Sie sind auf Seite 1von 76

SuperLab 4.

5 Manual

Copyright Information
Copyright 1990-2011 Cedrus Corporation. All Rights Reserved. Under the copyright laws, this manual or the software described within may not be copied, in whole or in part, without the written consent of the manufacturer, except in normal use of the software to make a backup copy. The same proprietary and copyright notices must be afxed to any permitted copies as were afxed to the original. This exception does not allow copies to be made for others, whether or not sold, but all of the material purchased (with all backup copies) may be sold, given, or loaned to another person. Under the law, copying includes translating into another language or format. The information contained in this publication is subject to change without notice. Cedrus Corporation makes no warranty of any kind with regard to this written material and shall not be held liable for errors contained herein or for incidental or consequential damages that may occur in connection with the use of this manual or the software, even if advised of the possibility of such damages. See also the License Agreement at the end of the manual. Cedrus and SuperLab are registered trademarks of Cedrus Corporation. All other products and brand names are trademarks of their respective holders.

Cedrus Corporation ! P.O. Box 6309! San Pedro, CA 90734! United States !

Toll Free: ! Other: ! Email: ! Web: !

1-800-CEDRUS1 (1-800-233 7871) 1-310-548-9595 superlab@cedrus.com www.cedrus.com

Rev. Feb 23, 2011

About Software Copying

Thank you for purchasing SuperLab, an experimental tool we hope will serve you well. SuperLab is the result of long, long hours of work. Weve put a lot of time into it so you wont have to. And wed like to continue improving it and implementing your ideas and suggestions. So please, we plead with you, if you must show SuperLab to a friend or colleague, do NOT give them the fully functional version; instead, give them a demonstration version or our contact information to send them a free demo. Giving away the fully functional version, in addition to being an unethical and a punishable crime, results in fewer sales and discourages us from continuing to support and improve the product. Thanks again for choosing SuperLab. We wish you success in your work.

- ii -

Table of Contents
1. Introduction!
Installation on Mac OS X Installation on Windows System Requirements The Community Forums Cedrus Support Whats Not Covered in This Manual Thank You

1
1 1 1 2 2 3 3

2. SuperLab Tutorial!
Creating a Block and a Trial Saving the Experiment Creating an Event Linking Participant Input and Input Devices Ready to Run? Dening Responses Adding Cues Stimulus Lists Creating/Importing Stimuli Creating the Blocks Randomization

5
6 7 7 8 8 9 9 10 11 13 15 16

3. More In-Depth Features!


Delivering a Stimulus or Event Marker Controlling Participant Input Determining the Correct Response Feedback Codes and Code Values Creating the Codes Using the Codes Trial Variable Editor Participant Groups Editing Groups Using Participant Groups

19
19 19 20 21 22 22 23 24 24 24 26

- iii -

4. Conditional Branching!
Features Needed Introduction to Expressions More On Expressions Looking at a Subset of Trials Introduction to Macros Using Code Values for More Capable Expressions Trial-Level Expressions

27
27 28 29 30 32 33 39

5. Timing!
Resetting The Timer Reaction Time Accuracy Presentation Accuracy

41
41 41 43

6. How Do I...!
Presenting Two or More Visual Stimuli at the Same Time Providing Feedback to Participants Presenting a Sound and Visual Stimulus at the Same Time Giving Participants a Break After N Trials Randomize a Stimulus List Moving Experiments Between Macs and PCs

47
47 48 50 50 53 55

7. The Cedrus Data Viewer!


Adding and Viewing Files Examining and Filtering the Data Other Features

57
58 59 59

- iv -

Also From Cedrus

Response Pads
Cedrus offers two series of response pads including one for fMRI. Both series offer a built-in reaction time timer and an Accessory Connector that provides six lines of I/O. Visit our web site (http://www.cedrus.com) for more details.

RB Series Response Pads


These response pads feature replaceable key tops, four different button layouts, and a USB interface. Continuously improved, the current RB-x30 models represent the fourth generation in this series. Each pad is shipped complete with white, red, green, blue, yellow, and clear key tops.

Lumina fMRI Response Pads


In use worldwide, these reliable fMRI pads are built using plastic and ber optics. Features include an input for the scanner trigger and new, quiet keys with a short travel distance (1.5mm) and tactile feedback. A Lumina system ships complete and ready to install, including all cabling.

Rev. Feb 23, 2011

Also From Cedrus

StimTracker
Using EEG/ERP in your experiment? StimTracker makes it easy to send event markers to your recording equipment. It supports:

Event markers sent from the computer via USB Up to four photocells Audio Microphone input Six digital input lines

Cedrus works closely with several manufacturers. StimTracker ships with a kit that includes an adapter board and cable. The kit makes it instantly compatible with:

ADInstruments ANT Neuro Biopac BioSemi Brain Products MindWare


Alternatively, it can come with a kit that lets you connect to a PCs parallel port or to a terminal block. StimTracker ships complete including a AC adapter, two photocells, USB cable, and a kit of your choice. Visit our web site (www.cedrus.com) for more details.

- vi -

Why is our memory good enough to recall to the last detail things that have happened to us, yet not good enough to recall how often we have told them to the same person.

- Franois de La Rochefoucauld, French writer, (1613-1680).

- vii -

C HAPTER 1

Introduction

Thank you for purchasing SuperLab, your tool for building experiments. With SuperLab, you can present images, text, and video, and collect responses accurately from participants. Your CD-ROM includes the following:

SuperLab application Cedrus Data Viewer application Letter Rotation sample experiment
The Cedrus Data Viewer lets you view collected data les one by one or in combination, transpose them, and merge them.

Installation on Mac OS X
Copy the SuperLab and Cedrus Data Viewer applications to your Applications folder.

Installation on Windows
Run the setup.exe le found on the CD-ROM. To run SuperLab, click on the Start button in the bottom left corner of your monitor, then click on All Programs, then SuperLab 4.5.

System Requirements On the PC: Windows 2000, XP, Vista, or Windows 7.


For using movies, you need DirectX 9.0 and a video card that works with DirectX 9.0, including support for hardware acceleration and VMR. (Note: DirectX 9.0 will require Windows XP or later.)

On the Mac: Mac OS 10.4 or later is required.


Both versions of SuperLab can play movies on a single core or single processor computer, but playing movies smoothly requires a dual core computer. We recommend 1 GB for memory. Having a very good video card is more important than the processing speed of the CPU.

The Community Forums


The next logical step is to register and setup your own account at the Cedrus community forums. It takes about one minute. The link is http://community.cedrus.com. Anyone can search the forums for answers or pre-built experiments. But only registered users can post questions and download experiments, including some sample ones that are needed for going through the conditional branching chapter in this manual. To get started:

Click on the Register link, shown surrounded above by a red rectangle. The forums software will ask you to enter some basic information and to chose a user
name and password.

When done, the forums software will send a link to the email address that you have
entered.

Click on the link that was sent to you. This step is necessary to reduce the amount of
spam. Important: the registration process will not be complete until you click on the link that was sent to your email address.

Cedrus Support
Cedrus provides several ways in which you can get support, report bugs, or to give us feedback on how we can improve SuperLab. Via Community Forums The forums are a very good place to start, rst by searching the existing posts to see if your question has been already been asked and answered. You can always post your own question.

1 - Introduction

PAGE 2

By Email Due to spam, we no longer publish an email address. Tech support emails can be sent to Cedrus by visiting this page: ! http://www.cedrus.com/misc/send_email.htm

By Phone The hours are 9 AM to 4 PM Pacic Standard Time (PST) time. The number is 1-800-CEDRUS1 (thats 1-800-233 7871). It is available toll free anywhere in the United States and Canada. If you are calling from overseas or from the Los Angeles area, the number is 1-310-548-9595.

Whats Not Covered in This Manual


This manual is intended to serve mainly as a tutorial, not a reference. Additional reference information is included within SuperLab itself. It can be accessed by clicking on the Help menu in SuperLab and then Help Contents. Here are some of the topics that can be found in the help le that are not included here:

Command line information Reference information on the different input devices that are supported A description of the collected data le Thank You
As of this writing, SuperLab is in its 16th year and has been sold in over 60 countries. This would not have been possible without you, our software users. So from the bottom of our hearts: thank you.

1 - Introduction

PAGE 3

C HAPTER 2

SuperLab Tutorial

In this tutorial, we will build a word/non-word experiment that will present instructions to the participant followed by six trials. Each trial will consist of a cue and stimulus. This experiment is designed to illustrate a few techniques and features of SuperLab:

Creating blocks, trials, and events, then linking them together Taking advantage of stimulus lists Dening the responses Randomizing trials
When developing your own experiments, it is important to plan carefully and take advantage of the stimulus lists and code values features, if possible. This can save you a lot of development time as well as make it easier to see the structure of your experiment and maintain it.

Going through this tutorial takes less than an hour. What you need to type is shown in this color.

Creating a Block and a Trial


The main experiment window where all trials are created is shown on the previous page. The smallest unit of an experiment is the event, usually a stimulus or an event marker. A trial is a set of events and a block is a set of trials. To run an experiment, you must have at least one block, one trial, and one event that are linked together. Note the blue rectangle drawn around the list of blocks. To familiarize yourself with the SuperLab workspace, try the following steps:

Click on the Edit menu. A menu will appear. Choose the menu item One New Block. This will cause a dialog (screen) to appear. This
is the Block Editor.

