How to Emulate Human Brain Functionality

The development of artificial neurons capable of communicating with living cells marks a groundbreaking milestone in neuroscience and bioelectronics. This innovation bridges the gap between biology and technology, opening new frontiers in medical science and human-machine integration. These synthetic neurons are designed to mimic the electrical signaling of natural nerve cells, enabling seamless interaction with biological tissues. Such advancements could revolutionize treatments for neurological disorders, including paralysis, Parkinson’s disease, and spinal cord injuries. Researchers in Neuroscience and Biomedical Engineering are leveraging neuromorphic technology to replicate neural behavior. By emulating synaptic responses, artificial neurons can restore lost functions and enhance communication between damaged neural pathways. This breakthrough also accelerates progress in brain-computer interfaces, prosthetics, and cognitive computing. It paves the way for intelligent implants capable of restoring sensory functions such as vision, hearing, and movement. As innovation advances, artificial neurons could redefine the future of healthcare and human augmentation. By merging electronics with living systems, scientists are moving closer to a new era where technology seamlessly integrates with the human body to improve quality of life.

Sergiu P. Pașca at Stanford University has spent the last 15 years developing methods to create functional human brain tissue from stem cells. This work has led to the first clinical trial for a psychiatric disorder, Timothy syndrome, based entirely on human stem cell–derived brain models! In simple terms, his lab starts with skin cells from patients, reprograms them into induced pluripotent stem cells (iPSCs), and differentiates them into neurons. These neurons self-organize into 3D structures called organoids. Pasca’s group then assembles multiple organoids into functional circuits, so-called "assembloids", representing different brain regions or even brain-to-body pathways. The folks at Dr. Pașca's lab modeled cortical circuits with excitatory and inhibitory neuron interactions, observed GABAergic neuron migration in human-derived tissue, and built assembloids that replicate corticospinal and sensory pathways. Interestingly, these circuits can drive muscle contractions and respond to noxious stimuli in the dish. To date, the team has trained over 350 groups worldwide to use these methods. One of the important potential impact areas is shifting away from animal proxies. The work offers the tools for human brain development and circuitry, built and observed in real time. Some real cutting-edge stuff here, folks. Btw, I found an interesting read from 2018, "The rise of three-dimensional human brain cultures" (link in the comments), where Sergiu Pașca explains a lot of high-level stuff in this area. It could be a nice read for those starting in neurotech.

HOW IS LEARNING EARLY IN DEVELOPMENT PRESERVED AS BRAINS GROW? THE ROLE OF SPIKING NEURONS AND SPATIOTEMPORAL SELF-SIMILARITY For over 50 years, I and my many PhD students and postdocs have developed neural network models that provide principled and unifying explanations and quantitative computer simulations of brain processes that carry out the main functions needed for human intelligence, including auditory and visual perception, attention, learning, cognition, planning, emotion, navigation, and action in healthy individuals and clinical patients. Almost all of these models used RATE-BASED NEURONS and neural networks whose equations and networks could be simulated on available computers. On the other hand, most neurons are SPIKING NEURONS, whereby individual spikes, periodic spikes, or bursts of spikes travel non-decrementally down axons to recipient cells. Why, then, do rate-based models explain psychological and neurobiological data so well? I and my postdocs Yongqiang Cao and Praveen Pilly showed how ALL rate-based neural network models whose cells obey the membrane equations of neurophysiology, also called shunting laws, can be converted into spiking neural network models without any loss of explanatory power, and sometimes with gains in explanatory power. Starting in 2008, I and my PhD students and postdocs like Jasmine Leveille, Praveen Pilly, and Max Versace developed spiking neural models that exhibited experimental properties that went beyond those of rate-based models, including GAMMA OSCILLATIONS and ATTENTION during good enough bottom-up and top-down matches, and BETA OSCILLATIONS and RESET and HYPOTHESIS TESTINGduring sufficiently bad mismatches. I also proposed, starting in 1971, an important property that it seems ONLY SPIKING NEURONS can achieve. Namely, how are SPATIAL PATTERNS of short-term memory (STM) traces and long-term memory (LTM) traces that are learned during childhood preserved as our brains grow and deform until adulthood, without having to be relearned at every developmental stage? Spiking neurons can preserve spatial patterns as a child’s brain grows and deforms while it develops towards adulthood because, as the axons that carry the spikes grow LONGER, they also grow WIDER, and can carry spiking signals proportionally FASTER, thereby preserving signal patterns at recipient cells. This property of SPATIOTEMPORAL SELF-SIMILARITY is one way that spiking neurons, working together in neural networks, contribute to the development of large-scale neural networks and architectures throughout life. My article: Spiking Neural Network Models of Neurons and Networks for Perception, Learning, Cognition, and Navigation: A Review will be published soon to review the history of rate-based and spiking neurons and neural networks.   #mind #brain #learning #spikes #neurons #neuralnetwork #neuralnetworks #resonance #perception #attention #cognition #emotion #navigation #planning #action #neocortex #forgetting #ai #google

