Figure 1.
Schematic experimental paradigm.
Each trial began with a cue and ended with the response. Once a response to a previous trial is completed, the cue for a current trial is presented with a designated duration, and it is then followed by a colored and oriented triangle stimulus. Cue-stimulus interval (CSI), response-cue interval (RCI), and response-stimulus interval (RSI) were defined as the interval between current cue and stimulus, previous response and cue, previous response and stimulus, respectively.
Figure 2.
The parameters of RCI, CSI and RSI.
Two levels of RSI and CSI were combined to produce 4 sessions: short RSI and short CSI (SS), short RSI and long CSI (SL), long RSI and short CSI (LS), and long RSI and long CSI (LL). In these four conditions, RCI included short (150 ms), long (600 ms), and longer (1050 ms) intervals.
Figure 3.
Examples of repeat and switch trials.
Repeat trials were defined when a cue in the current trial was the same as the one in the previous trial, indicating that the same task should be performed (top row in the figure). Switch trials were defined when a cue in the current trial was different from the previous trial, indicating that the task should be switched (bottom row in the figure).
Figure 4.
(A) Average switch costs of behavioral reaction time are shown for the 14 subjects in the four conditions. Switch cost was smaller in the long CSI condition. (B) Average switch costs of behavioral error rates are shown for the 14 subjects in the four conditions. Subjects responded more accurately in the SL and LL conditions (long CSI condition). Switch cost was smaller in the long CSI. Error bars show SEM.
Figure 5.
Cue-related ERP waveforms and brain topographies for short CSI conditions.
(A) Grand-average cue-related ERP waveforms at midline sites (frontal, Fz, central, Cz, and parietal, Pz) are shown for the 14 subjects in the short CSI conditions across both repeat and switch trials. A large sharp parietal positivity emerged at about 400 ms after cue onset. Switch trials evoked larger amplitude than that of repeat trials. (B) Brain topographies in the short CSI conditions across repeat (mid) and switch trials (top) and the difference topographies between switch and repeat trials (bottom) are shown. There was more activation for switch trials in parietal cortex.
Figure 6.
Cue-related ERP waveforms and brain topographies for long CSI conditions.
(A) Grand-average cue-related ERP waveforms at midline sites (Fz, Cz, Pz) are shown for the 14 subjects in the long CSI conditions across both repeat and switch trials. A slow parietal positivity emerged at about 400 ms after cue onset. Switch trials evoked larger amplitudes than that of repeat trials. (B) Brain topographies in the long CSI conditions across repeat (mid) and switch trials (top) and the difference topographies between switch and repeat trials (bottom) are shown. There was more activation for switch trials in parietal cortex.
Figure 7.
Switch cost of cue-related ERP positivity.
Switch cost of ERP positivity was evaluated by subtracting amplitude of the switch-related positivity for repeat trials from that for switch trials. Long RSI resulted in a smaller SC of ERP positivity.
Figure 8.
Stimulus-related ERP waveforms and brain topographies on short CSI condition.
(A) Grand-average stimulus-related ERP waveforms at midline sites (Fz, Cz, Pz) are shown for the 14 subjects in short CSI conditions across both repeat and switch trials. In all the repeat and switch trials, a large sharp central and frontal negativity emerged at about 400 ms after stimulus onset. (B) Brain topographies in the short CSI conditions across repeat (mid) and switch trials (top) and the difference topographies between switch and repeat trials (bottom) are shown.
Figure 9.
Stimulus-related ERP waveforms and brain topographies on long CSI condition.
(A) Grand-average stimulus-related ERP waveforms at midline sites (Fz, Cz, Pz) are shown for the 14 subjects in long CSI conditions across both repeat and switch trials. In all the repeat and switch trials, a slow central and parietal negativity emerged at about 300 ms after stimulus onset. (B) Brain topographies in the long CSI conditions across repeat (mid) and switch trials (top) and the difference topographies between switch and repeat trials (bottom) are shown.
Figure 10.
Switch cost of stimulus-related ERP negativity.
Switch cost of ERP negativity was evaluated by subtracting the amplitude of the switch-related negativity for repeat trials from that for switch trials across the four conditions (SS, SL, LS, and LL). CSI had main effects for SC. SC of neural activity was smaller in the long CSI condition.
Table 1.
Switch-related brain regions in cue-related ERP positivity.
Table 2.
Switch-related brain regions in stimulus-related ERP negativity.