Home » Welcome » Research on VMI

Research on VMI

Research into Video-Mediated Interpreting has analysed the quality of the interpreter’s performance and a range of psychological and physiological factors associated with this method of interpreting; the dynamics of participant interaction; and the strategies that interpreters develop in relation to video-mediated interpreting. In addition, Video-Mediated Interpreting has been investigated in terms of efficiency gains compared to onsite interpreting. Most studies have focused either on Video Remote Interpreting (VRI) or on Videoconference Interpreting (VCI) (see What is Video-Mediated Interpreting for definitions). This page page is therefore divided into research on VRI and research on VCI.

Full bibliographical details of the works cited in this section are available in the bibliography.

Research on Video Remote Interpreting (VRI)

VRI in supranational institutions: The earliest documented experiment with VRI was organised by UNESCO in 1976, to test the use of the Symphonie satellite. It linked the UNESCO headquarters in Paris with a conference centre in Nairobi, with the interpreters being situated in Paris. Similar experiments were organised by the UN in the 1970s and 1980s. When ISDN-based videoconferencing became available in the 1990s, feasibility studies were conducted in many supranational institutions, using the simultaneous mode of interpreting (see Moser-Mercer 2003; Mouzourakis 2006; Roziner & Shlesinger 2010), although ISDN connections were incompatible with the ISO 2063 standard in terms of sound quality and were therefore considered to be unacceptable for simultaneous interpreting by the AIIC (AIIC 2000/2012). The feasibility studies used a variety of technical conditions and revealed physiological and psychological challenges which recurred in different technical conditions, making it difficult to attribute them to a particular technical setup (Mouzourakis 2006). Two studies in particular addressed physiological and psychological variables, as well as the quality of VRI: the studies conducted by the International Telecommunications Union (ITU) in collaboration with the École de Traduction et d’Interprétation (ETI) (Moser-Mercer 2003), and by the European Parliament (EP) in 2004 (reported in Roziner & Shlesinger 2010). As well as investigating the performance of the participating interpreters, the studies also elicited the interpreters’ emotional responses to VRI, and measured stress indicators and aspects of the working environment. The outcomes of the two studies differ in several ways. For example, the ITU/ETI study found that the interpreters’ performance in VRI declined faster than their on-site performance, whilst the EP study found no significant differences in VRI and on-site performance. However, both studies highlight a sense of discomfort with VRI on the part of the interpreters which is hard to account for by objective measures. The most striking result of research on VRI in this setting thus seems to be the discrepancy between objective findings and subjective perception.

VRI in legal settings: In legal settings VRI has been used to cope with a shortage of qualified interpreters, a lack of time and the short duration of many assignments, which make the interpreter’s travel and physical presence particularly uneconomical. The practice of VRI in this field goes back to the 1980s, when VRI by telephone was introduced in the US. Over time, this has gradually been replaced by video VRI. A well-known example is the Ninth Judicial Circuit Court of Florida, which introduced a central video interpreting hub in 2007. The interpreters’ workstations in the hub are configured to allow a combination of consecutive and simultaneous interpreting. The Metropolitan Police Service in London introduced VRI in 2011, with interpreters working in consecutive mode from centralised hubs linked to London police stations. The European Directive on the right to interpretation and translation in criminal proceedings (2010/64/EU) explicitly refers to the possibility of using VRI to gain access to qualified legal interpreters. The first studies to address VRI in legal proceedings were conducted in the European AVIDICUS projects. Based on the outcomes of a survey designed to identify problems and needs, AVIDICUS 1 (2008‒2011) compared the quality of on-site interpreting and VRI (and VCI). The findings of these experiments reveal a significantly higher number of problems and, like Moser-Mercer’s (2003) data, a faster decline of interpreting performance in VRI (Braun 2013, 2014; Braun & Taylor 2012).  AVIDICUS 2 (2011‒2013) replicated the experiments after providing the interpreters with short-term training, and using better equipment. The findings yield a complex picture, making it impossible to say without reservation that training, familiarisation and the use of better equipment led to a clear improvement in performance (Braun & Taylor forthcoming). It is, however, possible to identify strategic attempts by the interpreter to overcome the lack of co-presence and the ensuing VC-induced problems (Braun 2016). AVIDICUS 3 (2014‒2016) currently assesses videoconferencing facilities in legal institutions in Europe, in terms of their fitness for interpreter-mediated communication.

VRI in healthcare: In healthcare settings VRI is used with similar motivations to those in legal settings, that is, optimising access to interpreters and achieving efficiency gains. VRI in healthcare is often delivered by telephone, but this has been changing with the advent of mobile videoconferencing devices (Locatis et al. 2011). A number of smaller, mostly survey-based studies of VRI in medical encounters using telephone and video link have been carried out. However, their findings are difficult to compare due to highly variable conditions. In a review of nine studies conducted between 1996 and 2003, Azarmina and Wallace (2005) find evidence that VRI is at least as acceptable as on-site interpreting to patients, doctors and (to a lesser extent) interpreters. Although none of these studies included an actual assessment of the interpreters’ performance, the authors also conclude that VRI “appears to be associated with levels of accuracy at least as good as those found in physically present interpretation” (2005: 144). They do, however, note that interpreters generally preferred on-site interpreting to VRI, and video to telephone. This is corroborated by more recent studies comparing the three modalities (Locatis et al. 2010; Price et al. 2012).

