ePoster

CONTEXT‑DEPENDENT MODULATION OF CEREBELLAR ASSOCIATIVE PLASTICITY

Javier Ubagoand 5 co-authors

University of Granada

FENS Forum 2026 (2026)
Barcelona, Spain
Board PS05-09AM-653

Presentation

Date TBA

Board: PS05-09AM-653

Poster preview

CONTEXT‑DEPENDENT MODULATION OF CEREBELLAR ASSOCIATIVE PLASTICITY poster preview

Event Information

Poster Board

PS05-09AM-653

Abstract

Adaptive motor behaviour depends on detecting when internal neural models fail and updating learning strategies accordingly. Whilst traditionally viewed as supervised error correction, cerebellar learning is typically understood to depend on neuromodulators that regulate plasticity under uncertainty. In this context, we asked whether novelty-driven activity in the locus coeruleus (LC) acts as a gating signal for vestibulo-cerebellar adaptation during vestibulo-ocular reflex (VOR) phase-reversal learning. To investigate this, we developed a closed-loop spiking model of VOR control that incorporates standard and phase-reversal VOR paradigms, where inferior olive error-related signals induce synaptic plasticity whereas LC dynamically modulates the underlying learning rules. In the model, unexpected increases in task error engage the LC, eliciting noradrenaline (NA) and dopamine (DA) release in the medial vestibular nuclei (MVN). This co-release enhances long-term potentiation at inhibitory Purkinje cell-to-MVN synapses and adjusts MVN excitability through changes in membrane resistance. As a result, rapid re-learning within the cerebellar cortex is prioritised over the slower consolidated learning in downstream nuclei. Simulations indicate that cerebellar adaptation depends on a balance between NA/DA; NA dominates under high novelty, accelerating adaptation, whereas DA becomes more influential when novelty is low, thus stabilising performance. These results support a meta-learning role for LC signals in dynamically regulating VOR plasticity and offer testable predictions for neuromodulatory manipulations during cerebellar reversal learning.

Recommended posters

Cookies

We use essential cookies to run the site. Analytics cookies are optional and help us improve World Wide. Learn more.