Information, certainty, and learning methods
Aim. Evidence-backed execution summary for Information, certainty, and learning methods from Information, certainty, and learning.
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?
rat
Subject model for the experiment.
- Use
- confirm full cohort details in the source paper
Apparatus
Rats were trained and tested in 32 Med Associates conditioning chambers distributed equally across four rooms. Twenty-four chambers (Set A) measured 28.5 × 30 x 25 cm (height × length × depth) and the other eight (Set B) were 21 × 30.5 × 24 cm (height × length × depth). Each chamber was...
- Use
- Rats were trained and tested in 32 Med Associates conditioning chambers distributed equally across four rooms. Twenty-four chambers (Set A) measured 28.5 × 30 x 25 cm (height × length × depth) and the other eight (Set B) were 21 × 30.5 × 24 cm (height × length × depth). Each chamber was...
Additional information
Conceptualization, Data curation, Software, Formal analysis, Visualization, Writing - original draft, Writing - review and editing.
- Use
- Conceptualization, Data curation, Software, Formal analysis, Visualization, Writing - original draft, Writing - review and editing.
Trials from onset of responding to peak responding
Software used for acquisition, scoring, statistics, or reporting.
- Use
- The above analyses reveal an overall tendency for the response rate to increase approximately linearly up to the point where the peak response rate is reached. To test how the rate of this increase varied across groups, we calculated the correlation coefficient between the slope of the line for each group, as shown...
Parsing response rates
Software used for acquisition, scoring, statistics, or reporting.
- Use
- Our parsing algorithm recursively extends the length, n e \begin{document}$n_{\mathrm{e}}$\end{document}, of the vector of inter-poke intervals one interval at a time. After each extension, it compares the rate estimate for each successively longer sub-sequence to the rate estimate for the full sequence, using the...
Additional analyses of trials to criterion
Software used for acquisition, scoring, statistics, or reporting.
- Use
- The final method we have used to analyse our data followed the method described by. This was based on a discrimination ratio calculated as the number of responses (R CS ) made during a brief time window (of length 2/15 ths of T ) in the middle of the CS-US interval divided by the same response count plus a baseline...
Before you run
What should be confirmed before execution?
First confirmation
Equipment is listed but no product mappings are linked.
Confirm before execution
This page is backed by a publishable Replication Data Ledger package with zero critical source-verification issues.
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?
Rate of responding as a function of reinforcement rate
After the initial point at which each rat had begun responding to the CS, response rates typically increased as conditioning continued, but there were large differences in how much responding increased. By the end of the experiment, response rates varied by more than three orders of magnitude, from as low as once every several minutes to as high as several times per second. From our analyses (see Supplementary Materials), we realised that the conventional way to compute response rates seriously underestimates the high poke rates observed during some CSs. The conventional calculation of response rate simply divides the poke count by the time interval across which pokes are counted. This implicitly assumes that the durations of the pokes themselves-during which another response cannot be initiated-constitute a negligible fraction of the duration in the denominator. However,...
Rate of responding as a function of reinforcement rate
Our analysis of response rates across the last five sessions uncovered several important findings (very similar results are obtained when using data from the final 10 sessions). First, the response rate during the CS scaled with the reinforcement rate of the CS. This is shown in, where each red dot shows one rat's response rate during the CS plotted against the CS reinforcement rate (1 /T ) for that rat. To extend the evidence for this relationship between response rate and reinforcement rate, the black dots in show each rat's response rate during the ITI plotted against the overall reinforcement rate (1 /C ) for that rat. This reveals that the ITI response rate scales with the overall reinforcement rate and, strikingly, the scaling is similar to that observed for the CS response rate and reinforcement rate. Indeed, a single regression line (solid black line in ) plotted...
Trials from onset of responding to peak responding
Having established the relationship between response rate and reinforcement rate, we next analysed how response rate increased over trials towards its maximum value. Our first analysis assumes that there is a consistent (monotonic) increase in response rate starting from the initial point of acquisition. This analysis followed a method recently described ( ) that uses the slope of the cumulative response rate over trials to identify the trial on which the response rate had reached each decile (from 10 to 90%) of the peak response rate. Based on our earlier analysis (see ), our measure of the CS response rate was calculated by dividing the response count (excluding the first response in each CS presentation) by the total time out of the magazine during the CS (and excluding the latency to the first response). As shown in, the response rate of an individual rat varies greatly from tria...
