【问题标题】:Python script doesnt work?: Playing sound, measuring reaction timePython 脚本不起作用?:播放声音,测量反应时间
【发布时间】:2016-12-06 01:34:57
【问题描述】:

我想制作一个脚本,仅当参与者从声音文件中听到 2/30 的声音时,才能测量参与者按 Enter 键或空格键的速度。 所以有时用户不需要按任何东西,脚本仍然会转到下一个声音文件。我该怎么做呢?我现在拥有的是:(而不是声音文件,我有文本 atm。):

# Grounding of Words Experiment #

#Import libraries
import re
import glob
from psychopy import sound, visual, event, data, core, gui # imports a module for visual presentation and one for controlling events like key presses


# ID, age, gender box display
myDlg = gui.Dlg(title="Experiment") #, pos=(400,400)
myDlg.addField('ID:')
myDlg.addField('Age:')
myDlg.addField('Gender:', choices = ['Female', 'Male'])
myDlg.show()#you have to call show() for a Dlg
if myDlg.OK:
    ID = myDlg.data[0]
    Age = myDlg.data[1]
    Gender = myDlg.data[2]
else:
    core.quit()

trial=0

#Creates the outfile, that will be the file containing our data, the name of the file, as saved on the computer is the filename
out_file="Grounding_experiment_results.csv"
#Creates the header for the data
header="trial,ID,Gender,Age,Word,rt,SpaceKlik\n"
#opens the outfile in writemode
with open(out_file,"w") as f:
    f.write(header)#writes the header in the outfile


# define window
win = visual.Window(fullscr=True) # defines a window using default values (= gray screen, fullscr=False, etc)

# Instruction box display

def instruct(txt):
    instructions = visual.TextStim(win, text=txt, height = 0.05) # create an instruction text
    instructions.draw() # draw the text stimulus in a "hidden screen" so that it is ready to be presented 
    win.flip() # flip the screen to reveal the stimulus
    event.waitKeys() # wait for any key press

instruct('''
Welcome to the experiment!

You will be hearing different words.
Whenever you hear the word "Klik" and "Kast" please press the left mouse button.
Whenever you hear any other word - do nothing.
Try to be as fast and accurate as possible.
Please put on the headphones. 
The experiment will take 5 minutes. 

Press any key to start the experiment''')

# Play sound

# Function that makes up a trial

trial(word):
    global trial
    trial += 1
    if word in ["Klik", "Press", "Throw"]:
        condition = "press"
    else :
        condition = "no_press"
    event.clearEvents()
    for frame in range(90):
        text = visual.TextStim(win, text=word, height = 0.05)
        text.draw() # draw the text stimulus in a "hidden screen" so that it is ready to be presented 
        time_start=win.flip()
    try:
        key, time_key=event.getKeys(keyList=['space', 'escape'], timeStamped = True)[0] # wait for any key press
    except IndexError:
        key = "0"
        rt = "NA"
    else:
        if key=='escape': 
                core.quit()
        rt = time_key - time_start
    if key == "space" and condition=="press":
        accuracy = 1
    elif key == "0" and condition=="no_press":
        accuracy = 1
    else:
        accuracy = 0
    with open(out_file,"a") as f:
        f.write("{},{},{},{},{},{},{},{}\n".format(trial,ID,Gender,Age,word,accuracy,rt,SpaceKlik))

# s = sound.Sound('sound.wav') 
# s.play()

# Register space bar press or mouse click

# Measure reaction time

# Check to see if answer is correct to sound - certain sound files are "klik". Others "kast", "løb", "sko" and so on

# Write csv logfile with coloumns: "ID", "Gender", "Word", "Correct/incorrect", "Reaction time", "Space/click"

我最终都会在 PsychoPy 中运行。预先感谢您的友好帮助。

【问题讨论】:

  • 作为一般提示,请避免将trial 作为全局变量。只需在 trial() 函数之外递增它并将其作为第二个参数传递给函数。同样,不要在每次运行指令函数时创建文本刺激(这是一个耗时的操作)。创建一次,并将文本刺激与新的文本内容一起传递给函数。这是您的试验功能中的一个更大的问题,其中文本刺激在每一帧上都重新实例化。这是非常省时的,可能会导致时间问题。

标签: python loops audio measure psychopy


【解决方案1】:

您的缩进存在问题,因此并非所有内容都在适当级别的试用功能中运行。您可能可以放弃异常处理以更直接地检查是否做出了响应。然后在for 循环中使用else: 子句来处理如果循环完成但没有按下任何键会发生什么。这避免了必须处理在任何给定帧上发生的无响应的逻辑(如果响应很快就会发生,这可能没有任何意义)。

类似这样的通用伪代码:

# define window
win = visual.Window(fullscr=True)

# function to draw instructions (probably overkill unless run more than once)
def instruct(instructions = 'xxx'):
    # as you have above

# function to run each trial
def trial(number = -999, sound_name = 'A'):

    sound = Sound.sound(sound_name)

    if sound_name in ["Klik.wav", "Press.wav", "Throw.wav"]:
        condition = "press"
    else:
        condition = "no_press"

    event.clearEvents()

    sound.play()

    for frame in range(90):

        time_start = win.flip()

        keys = event.getKeys(keyList=['space', 'escape'], timeStamped = True)
        if keys: # if a non-empty list returned:
            key, time_key = keys[0]
            rt = time_key - time_start

            sound.stop() # ready to start next trial immediately

            if key == 'escape': 
                core.quit()

            if condition == "press":
                return {'accuracy':1, 'rt':rt}
            else:
                return {'accuracy':0, 'rt':rt}

    else: # the loop ended without a key press
        if condition == "press":
            return {'accuracy':0, 'rt':'NA'}
        else:
            return {'accuracy':1, 'rt':'NA'}

