Cued spatial attention drives functionally relevant modulation of the mu rhythm in primary somatosensory cortex

Abstract

Cued spatial attention modulates functionally relevant alpha rhythms in visual cortices in humans. Here, we present evidence for analogous phenomena in primary somatosensory neocortex (SI). Using magnetoencephalography, we measured changes in the SI mu rhythm containing mu-alpha (7-14 Hz) and mu-beta (15-29 Hz) components. We found that cued attention impacted mu-alpha in the somatopically localized hand representation in SI, showing decreased power after attention was cued to the hand and increased power after attention was cued to the foot, with significant differences observed 500-1100 ms after cue. Mu-beta showed differences in a time window 800-850 ms after cue. The visual cue also drove an early evoked response beginning ∼70 ms after cue with distinct peaks modulated with cued attention. Distinct components of the tactile stimulus-evoked response were also modulated with cued attention. Analysis of a second dataset showed that, on a trial-by-trial basis, tactile detection probabilities decreased linearly with prestimulus mu-alpha and mu-beta power. These results support the growing consensus that cue-induced alpha modulation is a functionally relevant sensory gating mechanism deployed by attention. Further, while cued attention had a weaker effect on the allocation of mu-beta, oscillations in this band also predicted tactile detection.

Figures

Figure 1.

Experimental design for cued detection runs. See Materials and Methods.

Figure 2.

Impact of cued attention on SI mu-alpha and mu-beta. A, Two examples of the estimated SI ECD localizations (blue dots) overlaid on the subjects' structural MRI brain images. Response evoked by a suprathreshold tactile stimulus to the left hand, third digit, was localized to the SI hand representation in area 3b, confirmed by proximity to the Ω shape (marked in red bottom panel), in the anterior bank of the contralateral postcentral gyrus. B, Continuous postcue temporal evolution of the hand area SI mu-alpha (7–14 Hz) and mu-beta (15–29 Hz) activity in attend-in and attend-out conditions (avg. n = 12 Ss). C, Corresponding continuous prestimulus evolution of mu-alpha mu-beta. Asterisks, Significant difference between conditions (p < 0.05).

Figure 3.

Impact of cued attention visual cue and tactile stimulus SI ERs. A, Average hand area SI broadband ER from the visual cue in attend-in and attend-out conditions (mean n = 12 Ss). B, Corresponding average SI broadband ER from subsequent threshold-level tactile stimulation to the hand in attend-in and attend-out conditions. Asterisks, Significant difference between conditions (p < 0.05).

Figure 4.

Impact of prestimulus mu-alpha and mu-beta power on detection probabilities. A, Tactile detection probabilities, measured as percentage change in hit rate from the mean, as a function mu-alpha power sorted into 10 power percentile bins. Bold traces, Mean and SE across subjects (n = 10 Ss); thin traces, linear fit from linear regression analysis (R2 = 0.65, p < 0.05). B, Analogous traces as a function of mu-beta power (R2 = 0.84, p < 0.05). This analysis was performed on data from the Jones et al. (2007, 2009) studies (see Materials and Methods).