Computational modeling reveals cognitive processes in simple rodent depression tests methods
Show snapshot details
On this page
This experiment, in seven questions
Jump straight to the part of the recipe you need. Data and provenance labels stay close to the action they support.
Shopping and prep list
What do I need before I start?
Biological model pending
Subject model for the experiment.
- Use
- confirm full cohort details in the source paper
Before you run
What should be confirmed before execution?
First confirmation
Species or subject information is missing.
Confirm before execution
Protocol does not list structured equipment.
Confirm before execution
This page shows passages quoted from the paper. Equipment, materials, times, and temperatures are hidden until the page is rebuilt from the full paper and checked by a reviewer. Check the paper before using this method.
Confirm before execution
Open the source paper before finalizing run-specific details.
Procurement checkpoint
Use source-stated vendors where present. Treat mapped products as sourcing options unless the page marks an exact source match.
Open quote workflowStep-by-step procedure
What do I do, in order?
Data and code availability
Any additional information required to re-analyze the data reported in this study is available from the upon request.
Behavioral trajectories in the FST and TST
To induce depression-like behaviors, we employed two experimental manipulations: chronic restraint stress ([CRS] A) and chemogenetic inhibition of the medial prefrontal cortex (mPFC) pyramidal neurons,,,, ( H). Average behavioral trajectories for CRS are presented in B and 3C, while those for mPFC inhibition are shown in I and 3J. Notably, the probabilities of swimming and struggling typically declined after test onset, with swimming showing a rebound primarily in the FST, whereas late-stage variability increased in both the FST and TST. In the FST, both CRS and mPFC inhibition led to reduced swimming behavior in the later stages compared to controls ( B and 3I). In the TST, CRS produced a reduction in struggling behavior during the early stages ( C). Conventional analyses of immobility times corroborate these findings and align with the existing literature ( D-3G and 3KR...
Behavioral trajectories in the FST and TST
(A) Experimental setup for CRS. Twenty-nine C57BL/6 mice were divided into two groups; the CRS group ( n = 13) underwent daily restraint stress (6-8 h/day) for 21 days, while controls ( n = 16) experienced no stress. Behavioral assessments were conducted on days 22 (FST) and 24 (TST).
Behavioral trajectories in the FST and TST
(D-G) Immobility times for the FST (D and E) and TST (F and G) in the CRS experiment are shown for the entire testing period (D and F) and the final 4 min (E and G). Blue and orange bars represent the control and CRS group, respectively, while black dots indicate individual animal data. Error bars denote the SEM.
Behaviors in the FST and TST are driven by different reinforcement learning processes
(C-F) Model-predicted vs. observed proportions of immobility (CRS experiment). Predictions for the (C and D) FST and (E and F) TST. Each pair compares model predictions with observed data over the full testing period (C and E) and the final 4 min (D and F). Colored bars indicate median predictions from the winning models, with error bars denoting the 95% highest density interval (HDI). Red dots present the observed group means, showing close alignment between model predictions and empirical data.
Behaviors in the FST and TST are driven by different reinforcement learning processes
To evaluate the predictive power of these winning models, we conducted posterior predictive checks. As shown in C-4F and 4M-4P, both models successfully captured the observed immobility proportions across all groups, both over the entire task and during the final 4 min. Nearly all individual immobility times fell within the predicted 95% highest density interval ( G-4J and 4Q-4T), with strong correlations between predicted and observed immobility times across all experimental conditions and time windows (r > 0.95, p < 10 -5, Pearson correlation). K, 4L, 4U, and 4V demonstrate that the models reliably predicted each mouse's latency to immobility (r > 0.71, p < 0.001, Pearson correlation). Moreover, the winning models accurately replicated the mice's behavioral trajectories ( W-4Z), capturing the dynamic evolution of behavior over time. C...
