Skip to main content
NCBI home page
Human Brain Mapping logoLink to Human Brain Mapping
. 2007 Mar 27;29(2):207–221. doi: 10.1002/hbm.20384

Cortical network for vibrotactile attention: A fMRI study

Harold Burton 1,2,✉, Robert J Sinclair 1, Donald G McLaren 1
PMCID: PMC2593407  NIHMSID: NIHMS79074  PMID: 17390318

Abstract

We used fMRI to identify brain areas activated during tactile attention tasks. Participants detected the interval containing target stimulation of higher vibrotactile frequency or longer duration. Attributes were selectively or neutrally cued. A control backwards‐counting task included concurrent, but irrelevant corresponding vibrotactile stimulation. Group analyses of average F‐statistic maps, participant conjunction maps, and estimated time courses utilized data mapped to a standard average surface atlas (PALS B12). Repeated‐measures, random‐effects MANOVA examined blood oxygenation level‐dependent (BOLD) signal modulation differences amongst tasks in defined regions, where significant responses occurred in at least 50% of the group. Greater than 0.1% increase in BOLD responses were found during at least one of the tactile attention tasks in contralateral parietal opercular OP1, BA 4 finger region, frontal eye field, dorsal premotor, anterior and posterior BA 7, and bilaterally in superior temporal sulcal cortex (BA 22), ventral premotor, supplementary motor area, and frontal operculum/insula. The same tasks suppressed activity in ipsilateral OP4. The BA 22 ROI showed larger responses during neutral cuing. The control task suppressed BOLD in ipsilateral OP1 and OP4 and bilaterally in BA 40, but significantly enhanced responses in dorsal parietal–frontal regions compared with tactile attention tasks. No regional differences were found between selectively cued frequency and duration tasks. Tactile attention effects were most prominent in OP1. Posterior parietal responses possibly reflected the visual attention required for backwards‐counting, whereas the frontal regions potentially related to goal‐directed behavior when identifying target stimulation. Hum Brain Mapp 2008. © 2007 Wiley‐Liss, Inc.

Keywords: attention, human, magnetic resonance imaging, somatosensory cortex/*physiology

INTRODUCTION

We examined the cortical network engaged when attending to the frequency or the duration of a vibrotactile stimulus applied to the right index fingertip. We used a filtering task in which two or more features of a stimulus can change. In such tasks, selective cuing to a feature decreases reaction time and improves discrimination accuracy compared with a nonpredictive neutral cue [e.g., Garner, 1974]. Selective cuing focuses attention resources for validly compared with neutrally or falsely cued trials [Jonides and Mack, 1984; Kahneman, 1973; Posner, 1986]. Performance measures in neutrally cued trials, in which attention is divided between features, falls between those of validly and falsely cued trials. A prior behavioral study found that changes in the tactile attributes of vibration frequency and duration were detected with greater accuracy when validly compared to invalidly or neutrally cued, indicating that these attributes could be separately analyzed [Sinclair et al., 2000]. The same task criteria were used in the present study with the exception of substituting a nontactile attention task for invalidly cued trials.

An important premise is that attention to stimuli within a modality modifies activity in the very same areas normally driven by that stimulation [Stoeckel et al., 2004]. Thus, tactile attention effects can be anticipated in primary (S1) and secondary (S2) somatosensory cortex. Several studies demonstrated greater S1 activity contralateral to an attended tactile stimulus [Johansen‐Berg et al., 2000; Meador et al., 2002; Van de Winckel et al., 2005]. Enhanced responses to attended tactile stimuli, however, were more often observed in higher order somatosensory areas located along the parietal operculum and inferior lateral parietal cortex in monkeys [Burton et al., 1997b; Hsiao et al., 1993] and humans [Backes et al., 2000; Burton et al., 1997a; Fujiwara et al., 2002; Hamada et al., 2003; Hämäläinen et al., 2002; Ledberg et al., 1995; Mima et al., 1998; Nelson et al., 2004; Reed et al., 2004].

Many visual attention studies describe significant task dependent activation of a dorsal parietal–frontal network [Corbetta and Shulman, 2002] that includes premotor and the intraparietal sulcal cortex. A question is whether similar regions are involved in tactile attention. However, a more generic role for these same “dorsal‐lateral and ventral prefrontal regions, insula, and SMA” [Fox et al., 2005, p 9675] is for any goal‐directed processes. Consistent with the latter interpretation is activity in a comparable cortical network during sensorimotor cognitive tasks involving hand actions like grasping, holding or touching objects, and visual/spatial transformations. Example tasks include: hand and arm movements [Binkofski et al., 1999a, b], discriminating passively applied Braille patterns [Harada et al., 2004], detecting differences in the lengths of oblongs [Roland et al., 1998], and palpating common objects [Reed et al., 2005]. In the sensorimotor tasks, projecting tactile sensations to an object might crucially engage visual/tactile transformations for appropriate interactions with three‐dimensional objects [Rizzolatti et al., 2002]. These same sensorimotor tasks, however, required attention to the manipulated object or limb movements. It is thus difficult to disambiguate motor/spatial from attention factors in the prior studies that involved somatosensory stimulation. Goal‐directed, stimulus‐response selection was required in both the visual attention and somatomotor tasks used in these studies, which plausibly indicates why seemingly diverse tasks engaged overlapping cortical regions.

The present study employed passively applied vibrotactile stimulation. Such sensory information involves egocentric tactile perceptions that do not need any visual/tactile transformations [Katz, 1989]. In contrast to earlier studies involving attention to somatosensory stimulation, there are few if any spatial/exocentric facets. Yet, a similar frontal–parietal regional network is predicted because the cognitive selection of cued vibrotactile attributes requires that attention be set on the sensory information and because response selection is required when identifying a target stimulus. As in previous visual attention studies, there was “cognitive selection of sensory information and responses” [Corbetta and Shulman, 2002, p 201].

Abbreviations.