Trials from onset of responding to peak responding
The slope of the cumulative function can be used to estimate the rat's response rate across conditioning to find when responding had reached a given proportion of the peak response rate. To analyse how response rates changed across the course of conditioning, we extracted a segment of each rat's conditioning data starting from the trial, t 1, on which the response rate during the CS became reliably greater than the ITI response rate and finishing at the trial, t end, on which the response rate reached its peak (according to a moving average with a window width of three sessions). The total change in responding across conditioning was calculated by subtracting the response rate at the start of this segment of trials, R 1, from the peak response rate, R max (at the end of the segment): Δ R = R max - R 1. To identify when responding had increased by 10% of [...
Trials from onset of responding to peak responding
One of us has previously argued that responding appears abruptly when the accumulated evidence that the CS reinforcement rate is greater than the contextual rate exceeds a decision threshold ( ). The new, more extensive data require a more nuanced view. Evidence about the manner in which responding changes over the course of training is to some extent dependent on the analytic method used to track those changes. One method we have used here suggests that responding rises steadily over trials. The other method we have used relies on an information-theoretic measure of divergence to identify discrete points of change (up or down) in the response record. This method suggests the first increment in responding can be large and is usually followed by further, often smaller, increments in responding. At the same time, there is marked within-subject variability in the response rate, character...
The Kullback-Leibler divergence and the n D KL
When there is no divergence, the n D KL is distributed gamma(.5,1) (for proof, see Appendix in ). Thus, we can convert the information-theoretic measure of the strength of the evidence for divergence to the more familiar p -value measure.
Methods
A total of 176 experimentally naive female albino Sprague Dawley rats (8-10 weeks of age) were obtained from the Animal Resources Centre, Perth, Western Australia. They were housed in groups of 4 in split-level ventilated plastic tubs (Techniplast), measuring 40 × 46 × 40 cm (length × width × height), located in an animal research facility at the University of Sydney. They had unrestricted access to water in their home tubs. Three days before commencing the experiment, they were placed on a restricted food schedule. Each day, half an hour after the end of the daily training session, each tub of rats received a ration of their regular dry chow (3.4 kcal/g) equal to 5% of the total weight of all rats in the tub. This amount is approximately equal to their required daily energy intake ( ) and took at least 2 hr to be eaten (but was usually finished within 3 hr). Ra...
Apparatus
Rats were trained and tested in 32 Med Associates conditioning chambers distributed equally across four rooms. Twenty-four chambers (Set A) measured 28.5 × 30 x 25 cm (height × length × depth) and the other eight (Set B) were 21 × 30.5 × 24 cm (height × length × depth). Each chamber was individually enclosed in a sound- and light-resistant wooden shell (Set A) or PVC shell (Set B). The end walls of each chamber were made of aluminum; the sidewalls and ceiling were Plexiglas. The floor consisted of stainless-steel rods, 0.5 cm in diameter, spaced 1.5 cm apart. Each chamber had a recessed food magazine in the center of one end wall, with an infrared LED and sensor located just inside the magazine to record entries by the rat. A small metal cup measuring 3.5 cm in diameter and 0.5 cm deep was fixed on the floor of each food magazine either in the center (Set...
Measurement outputs
What raw and processed outputs should exist?
After the initial point at which each rat had begun responding to the CS, response rates typically increased as conditioning continued, but there were large differences in how m...
- Raw artifact
- Per-sample or per-animal endpoint measurements collected during the experiment
- Processed artifact
- Structured table with cleaned measurements ready for comparison
- Reported as
- Summary statistics and between-group or across-timepoint comparisons
We have adopted a further correction to the estimate of response rates during the CS. This correction was warranted because the distribution of latencies to the first poke after...
- Raw artifact
- Per-sample or per-animal endpoint measurements collected during the experiment
- Processed artifact
- Structured table with cleaned measurements ready for comparison
- Reported as
- Summary statistics and between-group or across-timepoint comparisons
Our analysis of response rates across the last five sessions uncovered several important findings (very similar results are obtained when using data from the final 10 sessions)....
