some evidence of code-mixing insertion, the use of random words in English are inserted in her speech in Spanish: “or” and “my”. We can also see the lack of written accents (Spanish “tildes”) or wrongly marked in the acute form “discriminacíon” and the use of hybrid terms like “accent” or “ocultada”.

As for typifying code switching, we also find different approaches. Popelack (2000) talks of tag, inter-sentential and inta-sentential. Gumperz (1982) divides the phenomenon into namely, situational and metaphorical switching. For Clyne (2000) code switching is the alternative usage of two languages in conversation whether within a sentence or between sentences. Intra-sentential code-switching means within a sentence and inter-sentential between sentences, whereas extra-sentential will be used for tags in this study.

Not only sociolinguistic field is concerned about code switching but also translation and interpreting field. As we have mentioned before, no agreement has been reached to define code switching and code mixing, as to be the same concept or two different terms. As for code-switching’s definition is still put in question as different opinions converge by linguists, psycholinguists, philosophers, and anthropologists.

Cortabarria’s[14] research on hybrid English-Spanish context in the USA takes place in health and judicial contexts. Her results included that Hispanics would frequently had to switch or borrow terms from English due to practical reasons and insert them into their dominant language. The result of this process is a hybrid form of Spanish (like in Zimbabwe in the trials). This unique and non standard expression of Spanish becomes a challenge for professional translators and interpreters who need to communicate L1 into L2; finding themselves in a dilemma of using non-standard language in order to achieve communication and respond to ethical principles and not as some early research pointed out to respond to poor knowledge or lexicon.

For non standard language is understood several linguistic phenomena, including code-switching as Mendieta (1999) expresses borrowing in all its forms: pure loanwords, calques, loan blends, semantic extensions, hybrid creation.”

Cortabarria’s findings show incompatibility between being able to communicate in a proficient and professional way using non-standard language. The linguistic tautness leaves no choice to the translators and interpreters that to make a choice between the use of standard speech, having this way to educate the audience, or to use non-standard language in the community.

As for legal context Cortabarria explains, “ Interpreters can educate by avoiding the use of non-standard words, but active reformulation of terms and phrases in less common.” (422) Then, when can we avoid CS? To which extent is this possible? Is there a more appropriate language for a bilingual or multilingual to narrate traumatic events in Court? What are the criteria to follow to choose one or another? The research made by Victor Mugari et al.[15] gives an explanation to the patterns and interpretation of the use of CS in the courtroom discourse of chiShona and English bilinguals in rape cases in Zimbabwe.

In the legal context code switching is often used to fulfill a lexical gap in the language of interaction, using in this case English terminology for technical jargon and referential terms. One of the reasons for this to happen is having received training in English, perceiving this way an easier comprehension of the term in the original language learned. Mugari et al. research conclude that code switching in case of rape in Harare Magistrate Court takes place for a number of multiple reasons but a common point to magistrates, prosecutors, witnesses and accused is that the switch to English occurs in order to express a concept that in chiShona is taboo. The participant will try to avoid the use of sexually explicit language and instead will use euphemisms. This is proven to be a way to remove power from the ordinary people, in particular women.

Prosecutor: You claim that the accused raped her and then

threatened to rape you as well. Can you confirm to this

court the time and place in question.

Witness: uhh… it was on Friday around 7pm.

Prosecutor: OK, unetsamba yekumapurisa here? Can you

produce a police report that shows that you reported

this matter at all?

‘OK, do you have a police report?’

Witness: No, I did not go to the police.

Prosecutor: Well you have failed then to show to this court the

basis of your history, worse still, no police report. I would

suggest kuti murikunyepera dare rino.

‘I would suggest that you are lying to this court.’

Again, as we have seen in examples of DMWs, women would make use of code switching in order to avoid confronting their Shona women identity where explicit sexual subject remains unacceptable.

Prosecutor: … so how did you have sex naye? Usanyare hako

take your time we will wait for you to tell us.

‘… so how did you have sex with her? Do not be shy…’

Accused: uhm.. ndakamukanda pamubhedha and… uhm

ndobva nadamubvisa pant yake then… uhm I put chinhu

changu mune chake.

‘I threw her on the bed and… uhm I removed her

panties then I put my thing into hers’

Prosecutor: What are you referring to as chinhu? What is it

called, the place yawakaisa chinhu chako?

‘What are you referring to as “thing”? What is the

name of the entity where you put your “thing”?

