Back to Contents   |   Previous Section   |   Next Section

5     Running the RFV

Being a Java 2 program, the RFV is run using a launcher for Java 2 applications. The program has been bundled into a Java JAR file called RFV.jar. If we want to run the program with the sample input file, sample.rfv (which is available from the RFV website), then the program can be run from the command line using the command:

java -jar RFV.jar sample.rfv

The java command at the beginning indicates that we wish to run a Java program (remember, it must be software version 1.2 or higher). The -jar RFV.jar means the program is to be found in the JAR file called RFV.jar, and sample.rfv is the name of the input file we wish to use.

This command assumes that RFV.jar is in the current directory, and that sample.rfv is also in the current directory. If these files are located elsewhere, the directory path should also be included with the filename. The sample.rfv file will also assume that all the related images are in the current directory (these images are also available from the RFV website). If the images to be used are in a different directory from the one which the RFV will be run, the input file should contain the full directory path with each image filename.

When the command line given above is entered, the RFV setup window should appear is is shown in Figure 5.

The RFV setup window

Figure 5.   The RFV setup window.

The setup window contains a region to enter a subject ID label at the top. This label should consist only of alphabetical characters, digits and the underscore character. No spaces are allowed. The subject ID will become the data output filename (with a .rfvd extension). It is important to note that if a file already exists in the current directory with the same filename, it will be overwritten. There are also two lists, one of the available components (which are defined with the COMPONENT parameter in the input file), and one for the presentation order. By selecting one or more components from the available list, it is possible to add them to the presentation order list. Components will be inserted into the presentation order list immediately above the current component that is highlighted in this presentation order list. If more than one component in the available list has been selected, then it is possible to insert the components in either the order they appear in the available list, the reverse of this order, or in a pseudo-random order. This depends on which of the three insert buttons are used. Note however, that each component can only appear in the presentation order list once.

For example, we can present all of the components in a pseudo-random order by pressing the `Select All' button under the available components list, and then the `Insert in Random Order' button. Before we can run the experiment, a valid subject ID must also be entered. If we used a subject ID called `test', then the setup window should appear similar to Figure 6 when we are ready to run the experiment. Once the `Run Experiment' button is pressed, the setup window will disappear, and the RFV window will appear. The first block in the first component in the presentation order list will be loaded and displayed. Note that each block is not loaded until it needs to be displayed.

Setting up an experiment

Figure 6.   Setting up an experiment.

Beneath the `Run Experiment' button, there is a status region. In Figure 5 it contains the text `RFV Version 2.1'. This region is used to report any messages to the user. For example, if the user tries to insert same component in the presentation order list twice, or run the experiment with an invalid subject ID, this will be reported in the status region.

5.1     RFV Errors

The RFV program does error checking when processing the input file, to make sure the correct syntax has been used when specifying the parameters. However, each block is not checked until it is loaded, and therefore the listing of all of the available components in the setup window does not mean that they have been checked. It is important that the input file be checked thoroughly, and that it is tested by the experimenter before letting experimental participants use the RFV.

Other errors may also be reported by the RFV, such as system problems or, for example, if the RFV window is set to be too small to fit in all the visual elements defined in a block. The error messages will provide some information, but if this is insufficient to allow the problem to be resolved, send email to the address listed on the RFV website with details of the error and the system set up. Any feedback or suggested improvements for the RFV are also welcome.

5.2     Guidelines for Running the RFV

The RFV accesses the system clock for all functionality that involves timing. However, Java is an interpreted language and as such the workload of the system can affect how accurately the program carries out and records timed events. It is thus important to ensure that the performance of the RFV is not affected by other processes on the system on which it is running.

There are two ways in particular to help achieve this. Firstly, it is recommended that computer does not remain connected to a network while the RFV program is being used. If it is, the system may have to deal with requests that can take up system resources. Secondly, there should not be any other processes running at the same time as the RFV. Again, this is simply to ensure that all system resources can be dedicated to running the RFV program.

5.3     Participant Interaction with the RFV

The participant interacts with the RFV using a standard computer mouse. Movements of the mouse over the stimulus correspond to movements of the focus window. Note that when not in a stimulus region, the mouse cursor is simply a default pointer. However, in a stimulus region it is a small black dot. This dot allows the participant can keep track of where they are when they move to an empty part of a stimulus region.

The only other mouse function is to allow the participant to select a button response. This requires the participant to move the mouse cursor into the area of the button they wish to select, and single click with the mouse button. If no buttons are present, a mouse click anywhere in the RFV window will allow the participant to continue on to the next block. (For multiple button mice, click with the main mouse button, which is usually the left mouse button).

The participant therefore does not need access to the keyboard when using the RFV program. There is in fact only a single keyboard command. This is pressing the `Q' key in order to quit the program. Care must be taken, since this command will work at any time when the RFV is running. It is recommended that the keyboard be kept in a position such that only the experimenter has access to it during an experiment. Even when all blocks have been presented and the end of experiment message is being displayed, the RFV window will remain visible until the `Q' key is pressed.

Back to Contents   |   Previous Section   |   Next Section