How Neural Activity Shapes Our Every Experience
A revolutionary framework transforming our understanding of brain-behavior relationships
Have you ever wondered how the flutter of neurons in your brain translates into the rich tapestry of your conscious experience—the warmth of joy, the swiftness of reaction, the flow of thought? For centuries, scientists have grappled with the fundamental question of how our inner world relates to the physical brain. Today, a revolutionary framework is emerging that promises to transform our understanding: the three-dimensional model of neural activity and phenomenal-behavioral patterns. This groundbreaking approach doesn't just map the brain; it deciphers the very code of human experience, offering unprecedented insights into both normal brain function and the mysteries of neuropsychiatric disorders.
Understanding the fundamental dimensions of mental experience
At the heart of this new model is a deceptively simple idea: our complex mental experiences and behaviors can be understood through three fundamental dimensions—psychomotricity (movement), affectivity (emotion), and thought (cognition). Each of these dimensions arises from distinct but interconnected neural systems working in concert 1 6 .
The model proposes that neural activity is organized into three specialized units, each with specific circuits, functions, and behavioral manifestations:
| Neural Unit | Core Brain Circuits | Primary Function | Behavioral Manifestations |
|---|---|---|---|
| External Unit | Sensorimotor network | Processes external inputs/outputs | Psychomotor activity (movement, response to environment) |
| Internal Unit | Salience network | Processes internal/body inputs | Affective states (emotions, feelings, interoception) |
| Associative Unit | Default-mode network | Processes associative inputs | Thought processes (ideas, reasoning, cognition) |
Imagine these three units as an orchestra: the external unit interacts with the outside world, the internal unit monitors our bodily states and feelings, and the associative unit weaves these together with memories and ideas to create our conscious experience 1 6 .
The power of this model lies not just in these separate units, but in their dynamic interplay. Our moment-to-moment experience emerges from the complex dance between these three systems, each influencing the others in a continuous flow of neural computation.
In one of the most ambitious neuroscience projects ever undertaken, an international collaboration of 12 laboratories recently published a "brain-wide map" of neural activity during complex behavior 3 . Recording from an astonishing 621,733 neurons across 279 brain areas in mice performing decision-making tasks, researchers discovered that representations of visual stimuli transiently appeared in classical visual areas after stimulus onset, then spread to ramp-like activity in midbrain and hindbrain regions that also encoded choices 3 .
Perhaps most surprisingly, the study found that neural responses correlated with impending motor action "almost everywhere in the brain," while responses to reward delivery and consumption were also widespread 3 . This astonishing distribution of function across brain regions challenges traditional localized views of brain function and supports the three-dimensional model's emphasis on integrated networks.
In another breakthrough, researchers have developed what they term a "foundation model of neural activity" that can predict how neurons in the visual cortex will respond to arbitrary natural videos 2 . This artificial intelligence system, trained on responses from approximately 135,000 neurons across multiple visual cortex areas in mice, generalized to new animals with minimal training and successfully predicted responses across various new stimulus domains 2 .
Even more remarkably, this foundation model could predict anatomical cell types, dendritic features, and neuronal connectivity within the MICrONS functional connectomics dataset 2 . This represents a crucial step toward building comprehensive models that can bridge the gap between neural structure and function.
Examining a landmark study in computational neuroscience
To understand how scientists are unraveling the brain's complex code, let's examine the foundation model experiment in more detail—a landmark study that exemplifies the sophisticated approaches revolutionizing neuroscience.
The research team developed a sophisticated artificial neural network comprising four specialized modules 2 :
Used ray tracing and eye-tracking data to infer each mouse's precise view of the stimulus, accounting for different head positions and eye movements.
Transformed behavioral inputs (locomotion and pupil dilation) into dynamic representations of the mouse's behavioral and attentive state.
Contained the majority of the model's computational power, using 3D convolution layers and recurrent components to produce nonlinear representations of vision modulated by behavior.