- Raw artifact
- Per-sample or per-animal endpoint measurements collected during the experiment
- Processed artifact
- Structured table with cleaned measurements ready for comparison
- Reported as
- Summary statistics and between-group or across-timepoint comparisons
Having established the relationship between response rate and reinforcement rate, we next analysed how response rate increased over trials towards its maximum value. Our first a...
- Raw artifact
- Per-sample or per-animal endpoint measurements collected during the experiment
- Processed artifact
- Structured table with cleaned measurements ready for comparison
- Reported as
- Summary statistics and between-group or across-timepoint comparisons
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
The experiment described here attempts to elucidate the role of C / T and T in an appetitive Pavlovian conditioning paradigm with rats by distinguishing their impact on the emergence of responding from their effect on the level of responding subsequently acquired after extende...
from paperScoring or quantification
Quantify the primary readouts for this experiment: After the initial point at which each rat had begun responding to the CS, response rates typically increased as conditioning continued, but there were large differences in how m...; We have adopted a further correction to the estimate of response rates during the CS. This correction was warranted because the distribution of latencies to the first poke after...; Our analysis of response rates across the last five sessions uncovered several important findings (very similar results are obtained when using data from the final 10 sessions)....; Having established the relationship between response rate and reinforcement rate, we next analysed how response rate increased over trials towards its maximum value. Our first a....
from paperStatistical comparison
The experiment described here attempts to elucidate the role of C / T and T in an appetitive Pavlovian conditioning paradigm with rats by distinguishing their impact on the emer...; The above analyses reveal an overall tendency for the response rate to increase approximately linearly up to the point where the peak response rate is reached. To test how the r...; A model of the behaviour-generating process must also generate the distributions of reinforcements to acquisition, not simply their central tendencies. Looking at, we can see t...; Several different indices were used to identify when responding to the CS first appeared. The first index involved creating, for each rat, cumulative records of response counts...
from paperReporting output
Report representative outputs alongside summary comparisons for After the initial point at which each rat had begun responding to the CS, response rates typically increased as conditioning continued, but there were large differences in how m..., We have adopted a further correction to the estimate of response rates during the CS. This correction was warranted because the distribution of latencies to the first poke after..., Our analysis of response rates across the last five sessions uncovered several important findings (very similar results are obtained when using data from the final 10 sessions)...., Having established the relationship between response rate and reinforcement rate, we next analysed how response rate increased over trials towards its maximum value. Our first a....
inferred from protocolStructured statistical methods
The experiment described here attempts to elucidate the role of C / T and T in an appetitive Pavlovian conditioning paradigm with rats by distinguishing their impact on the emer...; The above analyses reveal an overall tendency for the response rate to increase approximately linearly up to the point where the peak response rate is reached. To test how the r...; A model of the behaviour-generating process must also generate the distributions of reinforcements to acquisition, not simply their central tendencies. Looking at, we can see t...; Several different indices were used to identify when responding to the CS first appeared. The first index involved creating, for each rat, cumulative records of response counts...
source structuredSource and audit
What supports the facts on this page?
Evidence quotes (8)
After the initial point at which each rat had begun responding to the CS, response rates typically increased as conditioning continued, but there were large differences in how much responding increased. By the end of the experiment, response rates varied by more than three orders of magnitude, from as low as once every several minutes to as high as several times per second. From our analyses (see Supplementary Materials), we realised that the conventional way to compute response rates seriously underestimates the high poke rates observed during some CSs. The conventional calculation of response rate simply divides the poke count by the time interval across which pokes are counted. This implicitly assumes that the durations of the pokes themselves-during which another response cannot be initiated-constitute a negligible fraction of the duration in the denominator. However, this is far from true when there is more than one poke per second. Our analysis showed that the mean duration of a nose poke is 0.5 s, which will curtail the number of responses that can be produced within a fixed interval. Therefore, in our further analyses of response rates, we have corrected the...
