Dendrites are the receiving ends of a nerve cell. They pick up incoming signals from neighboring neurons and carry that information toward the cell body, where it gets processed. A single neuron can have one dendrite or dozens, depending on its job and location in the nervous system. This branching design gives each nerve cell the ability to gather input from thousands of sources at once.
How Dendrites Receive Signals
When one neuron communicates with another, it releases chemical messengers into the tiny gap between them. These messengers land on specialized receptors studded along the dendrite’s surface. What happens next depends on which type of receptor catches the signal.
Some receptors act like gates. When a chemical messenger binds to them, a channel opens in the membrane and charged particles rush through. This creates a rapid but short-lived electrical change in that patch of the dendrite. Other receptors trigger a slower chain of internal chemical reactions that can either excite or quiet the neuron over a longer period. Dopamine, for instance, works through this slower pathway.
The electrical shift at the receptor site can go one of two ways. If positively charged particles flow in, the local voltage nudges closer to zero (the resting voltage of a neuron sits around negative 60 millivolts). This is an excitatory signal, making the neuron more likely to fire. If negatively charged particles flow in instead, the voltage drops further below resting level, making the neuron less likely to fire. That’s an inhibitory signal. Dendrites receive both types constantly, often simultaneously.
Combining Thousands of Inputs
A dendrite doesn’t just pass along one signal at a time. It collects excitatory and inhibitory inputs from many sources and blends them together before the combined result reaches the cell body. This blending process is called summation, and it works in two dimensions: space and time.
Spatial summation happens when signals arrive at different spots on the dendrite at roughly the same moment. If two excitatory inputs land close together, within about 10 micrometers of each other, they can amplify each other beyond what you’d expect from simply adding them up. This “supralinear” boost disappears when the inputs are spaced more than 15 micrometers apart. Temporal summation happens when signals arrive at the same spot in quick succession. Research on hippocampal neurons shows that inputs arriving within a 30-millisecond window of each other produce this same amplified effect.
The practical result: dendrites don’t just relay messages passively. They perform a kind of computation, weighting and combining inputs so the cell body receives a processed summary rather than raw noise. Only when the combined signal crosses a voltage threshold does the neuron generate an electrical impulse that travels down the axon to the next cell.
Dendritic Spines and Memory
Most excitatory connections don’t land directly on the main trunk of a dendrite. They connect to tiny mushroom-shaped bumps called dendritic spines, each one hosting a single synapse. These spines are far from static. They grow, shrink, and change shape in response to experience, and this physical remodeling is one of the brain’s primary mechanisms for learning and memory.
When a connection between two neurons gets used frequently, the spine at that junction tends to grow a larger head. Spine head size correlates directly with synaptic strength: a bigger spine means a stronger connection. Repeated stimulation also triggers the formation of entirely new spines, a process called spinogenesis, which has been observed both in isolated brain tissue and in living animals. These newly formed spines can mature into fully functional synapses within hours to days. Conversely, when a connection weakens from disuse, spines shrink and may retract entirely.
This constant remodeling means the architecture of your dendrites is shaped by your experiences. Learning a new skill, forming a memory, or adapting to a changed environment all involve physical changes to dendritic spines.
Active Processing, Not Passive Wiring
For decades, neuroscientists treated dendrites as simple cables that passively funneled electricity toward the cell body. That view has changed substantially. Dendrites can generate their own electrical events, including calcium spikes that last hundreds of milliseconds and spread across stretches of dendrite longer than 30 micrometers. These spikes are triggered by a specific type of receptor (the NMDA receptor) and cause large surges of calcium into the dendrite.
These calcium events do more than amplify incoming signals. Research in the motor cortex of mice has shown that dendritic calcium spikes drive lasting changes in synaptic strength during learning. Individual dendritic branches appear to function as semi-independent units for storing information, meaning a single neuron doesn’t just compute one thing. Different branches of its dendritic tree can participate in different circuits and undergo plasticity independently.
Signals travel through dendrites at measurable speeds. In rat neurons, excitatory signals move at roughly 0.07 meters per second, while in human neurons the speed is about 0.09 meters per second, around 26% faster. Backpropagating signals, electrical impulses that travel from the cell body back into the dendrites, move considerably faster: about 0.23 m/s in rats and 0.34 m/s in humans. These backpropagating signals help dendrites “know” when their neuron has fired, which is important for certain forms of learning.
Shape Determines Function
Not all dendrites look alike, and their shape directly determines what signals a neuron can collect. Pyramidal neurons, the most common excitatory cells in your cerebral cortex, have a distinctive tree-like branching pattern with one long apical dendrite reaching toward the brain’s surface and a cluster of shorter basal dendrites spreading out from the bottom of the cell body. In the hippocampus, the total length of a pyramidal neuron’s apical dendrites averages around 6,300 micrometers, with basal dendrites adding another 5,000 micrometers.
Purkinje cells in the cerebellum take a different approach entirely. They grow an enormous, flat, fan-shaped dendritic arbor that continues expanding as the animal grows. This massive surface area lets a single Purkinje cell receive input from hundreds of thousands of other neurons, which is essential for the cerebellum’s role in coordinating movement.
Even within a single brain region, dendrite complexity varies by cell type and develops on different timelines. In the human prefrontal cortex, some pyramidal neurons reach peak dendritic complexity by 16 months of age, while neighboring neurons in a different cortical layer remain relatively simple during that same period and continue growing for over a year afterward.
When Dendrites Break Down
Because dendrites are so central to how neurons communicate, damage to their structure shows up in neurological disease. In Alzheimer’s disease, dendrites and their spines undergo significant pathological changes. A protein called tau, which normally helps maintain the internal scaffolding of the cell, becomes chemically altered and detaches from that scaffolding. The misfolded tau migrates from the axon into the cell body and dendrites, where it clumps into tangles. This disrupts the dendrite’s internal structure, compromises its ability to maintain spines, and weakens synaptic connections.
Spine loss is one of the earliest structural changes seen in Alzheimer’s, often appearing before neurons die outright. Since each spine represents a synaptic connection, losing spines means losing the communication links between neurons. This helps explain why memory and cognition decline progressively: the dendrites responsible for receiving and integrating information are physically deteriorating.