Academia.eduAcademia.edu

Figure 3 – uploaded by Ester Torres-Simon

See full PDFdownloadDownload figure
Experiment design and distribution Figure 1: Answers to the reading comprehension test.

Figure 1 Experiment design and distribution Figure 1: Answers to the reading comprehension test.

Related Figures (33)

Table 1: Types of narration and character intervention (CI) in English novels
Table 1: Types of narration and character intervention (CI) in English novels
Figure 1: Pipeline of the methodology for creating TMX files from novels The most common error in the extraction phase of the English texts was due to problems in recognizing the marks when the extraction involved very long segments. These segments contained several character interventions as
Figure 1: Pipeline of the methodology for creating TMX files from novels The most common error in the extraction phase of the English texts was due to problems in recognizing the marks when the extraction involved very long segments. These segments contained several character interventions as
Table 3: Percentage difference between the total numbers of English and Spanish dialogues
Table 3: Percentage difference between the total numbers of English and Spanish dialogues
The general results show that of 34,081 analyzed segments, repetitions account for 5.9% (2,015 segments were repeated). The repetition feature of the translation memory means that there is one translation per repeated segment, if the translation memory system finds the exact segment in the translation memory file; this 5.9% can be automatically translated. This also accounts for the Exact Matches, which were 3.2% of the corpus (1,129 segments). i S a aoe Tn literary narrative, dialogues are one of the main guides for plot development, but the neighboring description is in charge of directing the references thereby limiting the dialogues to their function of continuity. This is reflected in the matching of subtitles with their extracted dialogues, especially when the character interventions in the subtitles and in the novel are the same. In these cases, there may be a 100% match, however, that match will not occur frequently along the text: we see that 9.1% was found to be repeated, rendering it difficult to rely on automatic translation as a resource to speed up the translation process. alignment was precise for about 90% of the cases, with an erroneous alignment rate of 10%.
The general results show that of 34,081 analyzed segments, repetitions account for 5.9% (2,015 segments were repeated). The repetition feature of the translation memory means that there is one translation per repeated segment, if the translation memory system finds the exact segment in the translation memory file; this 5.9% can be automatically translated. This also accounts for the Exact Matches, which were 3.2% of the corpus (1,129 segments). i S a aoe Tn literary narrative, dialogues are one of the main guides for plot development, but the neighboring description is in charge of directing the references thereby limiting the dialogues to their function of continuity. This is reflected in the matching of subtitles with their extracted dialogues, especially when the character interventions in the subtitles and in the novel are the same. In these cases, there may be a 100% match, however, that match will not occur frequently along the text: we see that 9.1% was found to be repeated, rendering it difficult to rely on automatic translation as a resource to speed up the translation process. alignment was precise for about 90% of the cases, with an erroneous alignment rate of 10%.
Table 2: Specific features from groups HoH and D.
Table 2: Specific features from groups HoH and D.
Groups Gender distribution Age interval
Groups Gender distribution Age interval
Table 3: Design and distribution. Tools, methods and procedure
Table 3: Design and distribution. Tools, methods and procedure
Regardless of the low number of participants in each group, all six possible sequences were tested in order to identify any possible problem in the order of presentation. I seated the participants in front of the monitor one by one, and calibrated the eye-tracker to their eyes. Once the calibration process was successfully completed, I started the video playback. I assigned a sequence to each participant and showed them the three excerpts, alternating the three conditions, as shown in Table 2. After each clip, they orally answered the on-screen questions and recorded their answers with a microphone. Immediately after this session, they took part in the interview. Prior to the experiment, I wrote a protocol for the eye-tracking and interview session. The experiment took place in the Aula d’Analisi de la Parla at the Universitat Rovira i Virgili, during two weeks in December 2012. I arranged individual appointments with the participants. They selected a suitable time and date for the session and came to the laboratory individually. Upon arrival, I verbally informed them regarding their participation in the study and thei rights, and they signed the appropriate consent form.
