Alpha-band oscillations reflect external spatial coding for tactile stimuli in sighted, but not in congenitally blind humans

Abstract

We investigated the function of oscillatory alpha-band activity in the neural coding of spatial information during tactile processing. Sighted humans concurrently encode tactile location in skin-based and, after integration with posture, external spatial reference frames, whereas congenitally blind humans preferably use skin-based coding. Accordingly, lateralization of alpha-band activity in parietal regions during attentional orienting in expectance of tactile stimulation reflected external spatial coding in sighted, but skin-based coding in blind humans. Here, we asked whether alpha-band activity plays a similar role in spatial coding for tactile processing, that is, after the stimulus has been received. Sighted and congenitally blind participants were cued to attend to one hand in order to detect rare tactile deviant stimuli at this hand while ignoring tactile deviants at the other hand and tactile standard stimuli at both hands. The reference frames encoded by oscillatory activity during tactile processing were probed by adopting either an uncrossed or crossed hand posture. In sighted participants, attended relative to unattended standard stimuli suppressed the power in the alpha-band over ipsilateral centro-parietal and occipital cortex. Hand crossing attenuated this attentional modulation predominantly over ipsilateral posterior-parietal cortex. In contrast, although contralateral alpha-activity was enhanced for attended versus unattended stimuli in blind participants, no crossing effects were evident in the oscillatory activity of this group. These findings suggest that oscillatory alpha-band activity plays a pivotal role in the neural coding of external spatial information for touch.

Figure 1

Experimental paradigm and behavioral results. (A) Schematic trial. At the beginning of each trial (t = −1000 ms) an auditory cue indicated the task-relevant hand to the participants. After 1000 ms (t = 0 ms) a tactile stimulus was presented to either hand. Participants had to detect rare tactile targets the cued hand only and to ignore frequent standard stimuli at either hand. (B) Behavioral results. Hand posture influenced performance in the sighted group only (black circles) with higher d’-scores with uncrossed (left) than crossed hands (right). In the congenitally blind group (gray triangles) performance did not significantly differ between postures. Whiskers represent the standard error of the mean. Figure adapted from ref..

Figure 2

Attention-related alpha-band power modulations (10–14 Hz) relative to pre-trial and pre-stimulus baselines for sighted and congenitally blind groups. All data have been recoded as if stimulation always occurred on the anatomically right hand, so that the left hemisphere is contralateral to tactile stimulation in a skin-based reference frame, independent of posture. Left two columns, contra- and ipsilateral alpha-band power, pre-trial baseline (marked with gray rectangle); right two columns, contra- and ipsilateral alpha-band power, pre-stimulus baseline (marked with gray rectangle). (A–H) Difference of attended minus unattended alpha-band power for uncrossed (red) and crossed (blue) hands at central electrodes, as indicated in the insets in (E–H). (I–P) Difference of attended minus unattended alpha-band power for uncrossed and crossed hands at parietal electrodes (see insets (M–P). Contralateral power is depicted on the left hemisphere (A,C,E,G, I,K,M,O), and ipsilateral power on the right hemisphere (B,D,F,H,J,L,N,P). Shaded areas around power traces represent the standard error of the mean. Parts I-L used with permission from ref..

Figure 3

Alpha-band power modulations in the sighted group. (A) Topographies of alpha-band activity (10–14 Hz, 400 to 600 ms, marked with black rectangle in (B) with uncrossed (a,b) and crossed hands (d,e) following attended (a,d) and unattended (b,e) stimuli. (c,f,g,h) Difference topographies for attention effects with uncrossed (c) and crossed (f) hands, and for posture effects following attended (g) and unattended (h) stimuli. (i) Topography of the interaction between attention and posture. Data are displayed as if stimuli always occurred on the anatomically right hand, so that the left hemisphere is contralateral to tactile stimulation in a skin-based reference frame, independent of posture. (B) Time-frequency representation of the electrode showing the largest interaction between posture and attention (marked with an asterisk in (A) approximately P3/4 in the 10–10 system) with uncrossed (a,b) and crossed hands (d,e) following attended (a,d) and unattended (b,e) stimuli. Unmasked areas in (c,f,g,h) and (i) indicate significant differences between attention conditions with uncrossed (c) and crossed hands (f), between posture conditions following attended (g) and unattended stimuli (h), and a significant interaction between posture and attention (i) (cluster-based permutation test, p < 0.05). (C) Neural sources of alpha-band activity. Alpha-band activity (12 ± 2 Hz, t = 400 ms) with hands uncrossed (a,b) and crossed (d,e) following attended (a,d) and unattended (b,e) stimuli. Source statistics are shown for the interaction effect between posture and attention (i), for effects of posture following attended (g) an unattended (h) stimuli, and for effects of attention with uncrossed (c) and crossed (f) hands. Significant clusters in (c,f,g,h) and (i) are unmasked. The left (right) hemisphere is contralateral (ipsilateral) to the stimulated hand. The white dashed line denotes the central sulcus. Used with permission from ref..

Figure 4

Induced alpha power (total power minus stimulus-locked power) of the sighted group. Time-frequency representation of the electrode showing the largest interaction between posture and attention (marked with an asterisk * in Fig. 3A) with uncrossed (a,b) and crossed hands (d,e) following attended (a,d) and unattended (b,e) stimuli. Unmasked areas in (c,f,g,h,i) indicate statistically significant differences, as identified with cluster-based permutation testing at p < 0.05, between attention conditions with uncrossed (c) and crossed hands (f), between posture conditions following attended (g) and unattended stimuli (h), and a statistically significant interaction between posture and attention (i) Note, the figure is arranged identically to Fig. 3B, which depicts total power, so that the two figures can be compared directly.

Figure 5

Alpha-band activity in the blind group. (A) Topographies of alpha-band activity (10–14 Hz, 400 to 600 ms, marked with black rectangle in (B) with uncrossed (a,b) and crossed hands (d,e) following attended (a,d) and unattended (b,e) stimuli. (c,f,g,h). Difference topographies for attention effects with uncrossed (c) and crossed (f) hands, and for posture effects following attended (g) and unattended (h) stimuli. i. Topography of the interaction between attention and posture. Data are displayed as if stimuli always occurred on the anatomically right hand, so that the left hemisphere is contralateral to tactile stimulation in a skin-based reference frame, independent of posture. Note that, although no effects of posture were evident, topographies are split according to attention and posture to allow direct comparison to sighted participants’ data in Fig. 3A. (B) Time-frequency representation (TFR) of the electrode marked with an asterisk in (A) (approximately FC3/4 in the 10–10 system) following attended (a) and unattended (b) stimuli and time-frequency representations of the statistical difference between attention conditions (c) with significant clusters (p < 0.05) being unmasked. (C) Source reconstruction of alpha-band activity elicited by attended (a) and unattended (b) stimuli and the attention effect (c), view from above (left) and lateral view of the contralateral hemisphere (right), significant clusters are unmasked (CBPT: p = 0.005). The white dashed line denotes the central sulcus. The left (right) hemisphere is contralateral (ipsilateral) to the stimulated hand in all panels. Parts B and C used with permission from ref..

Figure 6

Beta-band activity of sighted and congenitally blind participants. Topographies of beta-band activity (16–24 Hz, 400 to 600 ms) in the sighted (A) and blind (B) group, with uncrossed (a,b) and crossed hands (d,e) following attended (a,d) and unattended (b,e) stimuli. (c,f,g,h) Difference topographies for attention effects with uncrossed (c) and crossed (f) hands, and for posture effects following attended (g) and unattended (h) stimuli. (i) Topography of the interaction between attention and posture. Data are displayed as if stimuli always occurred on the anatomically right hand, so that the left hemisphere is contralateral to tactile stimulation in a skin-based reference frame, independent of posture.