Unlocking AI's Potential with Continual Learning Memory: Meet HippoRAG ... One of the most significant challenges is enabling AI systems to efficiently integrate and retrieve vast amounts of information after the training phase, much like the human brain. A new research paper, "HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models," introduces an innovative approach to tackle this challenge head-on. 🧠 Drawing Inspiration from the Human Brain The hippocampal memory indexing theory proposes that human long-term memory relies on interactions between three key components: 1. Neocortex: Processes and stores actual memory representations.  2. Hippocampus: Holds an index of interconnected pointers to memory units in the neocortex and stores associations between them. 3. Parahippocampal regions: Connect the neocortex and hippocampus. Here's a simplified example of how this works in the human brain: When you meet a new person named "John" at a "conference", your neocortex processes and stores information about John's appearance, voice, the conference details, etc. The hippocampus then creates an index entry for "John" and associates it with other related index entries like the "conference". Later, when you think of the "conference", the hippocampus activates John's index entry, allowing you to recall information about him from your neocortex. The parahippocampal regions facilitate this process of encoding and retrieving memories. 👉 HippoRAG draws inspiration from this theory using artificial counterparts: 1. Neocortex: An LLM processes passages into discrete representations (entity triples) that are easy to index and manipulate. 2. Hippocampus: The resulting knowledge graph (KG) acts as an index, with nodes representing entities/concepts and edges representing their relationships. 3. Parahippocampal regions: Dense retrieval encoders link similar concepts in the KG. During retrieval, HippoRAG's LLM "neocortex" extracts relevant entities from a query. These are linked to nodes in the KG index by the dense encoder "parahippocampal regions". Starting from these nodes, the Personalized PageRank algorithm activates associated nodes in the KG "hippocampus", similar to following a trail of memories. Finally, activated nodes are mapped back to the original passages for retrieval. 👉 Efficiency and Effectiveness in Action The research team behind HippoRAG has demonstrated its superiority over existing retrieval-augmented generation (RAG) methods in multi-hop question answering tasks. Not only does HippoRAG achieve higher accuracy, but it also offers faster and more cost-effective retrieval, making it a promising solution for real-world applications. 👉 Revolutionizing Knowledge-Intensive Industries The implications of HippoRAG extend far beyond the realm of AI research. Its ability to efficiently integrate new knowledge has the potential to revolutionize various knowledge-intensive industries

Neuromorphic computing may be quietly reshaping how we think about machine learning, and it’s worth some attention. A recent prototype from researchers at UT Dallas shows how machines might learn the way humans do; by observing patterns and adapting over time, without needing massive datasets or energy-intensive training. Inspired by Hebbian learning and built on magnetic tunnel junctions, this system mimics the brain’s ability to strengthen connections through experience. It’s a shift away from brute-force algorithms toward something more elegant, efficient, and biologically grounded. The implications are far-reaching. Imagine AI that learns locally, adapts in real time, and runs on low-power devices; no cloud, no retraining, no environmental toll. This approach could unlock smarter wearables, privacy-preserving medical tools, and edge devices that truly understand context. It’s not just a technical breakthrough; it’s a philosophical one. If we want machines to think more like us, perhaps we should start by letting them learn like us. #NeuromorphicComputing #AIethics #MachineLearning #EdgeAI #SustainableTech #BrainInspiredAI #TechInnovation