Type Instructions. Click on the OK button. The Block Editor will close and Instructions will appear in the
list of blocks Now we will create the rst trial:

In the main experiment window, click on the list of trials, currently empty. The blue
selection rectangle will move from the list of blocks to the list of trials.

Click on the Edit menu again. Note how some menu items have changed. The item that used to read One New Block is
now One New Trial. This is because the Edit menu is dynamic, meaning that its items change depending on your actions.

Release the mouse button to make the Edit menu close. We still want to create a trial but
we will do so using the toolbar.

In the toolbar, click on the New Trial icon shown below. The Trial Editor will appear.

Notice how SuperLab displays yellow "tooltips" as you move the mouse cursor over the icons. The tooltips for the New Trial, Edit Trial, and Delete Trial icons are also dynamic and will change depending on where the blue selection rectangle is.

Type Instructions.

2 - SuperLab Tutorial

PAGE 6

Click on the OK button. The Trial Editor will close and Instructions will appear in the
list of trials. You can also press the Tab key to move the blue selection rectangle.

Saving the Experiment


We may not have created much of an experiment yet but its a good idea to save and save often:

Click on the Save Experiment icon in the toolbar. You can also click on the File menu and
select Save.

SuperLab will ask you to provide a name for your experiment. The default le name
extension is SL4. For example, if you type tutorial, SuperLab will save a le as tutorial.sl4.

Creating an Event
Back to the experiment, you are now ready to create your rst event. We want to complete the instructions part of the experiment.

In the main experiment window, click on the list of events. The selection rectangle will
move from the list of trials to the list of events.

In the toolbar, click on the New Event icon. The Event Editor will appear.
As mentioned earlier, an event is the smallest unit of an experiment. It is where you tell SuperLab what to present to the participant, what is a correct response, and how to provide feedback based on the participants response.

For the event name, type Instructions. Click on the Stimulus tab to select it if its not selected already. Click on the Event Type popup menu and select Text. In the large text editing eld, type:
You are about to be presented with some words. Press the letter Y if you recognize the word or the letter N if you do not. Press any key to start the experiment. Click on the OK button. The Event Editor will go away and Instructions will also appear in the list of events.

2 - SuperLab Tutorial

PAGE 7

Linking
In SuperLab, various elements of an experiment are created separately and then "linked" together. To link the block, trial, and event that you just created, follow these steps:

Click (once only) on the block "Instructions" to select it. To tell SuperLab that the trial "Instructions" is part of the selected block, click on the blue
checkbox that's on left of the trial name.

The block and trial are now linked. Let's do the same for the trial and event:

Click on the trial "Instructions" to select it (the trial name itself and not its checkbox). Click on the blue checkbox that's on the left of the event name.
Click on the Save Experiment icon.

Participant Input and Input Devices


Most experiments require some sort of response from the participant. Prior to running an experiment, we need to select the device that participants will use to respond:

Click on the Experiment menu and select Participant Input. A dialog opens. We will use the keyboard for this experiment. SuperLab offers two keyboard options.
The rst is used when the response is expected to be a single key press or key release, e.g. Y or N. The second is used when the response is expected to be words or sentences, e.g. Apple.

Click on the checkbox to the left of "Keyboard-Single Keys" Click on the Done button
The following is a list of input devices that are supported by SuperLab:

Keyboard Mouse or Touch Screen Microsoft Serial Mouse Microphone Input RB-x30 Series Response Pads RB Series Earlier Model Pads

2 - SuperLab Tutorial

PAGE 8

SV-1 Smart Voice Key Lumina fMRI Response Pads Measurement Computing I/O Card PST Serial Response Box Ready to Run?
You now have enough to run an experiment. Click on the Run Experiment icon shown on the right of the toolbar. SuperLab will present the Run dialog:

Since we're still early in the development stage, there is no need to enter the participant's name or collect data:

Click on the Save collected data checkbox to turn it off Click on the Run button
SuperLab will open a window that covers the entire screen, display the instructions that you typed in Step 4 earlier, and wait for you to press any key on the keyboard. You can also press the Esc key to cancel an experiment in progress.

Defining Responses
As mentioned in the instructions, the participant is expected to respond by pressing the keys Y and N. We need to create two corresponding responses. Note that SuperLab will always record what the user presses regardless of whether you dene responses or not, but by doing so you can later tell the Event Editor which is the correct response for a particular stimulus. This has the following benets:

You can chose to end an event only after a correct response instead of any response You can provide feedback to the participant based on whether his or her response is
correct or not

2 - SuperLab Tutorial

PAGE 9

The responses in the collected data le will automatically be coded for correctness
To create the responses:

Click on the Experiment menu and select Participant Input. A dialog opens. The
"Keyboard-Single Keys" input device is already enabled (checked) and selected.

Click on the Responses tab if it's not selected already. Click on the New... button to create a response. Another smaller dialog opens.

For the response name, type "Y key (is word)". It is a good idea to type a descriptive
name of the key instead of simply "Y".

Type "y" for the actual key. Click on OK to close the response editor.
We repeat the steps above for the N key:

Click on the New... button to create a response. The response editor opens. For the response name, type "N key (is not word)". Type "n" for the actual key. Click on OK to close the response editor. Click on Done to close the Participant Input dialog.
Remember to save; click on the Save Experiment icon.

Adding Cues
We are now ready to build the core of the experiment. Recall that we want to present six trials, each consisting of a cue and a stimulus. Let's start with the cue:

2 - SuperLab Tutorial

PAGE 10

Press Tab until the blue selection rectangle surrounds the list of events. Make sure that the existing event called "Instructions" is not selected. You can do this by
clicking anywhere on the white space under the event name, or by choosing Select and then None from the Edit menu.

When running an experiment, SuperLab presents events in the order in which they
appear in the main experiment window.

Click on the New Event icon on the toolbar (shown on the right). The Event Editor
appears.

For the event name, type Cue. Click on the Stimulus tab if it's not selected already. The Event Type should still be set to Text. In the large text editing eld, type "+".
A cue is typically presented for a brief moment to direct the participant's attention to a particular location on the screen:

Click on the Input tab to select it. Click on the ...or a time limit, whichever happens rst checkbox to turn it on. In the edit eld below the checkbox, type 500 to have the cue displayed for half of a
second.

Click on the After any response from the participant checkbox to turn it off. A participant response during a cue is typically ignored. To reduce the clutter in the
collected data le, click on the Record and save response checkbox to turn it off. We're done with the cue. Click on OK to close the Event Editor.

Stimulus Lists
The next step is to create the events that present the stimuli. This is similar to creating the instruction and cue events except that we do it six times, one for each stimulus. Here, we introduce the all important stimulus lists feature which will allow us to create a single event instead of six:

Click on the Experiment menu and select Stimulus Lists. A dialog opens. Click on the New icon. A second dialog, the Stimulus List Editor, opens. You can create lists of les or lists of single line sentences. Click on the List Type popup
menu in the upper right corner of the dialog and change File to Text.

2 - SuperLab Tutorial

PAGE 11

For the list name, type "My Word-NonWord list". In the large edit eld, enter three words and three nonwords by typing:
apple plane train neibs quighs chortz

Click on OK to close the Stimulus List Editor. Click on Done to close the Stimulus Lists dialog.
The list is ready. Before we proceed to using it, it's worth taking a minute to emphasize the advantages of using stimulus lists:

You can create a single event. This becomes more and more important as the experiment
gets larger.

When you have a few events that use stimulus lists, it is easier to see the structure of the
experiment in the main window.

It is easier to maintain the experiment. You can increase the size of the experiment by simply modifying the list and without
touching the structure of the experiment. Trying to do so without using stimulus lists is more error prone and costly.

2 - SuperLab Tutorial

PAGE 12

Click on the Save Experiment icon.

Creating/Importing Stimuli
You now have everything needed to create the stimuli using a single event:

Press Tab until the blue selection rectangle surrounds the list of events. Make sure that none of the existing events are selected. You can do this by clicking
anywhere on the white space under the events.

Click on the New Event icon on the toolbar. The Event Editor appears. For the event name, type Stimulus. Click on the Stimulus tab if it's not selected already. The Event Type should still be set to Text. Click on the Use Text option and select Text Chosen From the List:. The name of the list
that you created in the previous step appears. That's it. You have just taken care of presenting your stimuli with a couple of mouse clicks. We still need to tell SuperLab what is the correct response for a given list item.

2 - SuperLab Tutorial

PAGE 13

With the Event Editor still open, click on the Correct Response tab. The items in your
stimulus list are visible in this tab.

Select the rst three words: apple, train, and plane.


In Windows, you can select all three simultaneously by hold down the Shift key while you click with the mouse. On the Mac, you can select all three simultaneously by hold down the Command () key while you click with the mouse.

Click on the "...the correct response is" popup menu (in the upper right part of the dialog)
and select the One or more of the following option.

Click on the checkbox to the left of "Y key (is word)" to turn it on.
We're done with the list of words. For nonwords:

Select the three nonwords: neibs, quighs, and chortz. Click on the "...the correct response is" popup menu and select the One or more of the
following option.

