Time, control, and the nervous system 2 2 2 After Time, communication, and the nervous system , Norbert Wiener—(1948)

Caroline Haimerl, Filipe S. Rodrigues, Joseph J. Paton · 2025

Read Time, control, and the nervous system 2 2 2 After Time, communication, and the nervous system , Norbert Wiener—(1948) by Caroline Haimerl, Filipe S. Rodrigues, Joseph J. Paton (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-2503-22917.txt.

Introduction

Caroline Haimerl, Filipe Rodrigues, and Joseph Paton wrote a 2025 review on time, control, and the nervous system. It is tempting to treat time as a sense like sight or sound, since timing obeys some of the same laws. Yet no receptor catches time itself: the brain must derive it from other signals.

The authors frame time as part of a wider task. The world is a layered, shifting system, and the flood of outer and inner signals is the fabric the brain uses to steer acts. Seen that way, timing is a core computation, not one more item in felt life. The review uses timing as a door into how brains build adaptive behavior.

Catalog record

CREATORCaroline Haimerl, Filipe S. Rodrigues, Joseph J. Paton
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE62 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-22917

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Time, control, and the nervous system 2 2 2 After Time, communication, and the nervous system , Norbert Wiener—(1948) Caroline Haimerl, Filipe S. Rodrigues, Joseph J. Paton Abstract Because organisms are able to sense its passage, it is perhaps tempting to treat time as a sensory modality, akin to vision or audition. Indeed, certain features of sensory estimation, such as Weber’s law, apply to timing and sensation alike ( Gibbon, 1977a ; Pardo-Vazquez et al., 2019a ) . However, from an organismal perspective, time is a derived feature of other signals, not a stimulus that can be readily transduced by sensory receptors. Its importance for biology lies in the fact that the physical world comprises a complex dynamical system. The multiscale spatiotemporal structure of sensory and internally generated signals within an organism is the informational fabric underlying its ability to control behavior. Viewed this way, temporal computations assume a more fundamental role than is implied by treating time as just another element of the experienced world ( Paton & Buonomano, 2018a ) . Thus, in this review we focus on temporal processing as a means of approaching the more general problem of how the nervous system produces adaptive behavior. Keywords— timing, control theory, basal ganglia, cerebellum, cerebral cortex, behavior Corresponding authors: [email protected] [email protected] [email protected] Introduction Much of biology can be viewed through the lens of control. Our prokaryotic forbearers, for example, needed to control osmotic balance. Calcium entering a cell through a tear in its membrane is thought to have triggered two processes: ( a ) contraction of a ring of actomyosin to pull the edges of the tear together, and ( b ) exocytosis of vesicles to add new membrane required to patch it ( Brunet & Arendt, 2016a ) . The coupling of a damage signal, rising calcium levels, to the repair response represents a simple form of feedback control, fundamentally reactive in nature. However, if the cell could somehow anticipate membrane damage, it might preempt it. The evolution of mechanosensitive calcium channels that open in response to membrane tension appears to have enabled such anticipatory ability ( Martinac et al., 1990a ) . Membrane stress opens these channels, producing calcium influx, translating what was a sign of membrane damage to an earlier point in time. This can be seen as a logic system that implicitly embeds the prediction that membrane tension leads to membrane damage, hardcoded into the genome by virtue of the stability of the temporal relationships that it exploits. We highlight this example to emphasize that the benefits of predictive control extend beyond nervous systems, representing a general principle in biology ( Baluška & Levin, 2016a ) . Most biological systems are composed of layer upon layer of such mechanisms, involving individual cells, nervous systems, or organisms as a whole. Controllers couple observations to commands to achieve objectives (Figure 1 A). Nervous systems control behavior through interaction with a variety of sensor and effector systems, and multiple lines of research have sought to identify control objectives as a normative means of explaining aspects of behavior, including continuous motor control ( Wolpert et al., 1995a ; Todorov, 2004a ) , discrete action selection ( Daw et al., 2005a ; Cisek, 2011a ) , or behavior in general ( Gibson, 1977a ; Friston, 2010a ) . However, even if the brain as a whole is optimized for a global control objective (e.g., maximizing cumulative expected reward, minimizing free energy), it is clear that this is achieved through the concerted action of distinct interacting modules. We argue that existing computational theories for adaptive behavior either are too general to account for this modularity or focus on only specific neural circuit mechanisms and hence are too narrow to explain the breadth of computations required to account for organismal behavior as a whole. A means of tying together individual hypotheses regarding components of nervous systems into more cohesive, integrative explanations for the neural basis of behavior is currently lacking. Figure 1: A control-oriented view of modular brain architecture highlights the centrality of temporal processing for behavior. A) Canonical control diagram wherein a controller emits time-varying commands that operate through effectors to produce actions. Those actions influence the temporal evolution of the state of the controlled dynamical system. That system evolves as a function f() of its current state x , the actions of the controller, a , and any other external perturbations, k , the system is subject to. The state of the system x passes through sensors to give rise to the controller’s observations, y , of the system. Controllers output commands according to a control law g() that is a function of the controller’s current output commands, c , system observations, y , and inputs defining the current objective or target for the controller, \lambda . We note that controllers are fundamentally constrained by the dynamics of the system they aim to control, effectively embedding information about the temporal structure of the dynamical systems in which they operate into their control laws. B) A traditional windmill serves as a metaphor for logic systems. The rigid coupling of the tail vane to the rotor blades embeds knowledge of the fact that optimal harnessing of the available wind flow into useful mechanical action (e.g., pumping or grinding) is achieved by minimizing side winds hitting the vane. C) A control diagram for the muscle stretch reflex. In this case, the relevant state variable corresponds to muscle length, which can be influenced by either perturbations (the physician’s hammer on the patellar tendon) or actions that result from activation of spinal motor neurons (i.e., muscle contraction). Muscle length is transformed into observations the spinal reflex circuit can receive through muscle spindle organs, where they feed into the control law implemented by the reflex circuit. The parameters of this control law can be influenced by inputs, such as thresholds for activation, that ultimately define some behaviorally relevant objective for the reflex (e.g. your joint is at a particular angle so you need to activate the reflex at a particular range of muscle lengths accordingly). D) Ultimately, nervous systems are composed of layer upon layer of interconnected control schemes such as that described in panel C for the muscle stretch reflex. In some cases, the control laws are extremely flexible, subject to highly complex, modifiable, and experience-dependent parameterization, as is the case with learning systems, but stereotyped circuit architectures within each brain system can be seen as laying down the overall structure and optimality conditions for the control laws contained within. Depending on the brain structure, the upstream supplier of observations regarding the state of the target system might be sensory organs (proprioceptive, exteroceptive, or interoceptive), or nervous transmission from another brain circuit. On the output side, commands may influence motor neurons to control some effector relatively immediately, or provide objectives for downstream controllers, setting their parameters so as to achieve a goal state. In the current review, we explore the idea that time may provide a path towards more unified views of brain function, because both the nervous system as a whole and its constituent parts are controllers that evolved to achieve target states within a dynamical system. A system’s dynamics specify how its

Read the complete text