Context association in pyramidal neurons through local synaptic plasticity in apical dendrites

Maximilian Baronig, Robert LegensteinFrontiers in Neuroscience 2024 · DOI

Abstract

The unique characteristics of neocortical pyramidal neurons are thought to be crucial for many aspects of information processing and learning in the brain. Experimental data suggests that their segregation into two distinct compartments, the basal dendrites close to the soma and the apical dendrites branching out from the thick apical dendritic tuft, plays an essential role in cortical organization. A recent hypothesis states that layer 5 pyramidal cells associate top-down contextual information arriving at their apical tuft with features of the sensory input that predominantly arrives at their basal dendrites. It has however remained unclear whether such context association could be established by synaptic plasticity processes. In this work, we formalize the objective of such context association learning through a mathematical loss function and derive a plasticity rule for apical synapses that optimizes this loss. The resulting plasticity rule utilizes information that is available either locally at the synapse, through branch-local NMDA spikes, or through global Ca<sup>2+</sup>events, both of which have been observed experimentally in layer 5 pyramidal cells. We show in computer simulations that the plasticity rule enables pyramidal cells to associate top-down contextual input patterns with high somatic activity. Furthermore, it enables networks of pyramidal neuron models to perform context-dependent tasks and enables continual learning by allocating new dendritic branches to novel contexts.

Introduction

Published in Frontiers in Neuroscience (2024), this research introduces a mathematically grounded plasticity rule that explains how pyramidal neurons—the "workhorses" of the cortex—integrate sensory data with behavioral context. By deriving a learning rule from a formal loss function, the authors demonstrate how local dendritic computations allow neurons to solve complex, context-dependent tasks and resist "catastrophic forgetting" during continual learning.

Neocortical pyramidal neurons are structurally segregated into two main compartments: basal dendrites, which receive bottom-up sensory input, and apical dendrites, which receive top-down contextual information like attention or expectations. While neuroscientists have long hypothesized that these cells "associate" these two streams, the exact synaptic mechanism for this learning has remained a mystery. Existing models are often phenomenological, describing what happens but not why from an optimization perspective.

The Context Association Learning (CAL) Rule

The researchers developed an abstract model of a Layer 5 pyramidal neuron with multiple apical branches and a single basal integration site. They defined a mathematical loss function ($L_{CAL}$) based on three objectives:

  1. Association: Contextual patterns co-occurring with strong basal activity should trigger strong apical activation.
  2. Clustering: Co-active synapses should cluster on the same dendritic branch to trigger local nonlinearities.
  3. Dissociation: Contextual patterns that rarely co-occur with basal activity should be suppressed.
Structural segregation and plasticity signals in the pyramidal neuron model.

By applying gradient descent to this loss function, they derived the CAL rule. This rule relies only on signals available locally at the synapse or globally within the cell:

  • Pre-synaptic activity.
  • Branch-local NMDA spikes (signaling local cluster activation).
  • Global Calcium ($Ca^{2+}$) spikes (signaling the "success" of the association across the whole tuft).
  • Somatic back-propagation (signaling the neuron's overall output).

Key Insights: From Local Spikes to Global Success

A central feature of the CAL rule is the "Success Signal": the global $Ca^{2+}$ spike acts as a natural regulator. Once a neuron successfully associates a context with a sensory input, the resulting $Ca^{2+}$ spike prevents further potentiation. This prevents "resource waste" where too many branches tune to the same pattern.

The rule also instantiates branch competition. One branch "wins" a specific context, while others are depressed, leading to highly efficient, sparse representations. This leads to structural protection: because the rule clusters information onto specific branches, new tasks can be assigned to "fresh," unused branches. This naturally protects previously learned associations from being overwritten, offering a biological solution to catastrophic forgetting.

Experimental Results

The authors evaluated the CAL rule using several synthetic benchmarks, most notably the Context-Dependent Feature Association (CDFA) task. In this multi-task benchmark, "object classes" are defined by combinations of features (e.g., "blue" + "star"). The network must output "1" only if the features in the sensory input match the class provided in the context.

  • High Accuracy: In the CDFA task with 100 object classes, the network achieved a classification error of only ~6.9%. When using a more powerful unsupervised rule for basal weights (Krotov+), the error dropped to 1.44%.
  • Continual Learning Success: In sequential task learning, the network maintained low error rates on early tasks even after learning 48 subsequent tasks.
  • Sparse Coding: The model naturally produced sparse activation; most contexts activated only one branch per neuron, mirroring biological efficiency.

Discussion

While the CAL rule provides a normative explanation for dendritic nonlinearities, the model currently uses a simplified morphology. Real pyramidal neurons have much more complex branching geometries. Furthermore, while the rule uses biologically available signals, experimental verification of the specific "stop-learning" effect of $Ca^{2+}$ spikes in vivo remains a vital next step. Future work will also look toward a unified plasticity rule for both basal and apical compartments.

Summary updated

Abstract

The unique characteristics of neocortical pyramidal neurons are thought to be crucial for many aspects of information processing and learning in the brain. Experimental data suggests that their segregation into two distinct compartments, the basal dendrites close to the soma and the apical dendrites branching out from the thick apical dendritic tuft, plays an essential role in cortical organization. A recent hypothesis states that layer 5 pyramidal cells associate top-down contextual information arriving at their apical tuft with features of the sensory input that predominantly arrives at their basal dendrites. It has however remained unclear whether such context association could be established by synaptic plasticity processes. In this work, we formalize the objective of such context association learning through a mathematical loss function and derive a plasticity rule for apical synapses that optimizes this loss. The resulting plasticity rule utilizes information that is available either locally at the synapse, through branch-local NMDA spikes, or through global Ca<sup>2+</sup>events, both of which have been observed experimentally in layer 5 pyramidal cells. We show in computer simulations that the plasticity rule enables pyramidal cells to associate top-down contextual input patterns with high somatic activity. Furthermore, it enables networks of pyramidal neuron models to perform context-dependent tasks and enables continual learning by allocating new dendritic branches to novel contexts.
BibTeX
@inproceedings{baronig2024context,
  author = {Maximilian Baronig and Robert Legenstein},
  title = {{Context association in pyramidal neurons through local synaptic plasticity in apical dendrites}},
  booktitle = {Frontiers in Neuroscience},
  year = {2024},
  doi = {10.3389/fnins.2023.1276706},
}
Figure