Regardless of the low number of participants in each group, all six possible sequences were tested in order to identify any possible problem in the order of presentation. I seated the participants in front of the monitor one by one, and calibrated the eye-tracker to their eyes. Once the calibration process was successfully completed, I started the video playback. I assigned a sequence to each participant and showed them the three excerpts, alternating the three conditions, as shown in Table 2. After each clip, they orally answered the on-screen questions and recorded their answers with a microphone. Immediately after this session, they took part in the interview. Prior to the experiment, I wrote a protocol for the eye-tracking and interview session. The experiment took place in the Aula d’Analisi de la Parla at the Universitat Rovira i Virgili, during two weeks in December 2012. I arranged individual appointments with the participants. They selected a suitable time and date for the session and came to the laboratory individually. Upon arrival, I verbally informed them regarding their participation in the study and thei rights, and they signed the appropriate consent form.
Table 3: Means for fixation length (in milliseconds) and number of fixations by AOIs (Subtitle and Image), Group and subtitle condition (PRO, NP1 and NP2) Number of fixations
Table 3: Means for fixation length (in milliseconds) and number of fixations by AOIs (Subtitle and Image), Group and subtitle condition (PRO, NP1 and NP2) Number of fixations
Percentage of gaze time Figure 1: Percentage of fixations on areas of (image and subtitle) by group (HLE and LLE) and type of subtitle (PRO, NP1 and NP2)
Percentage of gaze time Figure 1: Percentage of fixations on areas of (image and subtitle) by group (HLE and LLE) and type of subtitle (PRO, NP1 and NP2)
Comprehension questionnaire
Comprehension questionnaire
Table 4: Percentage of correct questions by type of question and condition
Table 4: Percentage of correct questions by type of question and condition
Table 5: Self-reported comprehension and translation difficulty by type of subtitle Discussion The eye-tracking data show a significant difference in the behavior of participants with different levels of English, with the participants in the LLE group spending around 56% of the time in the image area and participants in the HLE group spending around 77% in the image area. This difference coincides with comments made by the participants during the interviews, when some mentioned they normally use subtitles as an aid rather than as their main source of information. This also corroborates the initial assumption that participants engage differently with the content depending on their language skills. The participants displayed similar gaze behavior under all three subtitle conditions, which is consistent with the participants’ failure to notice differences between the subtitles during the interviews. co cy ores oe ce 2) a ae a ee en, Ce Sacer & ee ns |i ee fo ee ence See |
Table 5: Self-reported comprehension and translation difficulty by type of subtitle Discussion The eye-tracking data show a significant difference in the behavior of participants with different levels of English, with the participants in the LLE group spending around 56% of the time in the image area and participants in the HLE group spending around 77% in the image area. This difference coincides with comments made by the participants during the interviews, when some mentioned they normally use subtitles as an aid rather than as their main source of information. This also corroborates the initial assumption that participants engage differently with the content depending on their language skills. The participants displayed similar gaze behavior under all three subtitle conditions, which is consistent with the participants’ failure to notice differences between the subtitles during the interviews. co cy ores oe ce 2) a ae a ee en, Ce Sacer & ee ns |i ee fo ee ence See |
Figure 1: Frequency of interactions for the three subjects as absolute numbers. Considering the link between personality traits and personification, an analysis of the above-explained pilot tests suggests that:
Figure 1: Frequency of interactions for the three subjects as absolute numbers. Considering the link between personality traits and personification, an analysis of the above-explained pilot tests suggests that:
Subject 1
Subject 1
TAP analysis: Frequency of interaction types
TAP analysis: Frequency of interaction types
Subject 3
Subject 3
Table 3: Percentage of code-switching messages in code-switching messages of monolingua background women
Table 3: Percentage of code-switching messages in code-switching messages of monolingua background women
Table 2: Percentage of code-switching messages in code-switching messages of bilingua background men Table 3 and Table 4 show that the percentage of code-switching in the SMS messages of monolingual background women is 39% while it is 65% in the messages sent by women with bilingual backgrounds.
Table 2: Percentage of code-switching messages in code-switching messages of bilingua background men Table 3 and Table 4 show that the percentage of code-switching in the SMS messages of monolingual background women is 39% while it is 65% in the messages sent by women with bilingual backgrounds.
