Function and plasticity of the visual system

Our research is focused on understanding how neuronal circuits process visual information coming from the eye and integrate it with other signals to enable animals to interpret the visual world and interact with their environment. Furthermore, we investigate how these circuits change during learning and new experiences, allowing the brain to store new information and to adapt to changes in the environment.

To study these questions we use a wide range of multi-disciplinary methods: in vivo two-photon imaging of neuronal and synaptic function and structure, extracellular and intracellular electrophysiological recordings, animal behavior and theoretical modelling, together with molecular and genetic approaches to identify different cell types, record and manipulate their function and trace specific pathways.

Repeatedly imaged apical dendrite from a layer 5 neuron in vivo. The protrusions are dendritic spines which carry the majority of excitatory synapses. Spine changes are depicted by arrows, red: spine gained, blue: spine lost.

Changes in visual circuits during learning

Learning alters our perceptions, cognition and behavior by modifying neuronal circuits in the brain. Understanding how this happens is crucial for understanding normal brain function, and for devising therapeutic approaches for correcting disorders of information storage and retrieval such as dementia. Yet the mechanisms of learning in the intact brain are not well understood. Relatively little is known about how new information is stored in neuronal circuits and how new experiences, which are behaviourally relevant for the animal alter single cells, their connections and the flow of information through neuronal networks. One reason for our lack of knowledge is that it has long been impossible to repeatedly record activity from the same identified neurons and their connections over the course of days or weeks. The newest generation of genetically-encoded calcium indicators in combination with two-photon laser scanning microscopy now makes this possible. These indicators allow us to visualize the activity of neuronal networks with single-cell and even single-synapse resolution in the intact brain.

To study learning-related changes in the brain, we are developing different behavioral paradigms for mice in which they have to learn visually-guided tasks. These tasks enable us to measure the animals’ visual perception and to assess their learning progress. We are then using calcium indicators to follow directly how the function of neurons in different visual areas changes when animals make new associations during visually-guided learning. Furthermore, we are studying which circuit modifications underlie these functional changes, by following individual synapses of different pathways over the time course of learning.

Thalamic axons expressing the genetically-encoded calcium indicator GCaMP5 imaged in visual cortex and example traces of calcium transients from individual axons showing their activity in a behaving mouse.

The function of higher-order thalamic pathways during vision

Visual perception relies on information flow from the eye to the visual cortex, where it is relayed and transformed via a series of thalamic and cortical processing stages. In recent years it has become increasingly clear that the traditional hierarchical model of sensory processing, which is based mostly on feed-forward flow of sensory information, is incomplete. Prominent feedback projections from higher-order brain areas and information from parallel circuits involving the thalamus impinge on every cortical processing level. Such major alternative pathways may convey contextual information, such as the animal’s motor output, previous experience, expectations and stimulus relevance, which can strongly modulate visual responses and influence how feed-forward sensory information is interpreted in the context of an animal’s internal state and behavior. However, little is known about what information is conveyed through these different pathways and how it influences the processing of feed-forward sensory information to allow animals to actively perceive and interpret the environment based on their past experience.

We are studying the organization and function of one major pathway that might integrate visual and non-visual information but which is still very elusive – the input from higher-order thalamic nuclei into visual cortex. We are studying the organization of these thalamo-cortical circuits in the mouse using anatomical tracing methods and are investigating which information is conveyed to different cortical areas by higher-order thalamic signals in the behaving animal and how it influences the processing of visual information.