Click on the checkbox to the left of "N key (is not word)" to turn it on.
We would like the event to stay on the screen until the participant presses the correct key:

Click on the Input tab. Click on the After a correct response from the participant checkbox to turn it on.
Last but not least (as far as this event is concerned), we need to reset the reaction time timer. This is a very important detail: not doing so means that the reaction times that SuperLab collects in the data le will be measured relative to the onset of the cue instead of the onset of the stimulus itself:

Still in the Input tab, click on the Reset RT timer checkbox to turn it on. Click on OK to close the Event Editor.
Back in the main window, notice that the Stimulus event has a stacked checkbox ( ) instead of standard checkbox. This is a visual indication that the event uses a stimulus list.

Remember to save; click on the Save Experiment icon.

2 - SuperLab Tutorial

PAGE 14

Things get even easier from here. We still need to create one trial and one block, and link them together. We start with the trial:

Press Tab until the blue selection rectangle surrounds the list of trials. Click on the New Trial icon in the toolbar. The Trial Editor appears. Type Word-Nonword Trial. Click on the OK button. The Trial Editor will go away and Word-Nonword Trial will
appear in the list of trials. Click on it to select it.

Click on the checkboxes to the left of the events "Cue" and "Stimulus". This tells
SuperLab that trial "Word-Nonword Trial" consists of event "Cue" followed 500 milliseconds later by the stimulus.

Creating the Blocks


It is tempting to simply link the "Word-Nonword Trial" to the existing "Instructions" block. But this would be bad practice and, in the case of this particular experiment, not even an option. The reason is because we want to randomize the trials. If everything was part of a single block, the instructions would get randomized as well and presented at the beginning only by chance.

Press Tab until the blue selection rectangle surrounds the list of blocks. Click on the New Block icon in the toolbar. The Block Editor appears. Type Block 1. Click on the Randomize tab. Click on the Randomize the trials in this block checkbox to turn it on. Click on the OK button. The Block Editor will close and Block 1 will appear in the list
of blocks. Click on it to select it.

Click on the checkbox to the left of the trial "Word-Nonword Trial".

2 - SuperLab Tutorial

PAGE 15

Click on the Save Experiment icon. Congratulations! You have learned several concepts of SuperLab such as linking and using stimulus lists. You now have a ready-to-go experiment. Click on the Run Experiment icon on the toolbar to run it. You have now completed the SuperLab 4 tutorial.

Randomization
SuperLab offers several ways to randomize. This section summarizes the available options:

Trials in a block: This is by far the most commonly used randomization. To randomize
trials within a block, double-click on the block to invoke the Block Editor and see the options in the Randomize tab. This is also discussed in the chapter How Do I....

Location of a visual stimulus: When editing the settings for a visual stimulus, you can
specify that the position be variable instead of xed. After doing so, you can specify that the location/position be variable by clicking on the Experiment menu, clicking on Trial Variables, then editing the Location trial variable that you are using.

Text color: Events of type Text can also use a variable to chose the color. To edit the
range of colors and select randomization, click on the Experiment menu, click on Trial Variables, then edit the Color trial variable that you are using.

Time limit: The time limit provided for a participant to respond can also be made a
variable in the Event Editors Input tab. To edit the time range and select randomization, click on the Experiment menu, click on Trial Variables, then edit the Number trial variable that you are using.

2 - SuperLab Tutorial

PAGE 16

There is an additional option to randomize events within a trial. This is an option that is used in very few circumstances, e.g. to randomize which of two stimuli within the trial is presented rst. To use it, double-click on the trial to invoke the Trial Editor, and see the options in the Randomize tab.

2 - SuperLab Tutorial

PAGE 17

C HAPTER 3

More In-Depth Features

Delivering a Stimulus or Event Marker


Many experiments require that an event marker be sent to external EEG/ERP or other recording equipment. In SuperLab, this is done by creating an event and, in the Event Editors Stimulus tab, setting its Event Type to Digital Output. You will see additional options to select the output device and pulse duration. As of this writing, supported digital output devices are:

Cedrus Lumina fMRI response pads Cedrus RB-x30 response pads Cedrus StimTracker universal event marker interface Cedrus SV-1 voice key device Supported Measurement Computing I/O cards Supported National Instruments I/O cards Controlling Participant Input
Determining how the event ends is done in the Input and Correct Response tabs.

IMPORTANT: If any of your trials contain more than just one event, pay particular attention to the Reset RT Timer checkbox. If none of your events have this checkbox turned on, the reaction time will be measured relative to the onset of the rst event in the trial. If the rst event happens to be a cue and the second event is the stimulus, RTs will be measured relative to the onset of the cue -- probably not the desired outcome.

Determining the Correct Response


SuperLab provides three different ways to determine the correct response(s) to an event:

Based on a code value thats attached to the event or the stimulus list item
(recommended for most cases)

Based on a stimulus list item if the event uses a stimulus list For the event itself if it does not use a stimulus list
If your trials do have a correct response, it is a good idea to dene them in the Correct Response tab even if you chose to end an event After any response from the participant in the Input tab. The most obvious benet is that the participants responses will be coded for you in the data le as correct, incorrect, or no response. But another advantage is the ability to provide feedback to the participant based on correctness or to mark trials or events with a particular code value. The following screen snapshot shows a correct response dened based on a code value:

Other features:

You can dene multiple correct responses even if they are from different devices, e.g.
keyboard and response pad.

For Go/No-Go experiments, a correct response can be None, the participant must not
respond, as shown:

3 - More In-Depth Features

PAGE 20

Please refer to the following sections for more information:

Adding Feedback Collected Data File Codes and Code Values


See online Help Contents for the additional event types like Picture File, Text, and so forth.

Feedback
It is common in an experiment to provide feedback to participants. In SuperLab, feedbacks are handled at the event level and generally consist of "actions" performed when a "condition" happens. The supported conditions are:

Response is correct: how SuperLab interprets correctness depends on the option selected
in the Event Editor's Correct Response tab

Response is not correct No response within a certain amount of time: this time value is the same as the event's
time limit if the event imposes a time limit, otherwise it can be user specied

Reaction time is too slow Reaction time is too fast Always: this condition's actions are always executed. This is useful, for example, to mark
a trial as having been presented regardless of the participants response The following actions are supported:

Present Event: this is useful for displaying a text message, playing a beep, or sending an
event marker; you can have many Present Event actions per condition

Specify Where To Go: tells SuperLab what to do for any given condition: repeat the trial,
present the remaining events in the trial, or skip any remaining events in the trial and move to the next one; you can have only one Specify Where To Go action and it needs to be the last one

Change Code Value: lets you "mark" an event or trial; you can have many Change Code
Value actions

3 - More In-Depth Features

PAGE 21

See chapter 4, Conditional Branching, for an example of how to use the Change Code Value action, and chapter 6 for an example on how to provide feedback to participants.

Codes and Code Values


Codes in SuperLab serve three main purposes:

To "describe" a trial, stimulus list item, or event: They are intended to help you with the
post-processing of data collected with SuperLab. The idea is to create codes, provide them with values, and attach the values to trials, events, or stimulus list items. When you run the experiment, the collected data is saved as a plain text le with one line per response. Each code that you create will result in an additional column (eld) being added to the data le, with the contents of the column being the code values. With some planning, you can have your data saved already pre-coded.

To determine the correct response: Once you have attached a code value to an event or
stimulus list item, you can tell SuperLab what is the correct response based on a code value (see the Event Editor's Correct Response tab).

To help with conditional branching: The code value that's attached to a trial can be
changed while an experiment is running, e.g. to mark a trial as having been presented or to indicate that the participant's response to a trial was correct. Then, when building an expression in the Block Editor (either in the When to Run tab or in the Macros tab), you can tell SuperLab to evaluate trials that have a specic code value attached to them.

Creating the Codes


To create a code, click on the Experiment menu and select Codes. The following dialog appears:

The dialog snapshot above shows four codes already created. Click on a code to see its values; in the example above, the code "Front or Back" has two values.

To create a new code, click on the New button on the left hand side. Another dialog
appears:

3 - More In-Depth Features

PAGE 22

When creating a code, you must specify what you intend to use it with: trials, or events and stimulus list items.

To create a value for the code, select the code and then click on the New button on the
right hand side.

Click on one of the two Delete buttons to delete a code or a code value. If the code or
code value is being used elsewhere in the program, SuperLab will ask you to conrm.

Click on one of the two Edit buttons to edit a code or a code value. You can rename a
code or code value at any time without having to reassign it to a trial, event, or stimulus list item.

Using the Codes


Once you have created codes and values, you can use them in the following places:

You can attach code values to a trial in the Trial Editor's Code Values tab. You can attach code values to stimulus list items in the Stimulus List Editor. You can attach code values to events in the Event Editor's Code Values tab. This is
possible only if the event does not use a stimulus list.

You can change a code value using an "action" in the Event Editor's Feedback tab by rst
specifying the condition for which to execute the action, e.g. if response is correct, and then creating a new action.

When creating or editing an expression in the Block Editor, you can specify a subset of
trials that has a particular code value attached to them. For additional information regarding Codes, please see the following sections below:

Trial Editor Event Editor

3 - More In-Depth Features

PAGE 23

Trial Variable Editor


Trial variables provide the added exibility of varying an aspect of the experiment relatively easily. The following table summarizes the trial variable types that SuperLab supports, how they are used, and which event types they can be used with:

Variable Type Color

Usage Vary the color of text being presented Vary an events time limit

Used With Event Types Text

Number

All event types; this variable can be used in the Event Editors Input tab Movie, Picture File, RSVP, Self-Paced Reading, and Text. For these event types, click on the Settings button in the Event Editor's Stimulus tab.