Table 1: Percentage of code-switching messages in code-switching messages of monolingual background men Table 1 and Table 2 show that the percentage of code-switching in the SMS messages of monolingual background men is 32% while it is 28% in the messages sent by men with a bilingual background.
Table 1: Percentage of code-switching messages in code-switching messages of monolingual background men Table 1 and Table 2 show that the percentage of code-switching in the SMS messages of monolingual background men is 32% while it is 28% in the messages sent by men with a bilingual background.
Table 4: Percentage of code-switching messages in code-switching messages of women with bilingual backgrounds Table 5 and Table 6 show that the percentage of code-switching in the SMS messages of men and women with monolingual backgrounds is 37% while it is 48% in the messages sent by men and women with bilingual backgrounds.
Table 4: Percentage of code-switching messages in code-switching messages of women with bilingual backgrounds Table 5 and Table 6 show that the percentage of code-switching in the SMS messages of men and women with monolingual backgrounds is 37% while it is 48% in the messages sent by men and women with bilingual backgrounds.
Table 5: Percentage of code-switching messages in code-switching messages of women and men with monolingual background Table 6: Percentage of code-switching messages in code-switching messages of women and men with bilingual backgrounds
Table 5: Percentage of code-switching messages in code-switching messages of women and men with monolingual background Table 6: Percentage of code-switching messages in code-switching messages of women and men with bilingual backgrounds
From all of the above tables, it is obvious that subjects from a bilingual background code-switch more than those from a monolingual background; 39% and 65% for groups of women and 37% and 48% for groups of men and women. However, the findings suggest that men from bilingual backgrounds code-switch less than men from monolingual backgrounds: 32% and 28% respectively. These results confirm that code-switching in SMS messages is not the same among men and women, and that it differs according to the language background.
From all of the above tables, it is obvious that subjects from a bilingual background code-switch more than those from a monolingual background; 39% and 65% for groups of women and 37% and 48% for groups of men and women. However, the findings suggest that men from bilingual backgrounds code-switch less than men from monolingual backgrounds: 32% and 28% respectively. These results confirm that code-switching in SMS messages is not the same among men and women, and that it differs according to the language background.
Table 9: Percentage of code-switching messages in code-switching messages of bilingual
Table 9: Percentage of code-switching messages in code-switching messages of bilingual
Table 7: Percentage of code-switching messages in code-switching messages of bilingual men Table 8: Percentage of code-switching messages in code-switching messages of multilingual
Table 7: Percentage of code-switching messages in code-switching messages of bilingual men Table 8: Percentage of code-switching messages in code-switching messages of multilingual
Table 9 and Table 10 show that the percentage of code-switching in SMS messages of bilingual women is 47%, while it is 62% in the case of multilingual women.
Table 9 and Table 10 show that the percentage of code-switching in SMS messages of bilingual women is 47%, while it is 62% in the case of multilingual women.
Table 11: Percentage of code-switching messages in code-switching messages of bilinguals Table 12: Percentage of of code-switching messages in code-switching messages of multilinquals
Table 11: Percentage of code-switching messages in code-switching messages of bilinguals Table 12: Percentage of of code-switching messages in code-switching messages of multilinquals
Table 10: Percentage of code-switching messages in code-switching messages of multilingual TATO Man Table 11 and Table 12 show that the percentage of code-switching in the SMS messages of bilingual men and women is 47%, while it is 38% in the messages sent by multilingual men and women.
Table 10: Percentage of code-switching messages in code-switching messages of multilingual TATO Man Table 11 and Table 12 show that the percentage of code-switching in the SMS messages of bilingual men and women is 47%, while it is 38% in the messages sent by multilingual men and women.
Again, the findings show that there are clear gender differences in the SMS code-switching: multilingual women code-switch much more than multilingual men (62% and 7% respectively). On the other hand, the group of multilingual men and women code-switch less than the bilingual group (38%
Again, the findings show that there are clear gender differences in the SMS code-switching: multilingual women code-switch much more than multilingual men (62% and 7% respectively). On the other hand, the group of multilingual men and women code-switch less than the bilingual group (38%
Related topics:
Translation StudiesSociolinguisticsAudiovisual Translation (subtitling)
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved