Prediction error encoding in apical dendrites of pyramidal neurons

Determine whether prediction error signals are encoded in the apical dendrites of cortical pyramidal neurons and, if so, characterize the conditions and mechanisms under which such dendritic error computation occurs.

Background

Apical dendrites receive rich top-down input and support nonlinear events such as dendritic spikes and plateau potentials, making them plausible substrates for integrating predictions and bottom-up signals.

While several dendritic models propose apical compartments as loci of error computation, direct experimental evidence demonstrating prediction error encoding in apical dendrites is lacking, leaving a key mechanistic question unresolved.

References

Specifically, although this information could significantly help narrow down the space of biologically plausible dendritic networks, it remains unknown whether or not prediction errors are encoded in apical dendrites.

— Neural mechanisms of predictive processing: a collaborative community experiment through the OpenScope program  (2504.09614 - Aizenbud et al., 13 Apr 2025) in Section V.3 Divergence and convergence between experiments and theories

It also remains unclear whether dendritic error construction through local inhibition can scale to propagate useful error signals across more than two areas.

— Dendritic structure enables powerful plasticity  (2608.23251 - Hünerbein et al., 24 Aug 2026) in Section 3.1, subsection “Dendritic error construction through local inhibition”

How that silence would be read out downstream---plausibly by disinhibition, and reported as a burst---is a question the framework raises without answering (S3 Text).

— Axonal delay dispersion decides whether a neuron detects an event or a sequence, and predicts cortical column diameter  (2609.04195 - Bi et al., 3 Sep 2026) in Section Discussion, subsection “Predictive coding and Transformer attention”