Location

Vary a visual stimulus position on the screen

Participant Groups
The participant groups feature in SuperLab lets you perform two main things:

Specify which blocks are presented and the order that they are presented in. This gives
you a way of specifying counterbalancing sequences.

Control when trials within a block, events within a trial, or trial variable levels are
randomized. Participant groups can be used when you need to divide your population of participants into groups based on certain factors such as age or gender; or to test a dependent variable. Because of the way SuperLab works, at least one group is always required. Due to this, SuperLab automatically creates a new group when you create a new experiment. You can edit this group and rename it. But you cannot delete the last group in an experiment.

Editing Groups
In this example, we will create two participant groups, one for left handed participants and another for right handed ones. Prior to proceeding with this example, create the following four blocks: Instructions, All Participants, Left Handed Participants, and Right Handed Participants.

3 - More In-Depth Features

PAGE 24

In SuperLab:

Click on the Experiment menu and select Participant Groups. A dialog appears showing
the list of existing groups. If you havent created or edited groups yet, the dialog will show only one group named Main Group.

Click on the Edit button to edit Main Group. The Group Editor appears. In the Group Name edit eld, change Main Group to Left Handed Group. In the Blocks to Run tab, click on the Run only the following blocks radio button. The
list below it becomes enabled and all four existing blocks have their blue checkboxes turned on.

Click on the checkbox of the block named Right Handed Participants to turn it off. Notice how the Group Editor has an up arrow and a down arrow. These can be used to
re-arrange the order in which blocks are presented.

Click on the OK button to close the Group Editor dialog.

3 - More In-Depth Features

PAGE 25

Now lets create the second group:

Click on the New button. The Group Editor appears In the Group Name edit eld, type Right Handed Group In the Blocks to Run tab, click on the Run only the following blocks radio button Click on the checkbox of the block named Left Handed Participants to turn it off Click on the OK button to close the Group Editor dialog Using Participant Groups
When running an experiment, you can specify which group to use when running on the y:

When randomizing, you can specify that randomization be done on a per group basis in the following places in the program:

In the Randomize tab of the Block Editor In the Randomize tab of the Trial Editor In the Level Selection tab of a Trial Variable Editor

3 - More In-Depth Features

PAGE 26

C HAPTER 4

Conditional Branching

Conditional branching is not a single feature in SuperLab. Rather, it refers to a set of features in SuperLab that allow you to develop more sophisticated experiments. Here are some of the things that you can accomplish with conditional branching:

Pause the experiment after a certain number of trials have been presented in order to give
the participant a break

Present trials in a block until the moving average of the participants reaction time drops
under a certain threshold, e.g. in memory experiments

Prevent the experiment from proceeding to the next block until all the trials in the current
block have been answered correctly

Alter the order in which trials are presented based on the participants response Features Needed
The SuperLab features that you need in order to use conditional branching are:

Codes and code values Feedbacks in the Event Editor so that you can alter a trials code value at run time Expressions in the Block Editors When to Run tab Expressions in the Trial Editors When to Present tab Macros
Why Use Code Values It is possible to branch or perform certain actions based on time elapsed, or based on the number or percentage of trials presented. But you will often need to branch based on the participants action, e.g. whether a response was incorrect or the reaction time was too slow. This is where using codes and code values becomes necessary: doing so lets you mark or tag trials so that they are properly counted later on.

Introduction to Expressions
Expressions in SuperLab provide a key building block. An expression consists of one or more Boolean comparisons where a left hand side (LHS) is compared to a right hand side (RHS). The fastest way to make sense of expressions is via this example: we have a block that contains 100 randomized trials but we want to present the participant with only 40 of them. To accomplish this:

Double-click on the relevant block to edit it. The Block Editor will appear. If not already selected, click on the When to Run tab. Note the Present the trials in this block popup menu. It is set to Always by default. We
will come back to this menu in a little bit.

Click on the New + icon. The Expression Editor will appear.

4 - Conditional Branching

PAGE 28

The Expression Editor is divided into three main sections: LHS (left hand side), an operator (e.g. Is Greater Than), and RHS (right hand side). The result of a comparison is always true (1) or false (0). At present, the LHS consists of a few SuperLab-provided variables and the RHS is always a constant value.

In the LHS section, click on Trials Presented in This Block. A list of functions appears
automatically with one called Number of selected by default. This is the function that we need for this example. See important note below about this LHS variable.

Click on the operator popup menu and change to Is Less Than. In the RHS, set the value to 40. Click on OK to close the Expression Editor.
The new Boolean comparison that you created will now be showing in the Block Editor. One additional step is needed to complete our task: click on the Present the trials in this block popup menu and change it to While the condition expression is true. This is it. With a few mouse clicks, you have modied the block so that it will present 40 trials out of a set of 100. Note About Trials Presented in This Block The Trials Presented in This Block variable can be used with three functions: Number of, Percentage of, and Average Reaction Time of. These functions operate only on the trials presented within the current block. In fact, each block that you create in SuperLab is born with its own Trials Presented in This Block variable. The range of trials that the functions should operate on can be further ltered by specifying a subset (more on later in this chapter). The number of trials presented in a block is cumulative, e.g. if you present 10 trials in Block A, exit Block A to present trials in another block, and then go back to Block A, the number of trials presented in Block A is not reset -- it will still be 10. But SuperLab provides an action that can be invoked by a macro to reset this count to zero.

More On Expressions
Expressions can be more sophisticated than the example just presented. You can have multiple comparisons and combine them with logical and and or. Lets continue with the earlier example. In addition to presenting 40 trials out of a set of 100, we also want to limit the blocks duration to 40 seconds. To proceed:

If you are not in the Block Editor already, double-click on the relevant block to edit it. In the When to Run tab, click on the New + icon. The Expression Editor will appear. The LHS default is Time Since Start of Block. Its the one we want. Change the operator to Is Less Than Or Equal To.

4 - Conditional Branching

PAGE 29

In the RHS, type 40000. Click on OK to close the Expression Editor.


This is how the Block Editor will now look:

Notice how the two comparisons that were created are listed. The expression will be true only if both comparisons are satised. Now for the fun part: lets say you want the expression to be true if either of the comparisons are true. You can accomplish this by clicking on the blue and operator. Doing so changes it to or:

There are no limits on how many comparisons an expression can have or on how you combine them.

Looking at a Subset of Trials


It is typical for many types of experiments to rst have a Learning block before the real trials are presented. Trials within the learning block are presented until the reaction time

4 - Conditional Branching

PAGE 30

drops below a certain threshold, indicating that the participant has learned how to process the trials well enough. In the Expression Editor, SuperLab provides a way of looking at the average reaction time of trials presented in a block. Better yet: rather than look at all the trials in the block, SuperLab provides a way of looking at the RT of the last few trials only and hence obtain a moving average. Lets take an example:

Create a block and name it Learning. The Block Editor will appear. In the When to Run tab, click on the New + icon. The Expression Editor will appear. In the LHS (left hand side) section, click on Trials Presented in This Block. In the list of functions, click on Average Reaction Time of.
So far, the LHS of the expression will provide the average RT of all the trials in the block. To obtain a moving average by looking at, say, the last 5 trials only:

4 - Conditional Branching

PAGE 31

Click on the Subset button. The Subset of Trials dialog appears. Click on the radio button labeled Only the last. In the edit eld, type 5. Take a look at the bottom of the dialog. It is possible to limit the trials being looked at by
the code value that is attached to it. We will come back to this feature in a little bit.

Click on OK to close the Subset of Trials dialog.


If only 3 trials have been presented but the subset species 5 trials, SuperLab will use whats available: the average reaction time of those 3 trials.

Introduction to Macros
Once you know what an expression is, macros becomes easy to understand. A SuperLab macro consists of:

An expression that is identical to what has been described thus far in this chapter A list of actions to perform if the expression evaluates to true Another list of actions to perform if the expression evaluates to false
Some of the available actions are: presenting an event, going to a particular block, exiting the experiment, repeating a trial, and so forth. Advantages of Macros Using an expression in the When to Run tab of the Block Editor makes it simpler to implement loops that cause a block to keep repeating until a certain condition is met. But macros have their own advantages. One, a macro is an independent object in SuperLab with its own name. Once created, it can be used in more than one block. And two, macros allow you to do more. See an example of using macros in Chapter 6 in the Giving Participants a Break After N Trials section. Relative versus Absolute Jumps SuperLab provides four options that can cause a jump, meaning that the block will abandon the next trial that it was about to present to go somewhere else. One option is implemented as its own Repeat Current Block action. The remaining three options are part of the Go to Another Block action:

4 - Conditional Branching

PAGE 32

Note the option to Present block [Block 3]. It is called an absolute jump because it species which block to go to. Whether SuperLab happens to be presenting Block 1 or Block 10, the Present block option will always force SuperLab to go to Block 3. The remaining three options are called relative jumps because which block SuperLab goes to depends entirely on which block it is currently presenting. If possible, using relative jumps is preferred over an absolute jump because it allows a macro to be reused in different blocks. An example is a staircase procedure where you create a single reusable macro that goes to the next block if the expression is true or goes back to the previous one if the expression is false.

