Make experiments in your research simulator

Nowadays it’s very easy to make creative behavioural experiments in a research driving simulator. While these valuable tools for research used to be really expensive, you can buy the simulator software for prices below 3000 USD.  You only need a computer with a good graphics board, a couple of monitors, a playstation steering wheel and pedals and some creativity and you are up to go. Every psychology department of universities can afford one, because the total cost will be around 5000 to 6000 USD. That’s what a professor costs each month in most universities.

When you have good tools you can develop and perform experiments much faster and make experiments of better quality. In a good research driving simulator you have script language facilities to create simulations with traffic, make secondary tasks, measure all kinds of data, ranging from response measures such as brake reaction time or steer reaction time,  to average speed, standard deviation of lateral position as a measure for swerving, or determine the minimum time to collision, steering responses etc.

You can define experiment files for each subject-condition combination so that you have your data ready immediately after the test trials.

As an example, suppose you want to measure how steering behaviour deteriorates after alcohol consumption or after the intake of medication.

  1. Then you start by choosing one from an available set of virtual environments, modify an existing one, or make a new virtual environment.
  2. Then you make a simulation: you position your simulator vehicle inside that world at a start position, define a route, make voice instructions or graphical instructions that are displayed on a monitor, define where or after how many minutes the subject should stop. Create traffic that drives at predefined rules. And most importantly, define what data you want to sample or measure, when sampling starts and ends, and maybe do a real-time analysis of these data, for example to compute SD lateral position or percentage of time the lane boundaries are exceeded.
  3. Make startup files for each subject/condition combination. Suppose you do two measurements: one without alcohol and one with alcohol 5 BAC level. Then, for each subject you make two startup files where the name of the output datafile and the name of the simulation is included.
  4. Do the measurements for each subject to create the data files
  5. Import the data files in your statistical software package and do your analyses.
  6. Make graphs and write your article.

With a proper research driving simulator you can do many experiments per researcher per year and your scientific output can increase substantially.

Ofcourse, good research requires a lot of skills and creativity, but if you have those, good tools are indispensable. And with these kinds of prices, you can’t go wrong.

A few behavioural experiments into car driving can be found here. A driving simulator study into car following behaviour is decribed in full at that website. Or check this as an example of a driving simulator study into speed choice.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s