BA

Brodmann areas

BOLD

Blood oxygen–level dependent

C

Control, backwards counting task

CM

Conjunction map

COG

Center of gravity

Dc

Comparison vibration duration

Ds

Standard vibration duration

EPI

Echo‐planar imaging

Fc

Comparison vibrotactile frequency

FEF

Frontal eye field

Fs

Standard vibrotactile frequency

GLM

General linear model

IPS

Intraparietal sulcus

LH & RH

Left and right hemisphere

N

Divided attention, neutral cuing task

OP

Parietal operculum

PMd

Dorsal premotor area

PMv

Ventral premotor area

ROI

Region of interest

SD

Selective attention to duration task

SF

Selective attention to frequency task

SMA

Supplementary motor area

S1

Primary somatosensory cortex

S2

Secondary somatosensory cortex

TR

Repetition times

MATERIALS AND METHODS

Participants

All participants provided informed consent following guidelines approved by the Human Studies Committee of Washington University. Participants were free of neurological disease and had normal brain anatomy on MR structural images. Twelve volunteers (four female; mean age = 28.3 years, SD = 12.8) scored a mean right handedness of 92.8 (SD = 16.9) on a modified Edinburgh Handedness Inventory [Raczkowski et al., 1974].

Vibrotactile Stimulation and Tasks

An MRI compatible tactile vibrator [Burton et al., 2004] delivered sinusoidal vibrations to the glabrous tip of the right index finger. Trials of stimulation involved paired vibrations. Three sequential, closely spaced trials were cued to the same task; time between trials was ∼3 s. A no‐stimulation interval followed the three trials, which allowed for recovery to baseline activity. The three stimulation trials and no‐stimulation interval were analyzed collectively as an epoch (Fig. 1). Epoch duration of ∼28 s equaled the time for 10 successive volume acquisitions [e.g., repetition times (TR)]. Timing of trials within an epoch was synchronized to scanner pulses marking the beginning of the 1st, 3rd and 5th TRs. The interval from these scanner pulses to the first vibration in a trial varied based on stimulation duration parameters (see below), but was 1.25 s on average. The second vibration in a pair always began 500 ms after the first vibration ended and stopped 3.75 s after trial initiation.

Figure 1.

Figure 1

Open in a new tab

Task paradigm. Top: Epochs contained three successive trials each of which involved paired vibrations (vertical bars). Vibration duration was a variable that was set between 956 and 1,045 ms (Table I). The interval between pairs was fixed at 500 ms. A ∼3 s interval followed each trial for participant responses. Trial timing was synchronized to the 1st, 3rd, or 5th TR frames (black arrows). However, there were slight differences in the interval between the onset of the TR and the first vibration in order to accommodate a range of differences in vibration durations. Vibrotactile stimulation occurred during volume acquisitions for TR frames 2, 4, and 6, and given hemodynamic delays, also affected images collected during TR frames 3, 5, and 7. Middle: A visual cue was present during vibration stimulation. Bottom: Vertical lines mark the beginning of 10 successive TR frames.

Epochs were cued to four different tasks during the course of ∼15 min imaging runs. There were eight repetitions of each task per run and there were four imaging runs. Thus, there were 32 epochs per task. We balanced vibrotactile duration and frequency parameters across all tasks and separately randomized the order of epoch‐tasks per run.

Of the four tasks, there were three tactile attention tasks. Two of these involved selective cuing to the vibrotactile attributes of frequency (SF task) or duration (SD task). In the third task, attention was divided between frequency and duration attributes during trials with neutral cuing (N task). In all three tasks, participants judged whether the first or second vibration in a trial, respectively, had target stimulation with higher frequency or longer duration. During neutral cuing participants first identified the target attribute and subsequently the interval with the target. Participants responded immediately after each trial by covertly thinking “one” or “two” for whether, respectively, the first or second vibration was the target.

The fourth (control) task involved counting backwards (C task) by three from a displayed three‐digit number. Vibrotactile stimulation parameters presented during backwards‐counting and all tactile attention epochs were identical. A new randomly generated number appeared at the beginning of the 1st, 3rd, and 5th TRs. Participants counted covertly and continuously after seeing each number (e.g., stimulus “100”; respond “97, 94, 91, etc.”). Counting stopped when the last number disappeared.

Each participant distinguished between six different vibrotactile frequencies: two were standards (F s), and each was compared to a higher and lower frequency vibration (F c) (F s1 vs. F c1‐low or F c1‐high, and F s2 vs. F c2‐low or F c2‐high). Stimulation durations were in reference to a 1000 ms standard (D s) with one each of shorter (D c‐short) and longer (D c‐long) comparison duration. Prior training sessions established specific equal‐intensity functions for each participant and the vibrotactile parameters for performance accuracy of >75% on all tactile tasks. These participant‐specific stimulation values were used during imaging sessions. Table I lists the group averages for the frequency and duration stimulation standards and comparisons.

Table I.

Average stimulus frequencies and durations for 12 participants

Mean freq (Hz) SE freq Mean amp (microns) SE amp
Fs1 25 0 130 27.5
Fc1‐low 14 1.1 134 21.3
Fc1‐high 36 1.3 106 20.8
Fs2 180 11.3 30 2.6
Fc2‐low 136 12.6 38 5.2
Fc2‐high 238 14.2 29 3.5
Mean duration (ms) SE duration
Ds 1,000 0
Dc‐short 960.4 4.2
Dc‐long 1041.3 4.2
Open in a new tab

Cue words for different tactile attention tasks (e.g., “FREQ” for SF, “TIME” for SD, “BOTH” for N) or three‐digit numbers for the control task appeared in block white letters during vibrotactile stimulation (Fig. 1, bottom shaded row). Participants were instructed to fixate on a white cross while cue words appeared above this cross. As an additional cue, each word was shown against a different background screen color: green for SF, red for SD, top half green and bottom half red for N, and grey for C. The screen was black otherwise except for the white cross.

Image Acquisition

Magnetic resonance images were acquired using a 1.5 T Siemens Vision scanner (Erlangen, Germany) and a standard circularly polarized head coil. Functional image data were obtained with a custom, single‐shot asymmetric spin‐echo, echo‐planar imaging (EPI) sequence that had T2* evolution time of 50 ms from a 90° flip angle with in‐plane spatial resolution of 3.75 × 3.75 mm2. Whole brain coverage was provided with 20 or 21 6‐mm interleaved axial slices (TR = 135.2 ms/slice) that had been aligned parallel to a line along the anterior and posterior commissures; alignment was based on prior registration to an atlas representative target image using a coarse, sagittal structural image sequence [resolution = 2 × 2 × 2 mm3] [Mugler and Brookeman, 1990]. Four initial volume acquisition frames (TRs) were excluded to allow for magnetization stabilization. Additional T1‐ and T2‐weighted structural images facilitated subsequent functional image alignment to Talairach atlas space [Talairach and Tournoux, 1988] as described previously [Burton et al., 2006].

