Multi-Participant Studies

Description

In Labvanced, it is possible to create studies with multiple subjects participating in one experiment at the same time. For this purpose, the task editor provides a number of additional functionalities in order to enable synchronization and data exchange during a multi-participant experiment.

Enabling Multi-Participant Functionalities

By default, the necessary tools for creating multi-participant studies are hidden in the task editor. If you want to enable them, you can do this under the menu item Study Settings in the right column. By clicking on the Transform button, all relevant tools will be enabled and visible. The number of participants per experiment can then be set.

  • Note that once the Transform button is clicked, there is no possibility to undo this. If you later decide that you don't need these additional tools, just ignore them and make sure to remove all related actions and synchronizations manually.

Role IDs

Since multiple subjects are participating at the same time, it is necessary to distinguish between them without collecting personal information. This can be done with the help of the numeric system variable "Role_Id". As soon as a new multi-participant experiment starts, every participant gets a unique Role_Id in ascending order starting at one. By using requirement actions it is possible to execute actions depending on the Role_Id assigned to a specific subject. An example for the requirement action could be: if the Role_Id equals one, the subject will see image x and if the Role_Id equals two, the subject will see image y.

Frame Synchronization

The synchronous start of an upcoming frame can be arranged by activating the checkbox Sync Frame Start under the menu item Frame Settings. If activated, participants must wait before starting the target frame until all other participants have reached the synchronized frame.

While creating a study, make sure that all participants reach a synchronized frame. Otherwise, some participants might reach a dead end and wait endlessly because another participant skipped the synchronized frame (for example, due to a conditional jump).

Task Synchronization

The start of a new task can also be synchronized by activating the corresponding checkbox under the menu item Task Settings. Note that this option only synchronizes the task start and therefore the first frame of a task. Other frames within the tasks are not affected. As for frame synchronization, ensure that all participants reach a synchronized task.

The Distribute Value Action

With the help of the distribute variable action, the actual information exchange can be realized. Dependent on a trigger of an event, an arbitrary value can be sent from one participant to a target variable of one or more other participants of the same running experiment.

Action Options

Once the Distribute Variable action is selected as the action of an event there are some options to chose from:

  • The first option specifies whether the variable distribution should affect all or only selected participants.
  • The second option specifies the target variable together with a value. Note that you can use all variables as the target variable. The value could be a constant value, a current variable value, or a more complex arithmetic operation. It is important to point out that the arrow does not represent an equal sign, but rather a potential assignment of the target variable. In some cases the action might be refused and the target variable remains unchanged, which also holds for the initiating participant of this action.
  • The third option indicates whether the affected action should make use of the blocking system.

The Blocking System

Since distributions might affect the behaviour of other participants, if two or more participants want to distribute their values to the same target variable at almost the same time, the study designer must decide whether to allow all distributions or reject all but one until it's processed.

By activating the third distribute variable option, the blocking system is enabled. In this case the first incoming distribution will be executed. Other distribution attempts will be blocked until it is processed, which means that all participants which are affected by the action have confirmed the receipt of this action.

By deactivating the third option, all incoming distribution attempts will be executed unconditionally.