A WinVis experiment consists of one or more WinVis trials. Each trial contains one or more WinVis stimuli. In general, the experiment provides a trial with input parameters. The experiment determines the values for the input parameters based on subject input or an initial conditions list. Subject input can come from a keyboard, mouse, or joystick, and from a digital i/o board.

The WinVis trial receives the parameters from the experiment, and based on those parameters can do many different operations. Most trials will want to run a series of stimulus images and/or sprites, display a fixation mark, and a background image. All of these are handled through WinVis stimuli. Trials can also write text strings to the monitor, and handle the sprite driver/tracker.

The WinVis stimulus receives parameters from its parent trial and outputs image data. This image data can be processed by a look-up table, or just displayed to the monitor. WinVis stimuli can function as stimulus images, sprites, background images, and fixation marks.

Once the image data is displayed the subject may respond (or not) and that response is sent back to the experiment. The experiment then decides whether to continue the experiment, and if so, what parameters should be sent to the next trial.

Experiments may also have inter-trial periods, during which the experiment may request some subject data, perform data-analysis, or display some text or image data.
[top]

The WinVis system provides users a powerful, web-based interface to create Stimuli, Trials, and Experiments. Once the creation is complete, the WinVis server will compile the code into a file for download onto the user’s computer. This file can either be an executable application or a .dll file for use in WinVis for Matlab®.

In the case of an executable application, the user will download and run the application. The application presents the user with a menu of options. In this menu, the user can customize how the application will run, and change the values of any input parameters from their default values. The user then chooses to generate the stimulus, trial, or experiment. WinVis generates the output, taking into account the global variable values that have been set by the calibration utility.

For a Stimulus, the application outputs a bitmap image which the user can view using any bitmap viewer (such as paint). The user can then confirm that the image looks as it should, and is appropriate for use in her or his trials and experiments. When using a WinVis stimulus in a WinVis trial, other aspects of the trial can make the stimulus image look different than the bitmap image, for example by using lookup tables.

The process is similar trials and experiments. The main difference is, instead of creating a file which the user can then view, the application actually switches to video mode and runs the trial or experiment. When running a trial for the first time, the images are generated before the trial begins. Those images are saved in a quick-save file, so if the user subsequently runs the trial using the same parameters, the images will not be re-generated. This quick-save file allows the user to run the same trial several times with little or no lag time.
[top]

Creating a new stimulus with WinVis is easy, and can be done in just a few steps. Below is an outline of the steps for creating a basic stimulus. There are more options to explore, which you can learn about in the help index. The help index also contains much more detailed information about any of the steps below. For tutorials on stimulus creation, view the tutorial page.

Set up your stimulus
First you should enter the information about the stimulus you are planning. You will enter basic information, such as the author of the stimulus, any article reference, and identifiers such as bit-depth. (related: Basic Information, Article Information, Identifiers )

Enter Stimulus Parameters
The stimulus parameters allow for variability within the same stimulus. When the stimulus executable is run, you will be able to modify the values of each parameter. You can use these parameter values in any part of the stimulus. (related: Parameters, Add a Parameter )

Add Sub-stimuli
Adding sub-stimuli to your stimulus is easy to do using the integrated stimulus search tool. The editor guides you through the process of choosing a sub-stimulus, adding it to your stimulus, and setting the sub-stimulus parameters. Any stimulus can have no sub-stimuli, one sub-stimulus, or many sub-stimuli. (related: Substimuli, Add a Substimulus )

Writing the Stimulus Formula
This is where you put your sub-stimuli together into one coherent stimulus. WinVis uses special operators for combining stimulus images. There are many operators available, but the most common, and the simplest, is the overlay operator. This simply lays one stimulus on top of another. (related: Stimulus Formula )

Alternatively, if you wish to program a stimulus from scratch, you can use C++ code to assign values to every pixel in the stimulus. This is made easier using local methods. (related: Methods, Add a Method )

Saving and Compiling your Stimulus
Once you have finished the above steps, you can save your stimulus to your private database, and compile it for use on your pc.
[top]

The process of creating trials with the WinVis trial editor is similar to creating stimuli. There are more options and features to explore, allowing for more variability and power. The editor makes all these choices easy and manageable, providing help at every step of the way. Below is an outline of the steps for creating a simple trial. There are more options to explore, which you can learn about in the help index. The help index also contains much more detailed information about any of the steps below.

Set up your trial
Just like in the stimulus editor, you should enter the basic information about your trial. There is space to enter information such as relevant journal articles, keywords for this trial, and the creator of the trial. (related: Basic Information, Article Information )

Add trial parameters
Trials can have any number of input parameters. The values for these parameters are decided before the trial runs. When running the trial as a stand-alone application, the user enters values for the trial parameters. When the trial is running as part of a WinVis experiment, the experiment will determine which input parameters to use. (related: Parameters, Add a Parameter )

Set up the monitor
WinVis trials support multi-monitor systems, so the monitor is the basic functional unit of a trial. For a simple trial you will set up one monitor. This involves specifying things such as a default resolution and bit-depth. (related: Monitors, Add a Monitor )

Add stimuli
Each trial has one or more stimulus images. A stimulus image can be displayed in the foreground as a simple stimulus, as a static background image, or a fixation image. (related: Stimuli, Add a Stimulus )

Sequence the images
In this step you will determine when and for how long any stimulus image shall be displayed. The stimulus sequence, like most of the other options in the trial editor, can be entered as expressions that depend on variables such as user input, trial length, and random numbers. (related: Sequence the Stimuli )

Add optional sprites
Some trials will have sprites. Sprites are similar to normal stimulus images, but they can have transparency. Sprites also make animation and movement easy to add to your trial. (related: Sprites )

Add optional look up tables
For monitors running in 8-bit mode, the WinVis trial editor makes it easy to add look-up tables. Look-up tables can be specified using a loop of C++ code, or the values can be added individually. You may specify as many look-up tables as you wish, then sequence them to be used at different times in the trial. (related: Lookup Tables )

Other options
There are many other options available in the trial editor. Some of these options include:

• Trial Methods
• Text Strings
• Sprite Driver/Tracker
• Digital I/O (coming soon)

Save and Compile
Once your trial is complete, you can compile it into an executable file for download. At any time in the creation process you can save your trial to your private database and return to it later. The executable can then be run with any combination of parameter values.
[top]

The experiment editor is still in early development, but the process of creating experiments will be easy and intuitive. A WinVis experiment will essentially consist of a series of trials, where the user chooses how the experiment decides which trials to run and what parameter values to use based on subject response.

The target release for the experiment editor is late summer/fall 2002. Until then, users are encouraged to try WinVis for Matlab as a powerful and flexible platform for experimentation.
[top]