Imagine we are making a visual search task, where there is one target (the letter “L”) amongst a large number of letter “T”s. We could make one component for each letter, but depending on the size of our search array that could get tricky, and mean alot of manual labor in the long run (e.g. if we wanted to change the size of each letter, we would have to change the letter height in all the components)….
We have already learned how to draw something with code when we learned about coding experiments from scratch. This can now be used in builder view by using what we learned in code components!
Remember, we can export a builder experiment to coder at any point, but we emphasized this is a one-way street and highly recommend avoiding editing the underlying python file. But exporting a python file can be useful to study how to do something in python if we are not sure. Start by adding a simple textStim with the letter “T” in the Text field, call the component “myLetter”. Save it and export the python file then search for “myLetter”.
Hopefully you can see under the hood how to make a textStim component:
myLetter = visual.TextStim(win=win, name='myLetter',
text='T',
font='Open Sans',
pos=(0, 0), height=0.1, wrapWidth=None, ori=0.0,
color='white', colorSpace='rgb', opacity=None,
languageStyle='LTR',
depth=0.0))
Note
When we use these “pinched” bits of python code they might not always translate smoothly to JavaScript. If this happens, export your experiment to JavaScript and instead use the JavaScript version of the component.
We have learned about how to use loops to add to lists, Objects created from components can be treated the same as any other variable i.e. we can have a list of Objects (also known as “instances” in python - in python lingo, here we have an “instance” of the “TextStim” class).
To make a list of textStim, we could use:
myLetters = []
for i in range(10):
myLetter = visual.TextStim(win=win, name='myLetter',
text='T',
font='Open Sans',
pos=(0, 0), height=0.1, wrapWidth=None, ori=0.0,
color='white', colorSpace='rgb', opacity=None,
languageStyle='LTR',
depth=0.0)
myLetters.append(myLetter)
This would make a list of 10 letters, but because they would all be in the same position, we would only see one letter if we drew them.
To present stimuli in a set of positions we could do something like this:
myXList = [-.5, 0, .5]
myLetters = []
for x in myXList:
myLetter = visual.TextStim(win=win, name='myLetter',
text='T',
font='Open Sans',
pos=(x, 0), height=0.1, wrapWidth=None, ori=0.0,
color='white', colorSpace='rgb', opacity=None,
languageStyle='LTR',
depth=0.0)
myLetters.append(myLetter)
print(myLetters)
This time we would have 3 letters but each would have a different X position, so we would be able to see them if we draw them.
We have already learned that all classes in the visual module have a method called draw() and setAutoDraw() and we know that the second of these methods means the object is drawn every time the window flips until we declare setAutoDraw(False).
Add a routine called “search” and add a fixation point that lasts the duration of our search time, let’s give it 10 seconds. OK then we need a code component, let’s call it ‘drawAll’. In the ‘Begin Routine’ tab, type:
for letter in myLetters:
letter.setAutoDraw(True)
Then in the ‘End Routine’ tab type:
for letter in myLetters:
letter.setAutoDraw(False)
If you run your experiment, you should now see three letter T’s displayed on-screen.
OK so this works, but with three positions we might think “what’s the point”, let’s make this worth our while by making 20 letters all at random locations. Rather than using a fixed set of positions let’s change our code to the following:
myLetters = []
for x in range(20):
myLetter = visual.TextStim(win=win, name='myLetter',
text='T',
font='Open Sans',
pos=(random(), random()), height=0.1, wrapWidth=None, ori=0.0,
color='white', colorSpace='rgb', opacity=None,
languageStyle='LTR',
depth=0.0)
myLetters.append(myLetter)
Here the method random() is used to generate a random value between 0 and 1.
Note
random() is a python method, if you are working online you will need the JavaScript equivalent which is Math.random() (hopefully PsychoPy will do this for you soon though!)
Because these random values are all positive, you might notice all the letters are in the top right hand side of the screen, so let’s scale those positions to be pos=(random()-0.5, random()-0.5).
Finally, let’s add a target, because there is only one of these we can use a basic component!
Hint: remember the color field of a text stim can take rgb values e.g. [1, 0, 0] would be red