Inquisit--To execute experiments

Author: Dr Simon Moss

Overview

Inquisit is a software package, utilized to conduct experimental studies and other research designs. This package can be used to collect reaction time data, survey responses, and many other measurements.Stimuli can be words or pictures. The package is offered free for 30 days and for approximately $500 for a single license and is available from the website www.millisecond.com. To execute these studies, researchers need to write a script, like a simple computer code.

Overview of the script

In essence, researchers use the editor to construct a script.In particular, they need to specify the attributes or features of a series of elements, including the stimuli, questions, trials, instructions, and procedures. Next, they execute this script using the run command.

The script can be opened and saved, similar to a typical word processing file, with .exp as the extension.The data is then recorded in another file, with .dat, as the extension.In this file, each row of data corresponds to a specific trial in reaction time studies or a single participant in survey designs.

The scripts of some typical studies accompany the software - usually in the directory c:program filesmillisecond softwareInquisit 3 utorials.These scripts include research on implicit association tasks, surveys, subliminal priming, and covert attention tasks. These scripts can be adapted to fulfil the needs of a specific study. Alternatively, researchers can read the electronic tutorials to learn how to construct these scripts.

Example of a script: implicit attitude task

Consider a study in which various words appear on a screen. The participants must press one button if the word is pleasant or corresponds to conservative values. They must press another button if the word is unpleasant or corresponds to liberal values.

Later this task is adapted slightly;& they must one button if the word is pleasant or corresponds to liberal values. They must press another button if the word is unpleasant or corresponds to conservative values.

Suppose a participant responds more rapidly if the same button corresponds to pleasant and liberal words. This pattern of responses indicates they share more positive attitudes towards liberal, rather than conservative, values.

Stimuli

First, researchers might define the stimuli that will be used in the study - words, symbols, and pictures that might be presented during a trial including feedback or prompts during this trial. Researchers can define the pleasant words first, using the following code.

1. / items = pleasant
2. / 1 = "HAPPY"
3. / 2 = "GOLD"
4. / 3 = "FORTUNE"
5. / 4 = "LOVELY"
6. / 5 = "LIBERTY"

The researchers can then specify the remaining sets of words.

1. / items = unpleasant
2. / 1 = "DEAD"
3. / 2 = "ILLNESS"
4. / 3 = "VILE"
5. / 4 = "CATASTOPHE"
6. / 5 = "LONELY"
7. / items = conservative
8. / 1 = "CONSERVATIVE"
9. / 2 = "TRADITIONAL"
10. / 3 = "ESTABLISHED"
11. / 4 = "CONVENTIONAL"
12. / 5 = "STATUS QUO"
14. / items = liberal
16. / 1 = "LIBERAL"
17. / 2 = "UNCONVENTIONAL"
18. / 3 = "TOLERANT"
19. / 4 = "MODERATE"
20. / 5 = "REBELLIOUS"

In addition, during the experiment, at the bottom, researchers often include text that reminds participants of which buttons to press, for the various conditions.

2. / items = ("Press 'p' for pleasant or conservative")
3. / position = (75, 25)
4. / txcolor = (0, 0, 255)
6. / items = ("Press 'p' for pleasant or liberal ")
7. / position = (75, 25)
8. / txcolor = (0, 0, 255)
10. / items = ("Press 'q' for unpleasant or conservative ")
11. / position = (25, 25)
12. / txcolor = (0, 0, 255)
14. / items = ("Press 'q' for unpleasant or liberal ")
15. / position = (25, 25)
16. / txcolor = (0, 0, 255)

In this instance, position (25, 25) implies the text should appear in the upper right corner - 25% down and 25% across.The default is (50, 50).In addition, txcolor (0, 0, 255) represents blue. The default is black.The three numbers represent the extent to which the color comprises red, green, and blue respectively.

The final stimulus represents an error message.This stimulus will be presented whenever the participants presses the wrong key.

2. / items = ("ERROR")
3. / txcolor = (255, 0, 0)

Instructions

Second, researchers can then prepare the instruction pages, beginning with a welcome page.The "^" character represents an empty line.

2. ^^^^^^^ Implicit Association Test
3. ^^Welcome and thank you for participating.
5. In this study, you will receive a series of words.You need to press either a Q or P, depending on whether the word is pleasant or unpleasant - or whether the word corresponds to conservative or liberal values. Precisely which key to press will be specifie later.
7. Press Q if the word is pleasant or is synonymous with conservative values
8. Press P if the word is unpleasant or is synonymous with liberal values
10. Press Q if the word is pleasant or is synonymous with liberal values
11. Press P if the word is unpleasant or is synonymous with conservative values
13. You have completed the study
14. Thank you for your participation

Furthermore, some prompts are needed to show participants how to activate the next page of instructions - and how to return to a previous page.