Our analysis of response rates across the last five sessions uncovered several important findings (very similar results are obtained when using data from the final 10 sessions). First, the response rate during the CS scaled with the reinforcement rate of the CS. This is shown in, where each red dot shows one rat's response rate during the CS plotted against the CS reinforcement rate (1 /T ) for that rat. To extend the evidence for this relationship between response rate and reinforcement rate, the black dots in show each rat's response rate during the ITI plotted against the overall reinforcement rate (1 /C ) for that rat. This reveals that the ITI response rate scales with the overall reinforcement rate and, strikingly, the scaling is similar to that observed for the CS response rate and reinforcement rate. Indeed, a single regression line (solid black line in ) plotted through all the data accounts for 81% of the variance in log response rates. It is noteworthy that this regression has a slope very close to 1, and fixing the slope at 1 produces only a small loss of explanatory power (R 2 =0.80). Fixing the slope at 1 is theoretically justified because it means tha...
Having established the relationship between response rate and reinforcement rate, we next analysed how response rate increased over trials towards its maximum value. Our first analysis assumes that there is a consistent (monotonic) increase in response rate starting from the initial point of acquisition. This analysis followed a method recently described ( ) that uses the slope of the cumulative response rate over trials to identify the trial on which the response rate had reached each decile (from 10 to 90%) of the peak response rate. Based on our earlier analysis (see ), our measure of the CS response rate was calculated by dividing the response count (excluding the first response in each CS presentation) by the total time out of the magazine during the CS (and excluding the latency to the first response). As shown in, the response rate of an individual rat varies greatly from trial to trial. However, a clearer picture of the overall change in responding over trials can be obtained by plotting the cumulative response count against the cumulative opportunity to respond (cumulative time out of the magazine; ).
The slope of the cumulative function can be used to estimate the rat's response rate across conditioning to find when responding had reached a given proportion of the peak response rate. To analyse how response rates changed across the course of conditioning, we extracted a segment of each rat's conditioning data starting from the trial, t 1, on which the response rate during the CS became reliably greater than the ITI response rate and finishing at the trial, t end, on which the response rate reached its peak (according to a moving average with a window width of three sessions). The total change in responding across conditioning was calculated by subtracting the response rate at the start of this segment of trials, R 1, from the peak response rate, R max (at the end of the segment): Δ R = R max - R 1. To identify when responding had increased by 10% of Δ R, we estimated what the cumulative response count, cum R ', would be at each trial, t, if the rat maintained a fixed level of responding equal to the starting rate plus 10% of Δ R. Thus, cum R' t = R 1 +0.1·Δ R ·cum T t, where cum T t is the cumulative CS-US i...
One of us has previously argued that responding appears abruptly when the accumulated evidence that the CS reinforcement rate is greater than the contextual rate exceeds a decision threshold ( ). The new, more extensive data require a more nuanced view. Evidence about the manner in which responding changes over the course of training is to some extent dependent on the analytic method used to track those changes. One method we have used here suggests that responding rises steadily over trials. The other method we have used relies on an information-theoretic measure of divergence to identify discrete points of change (up or down) in the response record. This method suggests the first increment in responding can be large and is usually followed by further, often smaller, increments in responding. At the same time, there is marked within-subject variability in the response rate, characterized by large steps up and down in the parsed response rates following the initial increment, but this variability tends to decrease across further training, with fewer and smaller steps in both the ITI and CS response rates. We think that the initial large increment reflects an underlying decision...
When there is no divergence, the n D KL is distributed gamma(.5,1) (for proof, see Appendix in ). Thus, we can convert the information-theoretic measure of the strength of the evidence for divergence to the more familiar p -value measure.
A total of 176 experimentally naive female albino Sprague Dawley rats (8-10 weeks of age) were obtained from the Animal Resources Centre, Perth, Western Australia. They were housed in groups of 4 in split-level ventilated plastic tubs (Techniplast), measuring 40 × 46 × 40 cm (length × width × height), located in an animal research facility at the University of Sydney. They had unrestricted access to water in their home tubs. Three days before commencing the experiment, they were placed on a restricted food schedule. Each day, half an hour after the end of the daily training session, each tub of rats received a ration of their regular dry chow (3.4 kcal/g) equal to 5% of the total weight of all rats in the tub. This amount is approximately equal to their required daily energy intake ( ) and took at least 2 hr to be eaten (but was usually finished within 3 hr). Rats on this schedule do not typically lose weight (and never more than 10%) but gain weight only very slowly. All experimental procedures were approved by the Animal Research Authority of the University of Sydney (protocol 2020/1840).
