This repository holds the eyetracking experiments we used in the Learning To Talk lab. We used Eprime with a Tobii eyetracker, so the actual experiments require proprietary software to open.
TJ Mahr created this repository well after all the data had been collected so that he could keep straight which recordings were used with which experiments. We had never used a proper version control system on our experiments.
For version control efficiency, movie files and the sound effects for the puzzle-game have not been added to the repository even though they are a minor part of the experiments.
Experiment type:
- MP: Our two-image mispronunciation (MP) experiment.
- RWL: Our four-image "visual world" experiment. (RWL was short for Real Word Listening, because they did not hear any nonwords or mispronunciations).
Dialects:
- AAE: Experiment used a speaker of African American English.
- SAE: Experiment used a speaker of Mainstream American English.
The S was for Standard but we've moved away from that terminology, but our filenames have that vestigial abbreviation.
These are the first eyetracking experiments we did. They do not feature any gaze-contingent stimulus presentation.
The MAE Mispronunciation experiment here is the basis for Law and Edwards (2015).
This next batch of experiments incorporated an orienting stimulus (a small animated image at the center of the screen) to prevent the listener from already being fixated a named image when the word is produced. By controlling gazes in this way, we figured we could get more reaction time measurements.
These are the experiments from the first year of our longitudinal study. There was a timing glitch with the experiments for first few participants. This experiment used a gaze-contingent stimulus presentation, but it only required the child to look the screen instead of a specific region of the screen. We tried different tweaks to the experiment, including a version with a gray background to see if that improved eyetracking. We also added an option to use a different calibration procedure (with a "ducky" movie).
We used newly recording stimuli so that two dialect versions of the experiment had similar durations.
We got rid of the final attention getter phrase (e.g., "check it out!") from the end of each trial to speed up the experiment. We also replace the movies that would play every 6 or 7 trials with a gaze contingency game in a child's gaze would remove puzzle pieces to reveal an image.