Spatial Cognition
Which-Way-Back Game
Also called: egocentric orientation task, point back to the start task, return direction task, homing direction task
Measure orientation from one's own point of view in a game format: participants watch a short walk through a dark corridor, then point back to where it started.
6-9 minUntimed
Free to start: 10 free participants per account, then $0.99 per participant for academic research, or any yearly plan. Try it here first, with no account.
- Papers, 2021 to 2025
- 5
- Evidence grade
- C
- Registered clinical studies
- 3
Figures for the Route Directions Task, the published test this game is compared with.
How these are countedWhat it measures
A game-format measure of orientation from one's own point of view. Participants watch short walks through a dark corridor, seen from above in a view that turns with the walker, then point back to where each walk started. The main score is the mean pointing error in degrees. It is a new measure: its reliability and norms come from use.
Constructs
- Orientation from one's own point of view
- Keeping track of a start
- Spatial updating
Research uses
- Research on spatial orientation across adult ages
- Research on spatial orientation in older adults and in patients
- Within-study group comparison
How the paradigm works
The Which-Way-Back Game is a game-format measure of orientation from one's own point of view: keeping track of where a walk started. Each route opens with the walker on its start square, inside a white ring, for 2 seconds. The walk then plays by itself through a dark corridor of a grid world seen from above, one square a second, and the view turns with the walker at each turn, over one second, so straight ahead is always up. Only the squares next to the walker are drawn. At the end a ring appears round the walker with the question "Which way back to the start?". The participant taps anywhere to point an arrow from the walker toward the start, or turns it with the left and right arrow keys in steps of 15 degrees, then confirms with This way. Tasks in which people watch a walk from their own point of view and then say where the start lies are used in spatial orientation research (Tu and colleagues, 2015; Puthusseryppady and colleagues, 2022, whose tablet version shows 14 filmed walks). A walking version, in which people walk a route and return to its start, needs a tracked room of 3.5 by 3.5 metres (Howett and colleagues, 2019). This game is an original grid world built by ConductScience on the Wayfinding Game's world, and nothing is taken from those tasks' films or rooms.
More about the paradigm
A session has 2 practice routes of 1 turn, after which the true direction is shown, and 14 test routes of 1, 2 and 3 turns, easiest first, which show nothing. Each route walks 8 to 17 squares, so a walk takes 9 to 20 seconds. Each session runs one of two forms: the second mirrors every route, so each turn goes the other way. The form follows the person's earlier sessions of the game in the study, so a retest 7 to 28 days later, run as a visit of the same study, gets the other form; a retest run as a second study can repeat the first form.
The main score is the pointing error: the angle between the answer and the true direction of the start, from 0 to 180 degrees, averaged over the test routes, so lower is better and pointing at random averages 90. The median error and the routes scored are also reported. A route during which the tab was hidden is left out. Nothing rewards speed. A game look and a plain look draw the same marks and differ only in the ground and the wall surfaces, and each session records which look ran. Scores are research measures.
Scores
- Pointing Errordegrees
- The angle between each answer and the true direction of the start, from 0 to 180 degrees, averaged over the test routes scored.
- Lower is better; pointing at random averages 90
- Median Pointing Errordegrees
- The median of the same angles over the test routes scored.
- Lower is better
- Routes Scoredcount
- Test routes with an answer, out of 14. A route during which the tab was hidden is left out.
- Fewer than 10 invalidate the session
Use and evidence
5 papers from 2021 to 2025 name this task in their title or abstract. That ranks it 561 of 582 test families in our census. These counts belong to the Route Directions Task, the published test this game is compared with. This game is a new measure, so its own reliability and norms come from use. See the counts for every test.
Show the numbers
| Year | Papers |
|---|---|
| 2025 | 0 |
| 2024 | 1 |
| 2023 | 1 |
| 2022 | 2 |
| 2021 | 1 |
| 2020 | 0 |
| 2019 | 1 |
| 2018 | 1 |
| 2017 | 3 |
| 2016 | 1 |
| 2015 | 1 |
| 2014 | 1 |
| 2013 | 1 |
| 2012 | 1 |
| 2011 | 0 |
| 2010 | 0 |
| 2009 | 0 |
| 2008 | 0 |
| 2007 | 0 |
| 2006 | 0 |
| 2005 | 0 |
| 2004 | 0 |
| 2003 | 1 |
| 2002 | 0 |
| 2001 | 2 |
| 2000 | 0 |
Use grade E
Fewer than 30 papers from 2021 to 2025.
