Fig 1.
Task paradigm and behavioral results for Experiment 1.
(A) Each trial began with a cue display of the distractor, 1,200 to 1,600 ms followed by a search display. In 2 separate sessions, the cue display was fully predictive (with 100% validity) or not predictive (with 10% validity) of the specific location of the red distractor circle. Participants were instructed to indicate the orientation of the gray line inside the yellow target circle in the search array. (B) The spatial probability of the target and distractor occurring during subsequent visual search with respect to 2 cue sessions (yellow represents the target; red represents the distractor). Before each session, participants were informed of the cue validity (valid or invalid) and its corresponding spatial probability of the target and distractor. (C) The mean (top) and slope (bottom) of accuracy across DTD in the valid- (red) and invalid- (blue) cue sessions for Experiment 1. (D) The mean (top) and slope (bottom) of reaction time for Experiment 1. (E) The mean (top) and slope (bottom) of efficiency scores for Experiment 1. The slope of accuracy (F), reaction times (G), efficiency scores (H) in the valid- (red) and invalid- (blue) cue sessions for the behavioral control experiment. Violin plots depict the distributions of measurements in each session, with dots representing each subject. The solid and dotted lines indicate medians and quartiles, respectively. **p < 0.01, *p < 0.05 (see S1 Data for raw values).
Fig 2.
EEG results during the cue-distractor intervals from Experiment 1.
(A) The spatial direction of the distractor cue varied from trial to trial. The spatial distribution of alpha power was modeled by the channel tuning functions (CTFs) across 10 ideal channel offsets; right panel shows channel offsets and the centre channel if distractor cue point 180 degrees (red arrow). (B) Alpha-band CTFs across the cue-distractor intervals for valid-cue and invalid-cue sessions. The difference between the 2 sessions was also plotted. (C) The direction selectivity of the alpha-band CTF (measured as CTF slope) across time in valid- (red) and invalid- (blue) cue sessions. The different channel response curves at 5 sampled time points (gray vertical dashed lines) were plotted in both sessions. The red and blue dashed lines at the top indicate clusters where sessions differed significantly from zero (p < 0.05), and significant differences between sessions are marked by the black dashed line (p < 0.05). (D) The cueing effect on alpha-band CTFs (valid−invalid; averaged from 1,040 to 1,200 ms) is related to anticipation of the distractor. (E) Correlation between alpha CTF slope and efficiency scores (ESs) slope. The ES difference between valid and invalid cues (cueing effects) varies as a function of the alpha CTF slope difference between valid and invalid cues. Positive values of the y-axis indicate larger ES slope with valid cues than with invalid cues. Negative values of the x-axis indicate that the negative-going alpha CTF slope was larger with valid cues than with invalid cues (see S2 Data for raw values).
Fig 3.
Lateralized EEG results during the stimulus period from Experiment 1.
(A) Time course of the alpha modulation index in the posterior electrodes for valid- (red) and invalid- (blue) cue sessions. The red and blue dashed lines indicate a significant difference from zero, and the black dashed line indicates clusters with a significant difference between 2 sessions (p < 0.05). Shades of light color along with the dark color lines represent error bars (±1 SEM). (B) Grand averaged ERPs at contralateral and ipsilateral electrode sites relative to the distractor (averaged over P7 and P8) in valid- (red) and invalid- (blue) cue sessions. (C) The left panel shows the corresponding difference waves (contralateral minus ipsilateral activity) for valid- (red lines) and invalid-cue sessions (blue lines). Violin plots depict the PD amplitude (248- to 316 ms) in the 2 sessions, with the dots representing each subject. The solid and dotted lines indicate medians and quartiles, respectively. *p < 0.05. Con, contralateral to distractor cue; Ips, ipsilateral to distractor cue (see S3 Data for raw values).
Fig 4.
Task paradigm and EEG results for Experiment 2.
(A) Three types of cue displays and corresponding spatial probability of the target and distractor occurring during subsequent visual search. Note that spatial probability was conceptual and did not actually appear around the cue. (B) The slope of ES for high- (red), low- (blue), and null- (black) predictive validity trials. (C) Time course of the alpha MI in the posterior electrodes for high- (red) and low- (blue) predictive validity trials. Shades of light color along with the dark color lines represent error bars (±1 SEM). (D) Grand averaged ERPs at contralateral and ipsilateral electrode sites relative to the distractor (averaged over P7 and P8) in high- (red) and low- (blue) predictive validity trials. (E) The upper panel shows the corresponding difference waves (contralateral minus ipsilateral activity) for high- (red lines), low- (blue lines), and null-predictive validity. The red, blue, and black dashed lines indicate a significant difference from zero, and the dashed lines with 2 colors indicate clusters with a significant difference between the 2 conditions (p < 0.05). The lower panel shows scatter plot for PD amplitude, with the dots representing each subject. *p < 0.05. Con, contralateral to distractor cue; Ips, ipsilateral to distractor cue (see S4 Data for raw values).
Fig 5.
Relationship between alpha MI and PD in Experiment 2.
(A) Alpha MI during the cue period as a function of the subsequent distractor-elicited PD amplitudes during a visual search between participants in high-predictive validity trials. The diagrams along with the scatter plot are the frequency distributions of alpha MI and PD amplitude, respectively. (B) Averaged single-trial PD for each quartile at the within-subjects level in high-predictive validity trials. The trials were sorted according to cue-induced alpha MI and binned into quartiles. The PD amplitudes were normalized and then averaged over subjects. *p < 0.05. (C) Scatter plot for low-predictive validity trials. (D) Quartile plot for low-predictive validity trials. Con, contralateral to distractor cue; Ips, ipsilateral to distractor cue (see S5 Data for raw values).
Fig 6.
Task paradigm and EEG results for Experiment 3.
(A) The arrow was fully predictive of the side on which the distractor circle of the corresponding color would subsequently appear. (B) Time course of the alpha MI. (C) Grand averaged ERPs at contralateral and ipsilateral electrode sites relative to the distractor. (D) The scatter plot between cue-induced alpha MI (averaged over the time-frequency windows highlighted by black outlines) and distractor-elicited PD amplitudes between participants showed a significant correlation. The diagrams along with the scatter plot are the frequency distributions of alpha MI and PD amplitude, respectively. (E) Averaged single-trial PD for each quartile at the within-subjects level. Trials were sorted according to cue-induced alpha MI and binned into quartiles. PD amplitudes were normalized and then averaged over subjects. *p < 0.05. Con, contralateral to distractor cue; Ips, ipsilateral to distractor cue (see S6 Data for raw values).