On-Edge: Neuromorphic Computing for Psychiatric Biophysical Modeling. The article in the comments presents a brain-inspired platform for real-time dynamic computing of Spiking Neural Networks (SNNs) using asynchronous sensing in a neuromorphic chip. It highlights the growing need for edge-computing, i.e., processing data near the sensors. This approach, inspired by the biological nervous system, promises always-on processing of sensory signals, supporting on-demand, sparse, and edge-computing. The system emulates dynamic and realistic neural processing phenomena like short-term plasticity, NMDA gating, AMPA diffusion, homeostasis, spike frequency adaptation, conductance-based dendritic compartments, and spike transmission delays. The analog circuits implementing these primitives are paired with low-latency asynchronous digital circuits for routing and mapping events. This asynchronous infrastructure allows defining different network architectures and provides direct event-based interfaces to convert and encode data from event-based and continuous-signal sensors. The article also discusses the system’s architecture, characterizes the mixed-signal analog-digital circuits that emulate neural dynamics, demonstrates their features with experimental measurements, and presents a software ecosystem for configuring the system. The system’s flexibility to emulate different biologically plausible neural networks and its ability to monitor both population and single neuron signals in real-time allow for the development and validation of complex models of neural processing for both basic research and edge-computing applications. Neuromorphic computing has several potential applications in psychiatry, including real-time data analysis at the edge of the web, human-like cognitive computing, and the use of models for EEG signals to better understand brain activity patterns associated with mental health disorders. It can also be used in robotics to develop intelligent therapeutic robots that interact with patients empathetically. By processing, analyzing, and multiplexing large amounts of multimodal data, neuromorphic computing can help develop personalized treatment plans based on a patient’s unique genetic makeup, lifestyle, and environmental factors. As we gain deeper insights into the human brain and neuromorphic computing, we can expect more innovative applications in psychiatry. The work highlighted here demonstrates the use of advanced spiking neuron models for efficient data processing, emphasizing the potential of neuromorphic computing in advancing psychiatry and contributing to the broader field of neuromorphic computing, with the promise of achieving AI with lower energy needs.

U.S. scientists engineered synthetic neurons that behave like real brain cells — and talk to them too In a revolutionary fusion of electronics and biology, scientists at MIT and the University of California have created the world’s first synthetic neurons that can not only mimic the behavior of living brain cells, but also communicate directly with them in real time. These artificial neurons operate using ion-based signaling — the same chemical language used in the human brain. Unlike traditional computer chips that use electrons to transmit information, these new neurons use ionic currents — electrically charged particles that allow for far more brain-like communication. Built from soft polymer membranes, nanoscale channels, and conductive hydrogels, each synthetic cell can spike, rest, and modulate just like a living neuron. What makes this breakthrough so powerful is biological compatibility. When inserted into brain tissue slices, the artificial neurons formed synapse-like connections with nearby living cells, sending and receiving signals without rejection or inflammation. The brain didn’t treat them as implants — it treated them as one of its own. These hybrid circuits could lead to future bio-electronic brain patches that restore damaged neural networks after injury or stroke, or even augment memory and cognition. The team is already testing closed-loop feedback systems in mice that adjust firing rates based on behavior and stimulus. For the first time, man-made neurons are not just mimicking the brain — they’re becoming part of it.

A neuron in a test tube works like a real one Engineers at the University of Massachusetts have created an artificial neuron that is indistinguishable from a real one. It fires, learns, and responds to chemical signals exactly like a biological nerve cell. And most impressively — it consumes the same amount of energy as the neurons in our brains. The challenge was immense. All previous attempts to make artificial neurons ran into the energy problem — they required ten times more voltage and a hundred times more power than real brain cells. And then came the breakthrough. A team led by Shuai Fu built their neuron around a memristor — a resistor with memory. But the real magic isn’t in the memristor itself. The key lies in protein nanowires produced by a bacterium with the tongue-twisting name Geobacter sulfurreducens. This microscopic organism makes conductive nanowires that lower the switching voltage to just 60 millivolts. The current is only 1.7 nanoamperes — about the same amount consumed by the neurons in your brain right now as you read this text. But energy efficiency is only half the story. The artificial neuron was taught the full cycle of a real nerve cell: charge accumulation before firing, a sharp spike during activation, a return to resting state — and even the refractory period, a brief pause when the neuron “rests” after firing. Then the researchers added chemical sensors to detect sodium ions and neurotransmitters like dopamine. Now the artificial neuron responds to chemical cues from its environment just as a biological one changes its behavior under neuromodulation. And then came the most exciting part. The researchers connected their neuron to living human heart cells — cardiomyocytes. It worked! The artificial neuron interpreted biological signals in real time and even detected changes in cell activity after exposure to noradrenaline. Imagine wearable sensors that don’t need signal amplifiers. Or neural interfaces that “speak” the same language as your brain. Restoring damaged neural circuits? In theory, possible. For now, though, these remain laboratory experiments — clinical trials are still far away. https://lnkd.in/ebvwrgHC