Evidence grade C
At least 2 test-retest papers, 30 reliability or validity papers or 3 registered clinical studies.
- Papers on test-retest reliability
- 2
- Papers on reliability or validity
- 2
- Registered clinical studies using it as an outcome
- 3
Measured on ConductCognition
Pooled across the studies that feed our platform norms, by device group. A figure needs 50 or more people.
No pooled figure yet.
How the grades work
Use: A is 1,000 or more papers from 2021 to 2025, B 300 to 999, C 100 to 299, D 30 to 99 and E fewer than 30. Evidence: A means at least 25 papers report its test-retest reliability and at least 25 registered clinical studies use it as an outcome; B, at least 8 test-retest papers, and either 5 registered clinical studies or 150 reliability or validity papers; C, at least 2 test-retest papers, 30 reliability or validity papers or 3 registered clinical studies; D, little published measurement evidence found. The counts measure use. They include every paper that names the task in its title or abstract. Grades describe the published research on the Route Directions Task, the published test this game is compared with, not the game's own.
Source: ConductCognition test census 0.39.0, counted 2 November 2026 from OpenAlex and ClinicalTrials.gov.
Task design
The trial sequence as the test runs in ConductCognition.
- Routes
- 2 practice routes of 1 turn, then 14 test routes, the same for everyone and in this order: 4 with 1 turn, 5 with 2 turns and 5 with 3 turns. Each walks 8 to 17 squares of a corridor in the Wayfinding Game's grid world, in legs of 2 to 6 squares with 90 degree turns, so a walk takes 9 to 20 seconds.
- Walk
- The start shows under the walker, inside a white ring, for 2 seconds. Then the walk plays by itself at 1 second a square, and at each turn the view turns with the walker over 1 second, so straight ahead is always up. Only the squares within 1 square of the walker are drawn.
- Answer
- A ring appears round the walker with the question "Which way back to the start?". A tap anywhere points the arrow from the walker toward the tap, and the left and right arrow keys turn it by 15 degrees. This way, or Enter, confirms. A second finger or a palm on the screen never moves the arrow, and there is no time limit.
- Feedback
- After a practice answer a blue arrow shows the true direction. Test routes show nothing, and nothing rewards speed.
- Forms
- Two forms: the second mirrors every route, so each turn goes the other way. Each session takes its form from the participant's earlier sessions of the game in the same study, so a retest visit of that study gets the other form.
- Looks
- A game look, with grass on the ground and brick lines on the walls, or a plain look with flat surfaces. The walker, the rings and the arrows are drawn the same in both, and each session records which look ran.
- Data
- Each route's code, the answer, the true direction and the error in degrees, the taps and key presses that set the arrow, the time to confirm, the form and the earlier sessions. The scorer works each error out again from the route.
The Which-Way-Back Game is a new measure, and its own reliability and norms come from use. The tasks cited below show filmed or walked routes; their results are research findings about those tasks and say nothing about this game. Scores count degrees, not time, and support research.
Quality checks
Each session is marked valid, needs review or invalid, with the reason, before you see its scores.
Scored rows: which_way_back_game from the scored run, the 14 test routes, one per route; a route during which the tab was hidden is left out. A pointing answer is not right or wrong, so accuracy is not judged against chance, and routes have no time limit, so none can be missed. Fewer than 10 of 14 routes scored invalidates. A mean error at or above 90 minus 1.645 times 51.96 over the square root of the routes scored (67.2 degrees at 14 routes) is flagged as no better than pointing at random. 12 or more answers within 15 degrees of one direction are flagged. A hidden tab is flagged.
Why these rules
There is no correct option to compare with chance, so accuracy is not judged against chance. An answer is an angle, and pointing at random gives errors spread evenly from 0 to 180 degrees, with a mean of 90 and a standard deviation of 51.96; a mean error that is not below 90 by 1.645 standard errors is no better than chance at one-sided p < .05, so it is flagged and kept, as chance-level answers can be a real result. The start lies in every quarter round the walker on at least 3 routes, so answers bunched in one direction cannot follow the walks; they are flagged. A score over fewer than 10 of the 14 routes cannot be compared with a full one, so it invalidates. A route watched while the tab was hidden may not have been seen, so it is left out. These limits are ConductScience choices: the sources set none.
No percentile. Scores and quality flags only.
Running this test
Settings you can change
| Setting | Standard | Allowed |
|---|---|---|
| Look | Game | Game, Plain |
Every plan can change the look, which never changes how a result is scored.