Using Code Values for More Capable Expressions


In the examples weve seen so far, we created expressions that used the time since the start of a block or the participants reaction time. But how can we create an expression that is based on the participants response? First, we create a code and give it some code values. Then, when the participant responds, we use the feedback mechanism in the Event Editor to mark or tag the trial. And nally, we can create an expression that looks only at those tagged trials. As usual, it is best to use an example. This time, we will start with a pre-existing experiment that we can build on. It can be downloaded by visiting the Cedrus Community Forums (http://community.cedrus.com), then click on Experiments (under SuperLab), then click on Demos. Alternatively, you can type the following URL directly: ! http://community.cedrus.com/showthread.php?t=553

This experiment contains two stimulus lists, one for words and the other for non-words. The participant is supposed to press the y key if its a word, or the n key otherwise. Each list contains 5 items resulting in 10 trials being presented. Note the presence of an event called Exiting Block Message. This event is not linked to any trials. Well use it for informational purposes only.

4 - Conditional Branching

PAGE 33

The Goal We want to modify the experiment so that if the participant responds correctly to three consecutive trials, SuperLab will stop presenting the rest of the trials. Creating the Code and Code Values We start by clicking on the Experiment menu and selecting Codes. The Codes dialog appears.

Click on the New button found on the left. The Code Editor appears. You can enter anything as the code name. Type Answered Correctly. Make sure that the codes values will be attached to A trial. Click on OK to close the Code Editor.
The next step is to give this code two values:

Click on the New button found on the right. A small dialog appears. Type Yes. Click on OK to close the small dialog. Click on the New button found on the right again. The small dialog appears. Type No. Click on OK to close the small dialog.
Click on Done to close the Codes dialog.

4 - Conditional Branching

PAGE 34

Marking The Trials The next step is to mark a trial as having been answered correctly. Start by double-clicking on the event Word Stim to edit it. The Event Editor appears. Click on the Feedback tab. The Feedback tab has two lists. The one on the left is where we need to create the condition(s) that we want to provide feedback for. The list on the right is where we dene the actions that we want to see happen for each condition.

Click on the New + icon that is on the left side. The New Condition dialog appears. In the popup menu, select The response is correct then click on OK.

In the list of actions on the right, note how SuperLab automatically created an action that would cause any remaining events in the trial to be skipped. This action makes no difference for the purpose of our experiment, so well leave it alone.

Click on the New + icon that is on the right side. The New Action dialog appears.
If necessary, adjust the settings so that they look like this:

Click on OK to close the New Action dialog. Click on OK again to close the Event Editor.
4 - Conditional Branching PAGE 35

One last thing: repeat the same steps for the other event, Non-word Stim. Done! The next (and last) step is to take advantage of the work done so far. Creating an Expression Using a Subset of Trials (Method One) Were now ready to implement what we set out to do: creating a macro that prevents the remaining trials in the block from being presented if the participant responds correctly to three consecutive trials. To start, double-click on Block 1 to edit it; the Block Editor will appear.

In the When to Run tab, click on the New + icon. The Expression Editor dialog
appears.

In the LHS section, click on Trials Presented in This Block. A list of functions appears
automatically with one called Number of selected by default. This is the function that we need for this example.

Click on the Subset button. The Subset of Trials dialog appears. Click on the radio button labeled Only the last. In the edit eld, type 3. Click on the checkbox labeled Limit the trials used to those marked with a specic code
value to turn it on.

Make sure that the code value Yes is selected.

4 - Conditional Branching

PAGE 36

Click on OK to close the Subset of Trials dialog. Back in the Expression Editor:

Set the operator to Is Equal To. In the edit eld for the number of trials, enter 3. Click on OK to close the Expression Editor.
Now you should be back in the Block Editor. There are two more settings that we need to change:

Change the Present the trials in this block popup menu to Only if the condition
expression is false.

Change the Evaluate expression menu to Before each trial is presented.


The Block Editor should look like this next screen snapshot. Click on OK to close it. When you run the experiment, it should end after you have responded correctly to three trials in a row.

Creating an Expression Using a Subset of Trials (Method Two) Another approach is to use a macro instead of the When to Run tab of the Block Editor. This offers at least two advantages: you can have SuperLab perform more than one action when the expression is true, and you can reuse the macro in another block if needed. Here, we will show participants a message to let them know that they did well.

4 - Conditional Branching

PAGE 37

To start, double-click on Block 1 to edit it; the Block Editor will appear.

In the Macros tab, click on the New + icon. The Macro Editor dialog appears. Give the macro a name, e.g. Exit block after 3 correct responses in a row. In the Expression tab, click on the New + icon. The Expression Editor appears. Follow the same exact steps described earlier to create the expression. When done, click
on OK to close the Expression Editor and go back to the Macro Editor. Now we need to create the actions.

Click on the Macro Editors If True tab. Click on the New + icon. The New Action dialog appears. From the Do the Following menu, select Present Event. Note how the main contents
of the New Action dialog changes depending on the chosen action.

From the Select Event menu, select Exiting Block Message. Click on OK.
You should now have one action in the list.

In the Macro Editors If True tab, click on the New + icon again. The New Action
dialog appears.

From the Do the Following menu, select Go to Another Block. Make sure that the radio button Continue with the next block is selected. Click on OK.
Back in the Macro Editor, the dialog should look like this:

4 - Conditional Branching

PAGE 38

Important: the two actions that you created may not be in the same order as shown in this screen snapshot. If so, click on the Up or Down arrow to rearrange the order. Click on OK to go back to the Block Editor. One last step is needed: the macro has been created but will not be used by the block unless it is selected. Click on the blue check mark to select it. Finally, click on OK to close the Block Editor, save the experiment, then run it to see the results.

Trial-Level Expressions
The nal conditional branching feature that we will look at is found in the When to Present tab of the Trial Editor. This feature is found only in SuperLab version 4.0.7 or later. The conditional branching features we looked at so far have mostly dealt with trials within a block, e.g. the reaction time over a range of trials or a count of trials. Things are simpler at the trial level: the only concern is whether to present the current trial or not. Trial-level expressions are most powerful when combined with a macro at the block level. An example is when you want to present all the trials in the block until the participant has responded correctly to all of them. You can do this by:

Using the feedback feature and code values to mark a trial as having been answered
correctly

4 - Conditional Branching

PAGE 39

Using a macro to cause a block to be repeated Using a trial-level expression to check the code value of the trial thats about to be
presented Loop Guard: The Silent Feature Note that if you use a macro to cause a block to be repeated, you do not have to worry about your experiment causing an endless loop and seeming to hang. SuperLab implements a loop guard feature that works automatically in the background: if a block is repeated twice but without any trials actually being presented to the participant, SuperLab will automatically force that block to exit and continues with the next block.

4 - Conditional Branching

PAGE 40

C HAPTER 5

Ti m i n g

This chapter discusses the crucial issue of timing, how SuperLab does it, what you can do to help improve it, and other details related to using a particular input device or timer. Other topics that affect the accuracy of timing are also discussed here. At Cedrus, we designed SuperLab to be friendly yet powerful, and we are trying to keep its manual as clear as possible by eliminating the unnecessary computerese whenever possible. Unfortunately, when discussing timing, we have to introduce some computer concepts as well as some Mac- or Windows-specic information.

Resetting The Timer


It is impossible to make sense of the collected reaction time (RT) data unless you know when the timer is started. It is important to understand when the RT timer is reset by default and how you can override that. In SuperLab 4.0, there is an option to reset the RT timer in the Event Editors Input tab. When creating a new event, this option is turned off by default. If all the events in a trial have this option turned off, SuperLab will reset the RT timer at the onset of the rst event in the trial. Please read if you are upgrading: In versions prior to 4.0, SuperLab reset the timer at the onset of each event. It is important that you be aware of the difference. The onset of an event depends on the event type and its settings:

For events of type Digital Output, ISI, Picture File, Serial Output, or Text, the onset is
after the event presentation is complete, i.e. after the picture is drawn or the signal is sent.

For events of type Audio File, Movie, or Sound File, the onset depends on the settings. If
the Wait for audio to nish playing or the Wait for movie to nish playing option is turned on, the onset of the event is when the sound or movie has nished playing. Otherwise, the onset is considered to be when the sound or movie has started playing.

Reaction Time Accuracy


The accuracy of the measured reaction times depends on a number of factors, but mainly on the input device used in the experiment and the operating system. Before going to the list of supported input devices one by one, there are two important factors to be aware of.