Research in Videoconference Interpreting (VCI)

VCI in supranational institutions: The earliest documented experiment with videoconferencing and interpreting took place in UNESCO in 1976. It included tests of both VRI and VCI (see above under VRI in supranational institutions). At the UNISPACE conference in Vienna in 1982, communication from the Soviet cosmonauts on board the MIR space station was transmitted to the Vienna delegates by video link and interpreted for them by interpreters in the Vienna conference room. Reports about these early tests suggest that VRI was perceived to be challenging or unacceptable, whilst VCI seemed less problematic. This overall trend was not confirmed in VCI tests using ISDN-based videoconferences, e.g. in the European Commission in 1995 (see Mouzourakis 2006), where sound quality was found to be insufficient for simultaneous interpreting. However, the view that VCI is acceptable under defined circumstances, whilst VRI is not, is also reflected in the AIIC guidance on the use of technologies in interpreting (AIIC 2000/2012).

VCI in legal settings: Legal institutions have turned to videoconferencing to make legal proceedings more efficient, minimise security concerns arising from prisoner transport, and support cross-border judicial co-operation. This has led to a growing demand for VCI in legal proceedings, normally conducted in consecutive mode. In many English-speaking countries, ISDN-based videoconference facilities were installed in the 1990s to link courts to other courts (e.g. to hear remote witnesses) and prisons (e.g. for bail hearings). A worldwide spread of videoconference technology in legal proceedings began in the 2000s, following the availability of broadband technology. In some countries, all courtrooms use the same videoconference equipment and layout to facilitate the work of all involved, including the interpreter. Such approaches are likely to have contributed to relatively positive attitudes towards VCI among interpreters in these countries, whilst scepticism prevails in countries where videoconference equipment often still dates from the ISDN era (Braun & Taylor 2012b). Fowler (2007) notes problems with the interpreter’s positioning and access to the microphone, and with the quality of the video image, in English magistrates’ courts. She argues that these problems, together with the absence of specific protocols on VCI in court, lead to frequent disruptions, requests for repetition and misunderstanding. One question arising, regardless of such issues, concerns the location of the interpreter in VCI. This was also one of the questions addressed by a comprehensive survey of VCI in Canadian immigration proceedings (Ellis 2004). In the setting examined, the immigration judge, the refugee protection officer and the interpreter sat in the immigration office, whilst the refugee and his/her lawyer were in another city. The fact that the interpreter was not co-located with the refugee was thought to have weakened the personal rapport between the two. It also caused interactional difficulties and precluded whispered interpreting. Judges felt that consecutive interpreting was disruptive. The hearings by video link also tended to be longer and were considered to be more fatiguing than comparable face-to-face hearings. These findings were corroborated by the European AVIDICUS projects, which have focused on the viability of VCI and VRI in legal proceedings. In addition, experimental studies conducted in AVIDICUS 1 (2008‒11) showed that VCI (and VRI) affected the quality of interpreting and caused more interaction problems than on-site interpreting. Overlapping speech proved difficult to resolve and led to information loss (Braun & Taylor 2012c). Furthermore, qualitative analyses of the communicative dynamics in interpreter-mediated videoconference-based investigative interviews, court hearings and cross-border settlement cases, carried out in AVIDICUS 2 (2011‒13), suggest that VCI entails not only a reduction in the quality of the relations between the participants but also a greater fragmentation of the discourse (Braun & Taylor forthcoming). AVIDICUS 3 (2014‒16) currently assesses the implementation of videoconferencing facilities in legal institutions across Europe in terms of their fitness for VCI.

Other settings: The use of VCI in other settings is not very well documented, but some reports and interpreting service provider websites suggest that VCI is used across different segments of the interpreting market and that solutions in the commercial sector tend to be custom-made. They may also combine the use of the telephone and of videoconferencing to integrate interpreters into proceedings. One configuration that is likely to gain momentum is three-way videoconferencing, whereby the primary participants and the interpreter are each in a different location. In the late 1990s, the ViKiS project in Germany assessed this configuration (Braun 2004). Using a prototype system, problems as well as adaptation strategies developed by the participating interpreters in this (then) novel working condition were identified. As in other studies, participants found the communication fatiguing and had difficulty establishing a rapport with the other participants. The sound quality in the ISDN-based prototype was insufficient. The one aspect to which interpreters were able to adapt was the interaction. The strategies evolved from reactive to more proactive strategies. However, the interpreters felt that they had to moderate the interaction, which posed ethical problems and increased the coordination effort (Braun 2004, 2007). With regard to cognitive processing, Moser-Mercer (2005) outlines problems with multi-sensory integration in videoconferences, which she believes make it more difficult for interpreters to process information and build mental representations of the situation.

A more comprehensive overview of Video-Mediated Interpreting is available in Braun, S. (2015). Remote Interpreting. In Mikkelson, H, & Jourdenais, R (Eds.), Routledge Handbook of Interpreting (pp. 352-367). New York: Routledge.