Hi Gang,

Thank you so much for the help. The suggestions make sense to model each individual's D' score for each condition as a random effect. Yes, for the second model we are looking for the effect of condition with D' as a nuisance covariate.

Would the first model listed still allow all subjects to have the same value for D' for each condition and be modeled as a random effect? The goal is to take the average D' score for each condition to see which brain follow the forgetting curve across all subjects. i.e. the average D' score for each condition predicts brain activity.


For example:
-datatable
SS01 hour 2.283607767 decon_alltime_REML/SS01_decon_results_all_targets_REML+orig[100] \
SS02 hour 2.283607767 decon_alltime_REML/SS02_decon_results_all_targets_REML+orig[100] \
SS04 hour 2.283607767 decon_alltime_REML/SS04_decon_results_all_targets_REML+orig[100] \
SS01 day 2.010995168 decon_alltime_REML/SS01_decon_results_all_targets_REML+orig[103] \
SS02 day 2.010995168 decon_alltime_REML/SS02_decon_results_all_targets_REML+orig[103] \
SS04 day 2.010995168 decon_alltime_REML/SS04_decon_results_all_targets_REML+orig[103] \
SS01 week 1.149194158 decon_alltime_REML/SS01_decon_results_all_targets_REML+orig[106] \
SS02 week 1.477045692 decon_alltime_REML/SS02_decon_results_all_targets_REML+orig[106] \
SS04 week 1.477045692 decon_alltime_REML/SS04_decon_results_all_targets_REML+orig[106] \
SS01 month 1.149194158 decon_alltime_REML/SS01_decon_results_all_targets_REML+orig[109] \
SS02 month 1.149194158 decon_alltime_REML/SS02_decon_results_all_targets_REML+orig[109] \
SS04 month 1.149194158 decon_alltime_REML/SS04_decon_results_all_targets_REML+orig[109] \



Edited 1 time(s). Last edit at 01/23/2020 05:26PM by Catherine Tallman.