Table 3.
List of issues that need consideration when designing bioelectromagnetic studies on cognitive performance measures - Impact, importance and recommendations.
| Issue | Impact | Importance | Recommendation(s) |
|---|---|---|---|
| Task type | Detection of effect | High | Use tasks which previously revealed effects and/or address the same cognitive variable (i.e. 'selective attention', 'working memory'); select task related to exposed brain region |
| Input and response modalities | Comparability between studies | Medium | Use modalities applied in tasks that previously revealed effects |
| Learning effects | Error variance and type II error | Medium | Include practice sessions; apply conditions in a crossover design; use parallel test forms or random or pseudo-random sequences |
| Task specificity | Detection of effect | High | Select task related to exposed brain region; if appropriate correct for multiple comparisonsa |
| Timing of tasks, task order and task duration | Development of effect; comparability between studies; alertness and motivation | High | Apply tasks with varying difficulty repeatedly in the same order; chose task difficulty and duration wisely to obviate fatigue and/or motivational loss |
| Study population and sample size | Effect size and significance; external validity | High | Run power analysis to get an indication of sufficiently large sample size; use homogeneous group for small sample sizes; apply proper matching procedures for two or more study groups; characterize study population |
| Handedness | Comparability between studies; homogeneity of study group(s) | Low to Mediumb | Include either right or left handers |
| Inclusion criteria and exclusion criteria | Detection of effect; comparability between studies; homogeneity of study group(s) | Medium to High | Define clear criteria prior and within a study according to the subject of investigation |
| Confounding factors | Error variance and type II error | High | Control for confounding factors as much as possible experimentally; apply randomization; check compliance of participants to predefined requirements |
| Experimental design and blinding | Detection of effect | High | Use within-subject, cross-over design if possible; double blinding is mandatory |
| Exposure conditions | Detection of effect; effect size and significance; interpretation | High | Ensure standardized reproducible exposure conditions; document setup and technical specifications (including signal characteristics); determine exposed brain areas |
| Field conditions and dosimetry | Interpretation | High | Use field intensities close to exposure limits; provide clear definition of field conditions; dosimetry and sham condition (no field) mandatory |
| Exposure duration and carryover effects | Interpretation; detection of effect | High | Allow for sufficient exposure duration; consider potential carryover effects in a crossover design; allow for sufficient time interval ('washout period') between conditions |
It must be noted that Table 3 contains generalized recommendations only and reflects the main issues which should be considered when designing a new study investigating RF EMF effects on mental processing; however, due to the complexity of each issue the recommendations may not fit all purposes by default. For a more elaborate discussion of each issue and the corresponding implications the reader is referred to the respective chapter of the review.
a A substantial sample size is needed to adequately perform such a correction; b of high importance when assessing motor reaction times with a cognitive task.