Illuminating the role of the prefrontal cortex in cognitive control of behavior

Cognitive control of behavior is essential for successful goal-directed behavior. I present research aimed at various components of cognitive control, such as inhibitory control and attention. For instance, inhibitory control allows us to inhibit unwanted actions, and attention lets us be vigilant to specific cues in our environment. While these processes reflect coordinated activity between many brain regions, the prefrontal cortex is often a main player. In this thesis, I study the role of the mPFC is characterized by investigating meaningful subpopulations, based on position on the dorsoventral axis, or based on neuronal projection targets. It appears that the mPFC contains subpopulations of projection-specific neuron populations, which each have their own role. In Chapter 2, we show that the dorsomedial and ventromedial mPFC (dmPFC, vmPFC, resp.) have distinct roles in cognitive control. Rats were trained in the 5-choice serial reaction time task (5-CSRTT). Upon reaching threshold performance, we inhibited either the dmPFC or vmPFC. We found that both areas are required for task accuracy, and that the vmPFC is also involved in inhibitory control. Moreover, the timing of activity in each brain region appeared to be different. Dorsal mPFC activity was required for the entire delay period, while vmPFC activity was only necessary for the last portion. This suggests that both brain regions encode distinct behaviors and are part of distinct networks that underlie cognitive control. Next, we investigated mPFC activity during acquisition of the 5-CSRTT. We show in Chapter 3 that neuronal activity in dmPFC and vmPFC does follows distinct patterns, and that this activity develops differently. Activity in both brain regions is shaped towards faster activation, and becomes more biased towards earlier portions of the delay period. Dorsal regions appear to be more active than ventral regions, especially in the first half of the delay. Moreover, it appeared that prefrontal activity decreases as animals progress through the learning stages. Together, these findings indicate that prefrontal activity in dorsal and ventral regions develops during learning stages. In Chapter 4, we present a new version of the conventional 5-CSRTT. Conventional settings require extensive training periods, which may complicate chronic recordings or manipulations, and often necessitate food restriction protocols to motivate animals. Therefore, we developed the CombiCage, a homecage-based paradigm. Using a time-restricted approach, where animals could only start trials during a specific time window during the day, we found that behavioral performance was similar to conventional settings. Animals fully progressed through the learning stages within a week, and often earned enough food to preclude use of food restriction. Taken together, the CombiCage significantly reduces training time, requires less researcher interference, and reduces the need for food restriction. Finally, we further explore the role of the mPFC in cognitive control, by characterizing and investigating projection-specific neuronal populations during the 5-CSRTT in the CombiCage. We anatomically define four distinct and discrete mPFC projection populations to subcortical nuclei, such as the mediodorsal thalamus and striatum. Next, we show that these projections elicit distinct postsynaptic responses in target neurons. We then trained animals in the CombiCage, and chemogenetically inhibited each of the projections, unraveling distinct roles for each projection population. Besides distinct behavioral roles, each of these projections had a unique activation pattern during in 5-CSRTT trials. All in all, we show in this chapter that the mPFC orchestrates cognitive control of behavior through specific neuronal circuits. This thesis highlights the role and complexity of mPFC activity during cognitive control of behavior. Our findings suggest that prefrontal subregions, especially projection-defined populations, have distinct roles in behavior. These populations are likely part of distinct larger networks, whose interplay and concurrent activity controls our actions.