2. / nextkey = ("P")
3. / prevkey = ("Q")

That is, nextkey represents the key that participants must press to proceed, and prevkey represents the key that participants must press to restore a previous page.

class="header1">Creating trials

class="header1">

class="header1">Thus far, the researcher has defined the stimuli and written the instructions.Next, the researcher must define the various conditions or classes of trials - and the keys that need to be pressed.

2. / stimulusframes = [1=pleasant]
3. / validresponse = ("q", "p")
4. / correctresponse = ("q")
5. This code defines the trials that are called pleasant.Stimulusframes indicates that 1 frame will be presented, and the word that will be presented is pleasant, as defined earlier. Valid response indicates that only a q or p will be registered - but only the q is correct.The remaining code defines the other classes of trials.
7. / validresponse = ("q", "p")
8. / correctresponse = ("p")
9. / stimulusframes = [1=unpleasant]
11. / validresponse = (" q", "p")
12. / correctresponse = ("q")
13. / stimulusframes = [1=conservative]
15. / validresponse = ("q", "p")
16. / correctresponse = ("p")
17. / stimulusframes = [1=conservative]
19. / validresponse = ("q", "p")
20. / correctresponse = ("q")
21. / stimulusframes = [1=liberal]
23. / validresponse = ("q", "p")
24. / correctresponse = ("p")
25. / stimulusframes = [1=liberal]

Creating blocks

Next, researchers need to define the various classes of blocks - sequences of trials.For example, during practice blocks, only one class of trials might be used.This example does not include a typical practice block - in which only one category is sometimes included at a time,

2. / trials = [1-40 = noreplace(liberalleft, conservativeright, pleasant, unpleasant)]
3. / bgstim = (unpleasant_liberal, pleasant_conservative)
4. / preinstructions = (compatible)
5. / errormessage = (errormessage, 200)
6. / blockfeedback = (latency, correct)
8. / trials = [1-40 = noreplace(conservativeright, liberalleft, pleasant, unpleasant)]
9. / bgstim = (pleasant_liberal, unpleasant_conservative)
10. / preinstructions = (incompatible)
11. / errormessage = (errormessage, 200)
12. / blockfeedback = (latency, correct)

These blocks are called compatible and incompatible. These blocks each comprise 40 trials, randomly selected without replacement.Bgstim specifies the instructions that will coincide with each trial. Preinstructions specifies the instructions that will precede these trials.Errormessage represents the error message that will appear, for a duration of 200 ms.Finally, the blockfeedback indicates the mean latency and percentage correct will appear when the block is complete.

Experiment and variables

Next, after defining the blocks, the research defines the sequence of blocks - which in turn are sequences of trials.

2. / preinstructions = (intro)
3. / postinstructions = (end)
4. / blocks = [1=block1;& 2=block2]

In this instance, blocks indicates that two blocks will be presented.Often, although not in this instance, blocks of practice trials are included as well.Whether the first block will be the compatible or incompatible conditions will be defined next, when specifying the between-subject variables to counterbalance.

2. / group = (1 of 2) (block1 = incompatible, block2=compatible)
3. / group = (2 of 2) (block1=compatible, block2=incompatible)

Accordingly, all odd numbered participants will first receive the incompatible block.All even numbered participants will receive the compatible block.

Defaults

Researchers can also change some of the defaults, as specified below.

With that, the script is essentially complete. However, we'll do a little fine tuning by specifying some default settings using the defaults element.

2. / screencolor = (175, 175, 255)
3. / fontstyle = ("Courier New", 14pt)

Executing the study

To execute the experiment, simply select run from the experiment menu.

Other features

When Inquisit is used, the stimuli can also be pictures and photographs.Typical script might be:

2. / items = targetA
3. / size = (20%, 20%)
5. /1 = "Flowers"
7. /1 = "man1.jpg"
8. /2 = "man12.jpg"
9. /3 = "man3.jpg"

These jpeg files are stored in the same directory as the script file.The (20%, 20%) refers to the size relative to the screen.

Inquisit can also align with other packages, such as survey monkey, to ascertain to assess reaction times while completing surveys.Most survey packages enable researchers to forward to other websites - in this instance, the Inquisit launch page - after particular questions are completed.

Inquisit also presents information on how to:

1. Execute conditional branching - that is, enable the responses of participants to determine which blocks appear next
2. Interrupt an experiment
3. Present stimuli that are answers the participant provided earlier
4. Test an experiment
5. Execute Inquisit 3 over the web
6. and many other features