####
# run the experiment:
####

####
# show the instructions:
####
instruct('some instructions')

####
# run the trials:
####
for trial_num in range(10):

    # run each trial and get the results.
    # I'm not sure where you're getting your sound values
    # but they could be indexed from a list using the trial number:
    result = trial(number=trial_num, sound_name=sound_names[trial_num])

    # then save the values from the returned dictionary to a file

【讨论】:

    【解决方案2】:

    好的,很好 - 谢谢!现在我继续尝试开发你的脚本。我已经添加了一个试用列表,现在我不太确定该怎么做。我的脚本如下所示:

    # Grounding of Words Experiment #
    
    # -*- coding: utf-8 -*-
    
    #Import libraries
    import re
    import glob
    from psychopy import sound, visual, event, data, core, gui # imports a module for visual presentation and one for controlling events like key presses
    import ppc
    
    # ID, age, gender box display
    myDlg = gui.Dlg(title="Experiment") #, pos=(400,400)
    myDlg.addField('ID:')
    myDlg.addField('Age:')
    myDlg.addField('Gender:', choices = ['Female', 'Male'])
    myDlg.show()#you have to call show() for a Dlg
    if myDlg.OK:
        ID = myDlg.data[0]
        Age = myDlg.data[1]
        Gender = myDlg.data[2]
    else:
        core.quit()
    
    # define window
    win = visual.Window(fullscr=True)
    
    
    # function to draw instructions (probably overkill unless run more than once)
    def instruct(txt):
        instructions = visual.TextStim(win, text=txt, height = 0.05) # create an instruction text
        instructions.draw() # draw the text stimulus in a "hidden screen" so that it is ready to be presented 
        win.flip() # flip the screen to reveal the stimulus
        event.waitKeys() # wait for any key press
        # as you have above
    
    
    # function to run each trial
    def trial(number = -999, sound_name = 'A'):
    
        sound = Sound.sound()
    
        if sound_name in ["dog-howling.wav"]:
            condition = "press"
        else:
            condition = "no_press"
    
        event.clearEvents()
    
        sound.play(sound_name)
    
        for frame in range(90):
    
            time_start = win.flip()
    
            keys = event.getKeys(keyList=['space', 'escape'], timeStamped = True)
            if keys: # if a non-empty list returned:
                key, time_key = keys[0]
                rt = time_key - time_start
    
                sound.stop() # ready to start next trial immediately
    
                if key == 'escape': 
                    core.quit()
    
                if condition == "press":
                    return {'accuracy':1, 'rt':rt}
                else:
                    return {'accuracy':0, 'rt':rt}
    
        else: # the loop ended without a key press
            if condition == "press":
                return {'accuracy':0, 'rt':'NA'}
            else:
                return {'accuracy':1, 'rt':'NA'}
    ####
    # define triallist
    
    trial_list = []
    conditions = ["klik", "notklik"]
    sounds = ["dog-howling.wav", "sound2.wav", "sound3.wav"]
    for condition in conditions:
        for sound in sounds:
            # Add a dictionary for every trial
                trial_list += [{
                    'ID': ID,
                    'age': AGE,
                    'gender': GENDER,
                    'condition': condition,
                    'sounds': sound,
                    'rating': '',
                    'rt': ''
                        }]
    
    # Randomize order
    trial_list = sample(trial_list, len(trial_list))
    
    # Add trial numbers
    for i, trial in enumerate(trial_list):
        trial['no'] = i + 1 # start at 1
    
    # write file
    
    ####
    # run the experiment:
    ####
    
    ####
    # show the instructions:
    ####
    instruct('''Welcome to the experiment!
    
    You will be hearing different words.
    Whenever you hear the word "Klik" and "Kast" please press the left mouse button.
    Whenever you hear any other word - do nothing.
    Try to be as fast and accurate as possible.
    Please put on the headphones. 
    The experiment will take 5 minutes. 
    
    Press any key to start the experiment''')
    
    ####
    # run the trials:
    ####
    for trial_num in range(10):
    
        # run each trial and get the results.
        # I'm not sure where you're getting your sound values
        # but they could be indexed from a list using the trial number:
        result = trial(number=trial_num, sound_name=sound_names[trial_num])
    
        # then save the values from the returned dictionary to a file+
    
    #Creates the outfile, that will be the file containing our data, the name of the file, as saved on the computer is the filename
    #out_file="Grounding_experiment_results.csv"
    #Creates the header for the data
    #header="trial,ID,Gender,Age,Word,rt,SpaceKlik\n"
    #opens the outfile in writemode
    #with open(out_file,"w") as f:
    #    f.write(header)#writes the header in the outfile
    

    【讨论】:

    • 您确实需要查看 TrialHandler 类。将您的试用列表提供给它,它会自动处理跟踪试用编号、保存数据等。 PsychoPy 的 Coder 视图中的 Demos 菜单下有几个示例。
    猜你喜欢
    • 1970-01-01
    • 1970-01-01
    • 2019-03-18
    • 1970-01-01
    • 1970-01-01
    • 2023-01-23
    • 1970-01-01
    • 2012-10-01
    • 1970-01-01
    相关资源
    最近更新 更多