Key evidence & numbers
| Finding from the spatial-training synthesis | Interpretation |
|---|---|
| 217 studies examined | Broad collection of spatial interventions |
| Retained analysis: 206 studies, 1,038 effect sizes | Analysis underlying the post-exclusion estimate |
| Average g = 0.47 after outlier exclusion | Moderate standardized training effect |
| Transfer to other spatial tasks | Some generalization within the spatial domain |
| Varied tasks and control conditions | No universal personal training dose or gain |
The synthesis considers durability, moderators and transfer. It is broader than a mental-rotation-only intervention review. Its results support spatial malleability, while not establishing that every brief online game reproduces the average intervention or changes general intelligence. Uttal et al., 2013
What exactly can get better?
The classic mental-rotation experiment measures matching objects at different orientations. Other tests add more items, different distractors or strict time limits. Performance can reflect learning the task's rules and strategies as well as changes in solving efficiency. The outcome needs to specify accuracy, speed and the item set. Shepard et al., 1971 Vandenberg et al., 1978
Retest research documents gains across repeated cognitive tests, so a second attempt alone is not a controlled training experiment. A useful evaluation must distinguish ordinary familiarity from the additional effect of an intervention. Scharfen et al., 2018
Does practice help everyone equally?
The meta-analysis does not support identical personal gains. It examines differences between interventions and participants, including age and sex, but a pooled effect cannot predict one person's improvement. Similarly, historical sex-difference reviews describe task-dependent averages rather than fixed individual ceilings. Uttal et al., 2013 Voyer et al., 1995 Linn et al., 1985
What affects it?
Does improvement transfer beyond the game?
Transfer to untrained spatial tasks is a more informative outcome than improvement only on repeated items, and the synthesis reports such transfer. That remains narrower than improved performance on every real-world spatial activity or a rise in overall intelligence. The similarity between training and assessment matters to what can be claimed. Uttal et al., 2013
How do you avoid mistaking a strategy change for a new capacity?
Record the task version, timing and previous exposure. If you start using a different comparison strategy, that can be useful learning, but describe it as part of the change. A result on one set of shapes does not independently validate a general spatial score. Shepard et al., 1971 Vandenberg et al., 1978
Measurement noise also matters. The reliability-paradox study shows that classic task outcomes can be robust at group level while giving unstable individual differences. A small change between short sessions should not be overinterpreted as a permanent shift. Hedge et al., 2017
What is a realistic practice goal?
Work towards more accurate and efficient solving of comparable unfamiliar items, and keep both outcomes visible. The cited synthesis does not establish a universal daily duration, number of sessions or target score for this browser task. It supports the possibility of learning, not a personalised forecast. Uttal et al., 2013 Scharfen et al., 2018
Avoid interpreting early difficulty through a sex stereotype. The group literature and the broader review of similarities do not determine an individual's potential. Current performance and the ability to learn are different questions. Voyer et al., 1995 Hyde, 2005
The honest benefit claim is therefore specific: spatial training has evidence of average improvements, including some within-domain transfer, under studied conditions. Whether a particular game's practice produces a particular transfer outcome requires direct testing. Uttal et al., 2013
Try the related free test
Mental rotation test
Try the mental rotation test to experience the task. If you practise, record prior attempts and use comparable unfamiliar items when assessing change. The game does not promise IQ gains or the average effect reported across spatial interventions.
Common questions
Does g = 0.47 mean 47% better?
No. It is a standardized mean difference, with outliers excluded in the cited summary. Uttal et al., 2013
Will practice raise my IQ?
The spatial-training result does not establish that outcome for this game. Improvement and transfer must be described at the level actually measured. Uttal et al., 2013
How many minutes should I practise?
These sources do not establish an optimal universal dose for the browser task. A fixed schedule would add unsupported precision. Uttal et al., 2013 Scharfen et al., 2018
Sources
- Voyer et al. (1995). Magnitude of sex differences in spatial abilities: A meta-analysis and consideration of critical variables..
- Linn et al. (1985). Emergence and Characterization of Sex Differences in Spatial Ability: A Meta-Analysis.
- Shepard et al. (1971). Mental Rotation of Three-Dimensional Objects.
- Vandenberg et al. (1978). Mental Rotations, a Group Test of Three-Dimensional Spatial Visualization.
- Uttal et al. (2013). The malleability of spatial skills: A meta-analysis of training studies..
- Hyde (2005). The gender similarities hypothesis..
- Scharfen et al. (2018). Retest effects in cognitive ability tests: A meta-analysis.
- Hedge et al. (2017). The reliability paradox: Why robust cognitive tasks do not produce reliable individual differences.
Explore your daily routines in Helibrain
The app has short tests of reaction time, working memory, self-control, reasoning and creativity. Retake them whenever you like, see your trend, and find habits that may help.
Start free in your browser
