Whether or not to act is determined by distinct signals from motor thalamus and orbitofrontal cortex to secondary motor cortex

Eriko Yoshida, Masashi Kondo, Ken Nakae, Rie Ako, Shin-Ichiro Terada, Natsuki Hatano, Ling Liu, Kenta Kobayashi, Shin Ishii, M MatsuzakiNature Communications 2025 · DOI

TL;DR

This study reveals how the brain decides whether to act or not by showing that the secondary motor cortex (M2) integrates two opposing signals: one from the motor thalamus that signals the value of acting, and another from the orbitofrontal cortex that signals to inhibit action. This balance ultimately dictates the decision to move or remain still.

What Problem Does This Paper Address?

The fundamental decision of "to act or not to act" is a daily occurrence, yet the neural mechanisms underlying this choice, particularly how signals are relayed to the secondary motor cortex (M2) – a key area for initiating movement – remain unclear. While M2 is known to be involved in decision-making and motor initiation, how it receives and processes information about action value and potential inhibition to make this binary choice is not well understood. This paper aims to elucidate the distinct pathways that inform M2 about whether to execute an action or refrain from doing so.

Methods

The researchers employed a combination of behavioral tasks, optogenetics, and in vivo calcium imaging in head-fixed male mice.

  • Behavioral Task: Mice were trained on a two-tone lever-pull task where different tones (A and B) signaled varying probabilities of reward. Crucially, for tone B, mice sometimes chose to pull the lever and sometimes did not, creating a scenario for studying "act or not to act" decisions. A "value-shift task" was also introduced, where the reward probability for tone B was suddenly increased, allowing researchers to study how the brain adapts to changing action values.
  • Optogenetics: This technique was used to selectively activate or inhibit specific neural pathways. ChR2 (for activation) and eNpHR (for inhibition) were expressed in neurons projecting to M2 from the motor thalamus (mTh) and the lateral orbitofrontal cortex (LO). This allowed researchers to directly manipulate the influence of these pathways on lever-pulling behavior.
  • Two-Photon Calcium Imaging: This allowed for the visualization of neural activity in real-time. Researchers imaged the axons of mTh→M2 and LO→M2 pathways, as well as the activity of M2 neurons themselves, while the mice performed the behavioral tasks. This provided insights into how these pathways encode information related to action value and decision-making.
  • Computational Modeling: A behavioral state model was developed to quantify the "relative action value" (ΔQ) of pulling the lever in cue B trials, considering factors like reward history and potential costs.

Datasets

The study utilized data collected from head-fixed male mice performing the described behavioral tasks. No external datasets were used for training or evaluation. The data generated from the behavioral experiments, optogenetic manipulations, and calcium imaging constitute the primary dataset for the study.

Key Insights and Ideas

  • Opposing Signals to M2: The core finding is that M2 receives two distinct and opposing streams of information: a positive "go" signal from the motor thalamus and a negative "stop" signal from the orbitofrontal cortex.
  • Motor Thalamus Signals Action Value: The motor thalamus, receiving input from the basal ganglia, appears to convey information about the value of taking an action, regardless of whether the action is ultimately performed. This signal seems to be excitatory for action.
  • Orbitofrontal Cortex Inhibits Action: The lateral orbitofrontal cortex (LO) sends signals that generally suppress action. A significant portion of LO inputs to M2 showed decreased activity before and during action, suggesting a role in behavioral inhibition.
  • M2 as the Decision Hub: M2 acts as an integration center, weighing the positive action value signal from the thalamus against the inhibitory signal from the LO to make the final decision to act or not.
  • Independent Pathways: The study found no trial-to-trial correlation between the activity of the "action-promoting" VM→M2L1 axons and the "action-inhibiting" LO→M2L1 axons, suggesting these pathways operate largely independently.

Key Results

  • Optogenetic Manipulation: Activating mTh→M2 pathways increased lever-pulling behavior, while activating LO→M2 pathways decreased it. Conversely, inhibiting these pathways had the opposite effect.
  • Calcium Imaging of Axons: VM→M2L1 axons showed increased activity that correlated with the action value, particularly during the cue and response periods, irrespective of whether the mouse pulled the lever. LO→M2L1 axons, specifically a subset (cluster 3), showed decreased activity during lever-pulling trials, and this decrease was not directly tied to action value but rather to the decision to inhibit.
  • M2 Neuron Activity: M2 neurons themselves showed activity patterns that mirrored those of their input pathways, with some neurons exhibiting "pull preference" (akin to VM→M2L1) and others exhibiting "non-pull preference" (akin to LO→M2L1).
  • Value-Shift Task: In response to an increased reward probability for tone B, VM→M2L1 axons accurately reflected the updated action value, while LO→M2L1 axons did not show a similar correlation with the changing value.

Limitations and Future Work

  • Optogenetic Experiment Sample Size: The authors note that the optogenetic experiments had a limited number of animals, which precluded robust statistical tests in some instances.
  • Generalizability: The study focused on a specific two-tone lever-pull task. Further research is needed to determine if these findings generalize to other decision-making tasks, such as go/no-go or stop-signal tasks, and in other species.
  • AAV Transfection Efficiency: The estimation of anterograde vs. retrograde transfection efficiency for AAV1 relied on assumptions about transfection rates, which could introduce some uncertainty.
  • Future Directions: The authors suggest investigating the role of the LO→M2 pathway in other tasks involving inhibition and exploring the interplay between different subtypes of inhibitory neurons in M2. They also propose that the balance between VM and LO signals is crucial for the decision-making process in M2. Further work could also clarify how the action value signal is updated and transmitted from the striatum to the motor thalamus and then to M2.

Summary updated

Abstract

"To act or not to act" is a fundamental decision made in daily life. However, it is unknown how the relevant signals are transmitted to the secondary motor cortex (M2), which is the cortical origin of motor initiation. Here, we found that in a decision-making task in male mice, inputs from the thalamus to M2 positively regulated the action while inputs from the lateral part of the orbitofrontal cortex (LO) negatively regulated it. The motor thalamus that received the basal ganglia outputs transmitted action value-related signals to M2 regardless of whether the animal acted or not. By contrast, a large subpopulation of LO inputs showed decreased activity before and during the action, regardless of the action value. These results suggest that M2 integrates the positive signal of the action value from the motor thalamus with the negative action-biased signal from the LO to finally determine whether to act or not.
BibTeX
@inproceedings{yoshida2025whether,
  author = {Eriko Yoshida and Masashi Kondo and Ken Nakae and Rie Ako and Shin-Ichiro Terada and Natsuki Hatano and Ling Liu and Kenta Kobayashi and Shin Ishii and M Matsuzaki},
  title = {{Whether or not to act is determined by distinct signals from motor thalamus and orbitofrontal cortex to secondary motor cortex}},
  booktitle = {Nature Communications},
  year = {2025},
  doi = {10.1038/s41467-025-58272-w},
}
Figure