![]() Mouse-tracking in OpenSesame, you have two options: you can use the GUI to control stimuli, etc., and just use Python code to record the mouse data, or you can control the entire trial using Python code, which is what I’ll do here. More complicated designs can be created by placing Python code in the experiment (‘inline code’), and that’s what we’ll do here. Most of what you’ll need for most experiments can be built in OpenSesame using just the GUI. If you’re new to OpenSesame, I would recommend you take a look at this introduction. OpenSesame is an open source experiment builder, written in Python. I’ve also done some work on using HTML5 to run similar experiments in a web browser, which I’ll be posting about later. I got around this by programming my own mouse-tracking experiments in OpenSesame, and it’s this code that I’m going to share here. ![]() However, while immensely useful, MouseTracker is a little inflexible in what your experiment actually looks like, and so for more advanced or exotic experiments it can still be necessary to code them yourself, which is exactly what happened to me. Happily, Jonathon Freeman packaged and released his code as MouseTracker, allowing researchers to quickly and easily program mouse-tracking experiments, and analyse the huge dataset (for cognitive psychology) the paradigm produces. Originally, mouse tracking experiments were programmed using in-house software (mostly using PsyScope on OSX, I think), and so there was a barrier to curious researchers applying the method to their own field. Since then, the paradigm has been applied to a whole range of questions, including social categorisation, lies and deception, lexical decision, pragmatic inference, and learning. Spivey et al (2005) introduced the method, showing that phonologically similar words have partially overlapping representations: told to ‘Click the candle’, participants move as if they’re partially drawn towards clicking the ‘candy’ instead (but not if the alternative is ‘pickle’). The idea behind mouse tracking is simple: you records people’s mouse movements as they respond to a question, and use them to figure out how drawn they were to each response over the course of the decision. In particular, I’m interested in what this technique can tell us about category-based induction, and reasoning under uncertainty. In my PhD, I’m looking at what process-tracing measures like mouse tracking can tell us about what goes on during high-level reasoning and decision making.
0 Comments
Leave a Reply. |