Tracking Cognitive Control: How do we solve interference?

Selecting relevant information in the presence of distracting one is a core component of cognitive control, referred as interference resolution. This process has been often investigated through the Stroop task, where responses are longer when two stimulus features are incongruent, compared to when they are congruent (Stroop effect). Despite a large body of literature about this process, the mechanisms of interference resolution are still matter of debate. The present PhD project aimed at shedding light on the temporal dynamics of interference resolution and the related neural underpinnings. In Study 1 we focused on investigating the brain oscillations involved in this process during a spatial Stroop task, aiming at understanding if these correlates and their temporal course change across the lifespan by recruiting younger and older adults. We found age-related differences in theta and beta bands. Theta may represent an early mechanism signalling the need to exert control, which seems to be impaired with aging. Beta may correspond to the process of relevant information selection and older adults showed an over-recruitment of these frequencies. Previous evidence suggested that these results may be attributed to age-related differences in the use of proactive and reactive control, in line with the DMC (Dual Mode of Control) model. Proactive control is defined as an anticipatory attentional bias, whereas reactive control as a late correction mechanism. To study more in depth the different contribution of these control modes, we developed Study 2. We used the same task, manipulating the percentage of congruency (PC) at different levels, list or item, to elicit proactive and reactive control respectively. We also recorded computer mouse trajectories because the high temporal resolution of this tool can shed light on the underlying temporal dynamics of these control modes. Analysis of mouse-derived measures showed that the Stroop effect was present as costs in responding to incongruent trials, reflected in a greater attraction toward the irrelevant information, less smooth trajectories, and longer time to respond due to the updating and adjustments of the trajectories. We found that the magnitude of the interference varied as a function of the PC manipulations, with smaller Stroop interference for low-PC manipulations. Our results suggested also that reactive control may work faster than previously thought, possibly triggering a rapid attentional bias toward the relevant information similar to the one predicted for proactive control. To investigate further the role of proactive control it was necessary to study the time preceding stimulus appearance. Hence, we developed Study 3 in which we used the same mouse-tracking task, manipulating the PC at the list level to mainly elicit proactive control, and we recorded EEG signal to have a window on the brain dynamics before stimulus presentation. We found clear PC-dependent modulations of the interference, both at the behavioural and the neural level, for which we found smaller Stroop effect for blocks with low PC. Behavioural results generally replicated those of Study 2. EEG results showed PC-related modulations of interference, which mirrored the same pattern observed in mouse-derived measures and mainly involved theta and beta bands. This project provides confirmations and new suggestions in the study of interference resolution. We confirmed the involvement of theta in this process, interpreted as an early mechanism of interference detection that signals the need to exert control. We also found a main involvement of beta that may represent the imposition of early attentional biases toward the relevant information. We interpreted these results in line with the Cascade of Control and the DMC models. This project represents a first attempt to evaluate more deeply the temporal course of proactive and reactive control, taking advantages of two techniques with high temporal resolution.