First, all references to serial port mean just that: a real, physical serial port with a 9-pin or 25pin connector. If you are using a USB-to-serial port converter, then the USB driver is likely introducing a delay of at least 5 milliseconds, and probably a lot more. Second, all Cedrus input devices in production since 2006 are XID devices. XID stands for eXperiment Interface Devices. XID is the rmware (software running inside the hardware device) that provides some much needed functionality such as time stamping of the participants responses. The aim is to try to get around some of the operating systems timing issues. A current limitation is that SuperLab cannot time responses from the participant while a visual stimulus is being drawn. The response will still be recorded, but the recorded RT might be off. Keyboard USB keyboards are regularly sampled, providing information at intervals determined by the keyboard itself (e.g. an Apple keyboard samples with a precision of 8 milliseconds, while a Kensington keyboard samples at 16 milliseconds). This is independent of the operating system. The operating system choice does affect accuracy, with SuperLab on Leopard providing a standard deviation of 0.0334 milliseconds, whereas SuperLab on Vista has a standard deviation of 0.333 milliseconds. More information is available on this forum post: ! http://community.cedrus.com/showthread.php?t=513

The accuracy from PS/2 keyboards ranges from 16 to 35 milliseconds. Lumina fMRI Response Pads (XID) The LSC-400 controller supplied with Lumina response pads provides a one millisecond timing resolution when used with the serial port. RB-x30 Series Response Pads (XID) The RB-530, RB-730, RB-830, and RB-834 response pads are internally accurate to about 500 microseconds, meaning that it takes about half of a millisecond for the information about a key press to leave the response pad. However, they are USB devices and require a USB driver that introduces a delay of about 5 milliseconds. Older RB Series Response Pads Models prior to the RB-x30 all used a serial port. They are all internally accurate to about 500 microseconds, meaning that it takes about half of a millisecond for the information about a key press to leave the response pad. Being based on a serial port, the speed with which that information reaches the computer depends on the baud rate, ranging from about 4 milliseconds when the baud rate is 2,400 to about half of a millisecond for a 19,200 baud rate. However, the transmission time is very constant and thus easy to account for.

5 - Timing

PAGE 42

Microphone Input (Mac only) SuperLab for Macintosh supports the built-in microphone jack. All you need is a compatible microphone. In internal tests, we found the timing resolution to be about 20 milliseconds. Microsoft Serial Mouse SuperLab provides direct support for the Microsoft Serial Mouse (or compatible mice). This needs to be a second mouse in addition to the one used for pulling down menus or clicking on buttons. SuperLab provides a 1 millisecond resolution for this mouse by directly opening the serial port and listening to the Serial Mouse. In doing so, it bypasses the standard operating system driver which polls the mouse once every 10 to 18 milliseconds only, depending on the mouse driver and the operating system. The disadvantage is that SuperLab supports only the keys of the Microsoft Serial Mouse, i.e. it is not possible to determine where on the screen the mouse pointer is. In fact, if its a mechanical mouse, we recommend that the tracking ball be removed as this greatly reduces the amount of unused information being sent to the PC. Mouse or Touch Screen The accuracy of the standard mouse that you use can vary greatly between 10 and 30 milliseconds, depending on the mouse itself and the driver. When installed, nearly all touch screens emulate a mouse and can therefore be used with SuperLab. The main experiment design issue to be aware of is that the participant will not see a moving cursor as can be seen with a mouse. SV-1 Voice Key (XID) The SV-1 voice key device provides a one millisecond timing resolution when used with the serial port. PST Serial Response Box The PST Serial Response Box provides a one millisecond timing resolution when used with the serial port.

Presentation Accuracy
Another important timing consideration is the accuracy of presentation. Computer monitors come with some built-in limitations. It is important to understand these limitations and how they affect stimulus presentation.

5 - Timing

PAGE 43

How CRT Monitors Work A cathode ray tube (CRT) monitor contains a gun shooting electrons at the screen. The inside of the screen has a phosphorous coating. When hit by the electrons, the phosphorous coating produces light, and a pixel is seen turned on. By turning enough pixels on, a picture and/or text forms on the screen. This approach to displaying information provides great exibility, but there is one catch: if the gun inside the monitor stops shooting electrons at the pixel location on the screen, the phosphorous coating will no longer emit light and the pixel will fade away. To get around this, a computer instructs the electron gun to shoot at given locations on the screen several times a second, typically 60 to 120 times per second. This operation is known as refreshing, or video refresh. The higher the refresh rate, the less icker a monitor has. How LCD Monitors Work Unlike CRT monitors, liquid crystal display (LCD) monitors do not need to be refreshed. This is true for LCD monitors known as active matrix, or TFT, which is all thats been produced in the past few years. It is not true of passive matrix displays. Even though LCD monitors do not need to be refreshed in order to keep the stimulus displayed, software remains unable to change the state of a pixel at any time, let alone an entire image. Changes are allowed at regular intervals, typically 60 times a second. Implications Regardless of the monitor type, there is a refresh rate involved, expressed in Hertz and abbreviated as Hz. A monitor thats refreshed or updated 60 times per second is said to have a refresh rate of 60 Hz, or 60 cycles per second. The most important implication is that a stimulus cannot be displayed for less than one refresh cycle. This is a hardware limitation regardless of which operating system or computer you are using. To determine the minimum exposure time of a stimulus in milliseconds, divide 1000 by the refresh rate. For example, if your monitor and video cards refresh rate is 60 Hz, then each cycle (and hence the minimum exposure time) is 1000 divided by 60, or 16.666 milliseconds. Another implication is that the exposure time of a stimulus must be in multiples of the refresh cycle. Using a 60 Hz monitor as an example, the exposure time can only be 16.6 milliseconds, 33.3 milliseconds, 49.9 milliseconds, and so forth. This does not matter much when presenting a stimulus for a few hundred milliseconds or more, but is important with very brief exposure times.

5 - Timing

PAGE 44

SuperLab 4.0 and later synchronizes the presentation of stimuli with the precise start of a refresh cycle on both Windows and Mac OS X 10.4 or later; it does not synchronize on MacOS10.3.9. Specifying an exposure time for a stimulus is typically done by entering a time limit in the Input tab of the Event Editor. For brief exposure times, we recommend that the time limit specied in milliseconds be ! ! ! ( N minus 1 ) times duration of cycle plus 1 millisecond

where N is the desired exposure time expressed in cycles. The logic behind this is to take advantage of SuperLabs precise syncing with the start of each cycle by specifying one less cycle that you want, add a tiny bit to it, and then let SuperLab round up to the next refresh cycle. For example, if you want the exposure time to be three cycles on a 60 Hz monitor, specify the time limit as ! ! ! ( 3 - 1 )! ! ! ! 2 times 16.66 milliseconds!! plus 1 millisecond! ! ! ! ! equal to 2 yields 33.33 milliseconds yields 34.33 milliseconds

You can enter the time limit as 34 milliseconds in the Event Editor. Because a stimulus can only be presented in multiples of refresh cycles, SuperLab will wait for specied 34 milliseconds, and then wait some more for the start of the next cycle, hence giving you the three cycles that you wanted.

5 - Timing

PAGE 45

C HAPTER 6

How Do I...

This chapter describes how to accomplish some common tasks in SuperLab 4.0. More How Do I topics will be added over time in the help les that ship with the software. Additional how to help and sample experiments can be found on the SuperLab experiments forum at ! http:/ /community.cedrus.com/forumdisplay.php?f=9

Presenting Two or More Visual Stimuli at the Same Time


It is possible to present as many visual stimuli simultaneously as you like. This applies to events of type Picture File and Text. Lets use an example where four pictures need to be presented at the same time. The basic idea is that the rst three pictures must be kept invisible and the fourth one made visible, causing all four to appear at the same time. Here are step by step details. For the rst picture le, in the Event Editor:

Click on the Input tab and click on the Immediately after the event is presented
checkbox to turn it on.

Click on the Stimulus tab and then click on the Settings button. A smaller dialog
(screen) appears. Note: the Settings button will be grayed out if you have not selected a picture le yet.

In the Picture Settings dialog, click on the Presentation Options tab. Click on the Keep stimulus invisible checkbox to turn it on. Click on OK to close the Picture Settings dialog.

For the second and third events, the settings need to be identical to the rst event with one change: in the Picture Settings dialog, the Erase screen before presenting stimulus checkbox must be turned off. For the fourth and last event, in the Picture Settings dialog the Keep stimulus invisible checkbox must be turned off. All four events must be linked to the same trial for the simultaneous presentation to work. The same instructions apply if you are presenting words.

Providing Feedback to Participants


In SuperLab, anything that can be presented as a stimulus may also be used as a feedback. This includes showing an image or text message, playing a sound or movie, or sending a marker to an external device. Providing feedback is a two-step process: rst you create an event that will show the message or play the sound, then you use it. In this example, we create an event called Incorrect Feedback to present a text message:

Once created, edit the event that collects the participants input and click on the Feedback tab. This tab has two main lists: one on the left for the condition that you want to provide feedback for, and the other for the actions that you want done given that condition.

6 - How Do I...

PAGE 48

Start by clicking on the + icon on the left side of the dialog. A second, smaller dialog opens showing a list of the supported conditions:

For this example, we keep the default The response is not correct. Click on OK.
You can now specify the actions that you want done when the participants response is not correct. By default, SuperLab creates an action that repeats the current trial, but you can easily double-click on it to edit it. To show the Incorrect Feedback:

Start by clicking on the + icon on the right side of the dialog. A dialog opens showing a
list of the available actions.

Click on the popup menu and select Present Event. The bottom portion of the dialog
automatically changes to show a menu containing all the available events.

Select the event Incorrect Feedback that we created earlier.

6 - How Do I...

PAGE 49

Click on OK to close the New Action dialog. Click on OK again to close the Event Editor.
When you now run the experiment, the participant will see a message whenever the incorrect response is pressed.

Presenting a Sound and Visual Stimulus at the Same Time


Displaying a visual stimulus while a sound is playing is fairly simple in SuperLab. You will need two events, the rst one of type Sound File and the second one of any other type, e.g. Text or Picture File. The order is important: the sound event must come rst.