Dragon Hatchling: A Brain-Inspired AI Architecture Aiming to Close the Gap to AGI Introduction A new research prototype called Dragon Hatchling is drawing attention for its attempt to replicate key features of human cognition. Developed by AI startup Pathway, the architecture dynamically rewires itself as it learns — a fundamental departure from today’s transformer-based LLMs and a potential stepping stone toward more adaptive, general intelligence. Key Details A Model Built to Learn Like a Brain • Dragon Hatchling simulates how neurons strengthen or weaken through experience, mirroring biological learning. • Unlike transformers, which lock their parameters after training, this system continuously updates internal connections in response to new inputs. • The researchers describe it as the first architecture capable of “generalizing over time,” automatically reshaping its own structure as it processes information. Why This Matters for AGI • Current LLMs excel at pattern recognition but cannot generalize reasoning the way humans do. • Pathway’s team argues this limitation is the core barrier to artificial general intelligence. • Dragon Hatchling's design aims to bridge this gap by evolving its internal wiring much like a living neural network, allowing context, memory, and adaptation to emerge organically. How the Architecture Works • Instead of stacked transformer layers, the model functions as a constantly shifting web of “neuron particles.” • These particles exchange signals, adjusting and reorganizing synapse-like connections as new information arrives. • This produces a form of short-term working memory rooted in structural change rather than static context windows. • In tests, the prototype matched GPT-2 performance on language modeling and translation, noteworthy for a first-generation model. Broader Implications and the Road Ahead • If validated, this approach could enable AI systems that improve continuously while online — learning, adapting, and evolving without retraining cycles. • Such capability could accelerate progress toward AGI but also raises governance and safety considerations, given models that grow more capable autonomously. • The research remains early and unreviewed, yet it signals a rising wave of brain-inspired architectures pushing beyond the transformer paradigm. Conclusion Dragon Hatchling represents a bold attempt to reimagine AI around biological principles rather than incremental transformer scaling. By focusing on continuous learning, adaptive wiring, and human-like generalization, it provides a glimpse into what next-generation AI systems may look like. Whether it becomes a true “missing link” or a promising detour, it underscores that the race toward AGI is entering a new, more experimental chapter. I share daily insights with 33,000+ followers and 11,000+ professional contacts across defense, tech, and policy. Keith King https://lnkd.in/gHPvUttw

3D Printed Nerve Networks on Living Brain!! Researchers at #Monash University have made a significant breakthrough in bioprinting by creating 3D nerve networks using "bioinks" containing living nerve cells, enabling these networks to grow and transmit nerve signals. This development closely mimics the 3D structure of circuits in a living brain, allowing for the study of how nerves and nerve networks form, the impact of diseases on neurotransmission, and drug screening for nerve cells and the nervous system. To create these 3D nerve networks using bioprinting, the researchers at Monash University employed a combination of tissue engineering techniques and advanced bioink materials. Here's a breakdown of the scientific details on how it was accomplished: 1. **Bioink Formulation:** Bioinks are materials compatible with 3D bioprinters and can contain living cells including living nerve cells or neurons. 2. **Living Cell Integration:** Two types of bioinks were developed for this study. One #bioink contained living neurons, while the other consisted of non-cellular materials. This allowed the researchers to mimic the complex arrangement of gray matter (cellular regions) and white matter (acellular regions) found in the brain. 3. **Bioprinting Process:** The bioprinting process involves precisely depositing these bioinks layer by layer, much like traditional 3D printing. However, instead of using plastic or metal, the printer deposits the specialized bioinks containing living neurons. This process was used to construct 3D neuronal structures. 4. **Mimicking Neural Architecture:** The researchers aimed to replicate the intricate neural architecture of the brain. The neurons in the cellular layer extended processes called neurites to form connections between different layers of the cortex, closely resembling the way neurons grow and interact within the brain. 5. **Electrophysiological Measurements:** To confirm the functionality of these 3D neuronal networks, the researchers conducted electrophysiological measurements. This involved monitoring electrical activity within the networks. They observed not only spontaneous nerve-like activity but also responses to electrical and drug stimulation. 6. **Significance:** The presence of detectable electrical activity within the tissue-engineered 3D networks is a groundbreaking achievement. It suggests that these bioprinted networks can serve as a valuable platform for various applications, including the study of neural network formation, the effects of diseases on neurotransmission, and drug screening for nervous system-related conditions. This innovative approach closely replicates the complexity of neural networks in the living brain, opening up new possibilities for neuroscience and bioprinting research. Isn't it exciting!!! #neuroscience #access Ajay Nandgaonkar Amit Saxena Sanju Senthil Kumar Dr Taruna Anand read more : https://lnkd.in/gAsYaqaK