In this thesis I aim to address the different components of visually driven goal-directed behavior, to understand how the brain can “keep an eye on the prize”. I use the mouse as a model system for the following reasons: 1) the main components of the mouse and human brain are similar, 2) we can make use of genetic tools in order to disentangle circuits involved in goal-directed behavior more specifically, and 3) the small size of the mouse brain makes it easier to visualize many cortical areas simultaneously, which is especially an advantage in tasks involving sensory input (posterior areas) as well as motor action (frontal areas).
In Chapter 1 I introduce the (mouse) visual system and the concept that visual perception is modulated by context. I also introduce working memory as persistent activity during a delay, and how this is likely an emergent property of distributed functional networks. In addition, I explain how the basal ganglia are ideally suited to take up an orchestrating role as part of these functional networks.
In Chapter 2 ”Mouse visual cortex contains a region of enhanced spatial resolution”, my colleagues and I reveal that despite the absence of a fovea in the retina, aspects of the organization of the visual cortex of mice resemble the foveal organization seen in primate visual cortex. We describe a region with enhanced spatial resolution in the visual cortex (‘focea’) which is behaviorally relevant: the mouse can see sharper with the focea. These findings have important implications for using the mouse as a model for human visual perception.
In Chapter 3 “The essential role of recurrent processing for figure-ground perception in mice”, we find evidence that feedback signals from higher to lower visual areas are important to provide the visual context to cells with smaller receptive fields. This is useful, for example, for segregating a figure from a similar background. When this feedback signal is disturbed, both the neural correlate for separating figures from a background in the brain as well as figure detection performance decrease.
In Chapter 4 "The direct and indirect pathways of the basal ganglia antagonistically influence cortical activity and perceptual decisions”, we first confirm that the two pathways of the basal ganglia can antagonistically bias motor responses and cortical activity in a visual detection task. We then demonstrate in a similar task with a delay between the sensory input and motor action that the basal ganglia also antagonistically bias the working memory of visual input and motor actions in the future.
In the same task, we also observe interesting neuronal correlates of task parameters after a trial has finished. From computational models we know that these correlates, known as eligibility traces, are important for learning: they can be used to assign credit (or blame) after an outcome. Hence, we decided to investigate these possible eligibility traces further in Chapter 5 “Post-trial representations of task events throughout the dorsal cortex are modulated by behavioral context”.
Altogether, this thesis adds relevant and new information to the process of visually driven goal-directed behavior. I propose that the cortex functions like an orchestra, with the basal ganglia as its conductor. New developments in systems neuroscience will enable large-scale recordings of individual neurons and neurotransmitters to further decipher if this is indeed how the brain manages to behave in a goal-directed way and keep an eye on the prize.