Experiments
Overview
In VRmaze, the Experimentation node is a central node in the structure of a protocol, as is the Protocol node or the Test node. There can be no run without Experimentation.
Experimentation child nodes
Experimentation describes the general environment common to all trials. The life cycle of a node is the same for all of them: a node has the same or a shorter life span than its parent. In the case of an experimentation, all child nodes, except the trial list, have a lifetime equal to the experimentation itself, which implies that a node, whether visual or logical, will be instantiated at the beginning of the experimentation and will be destroyed at the end.
The interest of nodes in an experimentation is therefore to create logical entities common to all the trials. This makes it possible to lighten the protocol tree and to keep a simple structural logic: a node integrated once in the experimentation allows not to add an instance of this same node in all the trials.
Object lifetime
Some logical nodes, children of an experimentation, may have a shorter lifetime than the experimentation itself. These objects have a specific logic and an adapted behavior.
Animation of properties
It is quite possible to animate properties in a child node of an experiment. The relative time used will correspond to the experiment start timecode, instead of the trial start timecode.
A word about the trial list
Each experiment has at least one trial list. This trial list contains all the trials that will be executed during the run.
Random Trial List.
It is possible to shuffle the trial list, either in edit mode or at runtime. In the latter case, it is possible to force the instructions of the first trial to be copied to the first trial of the shuffled list.
Experimental entities
Entities are special nodes in VRmaze, since they are not visual and have only logical behavior. The case of entities in an experimentation is to be taken with a particular consideration.
Indeed, some entities, allowing the management of external peripherals in particular, open connections or communication ports to other systems. It is wise to leave these communication managers open throughout the duration of an experiment (or even a protocol), so as not to have to open and close a communication at the beginning and end of each trial.
The purpose of the entity list of an experimentation is therefore to manage communications so that the registration nodes, integrated in each trial, can obtain information without having to worry about initializing communications.
Case of audio recording
Audio recording, via the microphone requires special attention. Here, needs must take precedence over structural logic. If you want to get one audio file per trial, you need to add a Microphone Recorder entity on each trial. Logical optimization would tell us to place this entity as a child of the experiment, but the result would be a single audio file for the entire experiment, not one audio file per trial.
Controlling the execution (with the condition)
Logically, the experimentation runs sequentially trial by trial, or, in the case of a decision process, non-sequentially, with the order of trials controlled by the `trial list' entity. VRmaze proposes a logic of control of the experimentation in progress, making it possible to stop it under certain logical conditions:
- Stop when a key is pressed on the keyboard
- Stop based on a maximum experimentation time reached
- Stop when a number of successes/errors is reached
- Stop when a cumulative number of successes/errors is reached
Accessibility of controls
Some experimentation controls are only available for certain types of experimentation.
Experimentation Controls
The types of experimentation controls are restricted in the current versions of VRmaze.
Stopping experimentation
When stopping experimentation, by logical control, the current trial is run to completion.
Stop the protocol?
Experimentation control allows for a control of the experimentation in progress. If the experimentation is stopped logically, the protocol will run the next experimentation.