Bringing together neuroscience and computational perspectives, the 2026 MBCT symposium explores how recurrence may be a canonical neural mechanism for integrating past experience, present input, and future goals across multiple behavioral timescales.

Every given moment of our lives is some combination of past, present, and future. While we occupy the current moment, we integrate sensory information from the recent past, draw on relevant memories from more distant experience, anticipate upcoming events, and make goal-directed decisions that guide our actions. 

What are the neural mechanisms underlying our brain's ability to coordinate information from such different timescales?

In neuroscience, recurrent feedback has long been known to link the recent past and the present by allowing neuronal computations to be affected not just by current sensory input, but by recurrent patterns of activity from the recent past. To what degree is such recurrent computation a canonical motif of neural computation that allows our brains to integrate multiple timescales required for behavior? In this symposium, we explore emerging views of behaviors that unfold over different timescales, and how recurrent neural computation can support these behaviors.

Sign up to receive future updates about the symposium and additional events

---

MBCT Symposium 2026 - Featured SpeakersSpeakerInstitutionTalk Title

Janice Chen

Johns Hopkins University

Studying memory for natural events across multiple timescales

Hidehiko Inagaki

Max Planck Florida Institute for Neuroscience

Perturbation experiments to dissect attractor dynamics for flexible motor timing

Caroline A. Runyan

University of Pittsburgh

State-dependent population codes across cortexKohitij KarYork UniversityComputational model-guided insights into temporal dynamics of the macaque ventral stream