Science & method
MindIQ builds on established intelligence and learning research. Here we openly explain what we measure, how we score, which learning methods we use – and where the limits are.
1. What we mean by intelligence
The most widely accepted description of human cognitive abilities today is the Cattell-Horn-Carroll (CHC) model (Carroll, 1993; McGrew, 2009). It distinguishes a general factor (“g”) and broad abilities such as fluid reasoning, quantitative reasoning, visual-spatial processing and crystallised (verbal) knowledge. The MindIQ test mainly measures fluid reasoning – solving new problems – complemented by numbers, spatial ability and verbal analogies.
2. The tasks
The test consists of 20 originally developed tasks, balanced across six areas. The task types follow proven formats from intelligence research:
- Number sequences – spotting and continuing rules (quantitative and inductive reasoning).
- Logical reasoning – telling valid from invalid conclusions (deductive reasoning).
- Visual patterns and matrices – following the principle of progressive matrices (Raven, 1938).
- Spatial reasoning – imagining rotations and views (cf. mental rotation, Shepard & Metzler, 1971).
- Analytical problem solving – organising and combining information.
- Analogies and language – recognising relations between concepts.
During the test we deliberately do not show whether an answer is correct – feedback could influence later answers. Encouragement depends on progress only.
3. Scoring
We score with the Rasch model (Rasch, 1960), a method from item response theory. Each task has a difficulty; ability is estimated from the pattern of solved and unsolved tasks, so harder tasks count more than easy ones – unlike simply counting correct answers. The ability is mapped onto the usual IQ scale: mean 100, standard deviation 15 (deviation IQ after Wechsler, 1939). Scores range from 55 to 145.
The task difficulties of the current version (iq-v1) are provisional. Once enough results are available, we re-estimate them from real test data. Every score stores its version so results remain traceable.
4. Limits – honestly
- MindIQ is an unsupervised online test. Concentration, time of day, device and surroundings affect the result.
- Every measurement has a margin of error. With 20 tasks it is larger than in tests lasting several hours administered by professionals.
- The test has not yet been normed on a representative population sample. It does not replace psychological assessment.
5. Learning based on research
The courses use learning methods with well-documented effectiveness (overview: Dunlosky et al., 2013):
- Retrieval practice: recalling instead of re-reading clearly improves retention (Roediger & Karpicke, 2006).
- Spaced repetition: the daily review brings content back at growing intervals (Cepeda et al., 2006), using the flash-card box principle (Leitner, 1972).
- Guess first: a guess before learning helps you remember the answer afterwards (Richland et al., 2009).
- Interleaving: mixing task types improves telling them apart and applying them (Rohrer & Taylor, 2007).
- Generating and explaining: filling gaps, ordering and explaining in your own words anchor knowledge (Slamecka & Graf, 1978; Dunlosky et al., 2013).
- Listening and seeing: audio lessons and visuals use several channels (Mayer, 2009).
- Immediate feedback in exercises shows what you already know (Hattie & Timperley, 2007).
6. Brain games
Regular training improves the practised tasks and is fun. Research also shows, however, that broad gains in general intelligence from such training have not been demonstrated (Simons et al., 2016). We therefore do not promise an IQ increase from games.
Prices and conditions are in our Terms. Questions about the method? Write to us – we are happy to explain more.
References
- Carroll, J. B. (1993). Human Cognitive Abilities: A Survey of Factor-Analytic Studies. Cambridge University Press.
- Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., & Rohrer, D. (2006). Distributed practice in verbal recall tasks. Psychological Bulletin, 132(3), 354–380.
- Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students' learning with effective learning techniques. Psychological Science in the Public Interest, 14(1), 4–58.
- Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1), 81–112.
- Leitner, S. (1972). So lernt man lernen. Herder.
- Mayer, R. E. (2009). Multimedia Learning (2nd ed.). Cambridge University Press.
- McGrew, K. S. (2009). CHC theory and the human cognitive abilities project. Intelligence, 37(1), 1–10.
- Rasch, G. (1960). Probabilistic Models for Some Intelligence and Attainment Tests. Danish Institute for Educational Research.
- Raven, J. C. (1938). Progressive Matrices. H. K. Lewis.
- Richland, L. E., Kornell, N., & Kao, L. S. (2009). The pretesting effect. Journal of Experimental Psychology: Applied, 15(3), 243–257.
- Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning. Psychological Science, 17(3), 249–255.
- Rohrer, D., & Taylor, K. (2007). The shuffling of mathematics problems improves learning. Instructional Science, 35, 481–498.
- Shepard, R. N., & Metzler, J. (1971). Mental rotation of three-dimensional objects. Science, 171(3972), 701–703.
- Simons, D. J., Boot, W. R., Charness, N., et al. (2016). Do “brain-training” programs work? Psychological Science in the Public Interest, 17(3), 103–186.
- Slamecka, N. J., & Graf, P. (1978). The generation effect. Journal of Experimental Psychology: Human Learning and Memory, 4(6), 592–604.
- Wechsler, D. (1939). The Measurement of Adult Intelligence. Williams & Wilkins.