Analysis of Volume Data in Individuals

Functional image data was preprocessed for motion correction, interleaved slice intensity differences, and atlas registration [Burton et al., 2006]. Images were re‐sampled in atlas space to 2 mm3 isotropic voxels and spatially smoothed (4 mm FWHM) before statistical analyses.

Significant activity per voxel was determined for each participant using an F‐test1 based on a general linear model (GLM) [Miezin et al., 2000; Ollinger et al., 2001a, b] with 10 regressors for the time points in each task epoch (40 across the four tasks) and run‐specific regressors for baseline, linear drift, and a high‐pass filter. The F‐statistics were transformed to equally probable Z‐scores. Additionally, Z‐scores were corrected for multiple comparisons on the basis of Monte Carlo simulations [Forman et al., 1995]. Four corrected F‐statistic Z‐score maps were computed using different significance thresholds to capture both low and high amplitude signals in, respectively dispersed and focal clusters. The criteria for Z‐scores with P = 0.05 were z = 3.0, 3.5, 4 and 4.5 over, respectively, at least 45, 24, 12, and 5 face‐connected voxels. The different threshold z‐score maps were combined by retaining only the significant Z‐score from corresponding voxels, which thereby indicated the distribution of significant activity irrespective of cluster size. The combined task specific, thresholded Z‐score maps were binary coded and used to create conjunction maps. Thus, for each participant there were F‐statistic Z‐score maps for each of the tasks and corresponding binary coded maps.

Time course estimates were obtained for each imaging run using separate GLMs. Time course estimates were converted to percent signal change by dividing the difference between task and baseline signals by the baseline activity (e.g., run intercept).

Analyses of Volume Data Mapped to the Cortical Surface

All group analyses of response distributions by task and analyses of contrasts between responses to different tasks were done after mapping volume data to participant‐specific fiducial surfaces and subsequently to the PALS‐B12 atlas (http://brainvis.wustl.edu/caret) [Van Essen, 2005]. The following supervised steps were used to map the volume data to the PALS‐B12 surface: (1) Creation of participant‐specific fiducial cortical surfaces utilized atlas registered, 1 mm3 structural images that were segmented approximately along cortical layer four2 using SureFit (Van Essen, 2005). (2) The data values in voxels that intersected the cortical surface were directly mapped to the vertices of each participant‐specific fiducial cortical surface by using the intersection of enclosing voxels and nodes. (3) The nodes representing an individual hemisphere were deformed to the standard PALS‐B12 atlas sphere with 73,730 nodes using selective landmarks and spherical alignment [Van Essen, 2005]. (4) The nodal data values for Z‐scores and binary coding for each individual were directly mapped to the standard surface nodes using a nearest‐neighbor algorithm. (5) However, the time course estimate values were mapped using barycentric averaging for the nodes in the individual sphere to the nearest node in the standard sphere [Saad et al., 2004], which provided additional spatial smoothing for the time course estimates.

We assessed the distribution of activity on the cortical surface per task by averaging the uncorrected F‐statistic Z‐scores per node. A t test was used to assess whether these means significantly differed from a z‐score population mean of zero (i.e., Inline graphic) [Bosch, 2000]. A Bonferroni adjustment for 73,730 tests (number of nodes in the PALS‐B12 surface mesh), N − 1 degrees of freedom (i.e., d.f. = 11) and a two‐tailed α of 0.05 required a t‐value ≥10.1 [Sankoh et al., 1997].

We determined the consistency of activation across the group for each task through conjunction maps (CMs), which were computed as algebraic summations across the participant binary coded maps. Node values in CMs ranged from 0–12, which indicated respectively that 0–100% of the participants had multiple comparison corrected significant BOLD responses at a given node. We used a node threshold of ≥6 participants in the CMs. Clusters of suprathreshold nodes showed regions where ≥50% of the group showed significant activity.

Response differences between tasks were assessed using the time course estimates from the per run GLMs for each frame per task epoch and analyzed with a repeated‐measures random‐effects MANOVA (PROC GLM, Statistical Analysis System version 9.1, SAS Institute, Carey, NC) [Burton et al., 2006]. Significant differences between tasks were evaluated by the probability of the exact F‐statistic from Wilks' Lambda. Time course estimates for each participant were averaged per node within regions of interest (ROIs, see below). A separate MANOVA was computed by ROI for each pairing of tasks: SF vs. SD, SF vs. N, SF vs. C, SD vs. N, SD vs. C, and N vs. C. We chose a significance threshold of α = 0.008, which reflects a correction for six paired comparisons.

ROIs were generically defined in several steps. First, nodes in CMs for each task with a value ≥6 were coded as 1 and all other nodes coded as 0. These binary‐coded task CMs were then combined using a logical OR, which yielded a binary coded composite CM that showed where any task evoked significant activity in ≥50% of the participants. Next, a ROI was specified by the intersection (e.g., a logical AND) of nodes in the composite CM and nodes included in previously identified cortical areas (see below). Coordinates specified for each ROI were based on a center of gravity calculation (COG) using the PALS‐B12 average fiducial surface [Van Essen, 2005]. The surface area of each ROI was computed by determining the area enclosed by the nodes on the average fiducial surface and applying a topological distortion correction for the smoothing that occurs during surface averaging between individuals.

