|
nif:isString
|
-
The design used is illustrated in Fig 1 (see also description above). It was intended to approximate a natural communicative setting with question plus answer sequences.
Figure data removed from full text. Figure identifier and caption: 10.1371/journal.pone.0186685.g001 Trial example.Example of one trial in Experiment 1 in which the verb form har (‘has’) is elicited either as a lexical verb (full verb) or a grammatical verb (auxiliary), and the two verbs are compared in fully matched sentences (i.e. Det har lexical/grammatical hun også 'So does/haslexical/grammatical she'). Thirty sentences were created for each verb in each condition (lexical/grammatical). The list of sentences can be seen in S1 file. Ten male and ten female names were used in context sentences and questions in varying combinations, but always with the same gender for both context sentence and question. 120 filler sentences were constructed using two different neutral verbs which did not present a variant and added to the experiment. To ensure responses could not be planned in advance, complexity was increased by varying the polarity of the expected response. This was achieved by color-coding the question in blue or orange to elicit either a positive (6) or a negative (7) response. Participants saw both conditions of each item in either a positive or a negative frame. Four lists were created to balance which items were shown in a positive/negative frame, and which color was coded as positive/negative, across participants. (6)Det hargram/lexhun også. 'So has/does she. '(7)Det hargram/lexhun ikke. She has not/does not. 'Since the context sentences in the two conditions consisted of exactly the same words, the second condition presented to the participants may have been primed by the first. To control this, all lists were split in two, so half the items appeared on each sublist in one condition and on the other sublist in the other condition. The order of these sublists was balanced across participants as well. For each sublist the order of items was randomised for each participant. Each participant saw a total of 120 stimuli (60 grammatical and 60 lexical) and 120 filler trials. Filler trials consisted of sentences that contained different constructions with one of the two verbs: blive (‘become’) and være (‘be’). These verbs are comparable in term of frequency of use with har and får in Danish. Moreover, the auxiliary uses are clearly grammatical, while the copular uses are less clearly grammatical. The filler trials thus approximate the grammar-lexicon distinction found in the target trials. An overview of the overall design can be seen in Fig 2.
Figure data removed from full text. Figure identifier and caption: 10.1371/journal.pone.0186685.g002 Design of the experiment.Each sequence starts with a context sentence which allows for the proper status (grammatical or lexical) to be elicited in the coming response. The color (blue or orange) indicates whether the participant should answer positively or negatively to the question. The dotted arrows represent the three types of measures used for the analyses with first self-paced measure upon reading the context sentence, then reaction time measures and eventually duration of the entire message.
Twenty four Danish speaking undergraduate students (12 females; mean age: 27 years) voluntarily took part in the experiment. One participant was removed from the analyses due to the repeated use of a verbum vicarium in one of the two conditions (e.g. use of the verb “gøre” instead of the verb “får”). The study was approved by the Research Ethics committee at the Faculty of Humanities, University of Copenhagen.
The experiment was run in a quiet room on a laptop using Psychopy [34]. Prior to the experimental task, participants were given instructions and completed a practice session of four trials with the experimenter in the room to ensure the task had been understood. At the beginning of each trial, a context sentence was shown in the middle of the screen in black font on a grey background. The sentence remained on the screen until the participant clicked the spacebar. This was followed immediately by a question in either blue or orange font which was displayed for three seconds. Subjects were requested to provide either an affirmative or a negative response to the question based on the color of the font as quickly as possible. Participants wore a headset with a microphone to record the responses. Recording of the participants response began at question onset and lasted five seconds after which a fixation cross appeared on the screen for one second before the next trial. The experiment took approximately 40 minutes to complete including a short break after the first 120 trials.
|
![[This page as RDF]](http://data.gesis.org/somesci/static/rdf-icon.gif){kind=icon}