Subthalamic nucleus local field potential activity during the Eriksen flanker task reveals a novel role for theta phase during conflict monitoring

Abstract

The subthalamic nucleus (STN) is thought to play a central role in modulating responses during conflict. Computational models have suggested that the location of the STN in the basal ganglia, as well as its numerous connections to conflict-related cortical structures, allows it to be ideally situated to act as a global inhibitor during conflict. Additionally, recent behavioral experiments have shown that deep brain stimulation to the STN results in impulsivity during high-conflict situations. However, the precise mechanisms that mediate the "hold-your-horses" function of the STN remain unclear. We recorded from deep brain stimulation electrodes implanted bilaterally in the STN of 13 human subjects with Parkinson's disease while they performed a flanker task. The incongruent trials with the shortest reaction times showed no behavioral or electrophysiological differences from congruent trials, suggesting that the distracter stimuli were successfully ignored. In these trials, cue-locked STN theta band activity demonstrated phase alignment across trials and was followed by a periresponse increase in theta power. In contrast, incongruent trials with longer reaction times demonstrated a relative reduction in theta phase alignment followed by higher theta power. Theta phase alignment negatively correlated with subject reaction time, and theta power positively correlated with trial reaction time. Thus, when conflicting stimuli are not properly ignored, disruption of STN theta phase alignment may help operationalize the hold-your-horses role of the nucleus, whereas later increases in the amplitude of theta oscillations may help overcome this function.

Figures

Figure 1.

Flanker task. A, Task. Each trial began with a warning cue with onset 500 ms before arrows were shown. Arrows were shown for 200 ms, and subjects had 2.2 s to respond before the next warning cue. A 2:1 ratio of incongruent to congruent trials was used. B, Behavioral responses across subjects. To the left are reaction times for all 13 subjects for congruent and incongruent trials. There was a significant difference between mean congruent (green bar with SEM) and incongruent (purple) reaction times. When the incongruent trials were median split into the fastest half (blue) and the slowest half (red), there was no difference in reaction time between the fastest incongruent trials and the congruent trials. The slowest incongruent trials, however, had reaction times that were significantly slower than the congruent trials. C, Average reaction time histograms normalized to each subject's mean incongruent trial reaction time. Incongruent trial histogram reveals two peaks. Color denotes whether trials in a given bin were put in the fast-incongruent (blue) or slow-incongruent (red) groups for further analysis.

Figure 2.

Effects of congruency on LFP across all subjects. A–D, Imperative cue-aligned (t = 0) averages of induced spectral change. Both congruent (A) and incongruent (B) trials showed an increase in cue-aligned theta power, a decrease in beta power followed by a postresponse rebound, and an increase in gamma power. C, Difference between trial types masked at a 0.05 significance level corrected for multiple comparisons, showing the theta band difference. D, Cue-aligned theta (3–8 Hz) band average time series (mean ± SEM) for congruent (green) and incongruent (purple) trials. Significant difference between the two conditions is marked by black bar (p < 0. 05 corrected for multiple comparisons). E–H, Same as A–D but aligned to the response. Theta difference is weaker and only significant in the theta band average time series (H). Note that here and in ensuing time–frequency plots that frequency is given on a log axis.

Figure 3.

Differences between slow and fast trials following incongruent cues across all subjects. A–D, Imperative cue-aligned (t = 0) averages of induced spectral change. A, No power differences between the fast-incongruent trials and congruent trials (masked at p < 0.05 significance level after correcting for multiple comparisons). B, C, Slow-incongruent trials showed higher cue-aligned theta power than congruent trials (B) and fast-incongruent trials (C). D, Theta (3–8 Hz) band average time series for slow-incongruent (red), fast-incongruent (blue), and congruent (green) trials. Note that mean ± SEM values are shown except for congruent trials (where ±SEM values were shown in Fig. 2). Significant difference between trial types is marked by horizontal bars (p < 0.05, corrected for multiple comparisons). E–H, Same as A–D but aligned to the response. Fast-incongruent trials showed no significant difference from the congruent. Preresponse theta power was higher in slow-incongruent trials. Cong, Congruent.

Figure 4.

Cue-locked theta phase realignment is disrupted in slow-incongruent trials. A–E, Imperative cue-aligned (t = 0) averages of phase locking across trials. A, Mean wavelet PLV for all fast-incongruent trials averaged across all 26 STNs. B, Same as A for slow-incongruent trials. C, There is a significant reduction in cue-locked theta PLV when the slow-incongruent trials are compared with fast-incongruent trials (masked at p < 0.05 significance level after correcting for multiple comparisons). D, Theta band filtered Hilbert PLV also showed impaired cue-locked phase alignment in the slow-incongruent trials. Slow-incongruent (red), fast-incongruent (blue), and congruent (green) mean PLV time series are shown ±SEM. Significant differences are denoted by horizontal bars. E, Theta bandpass-filtered Hilbert phase for all trials in all subjects sorted by reaction time (smoothed across each 100 trials and 40 ms). Prominent phase alignment is stimulus locked and independent of reaction time. Dotted black track indicates stimulus onset, and solid black trace indicates trial reaction time. F–J, Same as A–D but aligned to response. Comparison of D and I suggests that there are two phase alignment periods: just after but time locked to stimulus onset (D, arrow pointing to large peak) and periresponse (I, arrow pointing to second peak centered on response at t = 0). Both peaks can be seen in the response-aligned Hilbert PLV of the fast-incongruent trials, which showed the least variable reaction time (I). Cue-locked phase alignment (D) is greater than response-aligned phase alignment (I) for all trial types except slow-incongruent trials, which show impaired cue-locked phase alignment. K, Cue-locked phase alignment (calculated across all correct congruent and incongruent trials) correlated with subject reaction time (r = −0.58, p < 0.05). L, Response-locked phase alignment did not correlate with subject reaction time (r = −0.11). Correlations show the results of linear regression and corresponding 95% confidence limits. Cong, Congruent.

Figure 5.

LFP in error trials across all subjects. A, Incorrect incongruent trials had a group mean reaction time (±SEM) that was faster than slow-incongruent trials but was not different from fast-incongruent trials. B, Induced LFP power changes during incorrect incongruent trials aligned to an imperative cue. C, Differences between incorrect and correct incongruent trials across subjects masked at p < 0.05 (corrected for multiple comparisons). D, E, Same as B and C but aligned to the response. Increases in postresponse delta (2–4 Hz) and low beta (10–20 Hz) power can be seen. Incong, Incongruent.