CRS and mPFC inhibition increase immobility by altering learning and consequence sensitivity
Within the framework of reinforcement learning, model parameters represent distinct cognitive components, thereby clarifying how specific factors contribute to increased immobility. In the FST ( A and 5B), both the CRS and hM4D groups demonstrated significant reductions in the learning rate (α) and the lower bound of consequence sensitivity adaptation (lb), alongside an elevated adaptation rate (β), compared to the controls. Additionally, the hM4D group exhibited a higher inverse temperature (τ). In the TST ( C and 5D), both CRS and hM4D groups displayed lower τ values than controls, while the CRS group further demonstrated increased scaling factor (κ) and attention updating rate (η), together with decreased consequence sensitivity (ρ). Figure 5 Uncovering cognitive processes driving immobility via model parameters
CRS and mPFC inhibition increase immobility by altering learning and consequence sensitivity
(A and B) Parameter distributions of the RW + ada model for the FST. This model comprises five parameters: the learning rate (α), which determines how rapidly animals update expectations; the inverse temperature (τ), which regulates the impact of expected values on decision-making; the lower bound of adaptation (lb), reflecting intensified negative experiences from repeated failed escape attempts; the upper bound of adaptation (ub), representing diminished negative perception after sustained immobility; and the adaptation rate (β), which dictates the speed of adaptation. Individual data points representing the mean of 2,000 samples per mouse, with group-level medians and interquartile ranges depicted by boxplots. Control, CRS, and hM4D groups are represented in blue, orange, and magenta, respectively.
Measurement outputs
What raw and processed outputs should exist?
Analysis plan
How should the outputs become interpretable results?
Acquisition
Collect raw experimental outputs with enough metadata to preserve sample identity, condition, and timing.
inferred from protocolPreprocessing / cleaning
Review raw outputs for quality, remove unusable captures, and organize the data into a comparison-ready table or image set.
inferred from protocolScoring or quantification
Quantify the primary endpoint described in the methods before moving into any statistical comparison.
inferred from protocolStatistical comparison
Statistical method not yet structured for this page.
inferred from protocolReporting output
Report representative outputs alongside the summary comparison used to support the conclusion.
inferred from protocolSource and audit
What supports the facts on this page?
Evidence quotes (8)
Any additional information required to re-analyze the data reported in this study is available from the upon request.
To induce depression-like behaviors, we employed two experimental manipulations: chronic restraint stress ([CRS] A) and chemogenetic inhibition of the medial prefrontal cortex (mPFC) pyramidal neurons,,,, ( H). Average behavioral trajectories for CRS are presented in B and 3C, while those for mPFC inhibition are shown in I and 3J. Notably, the probabilities of swimming and struggling typically declined after test onset, with swimming showing a rebound primarily in the FST, whereas late-stage variability increased in both the FST and TST. In the FST, both CRS and mPFC inhibition led to reduced swimming behavior in the later stages compared to controls ( B and 3I). In the TST, CRS produced a reduction in struggling behavior during the early stages ( C). Conventional analyses of immobility times corroborate these findings and align with the existing literature ( D-3G and 3K-3N), supporting the reliability of our behavioral trajectory data. Furthermore, the CRS group exhibited significantly shorter inter-struggle intervals in both the FST ( E) and TST ( G). In the TST, CRS also reduced struggle bout duration ( K), with a similar trend observed in the FST ( I). By con...
(A) Experimental setup for CRS. Twenty-nine C57BL/6 mice were divided into two groups; the CRS group ( n = 13) underwent daily restraint stress (6-8 h/day) for 21 days, while controls ( n = 16) experienced no stress. Behavioral assessments were conducted on days 22 (FST) and 24 (TST).
(D-G) Immobility times for the FST (D and E) and TST (F and G) in the CRS experiment are shown for the entire testing period (D and F) and the final 4 min (E and G). Blue and orange bars represent the control and CRS group, respectively, while black dots indicate individual animal data. Error bars denote the SEM.
(C-F) Model-predicted vs. observed proportions of immobility (CRS experiment). Predictions for the (C and D) FST and (E and F) TST. Each pair compares model predictions with observed data over the full testing period (C and E) and the final 4 min (D and F). Colored bars indicate median predictions from the winning models, with error bars denoting the 95% highest density interval (HDI). Red dots present the observed group means, showing close alignment between model predictions and empirical data.
