CCN meeting | Dora Matzke (Universiteit van Amsterdam, the Netherlands) and Andrew Heathcote (University of Tasmania)

05-12-2019 from 12:30 to 14:00
Henri Dunantlaan 2, room 4.3
Add to my calendar

Inhibiting difficult choices: From simple descriptive to cognitive process models of the stop-signal paradigm (Dora Matzke)

The cognitive concept of response inhibition is frequently investigated using the stop-signal paradigm where participants perform an easy response time task, such as responding to the direction of an arrow. Occasionally, this "go" task is interrupted by a stop signal that instructs participants to withhold their response. Stop-signal performance is typically formalized as horse race between two competing processes: a go process that is initiated by the primary task stimulus and a stop process that is triggered by the stop signal. If the go process wins the race, the primary response is executed; if the stop process wins the race, the primary response is inhibited. The race model allows for the estimation of the latency of the unobservable stop response. It does so, however, without accounting for accuracy on the go task. This restriction means that the race model may not be used to investigate response inhibition in the full range of tasks used in psychology, which can involve difficulties that result in non-negligible levels of errors, or where it is theoretically important to manipulate error rates. I discuss a Bayesian framework that addresses this limitation, and hence expands the scope of the stop-signal paradigm to the study of response inhibition in the context of difficult as well as easy choices. Using a mixture-likelihood approach, I combine the treatment of go errors with the ability to address two common contaminants in stop-signal data: failures to trigger the go and the stop racers. I propose various parametrizations of the framework, ranging from the descriptive ex-Gaussian distribution to the racing Wald evidence-accumulation architecture, explore the strengths and weaknesses of the different models, and illustrate their utility with clinical and experimental data.

Genuine inhibitory deficits in aging (Andrew Heathcote)

Non-parametric estimates of mean stop-signal response time (SSRT) – the speed with which a stop process can inhibit an ongoing action – have been used to infer inhibition deficits in various conditions, including attention-deficit hyper-activity disorder, people with schizophrenia and the aged. Recent work using parametric models of the stop-signal task indicates that some of these inferences were confounded by the occurrence of “trigger failures”, where deficits occur because the stop process fails to start rather than because it runs slowly. We investigate this issue in aging using a stop-signal paradigm with a go task requiring choices of varying degrees of difficulty and with varying base rates. These latter characteristics were modelled by evidence-accumulation processes while the stop process was modeled descriptively by an ex-Gaussian distribution. Results indicated that, if anything, older participants had a lower level of trigger failures but had a clearly slower stop process, indicating genuine inhibitory deficits.