Table 3.

Key Recommendations for Researchers

General recommendations

Provide precise definitions of key terms (e.g., cognitive training, active control group, near and far transfer).

Avoid piecemeal publication; when this is unavoidable, provide references to the articles sharing the results.

Avoid hyperbole and incorrect generalization.

Use well-specified theories (e.g., computational models) to derive predictions about the potential effectiveness of cognitive training.

Use detailed measures (e.g., eye movements, mouse clicks) to understand the detail of the cognitive mechanisms mediating potential cognitive transfer.

Understand the strategies used by the participants.

Test interventions in silico before testing them in vivo.

Carry out a task analysis of the tasks used in pretest and posttest as well as in training.

Focus on near transfer because far transfer is elusive.

Recommendations about statistics and data curation

Put the data, analysis code, and other relevant information online.

Report results correctly and objectively; do not capitalize on chance with suspect statistical practices.

Reply to requests from meta-analysts asking for summary data and/or the original data.

When estimating latent factors, use multiple measures for each factor.

Randomize the presentation order of the tasks.

Use meta-analytic evidence for assessing the plausibility of cognitive-training interventions.

Pay attention to true heterogeneity in the data for making informed conclusions.