Primate Thalamic Nuclei Select Abstract Rules and Shape Prefrontal Dynamics

Academic Background

Cognitive control is the ability to flexibly adjust behavior according to goals and contexts, and the prefrontal cortex (PFC) plays a crucial role in this process. Previous research suggested that the PFC reads out rules from high-dimensional representations of task variables to guide behavior. However, increasing evidence indicates that the thalamus may play a key role in rule selection and information transmission. Specifically, the strong connections between the thalamus and the PFC allow the thalamus to select relevant information from PFC inputs and feed it back to the PFC, thereby influencing subsequent task processing.

This study aims to explore how the thalamus selects abstract rules from PFC inputs and maintains these rule representations through interactions with the PFC. By investigating neural activity in primates during rule-based tasks, the researchers hope to uncover the specific role of the thalamus in cognitive control and elucidate its dynamic interaction mechanisms with the prefrontal cortex.

Source of the Paper

This paper was co-authored by Jessica M. Phillips, Mohsen Afrasiabi, Niranjan A. Kambi, and others. The research team is affiliated with multiple institutions, including the Department of Psychology at the University of Wisconsin-Madison, the Wisconsin National Primate Research Center, and the Department of Neurosurgery at the Renaissance School of Medicine, Stony Brook University. The paper was published on June 18, 2025, in the journal Neuron, titled “Primate Thalamic Nuclei Select Abstract Rules and Shape Prefrontal Dynamics.”

Research Process

1. Experimental Design and Task

The research team designed a Hierarchical Rule Task (HRT), requiring monkeys to sequentially apply two rules: an abstract rule and a concrete rule. The abstract rule specifies a class of subordinate concrete rules, which can be mapped to specific actions or targets. For example, the abstract rule could be “shape” or “orientation,” while the concrete rules are specific instructions under the abstract rule, such as “select the left target” or “select the vertical target.”

2. Neural Recording and Data Analysis

The researchers used Linear Microelectrode Arrays (LMAs) to simultaneously record neural activity in four brain regions while monkeys performed the HRT task: the prefrontal cortex (PFC), the ventroanterior thalamus (VA), the mediodorsal thalamus (MD), and other related regions. Through Diffusion Magnetic Resonance Imaging (Diffusion MRI) and Probabilistic Tractography, the researchers precisely located the recording sites in these brain regions.

3. Data Analysis Methods

The researchers employed multiple methods to analyze neural data, including: - Selectivity Index (SI): Used to quantify neurons’ selectivity for abstract and concrete rules. - Pseudopopulation Decoding: Simulated neural population activity to decode rule information. - Adaptive Granger Causality (AGC): Analyzed causal influences between the thalamus and the PFC.

4. Model Construction and Validation

To further validate the experimental results, the researchers constructed a PFC-Basal Ganglia-Thalamus Model, simulating the selection and maintenance of abstract rules. The model used Leaky Integrate-and-Fire Neurons to simulate neural activity and tested the impact of thalamic lesions on PFC rule representations.

Key Findings

1. The Dominant Role of the Thalamus in Abstract Rule Selection

The study found that abstract rule information first appeared in the ventroanterior thalamus (VA) before being transmitted to the prefrontal cortex (PFC). Specifically, VA neurons exhibited selectivity within 100 milliseconds after the abstract rule cue appeared, while PFC neurons showed selectivity at 250 milliseconds. This indicates that the thalamus plays a dominant role in abstract rule selection.

2. Dynamic Interaction Between the Thalamus and PFC

Through adaptive Granger causality analysis, the researchers found that the thalamus’s causal influence on the PFC significantly increased after the abstract rule cue appeared and remained high during the subsequent delay period. This suggests that the thalamus not only dominates rule selection but also dynamically interacts with the PFC in rule maintenance.

3. Representation of Concrete Rules

Unlike abstract rules, concrete rule representations first appeared in the PFC before being observed in the thalamus. This indicates that once the abstract rule is established, the PFC can effectively transform subsequent inputs into concrete rules and responses.

4. Impact of Thalamic Lesions on PFC Rule Representations

By simulating thalamic lesions, the researchers found that lesions in VA or MD significantly reduced the PFC’s ability to represent abstract rules. Specifically, VA lesions caused the decoding accuracy of rule information in the PFC during the delay period to drop to chance levels, while MD lesions affected the PFC’s ability to maintain rules during the delay period.

Conclusions and Significance

This study reveals the critical role of the thalamus in cognitive control, particularly its dominant position in the selection and maintenance of abstract rules. Through dynamic interactions with the PFC, the thalamus can select relevant information from high-dimensional cortical representations and feed it back to the PFC, influencing subsequent task processing. This finding not only deepens our understanding of the neural mechanisms of cognitive control but also provides new insights for the treatment of related neuropsychiatric disorders.

Research Highlights

  1. First to Reveal the Thalamus’s Dominant Role in Abstract Rule Selection: The study found that the ventroanterior thalamus (VA) exhibits the earliest selectivity in abstract rule selection, challenging the traditional view that the PFC is the center of rule selection.
  2. Dynamic Interaction Mechanisms Between the Thalamus and PFC: Through adaptive Granger causality analysis, the researchers quantified the dynamic interaction between the thalamus and PFC in rule selection and maintenance for the first time.
  3. Multi-level Model Validation: The PFC-Basal Ganglia-Thalamus model constructed by the researchers not only validated the experimental results but also further clarified the impact of thalamic lesions on PFC rule representations through simulations.

Other Valuable Information

The study also found that the thalamus plays a key role in the early representation of behavioral outcomes. Specifically, in error trials, the thalamus’s representation of abstract rules was significantly reduced, further supporting its importance in cognitive control.


Through multi-level experiments and model validation, this study provides an in-depth exploration of the thalamus’s critical role in cognitive control, offering new perspectives on how the brain selects and executes rules.