To evaluate the predictive power of these winning models, we conducted posterior predictive checks. As shown in C-4F and 4M-4P, both models successfully captured the observed immobility proportions across all groups, both over the entire task and during the final 4 min. Nearly all individual immobility times fell within the predicted 95% highest density interval ( G-4J and 4Q-4T), with strong correlations between predicted and observed immobility times across all experimental conditions and time windows (r > 0.95, p < 10 -5, Pearson correlation). K, 4L, 4U, and 4V demonstrate that the models reliably predicted each mouse's latency to immobility (r > 0.71, p < 0.001, Pearson correlation). Moreover, the winning models accurately replicated the mice's behavioral trajectories ( W-4Z), capturing the dynamic evolution of behavior over time. Collectively, these results confirm that the winning models robustly simulate mouse behavior.
Within the framework of reinforcement learning, model parameters represent distinct cognitive components, thereby clarifying how specific factors contribute to increased immobility. In the FST ( A and 5B), both the CRS and hM4D groups demonstrated significant reductions in the learning rate (α) and the lower bound of consequence sensitivity adaptation (lb), alongside an elevated adaptation rate (β), compared to the controls. Additionally, the hM4D group exhibited a higher inverse temperature (τ). In the TST ( C and 5D), both CRS and hM4D groups displayed lower τ values than controls, while the CRS group further demonstrated increased scaling factor (κ) and attention updating rate (η), together with decreased consequence sensitivity (ρ). Figure 5 Uncovering cognitive processes driving immobility via model parameters
(A and B) Parameter distributions of the RW + ada model for the FST. This model comprises five parameters: the learning rate (α), which determines how rapidly animals update expectations; the inverse temperature (τ), which regulates the impact of expected values on decision-making; the lower bound of adaptation (lb), reflecting intensified negative experiences from repeated failed escape attempts; the upper bound of adaptation (ub), representing diminished negative perception after sustained immobility; and the adaptation rate (β), which dictates the speed of adaptation. Individual data points representing the mean of 2,000 samples per mouse, with group-level medians and interquartile ranges depicted by boxplots. Control, CRS, and hM4D groups are represented in blue, orange, and magenta, respectively.
Machine-readable layer
[
{
"@context": "https://schema.org",
"@type": "HowTo",
"name": "Computational modeling reveals cognitive processes in simple rodent depression tests methods",
"description": "Protocol for Computational modeling reveals cognitive processes in simple rodent depression tests methods",
"step": [
{
"@type": "HowToStep",
"position": 1,
"name": "Data and code availability",
"text": "Any additional information required to re-analyze the data reported in this study is available from the upon request."
},
{
"@type": "HowToStep",
"position": 2,
"name": "Behavioral trajectories in the FST and TST",
"text": "To induce depression-like behaviors, we employed two experimental manipulations: chronic restraint stress ([CRS] A) and chemogenetic inhibition of the medial prefrontal cortex (mPFC) pyramidal neurons,,,, ( H). Average behavioral trajectories for CRS are presented in B and 3C, while those for mPFC inhibition are shown in I and 3J. Notably, the probabilities of swimming and struggling typically declined after test onset, with swimming showing a rebound primarily in the FST, whereas late-stage variability increased in both the FST and TST. In the FST, both CRS and mPFC inhibition led to reduced swimming behavior in the later stages compared to controls ( B and 3I). In the TST, CRS produced a reduction in struggling behavior during the early stages ( C). Conventional analyses of immobility times corroborate these findings and align with the existing literature ( D-3G and 3KR..."
},
{
"@type": "HowToStep",
"position": 3,
"name": "Behavioral trajectories in the FST and TST",
"text": "(A) Experimental setup for CRS. Twenty-nine C57BL/6 mice were divided into two groups; the CRS group ( n = 13) underwent daily restraint stress (6-8 h/day) for 21 days, while controls ( n = 16) experienced no stress. Behavioral assessments were conducted on days 22 (FST) and 24 (TST)."
},
{
"@type": "HowToStep",
"position": 4,
"name": "Behavioral trajectories in the FST and TST",
"text": "(D-G) Immobility times for the FST (D and E) and TST (F and G) in the CRS experiment are shown for the entire testing period (D and F) and the final 4 min (E and G). Blue and orange bars represent the control and CRS group, respectively, while black dots indicate individual animal data. Error bars denote the SEM."