Accused: Ndakaisa… uhm… penis yangu muvagina make.

‘I put… uhm… my penis into her vagina’

Another reason to switch in this case to the opposite switch to chiShona takes place when the prosecutor uses a solemn and serious tone to raise possibilities to win the case.

Prosecutor: Your Worship, nekuti the accused akamushvira

kamwe chete pasina wirirano, eh zvinotipa enough

evidence to sentence him in accordance to the

terms of Section 318 of the Criminal Procedure and

Evidence amendment number 8 of 1997.

‘Your Worship, because the accused rape her once,

he gives us enough evidence…’

Mugari et al. article concludes with a call for individuals in charge of the court proceedings to understand the nature of these patterns of language. This attention-grabbing gesture is in correlation with Montoya’s argument in La Raza Law Journal[16]: “The essay proposes that linguistics norms in law schools can be refashioned through pedagogical innovations to minimize their subordinating effects”(147). And that for that should contribute providing understanding and pedagogical innovation to the matter.

Josephine Chavez’s case when discussed in Montoya’s law classroom forgot to pay attention to cultural, linguistic or socioeconomic specific contexts that could have helped the students to conceive the case in its complexity, and keeping them in the surface of the capital importance of contextualization. Discussions in the legal field being controlled by certain linguistic and socio-cultural norms condemn Outsiders to perpetuate the same experiences, such as Josephine Chavez or more recently in Zimbabwean rape victims narratives.

In Chavez’s case, words and, more important, concepts should have been analyzed in Spanish. Chavez’s case needs to make visible a representation of the bicultural identity of the client and needs to show a kind of storytelling that “has been taboo in the traditional law school classroom” (150).

Thus, Chavez’s case should have implicated information on terms like familia, vergüenza and respeto. These three terms being encoded in Spanish are demonstrating two different meanings and also allow us to recuperate the authenticity of Chavez’s story and a form to transgress the traditional domination.

[1] Dewaele, Jean-Marc (2010). Emotions in multiple languages. Basingstoke: Palgave Macmillan.

[2] Panayiotou, Alexia (2004). Switching codes, switching code: Bilinguals’ emotional responses in English and Greek. Journal of Multilingual and Multicultural Development 25(3):124 – 39.

[3] Burr, Vivien (2015). Social constructionism. 3rd edn. London: Routledge.

[4] Dewaele, Jean-Marc (2016). Multi-competence and emotion. In Vivian Cook & Lei Wei (eds.), The Cambridge handbook of linguistic multi-competence, 461 – 77. Cambridge: Cambrige University Press.

[5] Bond, Michael Harris, & Tat-ming Lai (1986). Embarrassment and code-switching into a second language. The Journal of Social Psychology 126(2):170 – 86.

[6] Martinovic, Ines, & Jeanette Altarriba (2013). Bilingualism and emotion: Implications for mental health. In Tej K. Bhatia & William C. Ritchie (eds.)

[7] Pavlenko, Aneta (2004). ‘Stop doing that, la komu skazala!’: Language choice and emotions in parent-child communication. Journal of Multilingual and Multicultural Development 25(3):179 – 203.

[8] Hans J. Ladegaard (2018) Codeswitching and emotional alignment: Talking about abuse in domestic migrant-worker returnee narratives, The Hong Kong Polytechnic University, Language in Society 47, 693 – 714.

[9] Terahni, Noreen, & Sarah Vaughan (2009). Lost in translation: Using bilingual differences to increase emotional mastery following bullying. Conselling and Psychotherapy Research 9(1):11 – 17.

[10] Harvey, Mary R. ; Elliot G. Mishler; Karestan Konen; & Patricia A. Harney (2000). In the aftermath of sexual abuse: Making and remarking meaning in narratives of trauma and recovery. Narrative Inquiry 10(2):291 – 311.

[11] Margaret Montoya, Mascaras, Trenzas, y Grenas: Un/Masking the Self While Un/Braiding Latina Stories with Legal Discourse, 17 Harvard Journal of Law and Gender 185 (1994).

[12] Muysken, Pieter. Bilingual Speech : A Typology of Code-Mixing. Cambridge, UK ; New York : Cambridge University Press, 2000.

[13]Jacqueline Toribio A. Book Review: Bilingual speech: A typology of codemixing Pieter Muysken (2000) Cambridge, U.K.: Cambridge University Press, ISBN 0521-77168 -4, pp.xvi+ 306. 

