About n-Back Task – Working Memory Assessment Tools
n-Back Task – Working Memory Assessment Tools is a complete implementation of the n-back paradigm, the most widely used laboratory measure of working memory updating. It is built for researchers, students, serious-games developers and anyone who needs valid cognitive data rather than just a memory minigame.
Everything is configurable from the Inspector:
- Choose single, dual or triple n-back;
- combine any of seven stimulus channels — spatial position, letters, numbers, colors, shapes, custom images and audio;
- set your own timing, target ratio, block structure and instruction texts;
- swap in your own sprites and sound files to build a task in any theme, in any language.
The output includes:
- the full signal-detection picture per channel and per block: hits, misses, false alarms, correct rejections, hit and false-alarm rates,
- d-prime, criterion C, and reaction times with standard deviations,
- a per-trial CSV that drops straight into R, Python or SPSS.
The scientifically critical details are handled correctly:
- warm-up trials are excluded from scoring,
- the log-linear correction keeps d-prime finite,
- and empty conditions report as missing.
The asset includes ready-to-run scenes and prefabs for the common designs, so you can press Play within a minute of importing and then adapt the closest variant to your own study.
Optional features include adaptive difficulty that tracks each participant's capacity, n±1 lure trials with separate scoring, a practice block, reproducible sessions via a fixed random seed, and Unity Events for hooking the task into your own game logic.
Suitable for cognitive psychology and neuroscience research, student projects and theses, brain-training and serious games, VR research environments, and any project that needs a validated measure of working memory.
REQUIREMENTS
This package requires TextMeshPro, which ships with Unity; if the Editor prompts you to import TMP Essential Resources after import, accept. It uses Unity's legacy Input class, so Project Settings → Player → Active Input Handling must be set to "Input Manager (Old)" or "Both". Under the new Input System alone the task will throw at runtime.
Minimum Unity version is 2021.3 LTS. The package is render-pipeline independent and works with Built-in, URP and HDRP.
VR controller input is optional and requires the SteamVR Unity Plugin(not included; obtain it separately from Valve). Without it, the package compiles and runs normally with keyboard input.
Everything is configurable from the Inspector:
- Choose single, dual or triple n-back;
- combine any of seven stimulus channels — spatial position, letters, numbers, colors, shapes, custom images and audio;
- set your own timing, target ratio, block structure and instruction texts;
- swap in your own sprites and sound files to build a task in any theme, in any language.
The output includes:
- the full signal-detection picture per channel and per block: hits, misses, false alarms, correct rejections, hit and false-alarm rates,
- d-prime, criterion C, and reaction times with standard deviations,
- a per-trial CSV that drops straight into R, Python or SPSS.
The scientifically critical details are handled correctly:
- warm-up trials are excluded from scoring,
- the log-linear correction keeps d-prime finite,
- and empty conditions report as missing.
The asset includes ready-to-run scenes and prefabs for the common designs, so you can press Play within a minute of importing and then adapt the closest variant to your own study.
Optional features include adaptive difficulty that tracks each participant's capacity, n±1 lure trials with separate scoring, a practice block, reproducible sessions via a fixed random seed, and Unity Events for hooking the task into your own game logic.
Suitable for cognitive psychology and neuroscience research, student projects and theses, brain-training and serious games, VR research environments, and any project that needs a validated measure of working memory.
REQUIREMENTS
This package requires TextMeshPro, which ships with Unity; if the Editor prompts you to import TMP Essential Resources after import, accept. It uses Unity's legacy Input class, so Project Settings → Player → Active Input Handling must be set to "Input Manager (Old)" or "Both". Under the new Input System alone the task will throw at runtime.
Minimum Unity version is 2021.3 LTS. The package is render-pipeline independent and works with Built-in, URP and HDRP.
VR controller input is optional and requires the SteamVR Unity Plugin(not included; obtain it separately from Valve). Without it, the package compiles and runs normally with keyboard input.