ESI Systems Neuroscience Conference 2020

Causal Approaches in Neuroscience

no abstract available

Areas along the dorsal pathway in primate visual cortex are specialized for the feedforward (bottom-up) encoding of visual motion information and of aspects of the structure of our 3D-environment. But neuronal responses in these areas are also modulated by the attentional state of the organism, presumably via highly specific feedback (top-down) signals. Here I will report on studies we conducted in area MT of rhesus monkeys to test hypotheses about the neuropharmacology of this attentional modulation and its anatomical origin.

An increasing number of studies successfully applies optogenetics in the visual system of non-human primates (NHP) in order to modulate neural activation and influence visually guided behaviour. However, it remains uncertain how optogenetic stimulation affects neural circuit activity more globally in the richly interconnected cortical and subcortical areas of the visual system and whether a visual percept can arise from such circuit activation. Here we report on experiments in which the construct AAV9-hSyn-ChR2-eYFP was injected into the primary visual cortex (V1) of four macaque monkeys. The effect of optogenetic V1 stimulation on global brain activation was assessed with functional resonance imaging (fMRI) at 4.7 T. Significant BOLD modulation was found in three monkeys in higher-order areas of extrastriate visual cortex connected with V1. Supporting experiments using electrophysiology confirmed the neural basis of this fMRI effect in V1. Behavioural tests using two-alternative forced choice procedures in one monkey established that this optogenetic stimulation could evoke a weak visual phosphene. Taken together, our results show how recruitment of activation in higher order visual cortex from V1 optogenetic stimulation can drive visual perception

The brain shows marked plasticity across a variety of learning and memory tasks as well as during recovery after injury. Many have proposed to leverage this innate plasticity using brain stimulation to treat neural disorders. Implementing such treatments requires advanced engineering tools and a thorough understanding of how stimulation-induced plasticity drives changes in network dynamics and connectivity at a large scale and across multiple brain areas. In this talk, I will cover our efforts to investigate targeted stimulation of sensorimotor cortex to drive cortical plasticity towards functional recovery. We have developed a large-scale interface consisting of state-of-the-art electrophysiology and optogenetics to simultaneously record and manipulate activity from about 5 cm2 of sensorimotor cortex in awake behaving macaques.
Using this interface, for the first time, we have shown the feasibility of inducing targeted changes in sensorimotor networks using optogenetics. Furthermore, we have incorporated the capability of producing ischemic lesions in the same interface enabling us to stimulate the cortex around the site of injury and monitor functional recovery via change in blood flow, neurophysiology and behavior. Currently we are using these technologies towards developing therapeutic interventions for neurological disorders such as stroke.

Damage to cortical structures removes not only the intrinsic processing capacity of the damaged area, but also affects the excitability of regions that receive projections from the lesioned region. In our study, we used ibotenic acid to lesion parts of the feline posterior parietal cortex, which is densely connected to primary visual cortex. Before, during and after lesion, we recorded from electrode arrays positioned in the ipsilateral and contralateral primary visual cortex (A18). In my talk I will present the technique that was used to lesion an entire cortical area, histological as well as electrophysiological results.

Neuroanatomical evidence in primates show that different subdivisions of the mediodorsal thalamus (MD) are key components within wider cortico-thalamo-cortical networks. Accordingly, damage to, or disruption of MD function impairs reward-guided learning and adaptive decision-making in primates and is thought to contribute to the cognitive deficits associated with human neurological disorders like schizophrenia and fronto-temporal dementia. Despite the wealth of information implicating MD and interconnected cortex in higher order cognitive processes, the underlying mechanisms that support how the MD influences these processes are not known.
In my lab, one of the tasks we use to assess reward-guided learning is a visuo-spatial associative task. In this talk, I will discuss our assessments of cognitive performance in humans and monkeys with damage to the MD. In addition, I will describe some of the specific functional and structural changes we’ve recently identified, in monkeys using neuroimaging, after learning-to-learn this complex visuo-spatial task.

The retinal image is insufficient for determining what is 'out there', because an infinitude of possible real-world objects could produce any given retinal image. Thus the visual system must infer which external cause is most likely given prior knowledge that is either innate or learned via interactions with the environment. I will describe a general framework of 'Bayesian inference' that we have used to explore the role of cortico-cortical feedback in the visual system, and I will further argue that this approach to 'seeing' makes our visual systems prone to false perceptions, or hallucinations, in a variety of different ways.

A long-term goal of sensory neuroscience is to understand the nature of the neural code in sensory cortex. To demonstrate this understanding, we need to show that we can (1) “read” the code – i.e., use neural signals from a subject’s cortex to outperform the subject in a demanding perceptual task and to account for the variability in the subject’s perceptual decisions, and (2) “write” the code – i.e., substitute sensory stimulation with artificially evoked, perceptually equivalent, neural responses.
Distributed representations and topography are two key properties of primate sensory cortex. For example, in primary visual cortex (V1), the most localized stimulus can activate millions of V1 neurons that are distributed over multiple mm2, and neurons that are similarly tuned are clustered together at the sub-mm scale and form several overlaid topographic maps. The distributed and topographic nature of V1’s representations raises the possibility that in some visual tasks, the neural code in V1 operates at the topographic scale. If this were the case, then the fundamental unit of information would be clusters of similarly tuned neurons (e.g., orientation columns) rather than individual neurons, and to account for the subjects’ performance, it would be necessary and sufficient to consider the summed activity of the thousands of neurons within each cluster.
The long-term goal of our research is to test the topographic population code hypothesis. In this presentation, I will discuss two studies that indirectly test the topographic population code hypothesis, and describe our progress toward developing a bi-directional, read-write, optical-genetic toolbox for directly testing this hypothesis in behaving macaques.