Specific ROI involved exclusive intersection of the composite CMs and previously defined cortical areas. The latter included Brodmann areas [Drury et al., 1999; Van Essen, 2005] and more recently defined parietal opercular cytoarchitectonic subdivisions OP1‐43 [Eickhoff et al., 2006a, c]. Prior functional imaging data was also used to constrain the definition of some ROI to create subdivisions within the Brodmann areas. Thus, within the borders for BA 3 and 1, we restricted the ROI to a previously defined finger sector [Maldjian et al., 1999]. A previously determined visual attention region [Astafiev et al., 2003] that had been mapped to the standard PALS surface was included within our posterior BA 7 ROI. An anterior BA7 subdivision was anterior and lateral to the posterior subdivision and included the intraparietal sulcus and nearby superior supramarginal gyral cortex. BA6 was subdivided into dorsal (PMd), ventral (PMv) [e.g., Rizzolatti et al., 2002] and medial sectors [e.g., SMA Picard and Strick, 1996]. The left PMd region was further subdivided to include a frontal eye field (FEF) [Astafiev et al., 2003]. The cytoarchitectonic borders for BA 44 and 46 were grouped together to define a joint 44/46. A frontal operculum ROI was defined anatomically.

RESULTS

Regional Activation Patterns‐Z‐Score Maps

Representative results from a single participant indicated that different tactile attention tasks activated similar cortical regions. For example, corresponding volume slices in Figure 2A show activity during the N and SF tasks in the pre‐ and postcentral gyri and posterior parietal cortex, involving PMv, BA 1, BA 7‐anterior, and BA 7‐posterior, respectively. However, distinctions existed in the location and extent of activity in these and other regions that were better illustrated after conversion of volume data to the cortical surface. Figure 2 illustrates the conversion of volume data to the cortical surface and the retention of data values and anatomical specificity after mapping to the specific individual fiducial surface. The surface maps shown in Figure 2B,C indicate that in BA 7 the Z‐scores were lower and the areal extent was smaller for the SF compared with the N task; the areal extent was slightly larger over BA 1 in the postcentral gyrus during the SF compared with the N task; and there were few differences noted in PMv. The group analyses provided further evidence of regional differences and task contrasts.

Figure 2.

Figure 2

Open in a new tab

F‐statistic Z‐score maps for a single participant. Volume data and surface maps are shown from the same individual during neutral cuing (N) and selective cuing to frequency (SF) tasks. A: Horizontal slices (Z = 46) taken through regions marked on the surface rendered cortex. Coronal slices taken near the centers of four ROI: PMv, BA 1, Ant. 7, and Post. 7. B: Inflated view of surface was obtained after segmentation and surface rendering of participant's left hemisphere. Arrows point at centers of gravity (COG) for ROI: PMv, BA1, BA7‐anterior, and BA7‐posterior. Corresponding locations are marked by + on the coronal slices in A. COG based on composite conjunction maps for 12 participants (see text). C: The same regions are noted in a flattened model for the left hemisphere of this individual.

Seventeen ROI in the left and 12 ROI in the right hemispheres were analyzed. Visual cortex activity was not evaluated because it presumably reflected the visual cues. Selected ROI are shown in Figure 3. In traditional somatosensory responsive areas, these included relatively small ROI in the contralateral finger representations in BA 3 and BA 1. Results from the minimal responses in ipsilateral S1 showed no task distinctions and were not considered further. The parietal opercular ROI were OP1, 3, and 4 [Eickhoff et al., 2006a, c]; these ranged in size from 36 mm2 in left OP3 to 521 mm2 in left OP1 (Table II). Posterior to the OP1 is the BA 40 portion of the inferior supramarginal cortex that contained moderate sized (∼1,000 mm2) ROI. We identified several ROI previously included as components of a dorsal parietal‐frontal attention network [Corbetta and Shulman, 2002]. These ROI were the anterior and posterior portions of BA 7, dorsal and ventral premotor (PMv and PMd) and the frontal eye field (FEF) region. Additional frontal cortex ROI involved inferior frontal language areas (BA 44–46) and a region within the frontal operculum. In medial frontal cortex, the ROI involved SMA and pre‐SMA portions of BA 6 and 9 and 32, respectively [Picard and Strick, 1996]. In temporal cortex, a ROI in BA 22 overlapped a previously defined multisensory area [Beauchamp et al., 2004].

Figure 3.

Figure 3

Open in a new tab

Defined regions of interest (ROI) were painted onto a standard surface mesh of nodes (PALS‐B12 atlas). Top Row: Borders (green lines) drawn on inflated hemispheres indicate selected Brodmann areas and probabilistic parietal opercular regions (description and methods for creating inflated and flat maps can be found in Van Essen, 2005). Second Row: ROI painted onto a flattened representation of the PALS‐B12 atlas. Third Row: Enlargment shows the borders of the four parietal opercular regions defined previously (Eickhoff et al., 2006c) and the opercular ROI (black lines) defined within these borders in the current study.

Table II.