[14] Beatriz Cortabarria (2015) The role of translators and interpreters in hybrid English–Spanish contexts in the USA, Language and Intercultural Communication, 15:3, 407-423,DOI: 10.1080/14708477.2015.1015345

[15] Victor Mugari, Laston Mukaro, Lovemore Mutonga, Nhamo W Samasuwo & Maxwell Kadenge (2015) Code-switching among chiShona-English bilinguals in courtroom discourse: Rape cases in Zimbabwe, South African Journal of African Languages, 35:2, 207-214,DOI: 10.1080/02572117.2015.1113008.

[16] Montoya, Margaret, Law and Language(s): Image, Integration and Innovation (1994). Berkeley La Raza Law Journal, Vol. 7, No. 2, 1994.

Costs of Language Switching in Bilingual Sentence Processing

Switch cost modulations in bilingual sentence processing: evidence from shadowing
This study aims to determine the cognitive costs of language switching in bilingual sentence processing. It also wants to investigate whether the cognitive cost is different depending on whether the switch is from L1 to L2 or vice versa. Many studies have shown that there is a significant cognitive cost to switching languages, including slower reaction times and more errors (Monsell 2003). Through tests involving shadowing tasks, cognates are used to see if there is a decrease in cognitive cost due to the activation of networks from both languages. Previous studies on bilingual subjects have already shown that code switching is more frequent in cases where cognates are present; therefore it is hypothesized that due to the overlap of lexical representations, cognates not only increase frequency but also facilitate the switch. The amount of cognitive cost could also depend on which direction the switch is occurring. Data from previous studies has shown that the switch from L1 to L2 is more demanding than that of L2 to L1 (Proverbio, 2004). This is due to the idea that proficiency in a language plays a role in the ease of processing L2; the stronger the mental representation of a word is, the more cognate facilitation occurs.
The study included fifty bilingual participants aged 18-41, all of which were Dutch natives who also spoke English as a second language. All were highly proficient in English having learned it early on in school and often taking university courses in English.
The stimulus material included forty different sentences. The sentences vary in language (Dutch or English), sentence structure (SVO, XVSO, or VSO) as well as the presence of a cognate (Non-cognate vs. cognate). If a cognate verb occurred, a language switch often followed it. Sample sentences are presented in Table 1 and 2.
Participants were tested alone in a sound proof room with a computer and headphones. They were instructed to listen to sentences in which a Dutch-English language switch would occur and start shadowing it as soon as they started hearing syllables. Correct repetition of the phrases was stressed. Researchers were in the other room monitoring the performance. Between each set, instructions in either Dutch or English were presented to cue the following language.  After the experiment, they then performed the Simon task and the Operation Span task in order to asses their cognitive abilities as well as another test to test their proficiency in the English language.

The results demonstrated that there is a definite cost of switching language, regardless of the direction; however, the switch from L1 to L2 is more costly. This is most likely due to language dominance and the switch cost could just be attributed to the fact that the baseline for L1 is already quicker than L2. There was also a language effect demonstrated; shadowing in L2 was slower and more prone to errors than L1. However, this effect is not due to proficiency in L2 as the data showed that language proficiency was only significant in the onset of the sentence. Contrary to the hypothesis, there was not an effect of cognate presence, no matter the language or sentence structure. Previous studies suggest that this is because the facilitated processing that typically occurs is with noun cognates, not verbs. The lack of facilitation is in line with many previous studies of verb cognate effects in sentences. In some cases, there was more switch cost when a verb cognate was present. It is possible that no coactivation of both languages occurred because the participant suppressed one, making it harder to retrieve in a switch. Additionally, the auditory effects of repeating the sentences out loud may have contributed to the lack of effects. The phonological effect of shadowing may also contribute to increased costs in language switching; not only does the subject have to switch their lexical representations, they also have to change the manner of articulation (Phillip & Koch, 2011) In general, this study has shown that there are no modulation effects of verb cognates in sentence context. It does however show the effects of bilingualism in switch cost. If language is asymmetrical and one language is stronger than the other, switch costs are higher for the less proficient language while the more dominant language is processed with more ease. These results are not new findings, however they provide additional support for previous studies on verb cognates as well as language dominance.
Bultena, S., Dijkstra, T., & van Hell, J.G. (2015) Switch cost modulations in bilingual sentence processing: evidence from shadowing, Language, Cognition and Neuroscience, 30(5), 586-605, doi: 10.1080/23273798.2014.964268