Neural Heterogeneity Enables Adaptive Encoding of Time Sequences
Raphael Lafond-Mercier, Leonard Maler, Avner Wallach, André Longtin · 2025
Read Neural Heterogeneity Enables Adaptive Encoding of Time Sequences by Raphael Lafond-Mercier, Leonard Maler, Avner Wallach, André Longtin (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-14855.txt.
Introduction
Raphael Lafond-Mercier, Leonard Maler, Avner Wallach, and André Longtin build a theory of how brains encode time gaps in a paper from 2025. Animals judge intervals from split seconds to minutes for navigation, speech, and memory. One mechanism, found in fish thalamus cells, uses neural fatigue: a tired cell recovers at a rate that tracks elapsed time.
The theory lets recovery speed and past sensitivity vary from cell to cell. A Bayesian frame then asks how much timing information such a mixed population holds. Diverse cells, it turns out, make the code flexible across tasks. The result explains a strength of messy biology: variety itself can be the adaptation. The work links single-cell quirks to a general timing sense.
Catalog record
| CREATOR | Raphael Lafond-Mercier, Leonard Maler, Avner Wallach, André Longtin |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 61 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-14855
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Neural Heterogeneity Enables Adaptive Encoding of Time Sequences Raphael Lafond-Mercier, Leonard Maler, Avner Wallach, André Longtin Abstract Biological systems represent time from microseconds to years. An important gap in our knowledge concerns the mechanisms for encoding time intervals of hundreds of milliseconds to minutes that matter for tasks like navigation, communication, storage, recall, and prediction of stimulus patterns. A recently identified mechanism in fish thalamic neurons addresses this gap. Representation of intervals between events uses the ubiquitous property of neural fatigue, where firing adaptation sets in quickly during an event. The recovery from fatigue by the next stimulus is a monotonous function of time elapsed. Here we develop a full theory for the representation of intervals, allowing for recovery time scales and sensitivity to past stimuli to vary across cells. Our Bayesian framework combines parametrically heterogeneous stochastic dynamical modeling with interval priors to predict available timing information independent of actual decoding mechanism. A compromise is found between optimally encoding the latest time interval and previous ones, crucial for spatial navigation. Cellular heterogeneity is actually necessary to represent interval sequences, a novel computational role for experimentally observed heterogeneity. This biophysical adaptation-based timing memory shapes spatiotemporal information for efficient storage and recall in target recurrent networks. keywords neural dynamics, neural adaptation, time encoding, information theory, heterogeneity 1 Introduction Long before human-made tools Pappas (2023) , biological systems devised multiple molecular and electric mechanisms to keep time over a wide range of scales Merchant et al. (2013) . Much is known about the interaction of neural activity and molecular clocks that generates hours to days long circadian rhythms Reppert and Weaver (2002) . Auditory circuits in barn owls require tens of microseconds to pinpoint the direction of a sound Leibold et al. (2001) . Scales of hundreds of milliseconds to a second, together with neural plasticity, are important for speech Poeppel and Assaneo (2020) , song generation in birds Singh Alvarado et al. (2021) , and motor control Edwards et al. (2002) . The intermediate scale of seconds to minutes is relevant e.g. to memorizing and recalling a path in space, a conversation or executing sequences of movements Buhusi and Meck (2005) ; Mita et al. (2009) ; Jazayeri and Shadlen (2015) . An animal will memorize the time to move from a start point A to a landmark B, and from B to a target C, and thereby estimate the total time from A to C. Much effort is devoted to uncovering how the temporal structure of experience on these time scales, i.e. episodic memory, is stored for later recall and predictions of the future Tacikowski et al. (2024) ; Saponati and Vinck (2023) . Various mechanisms across brain regions have been proposed that enable this encoding Paton and Buonomano (2018) ; Tsao et al. (2022) , involving clocks, activity ramping and neural sequence storage Zhou and Buonomano (2022) . The storage of interval duration between current and future times, known as prospective timing, may engage neural timers that measure the passage of time Tsao et al. (2022) . During ramping, timing information is possibly encoded in the increasing firing rates of neurons between start and stop times. Sequence memory could use high-dimensional neural activity trajectories to represent times or time intervals. Analogs of time and ramping cells emerge in deep reinforcement learning models performing simulated interval memory tasks, but these models are agnostic to the actual mechanisms Lin et al. (2023) . In mammals, spatiotemporal information is available in the hippocampus, mainly area CA1 where time cells fire at given points in time, complementing place cells that fire when the animal is at given points in space Eichenbaum (2014) . Ongoing work seeks to explain how this information is integrated to produce sequences of time and place-specific activations of neural ensembles over hours and days Rubin et al. (2015) ; Haimerl et al. (2019) . Time cells can in principle represent intervals between contiguous events Itskov et al. (2011) ; MacDonald et al. (2011) ; Eichenbaum (2014) , although the associated encoding and decoding mechanisms are not well understood. There is also evidence that representation of time duration starts before the hippocampus in sensory cortex Reinartz et al. (2024) . A new paradigm of time coding has recently been identified in a thalamic structure in weakly electric fish known as the preglomerular complex (PG). It projects to dorso-lateral pallium Wallach et al. (2018) , a likely homologue of the mammalian hippocampus Rodríguez-Expósito et al. (2017) . PG cells use a counter-intuitive, starkly different algorithm and biophysical substrate, offering a new window into time coding in mammals as they have similar structures. PG cells receive electrosensory information about their murky nocturnal environment from the optic tectum Giassi et al. (2012) . The electrosense has much in common with vision Clarke et al. (2015) : the electroreceptors in the skin read out perturbations to the fish’s self-generated electric field caused by objects in the environment Jun et al. (2016) . Whenever the fish encounters an object, many PG cells produce brief bursts of firings, regardless of which body part came closest to the object. The number of spikes in a burst depends on the time since the last encounter, and thus the sequence of bursts can be mapped to the sequence of intervals between encounters Wallach et al. (2018) . Adaptation has been extensively studied for various stimuli Wark et al. (2007) and on a wide range of time scales Drew and Abbott (2006) . For example, stimuli activating visual circuitry give rise to adaptation to contrast, orientation, or motion of the stimulus Kohn (2007) ; Clifford et al. (2007) ; Webster (2015) . Closely related paradigms attempt to explain the role of adaptation Weber et al. (2019) , such as efficient coding, by which neurons optimize the representation of stimuli given limited resources Tring et al. (2023) . Another such paradigm is predictive coding, by which neural systems need to make inferences about possible future stimuli Mao et al. (2025) . This type of adaptive response is implicitly dependent on the time elapsed between stimuli. Here, we present an adaptation mechanism which explicitly encodes the time elapsed between identical stimuli, i.e. encounters with surrounding objects. This elegant, tantalizing yet puzzling mechanism relies on deterministic and stochastic dynamics. Here we develop its general theory. We specifically address how it can optimally handle the diversity of prior distributions of time intervals, and the computational benefits of the strong heterogeneity of parameters across PG cells. Two-thirds of adaptive PG cells are sensitive to the history of recent encounters, and accordingly, have recovery time constants of tens of seconds. The other third can encode only the last interval. Why are both present? These questions relate to the fundamental problem of optimal information processing in the face of available biophysical diversity. Heterogeneity has attracted much attention recently as an essential part of coding schemes Gast et al. (2024) ; Mejias and Longtin (2012) ; Marsat et al. (2012) . Our work presents a deep analysis of these questions, and proves that heterogeneity is in fact required in order to properly encode more than one interval in a sequence. We first explain the adaptive stochastic firing model and