Evidence for Simulators #

Every substantive claim on the Simulators page is checked against current research. Here is each claim, how well today’s evidence supports it, and the sources. The full, de-duplicated source list lives on the references page.

Supported · moderate evidence — Simulator-based practice reliably improves real-world performance, but the benefit is solid and moderate rather than dramatic, and it varies considerably with how the simulator is used.

The Hays, Jacobs, Prince & Salas (1992) meta-analysis found positive, moderate transfer of flight-simulator training to aircraft performance, with effects depending on training task and design. The rewrite softens the old page’s ‘powerful tools’ framing to a moderate, conditional effect, which remains the consensus in the training-transfer literature in 2026.

Sources: Hays, R. T., Jacobs, J. W., Prince, C., & Salas, E. (1992), Flight simulator training effectiveness: a meta-analysis, Military Psychology 4(2):63-74 — https://doi.org/10.1207/s15327876mp0402_1 · full reference ›

Supported · moderate evidence — For procedural skills, a basic mock-up made of simple objects can be as effective as, or better than, tapping keys on a PC, because what matters is reproducing the task-relevant cues and physical actions rather than how realistic the equipment looks.

Training-transfer research holds that functional/psychological fidelity (matching the cues, decisions and response demands of the real task) drives transfer more than physical or surface realism. The de-overstated wording (‘can be better for procedural skills’) replaces the old page’s flat ‘may be more effective’ with a properly qualified, supported claim.

Sources: Schmidt, R. A., & Bjork, R. A. (1992), New conceptualizations of practice, Psychological Science 3(4):207-217 — https://doi.org/10.1111/j.1467-9280.1992.tb00029.x · Hays, R. T., Jacobs, J. W., Prince, C., & Salas, E. (1992), Flight simulator training effectiveness: a meta-analysis · full reference ›

Supported · moderate evidence — A cheaper, lower-fidelity simulator that preserves the layout, sequence and decisions of the task can match a high-fidelity one for learning the underlying skill, while costing far less.

Across the simulator-fidelity literature, added physical fidelity often yields little extra skill learning while greatly increasing cost; well-designed low-fidelity practice frequently matches high-fidelity practice on the trained skill. This corrects the old page’s implicit ‘more realistic is better’ framing around full-scale simulators.

Sources: Hays, R. T., Jacobs, J. W., Prince, C., & Salas, E. (1992), Flight simulator training effectiveness: a meta-analysis, Military Psychology — https://doi.org/10.1207/s15327876mp0402_1 · full reference ›

Supported · moderate evidence — Higher simulator fidelity does not reliably produce more learning; added realism mainly adds cost rather than skill compared with a well-designed cheaper simulator.

Meta-analytic and transfer-of-training work finds the fidelity-learning relationship is weak and non-monotonic: beyond the cues the task actually requires, extra physical realism buys little additional transfer. This is the central de-overstatement applied to the full-scale-simulator section.

Sources: Hays, R. T., Jacobs, J. W., Prince, C., & Salas, E. (1992), Flight simulator training effectiveness: a meta-analysis, Military Psychology 4(2):63-74 · full reference ›

Mixed · moderate evidence — Used well, simulator practice may reduce the total training hours needed on real equipment, but this is a possible saving whose size depends on the task and the faithfulness of practice, not a guaranteed one.

Transfer-of-training studies sometimes express results as a transfer-effectiveness ratio (real-equipment hours saved per simulator hour), which is positive on average but variable and frequently well below 1:1. The rewrite replaces the old page’s confident ‘reduces the overall hours’ with an appropriately hedged claim.

Sources: Hays, R. T., Jacobs, J. W., Prince, C., & Salas, E. (1992), Flight simulator training effectiveness: a meta-analysis, Military Psychology · full reference ›

Supported · moderate evidence — Most simulation is part-task training, so one hour in the simulator does not equal one hour of training on the real equipment.

This caveat from the original page is well founded: simulators typically train a subset of the real task, and measured transfer effectiveness is generally less than one-to-one, consistent with the part-task and transfer-of-training literatures. Kept and retained as an honest qualifier.

Sources: Hays, R. T., Jacobs, J. W., Prince, C., & Salas, E. (1992), Flight simulator training effectiveness: a meta-analysis, Military Psychology · full reference ›

Supported · strong evidence — Practising under varied, representative conditions transfers to real situations better than repeatedly rehearsing a single idealised version, even though variability can depress performance during practice.

Variability of practice and other desirable difficulties reliably improve retention and transfer despite often reducing performance during acquisition; this is a robust and current finding that supports spending saved hours on task-variety and overlearning rather than on finishing sooner.

Sources: Schmidt, R. A., & Bjork, R. A. (1992), New conceptualizations of practice, Psychological Science 3(4):207-217 — https://doi.org/10.1111/j.1467-9280.1992.tb00029.x · full reference ›

Mixed · moderate evidence — Skills practised under one set of simulator cues can resurface unhelpfully under different real-world conditions such as stress, because responses learned in practice are partly bound to the practice context.

Follows from the specificity-of-practice and contextual-interference literatures: behaviour learned under particular conditions or supports may not transfer cleanly, and can re-emerge when conditions differ. Supports the original page’s ‘unlearn bad habits / habits return under stress’ caution, with magnitude depending on task and how the cue is used.

Sources: Schmidt, R. A., & Bjork, R. A. (1992), New conceptualizations of practice, Psychological Science 3(4):207-217 · full reference ›

Mixed · weak evidence — Adults cannot be assumed to learn more slowly than children simply because children use imagination; the broad ‘children learn far more in fifteen years than adults in thirty’ comparison is not well supported as a general claim.

This row flags and softens an overstated rhetorical claim in the original page rather than asserting a new positive finding. Child-versus-adult learning rate is domain-specific (children have advantages in areas such as first-language phonology, adults in others), and there is no clean evidence that imagination ‘causes’ faster childhood learning across the board. The rewrite keeps only the defensible, modest point that imaginative, exploratory practice makes learning less of a chore.

Sources: Schmidt, R. A., & Bjork, R. A. (1992), New conceptualizations of practice, Psychological Science 3(4):207-217 · full reference ›