Frontal theta oscillations and cognitive flexibility: Age-related modulations in EEG activity

Figures

• 

  Fig. 1. The IDED Task. A: Trial structure. After the presentation of the fixation cross for 800 to 1300 ms the stimuli and target appear. The participants signal with a righthand mouse click which stimulus matches the target either by colour or shape (here: left button press to indicate colour match). After button press or after a time of 4000 ms without response the stimuli and target disappear. An empty screen is visible for 1000 ms followed by a new trial; B: Stimuli exemplars showcasing the six possible colour (red, green, blue, beige, magenta, cyan) and shape combinations (circle, star, diamond, triangle, pentagon, cross) resulting in 36 possible stimuli; C: Types of intra-dimensional shifts (ID). ID shifts could either be colour-to-colour (e.g. blue-to-red, top row) or shape-to-shape (e.g. diamond-to-circle, bottom row); D: Types of extra-dimensional shifts (ED). ED shifts could either be colour-to-shape (e.g. blue-to-cross, top row) or shape-to-colour (e.g. diamond-to-green, bottom row); E: Paradigm structure. Here, an example of possible consequent trials is presented. In each trial the target matches one of the stimuli by colour or by shape. In ID or ED trials the target changes…

• 

  Fig. 2. Behavioural results of the IDED task. A: Mean log transformed reaction times (log_RT). There was a significant difference in log_RTs between all conditions in both age groups with ED trials reaching the longest log_RTs followed by ID and repeat (F1.3, 47.94 = 163.57, p < 0.001, ηp2= 0.82). Across conditions, older participants had longer log_RTs than younger participants (F1, 37 = 62.00, p < 0.001, ηp2= 0.63) (significance not depicted); B: Standard deviation of log_RTs (SD_log_RT). There was a significant effect of condition (F2, 74 = 11.49, p < 0.001, ηp2= 0.237) and a significant effect of age group (F1, 37 = 4.61, p = 0.038, ηp2= 0.111) (significance not depicted); C: Individual error rates. In both age groups there is a gradual increase in error rate from repeat to ID and ED (F1.32, 48.91 = 17.27, p < 0.001, ηp2= 0.32); D: Performance z-scores. There was a significant effect of condition (F1.56, 57.86 = 70.94, p < 0.001, ηp2= 0.66) and a significant effect of age group (F1, 37 = 29.51, p < 0.001, ηp2= 0.44) (significance not depicted).

• 

  Fig. 3. Mean theta power (dB) across all conditions and all participants. Occipital theta activations after stimulus presentation (0 to 200 ms) are followed by frontocentral theta activations in the time window between 200 to 500 ms after target presentation.

• 

  Fig. 4. Condition-dependent modulation of midfrontal theta power. A & C: Midfrontal theta power averaged across the channels F1, F2, Fz, FC1, FC2, and FCz after target presentation in the young (A) and older (C) age group, separated by conditions (repeat, ID, ED). The grey shaded area indicates the time period used to calculate mean theta power (250 to 500 ms); B & D: Average topographical distribution of mean theta power in the young (B) and old (D) age group 250–500 ms after target presentation for each condition. The frontocentral electrodes F1, F2, Fz, FC1, FC2 and FCz used for analysis are marked with an X; E: Theta power averaged across the aforementioned midfrontal electrodes 250–500 ms after target presentation for each condition. Here, only young individuals show a significant increase in theta power upon set-shifting compared to the repeat condition (interaction term of age group and condition F2, 74 = 14.51, p < 0.001, ηp2= 0.28); F: Theta power across the midline electrodes 250–500 ms after target presentation. The asterisks indicate the electrodes in which young participants showcase significant difference across all conditions (highest t-value: FCz; repeat vs ED: t37 =…

• 

  Fig. 5. Theta activity and performance A: Single trial z-scores obtained from single trial correlations of theta power and reaction time. Young participants have higher z-values overall (F1, 37 = 13.73, p < 0.001, ηp2= 0.27). Within the young individuals both the ED and ID conditions significantly differ from zero in the positive direction (ED: t19 = 2.70, padj = 0.043, d = 0.60; ID t19 = 3.31, padj = 0.022, d = 0.74); B: Estimated slopes of relationship between mean theta power and performance z-scores. In the ED condition, older adults have a significant positive slope which is significantly different from the slope obtained from young participants (b = 1.11, SE = 0.48, t71.5 = 2.30, p = 0.024).