Region analyses

Area (mm2) COGx COGy COGz N vs. C N vs. SD N vs. SF C vs. SD C vs. SF SD vs. SF
LH ROI
 BA3, fingers 126 −48 −19 35 0.8231 0.9695 0.8308 0.8441 0.9932 0.8514
 BA1, fingers 143 −51 −21 43 0.8041 0.8972 0.5016 0.6858 0.5848 0.4076
 OP1 521 −54 −27 19 0.0030* 0.6862 0.1756 0.0142 <0.0001 0.0947
 OP3 36 −39 −14 17 0.0215 0.8713 0.7415 0.0083 0.0338 0.5943
 OP4 229 −57 −12 14 0.1141 0.5835 0.3824 0.2873 0.4453 0.7458
 BA 40 949 −55 −52 26 0.0046 0.5534 0.6578 0.0229 0.0042 0.8091
 BA 22 406 −54 −42 3 0.0129 0.0165 0.0228 0.9597 0.8634 0.8332
 BA 7, anterior 1676 −44 −52 42 0.2785 0.0051 0.0457 0.0652 0.3839 0.2672
 BA 7, posterior 1892 −26 −63 48 0.0044 0.0729 0.0845 <0.0001 <0.0001 0.8149
 BA 4, fingers‐hand 1208 −48 −10 40 0.0858 0.5456 0.6459 0.0113 0.0133 0.8581
 PMv 1675 −43 −1 36 0.1957 0.0454 0.1972 0.0032 0.0152 0.3815
 PMd 275 −24 −9 57 0.6537 0.1189 0.0470 0.0802 0.0354 0.7138
 FEF 246 −30 −12 51 0.0858 0.0773 0.3233 0.0009 0.0050 0.3405
 Pre‐SMA 695 −8 12 45 0.2646 0.1127 0.9736 0.5849 0.3129 0.1416
 SMA 608 −8 −7 56 0.5965 0.2769 0.3825 0.1455 0.1961 0.7597
 BA 44, 46 1216 −39 22 35 0.2175 0.0782 0.1985 0.5669 0.9483 0.5007
 Frontal operculum 405 ‐34 8 10 0.1640 0.6788 0.5382 0.3153 0.3014 0.9099
RH ROI
 OP1 398 49 ‐27 25 0.0033 0.2300 0.7288 0.0294 0.0010 0.1410
 OP4 398 54 ‐11 17 0.0046 0.4560 0.6780 0.0255 0.0105 0.7268
 BA 40 1194 54 ‐37 37 0.0002 0.5943 0.8743 0.0034 0.0011 0.7372
 BA 22 877 50 ‐37 5 0.0005 0.0435 0.0793 0.0952 0.0467 0.7522
 BA 7, anterior 1278 40 ‐47 45 0.1612 0.0319 0.0335 0.3713 0.4054 0.9343
 BA 7, posterior 2151 25 ‐61 50 0.0281 0.0593 0.0984 0.0003 0.0004 0.7256
 PMv 2033 35 ‐2 46 0.6812 0.0818 0.2127 0.1898 0.4192 0.5823
 Pre‐SMA 598 6 12 47 0.1937 0.1248 0.3680 0.7063 0.6795 0.4522
 SMA 827 5 ‐8 54 0.361 0.3282 0.7858 0.9648 0.5029 0.4655
 BA 44 1162 45 12 24 0.0068 0.0453 0.1091 0.6889 0.2980 0.5886
 BA 46 1303 38 30 30 0.0810 0.0924 0.2499 0.9796 0.4884 0.5152
 Frontal operculum 348 31 17 8 0.1494 0.6203 0.7568 0.3937 0.0880 0.4460
Open in a new tab
*

Significance corrected for multiple comparisons based on six tests per ROI (shaded cells: ≤0.008).

The principal observation shown in group‐level t‐test of significant activity (Fig. 4) was the comparable distribution of activity across the different tasks. However, some regions, including posterior parietal BA 7 and frontal BA 6 regions (Fig. 4), showed higher mean Z‐scores that were distributed over a wider extent during N and C compared with both selective attention tasks. Significant participant responses were also more consistently observed in these same ROI as illustrated in Supplementary Figure 1. Most dorsal parietal–frontal, frontal language, SMA, and BA 40 ROI had node clusters that, on average, involved overlap from ≥75% of the participants during the N and C tasks. Additionally, BA 7 showed clusters of nodes with nearly 100% participant overlap during the N and C tasks. In comparison, only the left OP1 somatosensory region showed overlapping activation from ≥75% of the participants. These high proportions were found only during the SD and SF tasks (Supplementary Fig. 1). These differences were further evaluated by entering regional time courses into random‐effects MANOVA.

Figure 4.

Figure 4

Open in a new tab

Multiple comparison corrected group‐level significance maps per task. ROI boundaries are shown in black and are identified in Figure 3.

Regional Differences by Task—Time Course Analyses

Primary (S1) and possible secondary (S2) somatosensory areas within lateral parietal cortex [Burton, 2001; Eickhoff et al., 2006a] showed differences in the time courses of responses between the tasks. Left S1 ROI (contralateral to stimulation) showed no (BA 3) or slightly elevated (BA 1) responses during the stimulation trials but no activity distinctions by task (Fig. 5). There was a slight rebound response in the S1 ROI after TR 6 when stimulation ended.

Figure 5.

Figure 5

Open in a new tab

Regional time course graphs by task. Findings from left and right hemisphere ROI are shown, respectively, in the two left and furthest right two columns of graphs. Data at each interval represent the group mean; for clarity only upward standard error bars are shown. Vibration trial timing (see Fig. 1), scaled to the graph ordinates, is shown below each column of graphs.

The parietal operculum ROI showed a variety of responses. Contralateral to stimulation OP1 exhibited positive BOLD responses during all tactile attention tasks; those during the N and SF tasks were significantly larger than corresponding suppressed responses during the control, backwards counting task (Fig. 5 and Table II). Contralateral OP3 had an extended suppressed response only during the C task and slight positive activation during the tactile attention tasks that significantly differed only for the SD vs. C contrast (Table II). Contralateral OP4 (not shown) did not respond during stimulation, had rebound responses when stimulation ended, but showed no response differences between tasks. Consistent with the notion that OP2 is responsive to vestibular stimuli [Eickhoff et al., 2006d], we found no significant group‐level responses in this ROI.

The affected ipsilateral parietal opercular ROI (OP1 and OP4) primarily showed response suppression during the C task. OP1, however, had a small positive response to vibrotactile stimulation during the SF task (Fig. 5). OP4 showed response suppression during all tasks with significantly greater suppression during the control task when compared with those during the N task (Table II).

BA 40 ROI bilaterally showed responses that involved significant strong suppression during the C task (Table II) and no positive responses to vibrotactile stimulation during any tactile attention tasks (Fig. 5). The activity in BA 40 most resembled responses in ipsilateral OP1.

These results indicate that (1) S1 contributed minimally to tactile attention processing; (2) OP1 is a likely candidate S2 region associated with enhanced responses when attending tactile stimulation; and (3) an indirect component of tactile attention processing is response suppression in several parietal regions when vibrotactile stimulation is irrelevant to a performed task.

Anterior and posterior subdivisions of BA 7 showed different responses to the tasks. Bilaterally, posterior BA 7 responses were significantly greater during the control compared with nearly all tactile attention tasks (Table II). In contrast, anterior BA 7 ROI had larger responses during neutral cuing (Fig. 5). These were significantly greater than those evoked when selectively cued to vibrotactile duration on the left (Table II).

Responses bilaterally in BA 22 ROI were similar to those noted in anterior BA 7 in being larger during the N task (Fig. 5). These larger responses were significant in contrasts with the control task on the right and approached significance on the left (Table II).

These results indicate that (1) posterior BA 7 was strongly affected when visually attending to the numbers; (2) dividing attention between tactile attributes evoked greater activity in anterior BA 7; (3) anterior and posterior subdivisions of BA 7 make modality distinguishable contributions to attention processing; and (4) multisensory capabilities in BA 22 might extend to multi‐attribute processing when attention is divided between the frequency and duration parameters of vibrotactile stimulation.