Rats were trained and tested in 32 Med Associates conditioning chambers distributed equally across four rooms. Twenty-four chambers (Set A) measured 28.5 × 30 x 25 cm (height × length × depth) and the other eight (Set B) were 21 × 30.5 × 24 cm (height × length × depth). Each chamber was individually enclosed in a sound- and light-resistant wooden shell (Set A) or PVC shell (Set B). The end walls of each chamber were made of aluminum; the sidewalls and ceiling were Plexiglas. The floor consisted of stainless-steel rods, 0.5 cm in diameter, spaced 1.5 cm apart. Each chamber had a recessed food magazine in the center of one end wall, with an infrared LED and sensor located just inside the magazine to record entries by the rat. A small metal cup measuring 3.5 cm in diameter and 0.5 cm deep was fixed on the floor of each food magazine either in the center (Set A) or offset to the left of center (Set B). Attached to the food magazine was a dispenser delivering 45 mg food pellets (purified rodent pellets; Bioserve, Frenchtown, NJ). Illumination of an LED (Med Associates product ENV-200RL-LED) mounted in the ceiling of the magazine served as the CS. Experim...
Machine-readable layer
[
{
"@context": "https://schema.org",
"@type": "HowTo",
"name": "Information, certainty, and learning methods",
"description": "Evidence-backed execution summary for Information, certainty, and learning methods from Information, certainty, and learning.",
"totalTime": "PT24070M",
"step": [
{
"@type": "HowToStep",
"position": 1,
"name": "Rate of responding as a function of reinforcement rate",
"text": "After the initial point at which each rat had begun responding to the CS, response rates typically increased as conditioning continued, but there were large differences in how much responding increased. By the end of the experiment, response rates varied by more than three orders of magnitude, from as low as once every several minutes to as high as several times per second. From our analyses (see Supplementary Materials), we realised that the conventional way to compute response rates seriously underestimates the high poke rates observed during some CSs. The conventional calculation of response rate simply divides the poke count by the time interval across which pokes are counted. This implicitly assumes that the durations of the pokes themselves-during which another response cannot be initiated-constitute a negligible fraction of the duration in the denominator. However,..."
},
{
"@type": "HowToStep",
"position": 2,
"name": "Rate of responding as a function of reinforcement rate",
"text": "Our analysis of response rates across the last five sessions uncovered several important findings (very similar results are obtained when using data from the final 10 sessions). First, the response rate during the CS scaled with the reinforcement rate of the CS. This is shown in, where each red dot shows one rat's response rate during the CS plotted against the CS reinforcement rate (1 /T ) for that rat. To extend the evidence for this relationship between response rate and reinforcement rate, the black dots in show each rat's response rate during the ITI plotted against the overall reinforcement rate (1 /C ) for that rat. This reveals that the ITI response rate scales with the overall reinforcement rate and, strikingly, the scaling is similar to that observed for the CS response rate and reinforcement rate. Indeed, a single regression line (solid black line in ) plotted..."
},
{
"@type": "HowToStep",
"position": 3,
"name": "Trials from onset of responding to peak responding",
"text": "Having established the relationship between response rate and reinforcement rate, we next analysed how response rate increased over trials towards its maximum value. Our first analysis assumes that there is a consistent (monotonic) increase in response rate starting from the initial point of acquisition. This analysis followed a method recently described ( ) that uses the slope of the cumulative response rate over trials to identify the trial on which the response rate had reached each decile (from 10 to 90%) of the peak response rate. Based on our earlier analysis (see ), our measure of the CS response rate was calculated by dividing the response count (excluding the first response in each CS presentation) by the total time out of the magazine during the CS (and excluding the latency to the first response). As shown in, the response rate of an individual rat varies greatly from tria..."
},
{
"@type": "HowToStep",
"position": 4,
"name": "Trials from onset of responding to peak responding",
"text": "The slope of the cumulative function can be used to estimate the rat's response rate across conditioning to find when responding had reached a given proportion of the peak response rate. To analyse how response rates changed across the course of conditioning, we extracted a segment of each rat's conditioning data starting from the trial, t 1, on which the response rate during the CS became reliably greater than the ITI response rate and finishing at the trial, t end, on which the response rate reached its peak (according to a moving average with a window width of three sessions). The total change in responding across conditioning was calculated by subtracting the response rate at the start of this segment of trials, R 1, from the peak response rate, R max (at the end of the segment): Δ R = R max - R 1. To identify when responding had increased by 10% of [..."
