Why do we stop at a red traffic light? New research framework seeks to answer that question
An accelerating driver who suddenly brakes because a traffic light changes to red is an example of response inhibition, or the interruption of an on-going action. The question is which cognitive processes play a role here. Mathematical psychologist Dora Matzke received a Veni grant for the development of a statistical model framework that helps researchers conduct reliable research into response inhibition without the need for infeasibly large datasets.
Response inhibition is probably the most researched of all executive functions (the brain's complex control functions) in experimental psychology, clinical psychology and neuropsychology. It also happens to be one of the most difficult research areas, as Dora Matzke explains: 'Let's take the example of a traffic light that changes to red. Determining how fast someone stops is a challenge in itself because the time people require to stop is not directly observable. But we are also curious about the process that precedes it. What we would really like to know is which cognitive processes ensure that you actually decide to stop. That information is important for better understanding certain conditions, for example. People who suffer from schizophrenia or ADHD, for instance, cannot stop an ongoing action as fast as people without such a condition. We would like to know which processes are responsible for this: are these people more cautious (because they first want to consider everything carefully), and are therefore slower in reacting, or do they only see the traffic light later, or is there a fundamental difference in the way they process information?'
A battle between two processes
Mathematical psychology can be of great value in answering this question. Two process models were recently developed that relate to response inhibition. Both models represent inhibition as a battle between two processes, one of which is linked to the continuation response, while the other is linked to the stop response. 'As opposed to traditional models, these process models offer parameter estimates that are directly linked to cognitive capacities, such as the degree of caution or the speed of information processing. This means you can say with a good deal of certainty which cognitive process is damaged by which condition.
We often lack large quantities of data
One difficult aspect of working with these models, however, is that they require large quantities of data to arrive at accurate parameter estimates. 'Unfortunately, that just isn't realistic in practice - researchers often simply don't have access to that amount of data.' Matzke has therefore decided to apply what is known as the Bayesian framework to the models. Unlike classical statistics, Bayesian statistics does not work with the p-value. Moreover, Bayesian statistics can yield accurate parameter estimates with relatively little data.' Matzke's second objective is to develop a method for making a judicious selection between two process models and testing hypotheses with regard to the contribution of different cognitive processes.
'Ultimately, that should bring me to the main objective of my research: creating a comprehensive framework, including software, that allows researchers to use relatively small datasets to answer fundamental and applied research questions by means of state-of-the-art techniques for testing research hypotheses.'