All frontal cortical ROI showed positive BOLD responses during all tasks with larger responses often present during the control task (Fig. 5: PMv, FEF, BA 4, BA 44/46). Although not shown, responses in left PMd and bilaterally in SMA, pre‐SMA and frontal operculum were similar. Responses in SMA were larger than those in pre‐SMA. On the left, activity during the C task was significantly larger than during the SD task in PMv, during the SD and SF tasks in FEF, and during the N task in the right BA 44 (Table II). For these regions the results indicate (1) that no distinguishable selective vs. divided tactile attention effects occurred in any pre‐motor, motor or frontal language areas and (2) that significant effects primarily involved larger responses when viewing numbers for the backwards counting task.

DISCUSSION

The following discussion first considers possible contributions of activated somatosensory regions to tactile attention and asks whether a single region predominates. Next, we evaluate the probable equivalence in cognitive requirements in the present tactile attention and prior visual attention studies and examine whether such comparability explains widespread activation of the same dorsal posterior parietal and frontal cortex regions linked with visual attention [Corbetta and Shulman, 2002]. We also examine whether similar activation of dorsal parietal–frontal regions by tactile and visual attention tasks suggests the existence of a modality‐independent attention network or whether comparable activation principally indicates a network dedicated to goal directed behaviors [Fox et al., 2005]. We conclude with a discussion of regional distinctions between the different tactile attention tasks and ask whether the experimental design used for this fMRI study could resolve vibrotactile filtering mechanisms evident in prior behavioral studies [Sinclair et al., 2000].

Somatosensory Areas

The primary (S1) somatosensory cortical areas, by definition [Mountcastle, 2005], process tactile information and hence, are activated when attending to vibrotactile stimulation. Whether attention leads to greater activation in S1 is in doubt. In the present study, BA 3 showed few activity differences; BA 1 response distinctions were negligible. There were no significant differences in responses when the vibrotactile stimulation parameters were attended compared to when the same stimulation was not relevant during backwards counting. Selective or divided attention did not selectively activate any part of S1. Stimulating only a fraction of the right index fingertip might explain these negative results because the cortical representation of the affected skin is probably extremely circumscribed in S1, and yet localization of the second digit representation is variable in imaging studies [Kurth et al., 2000; Maldjian et al., 1999]. Group analyses might have missed a small and variable S1 representation, especially in BA 3. More probable, however, is that current findings concur with earlier reports of inconsequential tactile attention effects in S1 in both humans and monkeys [see citations in Burton and Sinclair, 2000a]. Others, however, reported attention effects in S1 [Johansen‐Berg et al., 2000; Meador et al., 2002; Van de Winckel et al., 2005]. Different experimental designs possibly underlie these disagreements, especially where control tasks in contrasting studies differ. In the current study the control was an attention demanding backwards counting task during which concurrent vibrotactile stimulation matched the parameters presented during the tactile attention tasks. Attention was therefore required during all tasks. In many studies the control was not a demanding attention task, but involved instructions to ignore the tactile stimulation. Additionally, contrasts between attention and control tasks have not always involved identical tactile stimulation parameters. Thus, prior reports of larger S1 responses when attending compared to control conditions can be alternatively explained and presumably do not support the conclusion that reported S1 response enhancements specifically depended on tactile attention.

Other candidate somatosensory areas were activated in the present study. Principally amongst these were parietal opercular regions defined as the second somatosensory area (S2) [Backes et al., 2000; Burton, 2001; Burton et al., 1997a; Disbrow et al., 2000, 2003; Eickhoff et al., 2006a; Fujiwara et al., 2002; Hamada et al., 2003; Hämäläinen et al., 2002; Ledberg et al., 1995; Mima et al., 1998; Nelson et al., 2004; Reed et al., 2004]. Previous studies noted mostly enhanced responses in S2 during tactile attention tasks [Backes et al., 2000; Burton and Sinclair, 2000b; Burton et al., 1999; Fujiwara et al., 2002; Hämäläinen et al., 2002; Johansen‐Berg et al., 2000; Macaluso et al., 2002; Young et al., 2004]. In the present study, OP1 contralateral to the stimulated finger was the only opercular parietal region that showed significantly enhanced responses during all tactile attention compared with the control task. There were no significant differences between the divided and selectively cued tasks despite larger responses during the former. Prior reports of larger S2 responses during tactile attention probably manifested activity in OP1. Accompanying enhanced activity in the contralateral OP1 was suppression of responses during the control task in ipsilateral OP1. Additionally, ipsilateral OP4 showed suppressed activity during all tasks with significantly greater suppression during the control task. Contralateral OP3 was slightly activated during the attention tasks. Thus, tactile attention processes engaged different opercular parietal regions in potentially distinctive tactile processes; attention did not just cause enhanced responses in OP1 although these were the most prominent. Being able to detect response differences between the various OP ROI was possibly a consequence of using surface alignment procedures because these optimized seeing functional correspondence across the group despite varied individual parietal opercular anatomy. The multiple and divergent OP ROI responses are consistent with prior cytoarchitectonic and functional distinctions within the parietal operculum (Eickhoff et al., 2006a, b, c) and further support the notion that it is misleading to retain the concept of a single S2 region. Prior regional distinctions, primarily described in animals, have not noted tactile attention effect differences in the identified regions [Burton et al., 1997b; Disbrow et al., 2003], which currently makes it unreasonable to suggest analogies between these and the human OP ROI.

A rostral segment of BA 40 was assigned to OP1 based on recent cytoarchitectonic parcellation of the parietal operculum in humans [Eickhoff et al., 2006c]. Prior imaging studies that described tactile attention effects in S2 potentially included this rostral segment of BA 40 in their analysis. However, in the current study lateral parietal BA 40, exclusive of OP1, showed no activity during the tactile attention tasks but had suppressed responses during the control task. It is unclear whether response suppression to a predominately visual processing task reflected involvement in tactile processing, especially in a region also known to be effected by visual stimulation [Andersen et al., 1997]. However, the superior part of BA 40 that is most closely associated with the inferior supramarginal gyrus is possibly homologous to area 7b in monkeys [Burton, 2001; Eidelberg and Galaburda, 1984], which contains neurons responsive to tactile and visual stimulation [Hyvärinen, 1982; Robinson and Burton 1980a, b]. These conflicting findings indicate that a role for BA 40 in tactile attention remains undefined and that BA 40 may have distinguishable functional subdivisions.