},
{
"@type": "HowToStep",
"position": 5,
"name": "Behaviors in the FST and TST are driven by different reinforcement learning processes",
"text": "(C-F) Model-predicted vs. observed proportions of immobility (CRS experiment). Predictions for the (C and D) FST and (E and F) TST. Each pair compares model predictions with observed data over the full testing period (C and E) and the final 4 min (D and F). Colored bars indicate median predictions from the winning models, with error bars denoting the 95% highest density interval (HDI). Red dots present the observed group means, showing close alignment between model predictions and empirical data."
},
{
"@type": "HowToStep",
"position": 6,
"name": "Behaviors in the FST and TST are driven by different reinforcement learning processes",
"text": "To evaluate the predictive power of these winning models, we conducted posterior predictive checks. As shown in C-4F and 4M-4P, both models successfully captured the observed immobility proportions across all groups, both over the entire task and during the final 4 min. Nearly all individual immobility times fell within the predicted 95% highest density interval ( G-4J and 4Q-4T), with strong correlations between predicted and observed immobility times across all experimental conditions and time windows (r > 0.95, p < 10 -5, Pearson correlation). K, 4L, 4U, and 4V demonstrate that the models reliably predicted each mouse's latency to immobility (r > 0.71, p < 0.001, Pearson correlation). Moreover, the winning models accurately replicated the mice's behavioral trajectories ( W-4Z), capturing the dynamic evolution of behavior over time. C..."
},
{
"@type": "HowToStep",
"position": 7,
"name": "CRS and mPFC inhibition increase immobility by altering learning and consequence sensitivity",
"text": "Within the framework of reinforcement learning, model parameters represent distinct cognitive components, thereby clarifying how specific factors contribute to increased immobility. In the FST ( A and 5B), both the CRS and hM4D groups demonstrated significant reductions in the learning rate (α) and the lower bound of consequence sensitivity adaptation (lb), alongside an elevated adaptation rate (β), compared to the controls. Additionally, the hM4D group exhibited a higher inverse temperature (τ). In the TST ( C and 5D), both CRS and hM4D groups displayed lower τ values than controls, while the CRS group further demonstrated increased scaling factor (κ) and attention updating rate (η), together with decreased consequence sensitivity (ρ). Figure 5 Uncovering cognitive processes driving immobility via model parameters"
},
{
"@type": "HowToStep",
"position": 8,
"name": "CRS and mPFC inhibition increase immobility by altering learning and consequence sensitivity",
"text": "(A and B) Parameter distributions of the RW + ada model for the FST. This model comprises five parameters: the learning rate (α), which determines how rapidly animals update expectations; the inverse temperature (τ), which regulates the impact of expected values on decision-making; the lower bound of adaptation (lb), reflecting intensified negative experiences from repeated failed escape attempts; the upper bound of adaptation (ub), representing diminished negative perception after sustained immobility; and the adaptation rate (β), which dictates the speed of adaptation. Individual data points representing the mean of 2,000 samples per mouse, with group-level medians and interquartile ranges depicted by boxplots. Control, CRS, and hM4D groups are represented in blue, orange, and magenta, respectively."
}
],
"tool": [],
"supply": [],
"isBasedOn": {
"@type": "ScholarlyArticle",
"headline": "Computational modeling reveals cognitive processes in simple rodent depression tests",
"datePublished": "2025",
"author": [
{
"@type": "Person",
"name": "Zhihan Li"
},
{
"@type": "Person",
"name": "Tianyu Lu"
},
{
"@type": "Person",
"name": "Jiaozhao Yan"
},
{
"@type": "Person",
"name": "Xiang Zhang"
},
{
"@type": "Person",
"name": "Yun-Feng Li"
}
],
"identifier": "10.1038/s41583-018-0054-8"
}
},
{
"@context": "https://schema.org",
"@type": "BreadcrumbList",
"itemListElement": [
{
"@type": "ListItem",
"position": 1,
"name": "Experiments",
"item": "https://replicatescience.com/experiments"
},
{
"@type": "ListItem",
"position": 2,
"name": "Computational modeling reveals cognitive processes in simple rodent depression tests methods",
"item": "https://replicatescience.com/experiments/computational-modeling-reveals-cognitive-processes-in-simple-rodent-depression-tests-methods-zhihan-li-pmc12859488/computational-modeling-reveals-cognitive-processes-in-simple-rodent-depression-t"
}
]
}
]