For the sound event, in the Event Editor:

Click on the Input tab and click on the Immediately after the event is presented
checkbox to turn it on

Click on the Stimulus tab and then click on the Wait for sound to nish playing
checkbox to turn it off Thats it.

Giving Participants a Break After N Trials


In long experiments, it is often desirable to give the participant a break after a number of trials has been presented. Separating trials into two different blocks works but only when not randomizing. If you need to randomize the trials as a single set, they must remain in one block. Providing the break can simply be done using the conditional branching feature. Here is a step by step example; we will start by creating the event that displays a message to the

6 - How Do I...

PAGE 50

participant. The example assumes that you already have an experiment setup and you only need to add the break.

In the main experiment window, click on the list of events; the selection rectangle will
move from the list of trials to the list of events

In the toolbar, click on the New Event icon; the Event Editor will appear For the event name, type Participant Break Click on the Stimulus tab to select it if its not selected already Click on the Event Type popup menu and select Text In the large text editing eld, type:
You can now take a break. When you are ready to continue, press any key to continue.

Click on the OK button. The Event Editor will close and Participant Break will appear
in the list of events. The next step is to create a macro:

In the main experiment window, double-click on your block to edit it Click on the Macros tab Click on the New button icon to create a new macro; the Macro Editor appears Name the macro Give a participant a break In the Expression tab, click on the New (+) button; the Expression Editor appears In the list of variables on the left, click on Trials Presented in This Block; three
"functions" appear automatically on the right side

Click on the Number of function to select it if it's not selected already In the RHS (right hand side) section at the bottom of the dialog, enter the number of trials
after which you want to introduce a pause, e.g. type 30 if you want to give the participant a break after 30 trials

6 - How Do I...

PAGE 51

Click on the OK button to close the Expression Editor (but not the Macro Editor). Only a few steps remain:

While still in the Macro Editor, click on the If True tab Click on the New button icon to create a new action; the New Action dialog appears Click on the "Do the Following" popup menu and select "Present Event" Click on the "Select Event" popup menu and select the "Participant Break" event that you
created earlier

Click on OK to close the New Action dialog Click on OK to close the Macro Editor dialog. You should now be back in the Block
Editor. You created a macro but you still need to tell SuperLab to use it:

6 - How Do I...

PAGE 52

Click on the blue checkbox to the left of the macro to select it Click on the OK button to close the Block Editor
That's it; the participant should now get the "Participant Break" event presented after 30 trials. You can reuse a macro in any number of blocks.

Randomize a Stimulus List


Stimulus list items in SuperLab can be randomized, with some restrictions. It is helpful to understand how SuperLab accesses stimulus list items in order to understand how randomization is accomplished. In the following example experiment, there is only one block, one trial, and two events:

The trial consists of a xation point followed by a stimulus. However, the second event the word uses a stimulus list that contains 10 items. When running the experiment, SuperLab creates internally 10 trials where each trial uses a single item from the stimulus list. You can visually see this by clicking on the Experiment menu and choosing Expand Events That Use Lists. This feature will create a new expanded experiment:

6 - How Do I...

PAGE 53

As you can see thus far, SuperLab does not pull an item from the stimulus list at run-time. Instead, it converts trials and events that use stimulus lists into many individual trials and events. Therefore, randomizing a stimulus list is no different than randomizing trials in a block. This is done when you edit a block, in the Block Editors Randomize tab. A word of caution: Any changes that you make in an expanded experiment will not apply to the original (source) experiment. We therefore recommend that you use an expanded experiment only for viewing how SuperLab will present trials at run-time. Multiple Lists When using more than one stimulus list within the same trial, the rst item from the rst list is always paired with the rst item from the second list. If you have list A that has 5 items and list B that has only 3 items, a total of ve trials will be presented using the following pairing:

6 - How Do I...

PAGE 54

A1 - B1 A2 - B2 A3 - B3 A4 - B1 A5 - B2 Note how, when SuperLab used all three items from list B, it went back to the top of list B and reused B1 and B2.

Moving Experiments Between Macs and PCs


The Mac OS and Windows versions of SuperLab 4.0 are generally the same. Experiments created with SuperLab 4.0 or later versions can be moved freely between computers using Mac OS and Windows, but there are two caveats that one needs to be aware of:

SuperLab for Windows does not support the PICT le format commonly used on the
Mac. If you are planning on using an experiment on both platforms, consider saving the pictures using JPEG, GIF, PNG, or BMP le formats.

MPG is the only movie le format that is supported on both platforms.


For events that use text (e.g. Text, Self-Paced Reading), SuperLab stores font information independently for each operating system.

6 - How Do I...

PAGE 55

C HAPTER 7

The Cedrus Data Viewer

SuperLab saves the collected data in plain text les, with elds/columns separated by tabs. These les can be opened using Excel, Numbers, SPSS, and any program that can open a text le. SuperLab creates one le for each experiment run. Cedrus provides a free Data Viewer program that offers several handy features. You can:

View one or several les simultaneously View summary data such as counts, percentages, and average RTs Flag records manually or have Data Viewer ag records that meet certain criteria Show just the agged records for a better understanding of the data Combine any number of data les into a single, merged le Transpose records in the merged le for easier handling in SPSS

Data Viewer works on Mac OS 10.3.9 or later and Windows 2000, XP, or Vista. The latest version can always be downloaded from ! http://www.superlab.com/dataviewer

Adding and Viewing Files


When you run Cedrus Data Viewer, you are greeted with a window that doesnt contain any les yet. You can add les by clicking on the icon on the bottom left corner. Data Viewer will prompt you to locate one or more les. Note that you can open as many les as you want to at the same time. After adding SuperLab data les, your screen may look like this one:

There are several things that you can do: ! Click on Raw Data or Summary Stats. ! Select which combination of les to look at. The Raw Data and Summary Stats views are updated quickly and automatically every time you change the selected les. ! The bottom of the window shows summary information. ! In the Summary Stats view, you can see counts, percentages, or average RTs for each block, trial, code value, or trial variable. ! Click on the Merge Selected Files button to have the Cedrus Data Viewer produce a single, merged data le.

7 - The Cedrus Data Viewer

PAGE 58

Examining and Filtering the Data


Cedrus Data Viewer 2.0 or later provides useful tools for examining the collected data, agging (marking) certain records manually or using one of the built-in methods, hiding records, hiding columns, and so forth. These tools do not affect the opened data les. They affect only the set of records that you are viewing, and if you so choose, the set of records that will be exported and merge. You can ag a record by simply clicking on it in the Flag column. Another click removes the ag. But far more useful is Data Viewers built-in methods for agging records based on one of the following criteria:

The correctness of the participants response The reaction times range or limits The reaction times standard deviation range or limits
To access these methods, click on the Rows menu and select Flag Special. The following dialog will appear:

An example of how this feature can be useful is searching for reaction times that are too large to make sense and then excluding them from the merged data les. Note that the way this feature was implemented allows for drilling down on data. In other words, you can ag an initial set of records based on one criterion and afterwards apply a second criterion but only to the already agged records.

Other Features
You can use row coloring with Cedrus Data Viewer to make it easier to spot agged records or incorrect responses. Row coloring is turned off by default but can be easily enabled by clicking on the Rows menu and selecting Row Colors.

7 - The Cedrus Data Viewer

PAGE 59

You can also move columns or hide them to suit your viewing preferences. You do so by right-clicking (or control-click on the Mac) on the column header; a menu will appear offering you these options.

7 - The Cedrus Data Viewer

PAGE 60

A PPENDIX A

Supported Hardware

Participant Input RB-400 / RB-600 RB-410 / RB-610 RB-420 / RB-520 / RB-620 / RB-820 RB-530 / RB-730 / RB-830 / RB-834 SV-1 Voice Key Lumina fMRI Response Pads Measurement Computing I/O devices (Windows only) Microphone Microsoft serial mouse PST Serial Response Box Standard Keyboard Standard Mouse Digital Output
SuperLab 4.0 supports the following Measurement Computing I/O cards for Windows only:

PC-CARD-DIO48 (for Laptop) PCI-DIO24


The following National Instruments I/O devices are supported on Mac and Windows:

DaqCard-DIO-24 PCI-DIO-96 USB-6008 USB-6009 USB-6501

A PPENDIX B

License Agreement

Nearly every software license sold is governed by a license agreement, sometimes called an End User License Agreement (EULA). Here is the EULA that governs the use of SuperLab, rst in plain English.

In Plain English
It's really simple: 1. 2. 3. If you have one license, no two persons may use SuperLab at any one time. If you have two licenses, no three persons can use SuperLab at the same time. Whenever in doubt, refer to item 1 above. SuperLab may not be installed on a virtual machine such as Parallels or VMware.

Our EULA offers more exibility than other vendors', many of whom insist that the software may be installed on one machine only.

Some Frequently Asked Questions


Can I develop an experiment on one computer and run it on others? No. You can run SuperLab to develop and collect data on only a single computer. If you want to collect data (but not edit or modify an experiment) on other computers, Cedrus offers inexpensive "run only" licenses. Can I install SuperLab on the computer in the ofce and the one at home? Yes, provided there is no chance that another person may use it while you're using SuperLab at home.