A portion of posterior superior temporal sulcal cortex is a multi‐sensory region that responds to tactile, auditory, and visual stimulation [Beauchamp et al., 2004; Burton et al., 2004, 2006; Macaluso and Driver, 2005]. Uniquely in the current study, activity in BA 22 ROI was greater when attention was divided between vibration frequency and duration attributes compared to when each attribute was selectively attended. An explanation originating from greater effort when dividing attention is not consistent with the absence of comparable distinctions in SMA or pre‐SMA, regions associated with task effort levels. Further studies need to discover whether multi‐modality processing in BA 22 extends to within‐modal attribute distinctions.

Dorsal Parietal‐Frontal Regions

The cognitive demands in the present study were similar to those in prior visual attention studies [Corbetta and Shulman, 2002]. Thus, selectively and neutrally cued tactile attention tasks involved cognitive selection of a vibration attribute (frequency or duration). The tactile paradigms required selection of sensory information that incorporated target search and detection of requisite vibrations. Additionally, response selection followed target capture. Backwards counting engaged visual and subsequent language selection, ignoring and possible repression of concurrent tactile information, and successive selection of number responses. Regions affected in fMRI studies of visual attention show BOLD responses during target search and detection. These responses rise to a peak during search and then persist until target detection [Shulman et al., 2003] when the BOLD signal rapidly decreases to pre‐search levels. The duration of these responses reflect termination of target searching [Corbetta and Shulman, 2002; Sapir et al., 2005]. In the current experiment, target search and detection and response selections extended across three closely spaced trials, which possibly elicited the sustained responses observed throughout most or all of the epoch intervals. The correspondence of target search and response selection between the current tactile attention and prior visual attention studies possibly explains the concurrence of activated ROI in premotor cortex, including the left frontal eye field and dorsal/posterior parietal regions.

The dorsal parietal and nearly all ROI in frontal cortex showed greater overlap in the distributions of significant activation across participants during neutral cuing and backwards counting. The consistency of activation across the group in these regions possibly implicates equivalent cognitive demands for these two tasks compared to those with selective cuing. Divided attention required greater effort and more short‐term memory because both frequency and duration had to be monitored and evaluated to identify a target. Backwards counting involved verbal memory; and response selection was also more effortful because initiation and execution changed for successive counts. In contrast, selective cuing tasks were less demanding because resources could be focused on particular targets and responses. These cognitive differences between tasks were apt to underlie the differences in regional response consistency, especially in those ROI required for more demanding tasks.

A variety of tactile tasks commonly affect dorsal parietal cortex [Binkofski et al., 1999a; Bodegård et al., 2001; Harada et al., 2004; Prather et al., 2004; Reed et al., 2005; Roland et al., 1998; Stoeckel et al., 2003, 2004; Van Boven et al., 2005; Zhang et al., 2005]. Participants actively explored attended tactile objects in most of these studies, which thereby added somatomotor and spatial perception variables to the tasks. These studies generally noted activation of a BA 7 site similar to our anterior BA 7 subdivision. The present study utilized exclusively passive tactile stimulation and covert responses, which precluded somatomotor/spatial factors. Generally our anterior BA 7 ROI showed negligible response differences between tasks except that responses were larger during neutral cuing when both tactile attributes had to be processed. The absence of significant task differences, especially between the tactile attention vs. control tasks, challenges the suggestion that anterior BA 7 cortex is necessarily dedicated to tactile attention per se.

These same studies also noted corresponding activity in prefrontal cortex. Visual attention similarly affects prefrontal cortex together with a posterior BA 7 subdivision [Astafiev et al., 2003; Corbetta, 1998; Corbetta and Shulman, 2002; Corbetta et al., 1991, 1993; Shulman et al., 2002, 2003]. Our tactile attention and backwards counting tasks activated equivalent prefrontal and posterior BA 7 regions. However, consistent with potentially greater involvement with visual targets [e.g., Astafiev et al., 2003], the more visually demanding backwards counting task elicited larger responses in all of these ROI. Collectively, prior and current results suggest that activity in these regions indicates resource mobilization for target or response selection, which are cognitive components whenever discriminating stimulation attributes, whether tactile or visual. A probable encompassing interpretation is that these are regions activated when performance is goal‐directed [Fox et al., 2005], which is a component of selective target search irrespective of modality or method of search (e.g., visual or haptic exploration). An earlier theory suggested that goal directed behaviors more critically involve appropriate response selection and not selective processing of stimulation when paying attention [Allport, 1989]. Response selection occurs even without overt actions. Hence, although in the present study stimulation was passively applied, the covert response selection might have required the same processes as were engaged during somatomotor tasks and haptic exploration. However, the current and nearly all prior attention studies do not dissociate target search from response selection variables.

Neutral versus Selective Cuing and TactileAttribute Filtering

We previously showed that attention could be directed separately to the duration and frequency of vibrotactile stimuli [Sinclair et al., 2000]. However, in the present study, no regions responded differently to either of the selectively cued attributes. Also, except for responses in anterior BA 7, contralateral to tactile stimulation, no regions showed significant response distinctions during divided and focused attention. Differences in experimental design may have affected these findings. In the current study, the same cue applied to sets of three trials. In the earlier study, individual trials were cued in random order [Sinclair et al., 2000]. Potentially revealing distinctions in cognitive dependent activity may occur while marshalling resources and may disappear once attention is focused. Unfortunately, pilot fMRI experiments using the single trial cuing procedure of our earlier behavioral study found almost no BOLD response differences anywhere. Because of the rapidity of the single‐trial cuing design, we speculated that BOLD activity related to the cognitive set required for one trial might partly persist into the next trial, blurring distinctions in BOLD responses between cuing conditions. Results of the current study, in which three trials were cued alike with long periods between sets, suggest the possibility that the same cognitive resources are utilized for detecting differences in the duration and frequency of vibrations. It remains possible that subtle differences in the way these resources are used for different tactile attributes may require greater temporal sensitivity than current BOLD technology. For example, there is a trend toward early higher responses in OP1 when selectively cued to frequency (see Fig. 5). Alternately, grouping cued trials might have reduced sensitivity to transitory activity related to altering cognitive state as directed by cue. These issues remain for future study.

