Radial Arm Maze Working Memory
Overview
The radial arm maze (RAM) working memory task evaluates spatial working memory by baiting all arms of an 8-arm radial maze and measuring the animal's ability to visit each arm only once to collect all rewards without re-entering previously visited arms. Each re-entry into a depleted arm constitutes a working memory error because the animal must remember within the session which arms it has already visited. This task places heavy demands on hippocampal-dependent spatial working memory because the animal must maintain a continuously updated representation of visited versus unvisited arms using distal spatial cues, and the memory load increases with each successive arm choice.
The primary dependent measure is the number of working memory errors (arm re-entries) committed before all 8 arms have been visited. The theoretical optimum is 8 choices to visit all arms, so errors are scored as total choices minus 8. Additional metrics include the choice sequence pattern, the number of correct choices in the first 8 entries, the spatial distribution of errors, and the delay tolerance when inter-choice intervals are imposed. Win-shift and win-stay response patterns can also be analyzed to determine whether animals are using spatial or non-spatial strategies. This task requires 5-10 days of food-motivated training before stable performance is achieved.
ConductMaze automates radial arm maze tracking through overhead video and infrared beam-break sensors positioned at the entrance of each arm. The software detects arm entries in real time, maintains the running list of visited arms, flags re-entries as errors, and computes working memory metrics as the session progresses. Configurable arm-locking features can block re-entered arms to implement correction procedures, and the software supports variable delay insertion between choices for delay-dependent working memory analysis.
Trial Flow
Arm Baiting
All 8 arms are baited with food reward at the distal end.
Central Release
Animal is placed in the central hub and allowed to freely choose among all arms.
Choice Recording
Each arm entry is recorded; reward is consumed on first visits and absent on re-entries.
Error Detection
Re-entry into a previously visited arm is flagged as a working memory error.
Choice Pattern Analysis
Sequential arm choices are analyzed for spatial strategy patterns and adjacency biases.
Completion Check
Session ends when all 8 arms are visited or the maximum number of choices is reached.
Metrics Computation
Total errors, correct in first 8, choice sequence, and strategy metrics are computed.
Session End
Animal is removed and data are exported; maze is cleaned for the next subject.
Parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
| Number of Arms | integer | 8 | Total number of arms in the radial maze (standard is 8). |
| Maximum Choices | integer | 16 | Maximum number of arm entries before the session is terminated. |
| Session Timeout | duration | 10min | Maximum session duration regardless of choices made. |
| Reward Type | enum | sucrose-pellet-45mg | Type and size of food reward placed at the end of each arm. |
| Food Restriction Target | percentage | 85% | Target body weight as percentage of free-feeding weight during testing. |
| Arm Entry Criterion | enum | all-four-paws | Criterion for scoring an arm entry: all-four-paws or beam-break at arm midpoint. |
| Arm Length | distance | 45cm | Length of each arm from the central hub to the reward location. |
| Pretraining Days | integer | 7 | Number of days of maze habituation and free-feeding reward exploration before testing. |
| Inter-Choice Delay | duration | 0s | Optional delay imposed between choices by briefly confining the animal in the center hub. |
Metrics
| Metric | Unit | Description |
|---|---|---|
| Working Memory Errors | count | Number of re-entries into previously visited arms before all arms are collected. |
| Correct in First 8 | count | Number of novel arm visits within the first 8 choices (max = 8). |
| Total Choices to Complete | count | Total number of arm entries required to visit all 8 arms. |
| Mean Choice Latency | s | Average time between consecutive arm entries. |
| Adjacent Arm Bias | % | Percentage of choices directed to arms adjacent to the previously visited arm. |
| Win-Shift Ratio | ratio | Proportion of choices that follow a win-shift (avoid previously rewarded location) strategy. |
| Time to Complete | s | Total time from session start to collection of all 8 rewards. |
Sample Data
| Subject | Group | WM Errors | Correct in First 8 | Choices to Complete | Time (s) |
|---|
Representative data for illustration purposes. Actual values will vary by species, strain, and experimental conditions.
Applications
- 1Hippocampal Working Memory — Definitive assessment of spatial working memory with parametrically increasing memory load across choices.
- 2Cholinergic System Studies — Evaluating the role of acetylcholine in spatial working memory using muscarinic and nicotinic antagonists.
- 3Delay-Dependent Analysis — Introducing inter-choice delays to probe the persistence of within-trial spatial working memory representations.
- 4Strategy Analysis — Dissecting spatial versus non-spatial foraging strategies to understand the cognitive basis of maze performance.
- 5Gene-Dose Studies — Detecting working memory impairments in heterozygous knockout mice with intermediate phenotypes.
Related Protocols
Compatible Products
Ready to Automate Your Behavioral Protocols?
Contact us for a demo and pricing information.
