Controling the timing of stimulus and response presentation is often critical in many experimental paradigms. In this article, we showcase FindingFive's capabilities of fine-tuning the presentation of stimuli and responses via manipulations of the "Barrier" property.
Barrier is a very useful property of stimulus in the FindingFive Study Grammar. As the name suggests, it allows a stimulus to block the presentation of other content (both stimuli and responses) on a trial until the stimulus' own presentation is complete. We will look at several common designs that can be implemented via barrier, and optionally by turning it off.
A typical experimental design is to have participants first watch a video clip, and then show a rating scale after the video clip finishes playing to solicit a rating response.
This design can be accomplished by having the following trial template:
Note that throughout this article, the names of stimuli and responses here are just examples. Please use the actual names of stimuli and responses in your study when implementing these designs.
"tt_rate_audiostims": {
"type": "basic",
"stimuli": ["stimulus_barrier_on"],
"responses": ["rating_scale"]
}
and a stimulus named "stimulus_barrier_on" defined as follows:
{
"type": "video",
"content": "video_clip1.mp4"
}
When the trial is presented, the rating scale will only show up after the entire video clip has played. Note that we didn't specify the barrier property anywhere in the code! This is because the barrier property of the video stimulus is set to true by default on FindingFive. Intuitively, stimulus types that have a duration, such as audio stimuli and video stimuli, are normally expected to act as a barrier. As a result, you don't need to specify the barrier property unless you want to make them a non-barrier, which we will look at in the next design.
Imagine a go/no-go task where participants need to press a key, as quickly as possible, if the audio clip being played on a trial contains a popping sound. Otherwise, they should just let the trial complete without taking any actions.
The key element in this design is that participants need to act as quickly as possible. In other words, we can't leave the default setting of barrier on for the audio stimuli, because that would prevent participants from responding (as in Design #1) before the audio stimuli finish playing. Instead, we want participants to be able to respond the moment they hear the popping sound and interrupt the audio clip being play on the trial.
To implement this design, let's first create a trial template like this one:
"tt_sound_detection": {
"type": "basic",
"stimuli": ["popping_sound", "no_popping_sound"],
"responses": ["detection_keypress"]
}
and for the audio stimuli "popping_sound" and "no_popping_sound", we turn off the barrier property by setting it to false:
{
"type": "audio",
"content": "popping_sound.mp3", // this is just an example for the popping sound
"barrier": false
}
and the keypress response can be defined as:
{
"type": "keypress",
"whitelist": ["g"],
"multiple": false
}
The resulting experiment will allow participants to press the "g" key at any point during the entire audio clip; the participant response will stop the audio clip and move onto the next trial.
Self-paced reading is a common task in psycholinguistics research, where a sentence gradually unfolds on the screen, one word at a time, as the participant presses a designated key. Self-paced reading tasks also often feature a comprehension question after each sentence is read. Crucially, the sentence of interest normally is removed from the screen when the participant is asked to respond to the comprehension question.
To achieve this design, we can again leverage the barrier property of the Tokenized Text Stimulus, which has built-in support for self-paced reading. First, we define the trial template as:
"tts_sentence_comprehension": {
"type": "basic",
"stimuli": ["tts_circles"],
"responses": ["tts_choice"] // we omit the definition of this response here. It can be any response supported by FindingFive
}
where the tokenized text stimulus "tts_circles" is defined as:
{
"type": "tokenized_text",
"alignment": "center",
"content": "Choose the picture where the large circle is to the left of the small circles",
"size": "30px",
"self_paced": true
}
Like audio and video stimuli, the barrier property of a tokenized text stimulus is also set to true by default, so we didn't need to specify that explicitlly here. As a result, participants must first read through the sentence word by word, without seeing what the comprehension question is. When the comprehension question appears, the original sentence would be gone from the screen, testing the participant's memory and comprehension of the sentence.
In this article, we have presented three common experimental designs that can be implemented via FindingFive's barrier feature. Did we help you brainstorm exciting ideas that you want to test right now on FindingFive? Dive into the details by checking out the rest of this Study Grammar documentation!