The Legal Text


END-USER LICENSE AGREEMENT FOR CEDRUS SOFTWARE IMPORTANT This End-User License Agreement (EULA) is a legal agreement between you (either an individual or a single entity) and the PUBLISHER from whom you acquired the software product SuperLab identied above (SOFTWARE). The SOFTWARE includes computer software, the associated media, any printed materials, and any online or electronic documentation. By installing, copying or otherwise using the SOFTWARE, you agree to be bound by the terms of this EULA. If you do not agree to the terms of this EULA, Cedrus Corporation (Cedrus) is unwilling to license the SOFTWARE to you. In such event,

you may not use or copy the SOFTWARE, and you should promptly contact Cedrus for instructions on return of the unused product(s) for a refund. Software LICENSE The SOFTWARE is protected by copyright laws and international copyright treaties, as well as other intellectual property laws and treaties. The SOFTWARE is licensed, not sold. 1. GRANT OF LICENSE. This EULA grants you the following rights:

Software. You may install and use one copy of the SOFTWARE on a computer. Storage/Network Use. You may also store or install a copy of the computer software
portion of the SOFTWARE on the computer to allow your other computers to use the SOFTWARE over an internal network, and distribute the SOFTWARE to your other computers over an internal network. However, you must acquire and dedicate a license for the SOFTWARE for each computer on which the SOFTWARE is used or to which it is distributed. A license for the SOFTWARE may not be shared or used concurrently on different computers.

Back-up Copy. You may make a single back-up copy of the SOFTWARE. You may use
the back-up copy solely for archival purposes. 2. DESCRIPTION OF OTHER RIGHTS AND LIMITATIONS.

Limitations on Reverse Engineering, Decompilation and Disassembly. You may not


reverse engineer, decompile, or disassemble the SOFTWARE, except and only to the extent that such activity is expressly permitted by applicable law notwithstanding this limitation.

Separation of Components. The SOFTWARE is licensed as a single product. Its


component parts may not be separated for use on more than one computer.

Single COMPUTER. The SOFTWARE is licensed as a single integrated product. The


SOFTWARE may only be used with a single computer.

Rental. You may not rent or lease the SOFTWARE. Software Transfer. You may permanently transfer all of your rights under this EULA
provided you retain no copies, you transfer all of the SOFTWARE (including all component parts, the media and printed materials, any upgrades, and this EULA), and the recipient agrees to the terms of this EULA. If the SOFTWARE is an upgrade, any transfer must include all prior versions of the SOFTWARE.

Termination. Without prejudice to any other rights, Cedrus may terminate this EULA if
you fail to comply with the terms and conditions of this EULA. In such event, you must destroy all copies of the SOFTWARE and all of its component parts.

Appendix B - License Agreement

PAGE 64

Virtual Machines. SOFTWARE may not be installed and used on a virtual computer or
any form of "guest" operating system or virtual "appliance". 3. UPGRADES. If the SOFTWARE is an upgrade from another product, whether from Cedrus or another supplier, you may use or transfer the SOFTWARE only in conjunction with that upgraded product, unless you destroy the upgraded product. If the SOFTWARE is an upgrade of a Cedrus product, you now may use that upgraded product only in accordance with this EULA. If the SOFTWARE is an upgrade of a component of a package of software programs which you licensed as a single product, the SOFTWARE may be used and transferred only as part of that single product package and may not be separated for use on more than one computer. 4. COPYRIGHT. All title and copyrights in and to the SOFTWARE (including but not limited to any images, photographs, animations, video, audio, music, text and applets, incorporated into the SOFTWARE), the accompanying printed materials, and any copies of the SOFTWARE, are owned by Cedrus or its suppliers. You may not copy the printed materials accompanying the SOFTWARE. All rights not specically granted under this EULA are reserved by Cedrus. 5. DUAL-MEDIA SOFTWARE. You may receive the SOFTWARE in more than one medium. Regardless of the type or size of medium you receive, you may use SOFTWARE on only one computer. You may not use or install the other medium on another computer. You may not loan, rent, lease, or otherwise transfer the other medium to another user, except as part of the permanent transfer (as provided above) of the SOFTWARE. 6. PRODUCT SUPPORT. Product support for the SOFTWARE is provided by Cedrus or its resellers. 7. U.S. GOVERNMENT RESTRICTED RIGHTS. The SOFTWARE and documentation are provided with RESTRICTED RIGHTS. Use, duplication, or disclosure by the Government is subject to restrictions as set forth in subparagraph (c)(1)(ii) of the Rights in Technical Data and Computer Software clause at DFARS 252.227-7013 or subparagraphs (c)(1) and (2) of the Commercial Computer Software Restricted Rights at 48 CFR 52.227-19, as applicable. Manufacturer is Cedrus Corporation/ 1121 S. Meyler Street/San Pedro, CA 90731. 8. LIMITATION OF LIABILITY. Under no circumstances and under no legal theory, tort, contract, or otherwise, shall Cedrus, its suppliers, or resellers be liable to you or any other person for any indirect, special, incidental, or consequential damages of any character including, without limitation, damages for loss of goodwill, work stoppage, computer failure or malfunction, or any and all other commercial damages or losses. In no event will Cedrus be liable for any damages in excess of

Appendix B - License Agreement

PAGE 65

the amount Cedrus received from you for a license to the SOFTWARE, even if Cedrus shall have been informed of the possibility of such damages, or for any claim by any other party. This limitation of liability shall not apply to liability for death or personal injury to the extent applicable law prohibits such limitation. Furthermore, some jurisdictions do not allow the exclusion or limitation of incidental or consequential damages, so this limitation and exclusion may not apply to you. 9. HIGH RISK ACTIVITIES. The SOFTWARE and the output produced by the SOFTWARE is not fault-tolerant and is not designed, manufactured or intended for use or resale as on-line control equipment in hazardous environments requiring fail-safe performance, such as in the operation of nuclear facilities, aircraft navigation or communication systems, air trafc control, direct life support machines, or weapons systems, in which the failure of the SOFTWARE could lead directly to death, personal injury, or severe physical or environmental damage ("High Risk Activities"). Cedrus and its suppliers specically disclaim any express or implied warranty of tness for High Risk Activities. 10. WAIVER OF RESPONSIBILITY FOR MISUSE. The SOFTWARE can process copyright material and has no way of knowing if such processing is in breach of copyright law. You agree that it is your responsibility to ensure that you only process material for which you own or have licensed the copyright and, further, to indemnify and hold harmless Cedrus from any claim whatsoever relating to your improper use of the SOFTWARE to process copyright materials. 11. MISCELLANEOUS. This EULA represents the complete agreement concerning this license. The acceptance of any purchase order placed by you is expressly made conditional on your assent to the terms set forth herein, and not those in your purchase order. If any provision of this EULA is held to be unenforceable, such provision shall be reformed only to the extent necessary to make it enforceable. This EULA shall be governed by American law in the State of California (except for conict of law provisions). The application the United Nations Convention of Contracts for the International Sale of Goods is expressly excluded.

Appendix B - License Agreement

PAGE 66

Index

A Accuracy keyboard 42 presentation 43-45 reaction time 22, 41 ADInstruments iv B Biopac iv BioSemi iv Blocks, creating 6 Brain Products iv C Codes creating 22, 34 using 23, 27, 29 values 20-22, 27, 29, 33-34, 39-40, 58 Color 16, 24 Conditional Branching 27-40, 50 Cues, adding 10 D Data Viewer 57-59 Digital output 19, 41 DirectX 1 E EEG / ERP iv, 19 I Input 2529, 46, 48, 52 Input Devices 3, 8, 41 I/O Card 9, 19 ISI 41 G Go/No-Go experiments 20 Groups (see Participant Groups) F Feedback 9, 20, 21 Fiber optics iii fMRI iii, 19, 42 Forums 2, 33 Event creating 7, 11, 13 marker iv, 6, 19, 21 Expand events 53-54 Experiment le format 55 moving between Macs and PCs 55 Expressions creating 28-29 subset of trials 29-32, 36-37 trial-level 39-40 using code values for 33

Index

PAGE 67

K Keyboard 8, 20 L Linking 8, 16 Lumina fMRI iii, 9, 20, 42 M Macros 32-39, 51 Measurement Computing 9, 19 Microphone Input iv, 8, 42 Monitors 43-44 N National Instruments 19 P Participant Groups editing 24-25 using 26 Participant Input 8, 10, 20 Presentation 41 accuracy 43, 44 two or more visual stimulus simultaneously 47 PST Serial Response Box 9, 43 Pulse duration 19 R Randomization 16-17, 26, 53 Randomize 15-16, 24, 26, 28, 50, 53, 54 events within a trial 17 Reaction time (see Timing) Responses collected 3, 11, 57 dening 9

Response pads iii, 8, 20, 42 Results le (see Data Viewer) S Save collected data 9, 57 experiment 7, 14, 50 Scanner trigger iii Stimulus Lists attaching codes to 23 creating/importing 11-15 randomizing 53-54 Subset (see Expressions) Support (technical) 2 System Requirements 1 T Text color 16 Timer (RT) 14, 20, 41 Timing 41-43 Touch Screen 8, 43, 48 Trial creating 6-7, 14 editor 15 randomizing 16, 17 Trial Variables 16, 24, 26, 58 Trigger input (see Scanner trigger) Tutorial 3, 5-17 V Video Card 1, 44 Visual stimulus 16, 24, 47, 50 X! ! ! !

XID devices 42

Index

PAGE 68