PreSti – Neuroscience stimuli presentation software
Petr Jaroš (Department of Computer Science and Engineering, University of West Bohemia, Pilsen, Czech Republic), Roman Mouček (Department of Computer Science and Engineering, University of West Bohemia, Pilsen, Czech Republic), Petr Háka (Department of Computer Science and Engineering, University of West Bohemia, Pilsen, Czech Republic)
Our research group at Department of Computer Science and Engineering, University of West Bohemia in cooperation with other partner institutions within Czech National Node for Neuroinformatics specializes in the research of attention, especially attention of drivers and seriously injured people. We widely use the methods of electroencephalography (EEG) and methods of event-related potentials (ERP).
There are several programs used for presentation of stimuli and for development of test scenarios. During our research we have collected requirements for software of this type and none of the existing software has covered them. It is necessary to create test scenarios during ERP experiments. With the currently most used software Presentation produced by Neurobehavioral System only programmers can create new scenarios. These scenarios are written using a custom programming language. One of our main requirements for ERP software was to allow designing and editing of test scenarios to people without programming knowledge such as laboratory assistants and doctors in hospitals.
In our approach the test scenario is designed by creating flowchart(s). This concept is called visual programming and its benefits are simplicity and good readability. A user is allowed to create scenarios by clicking mouse and specifying parameters. Visual programming itself has a disadvantage because it does not have such a big expressivity as traditional programming. A great focus was given in extensibility of flowchart diagrams to avoid this problem. It is possible to design new extensions which can contain flowchart symbols with any functionality (e.g. flow control, stimuli, etc.) So the user of PreSti software does not need to have programming skills to work on ERP experiments. When the user is missing some functionality he/she can get an extension from other users or ask a programmer to create one. This extension functionality appears in the program as a new symbol to the user and it is compatible with all existing symbols due to the visual programming concept.
The visual programming concept was designed as simple as possible. All stimuli, flow control mechanisms, inputs and outputs are represented by symbols. All symbols can communicate with others using four different ports. There are two types of connectors between ports: Trigger, which connects the event port of one symbol and the action port of the second symbol; and Connection, which connects getter and setter of two symbols. Connection is used for a transfer of a value between symbols. Generally, the flow of the program is from the symbols on the left side to the symbols on the right side and variables are falling down from the top. Thanks to this concept it is very easy to extend the functionality with new symbols. A default set of symbols covers basic stimuli presentation, interface interaction, synchronization signal creation, output conditions, branching and some more. In case of big test scenarios flowchart diagrams can grow up. It is possible than to split the diagram into small sub diagrams to make them well readable.
Because there is a request for this software to run on non-real-time operation systems it is necessary to guarantee minimal latency between presentation of stimuli and dispatch of corresponding synchronization signal. When the software starts it takes care about priority of all system processes to avoid lags originated in other processes.
PreSti is neuroscience stimuli presentation software accessing the design of test scenarios in ERP experiments to wider user audience – to the people who really think out experiments. Thanks to extensibility and visual programming language design the program does not suffer from reduction of possibilities during experiments development.