Lifeboat Foundation

New research, published in the journal Patterns and led by the University of Glasgow’s School of Psychology and Neuroscience, uses 3D modeling to analyze the way Deep Neural Networks—part of the broader family of machine learning—process information, to visualize how their information processing matches that of humans.

It is hoped this new work will pave the way for the creation of more dependable AI technology that will process information like humans and make errors that we can understand and predict.

One of the challenges still facing AI development is how to better understand the process of machine thinking, and whether it matches how humans process information, in order to ensure accuracy. Deep Neural Networks are often presented as the current best model of human decision-making behavior, achieving or even exceeding human performance in some tasks. However, even deceptively simple visual discrimination tasks can reveal clear inconsistencies and errors from the AI models, when compared to humans.

O.o! Circa 2019

The real issue raised is not about brain transplants or sci-fi fantasies of reanimation of corpses. It is about how we define life and death.

The mind is a powerful thing – it can generate both symptoms of illness and symptoms of healing. Here’s what this could tell us about consciousness.

Neural circuits underlying brain functions are vulnerable to damage, including ischemic injury, leading to neuronal loss and gliosis. Recent technology of direct conversion of endogenous astrocytes into neurons in situ can simultaneously replenish the neuronal population and reverse the glial scar. However, whether these newly reprogrammed neurons undergo normal development, integrate into the existing neuronal circuit, and acquire functional properties specific for this circuit is not known. We investigated the effect of NeuroD1-mediated in vivo direct reprogramming on visual cortical circuit integration and functional recovery in a mouse model of ischemic injury. After performing electrophysiological extracellular recordings and two-photon calcium imaging of reprogrammed cells in vivo and mapping the synaptic connections formed onto these cells ex vivo, we discovered that NeuroD1 reprogrammed neurons were integrated into the cortical microcircuit and acquired direct visual responses. Furthermore, following visual experience, the reprogrammed neurons demonstrated maturation of orientation selectivity and functional connectivity. Our results show that NeuroD1-reprogrammed neurons can successfully develop and integrate into the visual cortical circuit leading to vision recovery after ischemic injury.

Functional circuit impairment associated with neuronal loss is commonly seen in patients with brain injuries, such as ischemia. Though neural stem cells (NSCs) exist in the subventricular zone (SVZ) in the adult brain, they are found to differentiate mainly into astrocytes when they migrate to injured cortex (Benner et al., 2013; Faiz et al., 2015), and their neurogenesis capacity is too limited to compensate for the neuronal loss. Currently, it still remains a challenge to generate neurons in adults and functionally incorporate them into the local circuits. Several strategies have shown the capability to induce neurogenesis and lead to some behavioral recovery. One promising approach is to transplant stem cell-derived neurons or neural progenitor cells (Tornero et al., 2013; Michelsen et al., 2015; Falkner et al., 2016; Somaa et al., 2017). Yet, there are concerns about graft rejection and tumorigenicity of the transplanted cells (Erdo et al., 2003; Marei et al., 2018).

Live human brain tissue — generously donated by brain surgery patients with epilepsy or tumors — is yielding incredible #neuroscience insights. A study on cells… See More.

As part of an international effort to map cell types in the brain, scientists identified increased diversity of neurons in regions of the human brain that expanded during our evolution.

Effort to scan the entire Human Brain continues.

Summary: A new, non-invasive neuroimaging technique allowed researchers to investigate the visual sensory thalamus, a brain area associated with visual difficulties in dyslexia and other disorders.

Source: TU Dresden

Continue reading “Charting Hidden Territory of the Human Brain” »

“The use of organophosphate esters in everything from TVs to car seats has proliferated under the false assumption that they’re safe,” said Heather Patisaul, lead author and neuroendocrinologist at North Carolina State University. “Unfortunately, these chemicals appear to be just as harmful as the chemicals they’re intended to replace but act by a different mechanism.”

Summary: Exposure to even low levels of common chemicals called organophosphate esters can harm IQ, memory, learning, and brain development overall in young children.

Source: Green Science Policy Institute

Continue reading “Common Chemicals in Electronics and Baby Products Harm Brain Development” »

Intel has introduced Loihi 2 its second-generation neuromorphic research chip, featuring eight times the computational neurons compared to the earlier Loihi 1.

It probably didn’t feel like much, but that simple kind of motion required the concerted effort of millions of different neurons in several regions of your brain, followed by signals sent at 200 mph from your brain to your spinal cord and then to the muscles that contracted to move your arm.

At the cellular level, that quick motion is a highly complicated process and, like most things that involve the human brain, scientists don’t fully understand how it all comes together.

Continue reading “An ultra detailed map of the brain region that controls movement, from mice to monkeys to humans” »