|
The example experiment contained the following lines:
numberOfDifferentStimuli = 4
"heed" ""
"hid" ""
"hood" ""
"hud" ""
...
...
numberOfDifferentResponses = 5
0.2 0.3 0.7 0.8 "h I d" 40 "" "i"
0.3 0.4 0.5 0.6 "h E d" 40 "" "e"
0.4 0.5 0.3 0.4 "h A d" 40 "" "a"
0.5 0.6 0.5 0.6 "h O d" 40 "" "o"
0.6 0.7 0.7 0.8 "h U d" 40 "" "u"
For every stimulus, the buttons contained the same visible texts, such as "h I d" and "h A d".
But suppose you have an experiment about the perception of voicing in plosives. The first stimulus starts with an ambiguous /b/ or /p/, and you want the participant to choose between "bath" and "path". The next stimulus starts with an ambiguous /d/ or /t/, and you want the participant to choose between "dart" and "tart". You would go about like this:
numberOfDifferentStimuli = 6
"bpath1" "|bath|path"
"bpath2" "|bath|path"
"bpath3" "|bath|path"
"dtart1" "|dart|tart"
"dtart2" "|dart|tart"
"dtart3" "|dart|tart"
...
...
numberOfDifferentResponses = 2
0.2 0.4 0.7 0.8 "" 40 "" "left"
0.6 0.8 0.7 0.8 "" 40 "" "right"
In this case, the two response buttons show either "path" and "path", or "dart" and "tart".
In the ResultsMFC (see ExperimentMFC 2.9. How an experiment proceeds), the stimuli will be recorded as “bpath1|bath|path” and the like, not just as “bpath1”. Praat does this in order to be able to cope with balanced designs such as
"bpath1" "|bath|path"
"bpath1" "|path|bath"
In other words, the button ordering is considered part of the stimulus.
This trick can be combined with a stimulus-dependent run text:
numberOfDifferentStimuli = 32
"bpath1" "Throw a...|bath|path"
"bpath1" "Walk a...|bath|path"
"bpath2" "Walk a...|bath|path"
"dtart1" "Throw a...|dart|tart"
"dtart1" "Carry a...|dart|tart"
...
runText = ""
...
numberOfDifferentResponses = 2
0.2 0.4 0.7 0.8 "" 40 "" "left"
0.6 0.8 0.7 0.8 "" 40 "" "right"
© ppgb 20070926