},
{
"@type": "HowToStep",
"position": 5,
"name": "Trials from onset of responding to peak responding",
"text": "One of us has previously argued that responding appears abruptly when the accumulated evidence that the CS reinforcement rate is greater than the contextual rate exceeds a decision threshold ( ). The new, more extensive data require a more nuanced view. Evidence about the manner in which responding changes over the course of training is to some extent dependent on the analytic method used to track those changes. One method we have used here suggests that responding rises steadily over trials. The other method we have used relies on an information-theoretic measure of divergence to identify discrete points of change (up or down) in the response record. This method suggests the first increment in responding can be large and is usually followed by further, often smaller, increments in responding. At the same time, there is marked within-subject variability in the response rate, character..."
},
{
"@type": "HowToStep",
"position": 6,
"name": "The Kullback-Leibler divergence and the n D KL",
"text": "When there is no divergence, the n D KL is distributed gamma(.5,1) (for proof, see Appendix in ). Thus, we can convert the information-theoretic measure of the strength of the evidence for divergence to the more familiar p -value measure."
},
{
"@type": "HowToStep",
"position": 7,
"name": "Methods",
"text": "A total of 176 experimentally naive female albino Sprague Dawley rats (8-10 weeks of age) were obtained from the Animal Resources Centre, Perth, Western Australia. They were housed in groups of 4 in split-level ventilated plastic tubs (Techniplast), measuring 40 × 46 × 40 cm (length × width × height), located in an animal research facility at the University of Sydney. They had unrestricted access to water in their home tubs. Three days before commencing the experiment, they were placed on a restricted food schedule. Each day, half an hour after the end of the daily training session, each tub of rats received a ration of their regular dry chow (3.4 kcal/g) equal to 5% of the total weight of all rats in the tub. This amount is approximately equal to their required daily energy intake ( ) and took at least 2 hr to be eaten (but was usually finished within 3 hr). Ra..."
},
{
"@type": "HowToStep",
"position": 8,
"name": "Apparatus",
"text": "Rats were trained and tested in 32 Med Associates conditioning chambers distributed equally across four rooms. Twenty-four chambers (Set A) measured 28.5 × 30 x 25 cm (height × length × depth) and the other eight (Set B) were 21 × 30.5 × 24 cm (height × length × depth). Each chamber was individually enclosed in a sound- and light-resistant wooden shell (Set A) or PVC shell (Set B). The end walls of each chamber were made of aluminum; the sidewalls and ceiling were Plexiglas. The floor consisted of stainless-steel rods, 0.5 cm in diameter, spaced 1.5 cm apart. Each chamber had a recessed food magazine in the center of one end wall, with an infrared LED and sensor located just inside the magazine to record entries by the rat. A small metal cup measuring 3.5 cm in diameter and 0.5 cm deep was fixed on the floor of each food magazine either in the center (Set..."
}
],
"tool": [
{
"@type": "HowToTool",
"name": "Apparatus"
},
{
"@type": "HowToTool",
"name": "Additional information"
}
],
"supply": [],
"isBasedOn": {
"@type": "ScholarlyArticle",
"headline": "Information, certainty, and learning",
"datePublished": "2026",
"author": [
{
"@type": "Person",
"name": "Justin A Harris"
},
{
"@type": "Person",
"name": "Charles Randy Gallistel"
},
{
"@type": "Person",
"name": "Nathan Holmes"
},
{
"@type": "Person",
"name": "Justin A Harris"
},
{
"@type": "Person",
"name": "Charles Randy Gallistel"
}
],
"identifier": "10.7554/eLife.102155"
}
},
{
"@context": "https://schema.org",
"@type": "BreadcrumbList",
"itemListElement": [
{
"@type": "ListItem",
"position": 1,
"name": "Experiments",
"item": "https://replicatescience.com/experiments"
},
{
"@type": "ListItem",
"position": 2,
"name": "Information, certainty, and learning methods",
"item": "https://replicatescience.com/experiments/information-certainty-and-learning-methods-justin-a-harris-pmc12863717/information-certainty-and-learning"
}
]
}
]