Mapped these representations onto the activity of individual neurons by performing linear combinations of features generated by the core at each neuron's receptive field.
The researchers trained this system on approximately 900 minutes of natural video responses from 66,000 neurons across 8 mice and 6 visual areas, creating what they called a "foundation core" that captured shared latent representations 2 .
The foundation model demonstrated remarkable capabilities, outperforming previous models by 25-46% in predicting neuronal responses to held-out test data 2 . But its true power emerged when tested on completely different types of visual stimuli—the model successfully predicted responses to moving dots, flashing dots, Gabor patches, coherent noise, and static natural images that it had never encountered during training 2 .
| Visual Area | Model Performance | Key Characteristics |
|---|---|---|
| V1 (Primary Visual Cortex) | High | Basic feature extraction |
| LM (Lateromedial) | Similar to V1 | Increased complexity of neuronal tuning |
| RL (Rostrolateral) | Similar to V1 | Complex feature representation |
| AL (Anterolateral) | Similar to V1 | Higher-order visual processing |
| Stimulus Type | Model Performance | Scientific Significance |
|---|---|---|
| Natural Videos | High prediction accuracy | Validates model on training domain |
| Moving Dots | Successful generalization | Tests motion processing mechanisms |
| Gabor Patches | Successful generalization | Examines response to parametric stimuli |
| Static Images | Successful generalization | Challenges model with non-dynamic input |
Most strikingly, when the researchers transferred the foundation core to new mice, these foundation models achieved high levels of predictive accuracy with significantly less training data than individually trained models 2 . This suggests the model had captured fundamental principles of visual processing that generalize across individuals.
"This work represents a crucial step towards building foundation models of the brain" 2 .
The implications are profound: we're moving toward computational models that don't just describe neural responses to familiar stimuli but can predict how the brain will react to completely novel experiences.
Modern neuroscience relies on an increasingly sophisticated toolkit for observing, measuring, and interpreting neural activity. Here are some key technologies driving these advances:
These revolutionary electrodes can record from hundreds of neurons simultaneously across multiple brain regions, enabling the massive-scale mapping demonstrated in the brain-wide map project 3 .
Techniques like those used in the MICrONS project combine functional recording with detailed anatomical reconstruction, allowing researchers to relate neural activity to physical connectivity 2 .
Artificial intelligence systems, like the foundation model discussed, can simulate brain activity and generate testable predictions about neural computation 2 .
These molecular techniques allow precise manipulation of specific proteins involved in neural signaling, helping researchers establish causal relationships between molecular mechanisms and neural functions 5 .
Advanced in vitro systems including brain organoids, organs-on-chip, and 3D printed neural scaffolds enable researchers to create more biologically relevant models for studying neural development and dysfunction 4 .
Technologies like Single Molecule Counting (SMC®) allow detection of low-abundant biomarkers in blood and cerebrospinal fluid, opening new avenues for tracking neural health and disease .
The three-dimensional model of neural activity represents more than just another theory—it offers a comprehensive framework for understanding how the intricate dance of neural circuits gives rise to the full spectrum of human experience and behavior. By viewing brain function through the lenses of external, internal, and associative processing, researchers are developing unprecedented insights into both normal brain function and the alterations underlying neuropsychiatric disorders.
As these models become increasingly refined and integrated with large-scale neural recording technologies, we're moving toward what the BRAIN Initiative has termed "a comprehensive understanding of the brain in action, spanning molecules, cells, circuits, systems, and behavior" 8 .
The future of neuroscience lies in integrating these levels of analysis—connecting the firing of individual neurons to the grand symphony of conscious experience.
The journey to fully decipher the brain's code is far from over, but with powerful new models and technologies, scientists are increasingly able to trace the connections between neural activity and the phenomenal-behavioral patterns that make us who we are. As this research advances, it promises not only to illuminate one of science's greatest mysteries but to transform our approach to brain health and disease.