Need a setting this test does not have?
Devices and languages
- Untimed test: runs on desktops, laptops and tablets. Participant devices
- Languages: English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese.
- Available on: Every plan, including the free plan.
- No percentile. Scores and quality flags only. How norms work
- Recruiting online? See the setup guides for Prolific, Sona Systems, CloudResearch Connect and Qualtrics.
Cite and describe
Participants completed the Which-Way-Back Game in a web browser (ConductCognition task version 1.0.0, scoring version 1.0.0; Tu et al., 2015; Puthusseryppady et al., 2022), with the standard settings (look: Game). The main score was pointing error (the angle between each answer and the true direction of the start, from 0 to 180 degrees, averaged over the test routes scored).
ConductCognition versions: task version 1.0.0, scoring version 1.0.0. Every export records the versions each session used, for a methods section.
- Tu, S., Wong, S., Hodges, J. R., Irish, M., Piguet, O., & Hornberger, M. (2015). Lost in spatial translation: A novel tool to objectively assess spatial disorientation in Alzheimer's disease and frontotemporal dementia. Cortex, 67, 83-94. https://doi.org/10.1016/j.cortex.2015.03.016The method: walks through a virtual shop, filmed from one's own point of view, during which the person keeps track of the starting point (abstract read).
- Puthusseryppady, V., Morrissey, S., Spiers, H., Patel, M., & Hornberger, M. (2022). Predicting real world spatial disorientation in Alzheimer's disease patients using virtual reality navigation tests. Scientific Reports, 12, 13397. https://doi.org/10.1038/s41598-022-17634-wThe tablet version of that task: 14 filmed walks, after each of which the person shows the direction of the start. The results belong to that task and say nothing about this game.
- Coughlan, G., Puthusseryppady, V., Lowry, E., Gillings, R., Spiers, H., Minihane, A. M., & Hornberger, M. (2020). Test-retest reliability of spatial navigation in adults at-risk of Alzheimer's disease. PLOS ONE, 15(9), e0239077. https://doi.org/10.1371/journal.pone.0239077Retest of that task after about 18 months in 59 adults: an intraclass correlation of .72 for its orientation score, a figure about that task and not about this game.
- Howett, D., Castegnaro, A., Krzywicka, K., Hagman, J., Marchment, D., Henson, R., Rio, M., King, J. A., Burgess, N., & Chan, D. (2019). Differentiation of mild cognitive impairment using an entorhinal cortex-based test of virtual reality navigation. Brain, 142(6), 1751-1766. https://doi.org/10.1093/brain/awz116A walking version of returning to a start, which needs a tracked room of 3.5 by 3.5 metres; this game is an on-screen form.
Source papers
Questions
How long does the Which-Way-Back Game take?
About 6 to 9 minutes, as listed in the test library.
What device does a participant need?
Untimed test: runs on desktops, laptops and tablets. The browser must be Chrome, Edge, Firefox or Safari.
What scores does it produce?
Pointing Error, Median Pointing Error and Routes Scored. Each session is also marked valid, needs review or invalid, with the reason.
How widely is the Route Directions Task used in research?
These counts belong to the Route Directions Task, the published test this game is compared with. This game is a new measure, so its own reliability and norms come from use. 5 papers from 2021 to 2025 name the task in their title or abstract. That ranks it 561 of 582 test families in the ConductCognition test census (use grade E). Source: ConductCognition test census 0.39.0, counted 2 November 2026 from OpenAlex and ClinicalTrials.gov.
How much reliability evidence does the Route Directions Task have?
These counts belong to the Route Directions Task, the published test this game is compared with. This game is a new measure, so its own reliability and norms come from use. In the published research, 2 papers report the task's test-retest reliability and 3 registered clinical studies use it as an outcome (evidence grade C in the ConductCognition test census).
Does it report a percentile?
No. Scores and quality flags only.
Which settings can a researcher change?
Look (Game, Plain). Every plan can change the look, which never changes how a result is scored.
Which languages can participants use?
Participants can take this test in English, Spanish, French, German, Italian, Portuguese (Brazil), Chinese (Simplified) and Japanese. English is on every plan, and the other languages come with yearly plans. Have a native speaker read the machine-translated text before launch.
Is this a clinical test?
No. ConductCognition is for research use only. It runs, scores and exports the test; clinical interpretation is outside the platform.
Every test: the full test list.
Run this test in your next study
Start free with 10 free participants on any test, then use participant credits or a yearly plan.