Supporting information

This article contains supplementary material available via the Internet at http://www.interscience.wiley.com/jpages/1065-9471/suppmat

Acknowledgements

We are indebted to D. Van Essen, A. Snyder, M. McAvoy, D. Dierker, and J. Harwell for analysis software, J. Kreitler for design and construction of the tactile stimulator, and J. Wingert and S. Dixit for assistance with data analyses.

Footnotes

1

F‐statistics make no assumptions about underlying hemodynamic responses, but yield z‐scores “equivalent to the highest obtained with… cross‐correlation method(s)” (p. 223, Ollinger et al., 2001b). The null hypothesis is that the signal time series associated with a task is “noise” modeled by a constant plus a linear trend plus all other task signals. The alternative hypothesis is that the time series contains a “task related signal plus noise” and thereby has lower variance (http://afni.nimh.nih.gov/pub/dist/doc/manual/Deconvolvem.pdf) (Ward, 2006).

2

Through spatial smoothing, surface maps probably captured activity anywhere within the cortical grey matter despite segmentation through ∼Layer 4 in volume slices. Image data was collected in 6 × 3.75 × 3.75 mm3volumes, interpolated to 2 mm3and then smoothed to 4 mm at FWHM prior to GLM analyses. Next, significance was evaluated using multiple thresholds and face connected voxels; clusters varied from 5–45, which in a strict white‐to‐grey linear dimension exceeds cortical thickness.

3

Cytoarchitecture maps of 10 post‐mortem brains (provided by Eickhoff et al., 2006a) were linearly transformed to a Talairach atlas (Talairach and Tournoux, 1988) representative target template. A probabilistic volume for each operculum subdivision was created by summing across individuals. The probabilistic volumes were mapped to the cortical surface. Thresholds for each opercular region were manipulated to (i) achieve contiguous surface regions (for each region and group of regions) and (ii) produce volumes with similar sizes and spatial extent that matched previously published data (Eickhoff et al., 2006a). The areas for OP1 to OP4, respectively, for the left hemisphere are: 877, 162, 266, and 682 mm2. In the right hemisphere, the areas are: 1185, 166, 475, and 901 mm2.

REFERENCES

  1. Allport A ( 1989): Visual attention In: Posner MI, editor. Foundations of Cognitive Science. Cambridge: MIT; pp 631–682. [Google Scholar]
  2. Andersen RA, Snyder LH, Bradley DC, Xing J ( 1997): Multimodal representation of space in the posterior parietal cortex and its use in planning movements. Annu Rev Neurosci 20: 303–330. [DOI] [PubMed] [Google Scholar]
  3. Astafiev SV, Shulman GL, Stanley CM, Snyder AZ, Van Essen DC, Corbetta M ( 2003): Functional organization of human intraparietal and frontal cortex for attending, looking, and pointing. J Neurosci 23: 4689–4699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Backes WH, Mess WH, van Kranen‐Mastenbroek V, Reulen JPH ( 2000): Somatosensory cortex responses to median nerve stimulation: fMRI effects of current amplitude and selective attention. Clin Neurophysiol 111: 1738–1744. [DOI] [PubMed] [Google Scholar]
  5. Beauchamp MS, Lee KE, Argall BD, Martin A ( 2004): Integration of auditory and visual information about objects in superior temporal sulcus. Neuron 41: 809–823. [DOI] [PubMed] [Google Scholar]
  6. Binkofski F, Buccino G, Posse S, Seitz R, Rizzolatti G, Freund H ( 1999a): A fronto‐parietal circuit for object manipulation in man: Evidence from an fMRI‐study. Eur J Neurosci 11: 3276–3286. [DOI] [PubMed] [Google Scholar]
  7. Binkofski F, Buccino G, Stephan K, Rizzolatti G, Seitz R, Freund HJ ( 1999b): A parieto‐premotor network for object manipulation: Evidence from neuroimaging. Exp Brain Res 128: 210–213. [DOI] [PubMed] [Google Scholar]
  8. Bodegård A, Geyer S, Grefkes C, Zilles K, Roland PE ( 2001): Hierarchical processing of tactile shape in the human brain. Neuron 31: 317–328. [DOI] [PubMed] [Google Scholar]
  9. Bosch V ( 2000): Statistical analysis of multi‐subject fMRI data: Assessment of focal activations. J Magn Reson Imaging 11: 61–64. [DOI] [PubMed] [Google Scholar]
  10. Burton H ( 2001): Cerebral cortex regions devoted to the somatosensory system: Results from brain imaging studies in humans In: Nelson RJ, editor. The Somatosensory System: Deciphering the Brain's Own Body Image. New York: CRC; pp 27–72. [Google Scholar]
  11. Burton H, Sinclair R ( 2000a): Tactile‐spatial and cross‐modal attention effects in the primary somatosensory cortical areas 3b and 1–2 of rhesus monkeys. Somatosens Mot Res 17: 213–228. [DOI] [PubMed] [Google Scholar]
  12. Burton H, Sinclair RJ ( 2000b): Attending to and remembering tactile stimuli: A review of brain imaging data and single‐neuron responses. J Clin Neurophysiol 17: 575–591. [DOI] [PubMed] [Google Scholar]
  13. Burton H, MacLeod AM, Videen TO, Raichle ME ( 1997a): Multiple foci in parietal and frontal cortex activated by rubbing embossed grating patterns across fingerpads: A positron emission tomography study in humans. Cereb Cortex 7: 3–17. [DOI] [PubMed] [Google Scholar]
  14. Burton H, Sinclair R, Hong S, Pruett JJ, Whang K ( 1997b): Tactile‐spatial and cross‐modal attention effects in the second somatosensory and 7b cortical areas of rhesus monkeys. Somatosens Mot Res 14: 237–267. [DOI] [PubMed] [Google Scholar]
  15. Burton H, Abend NS, MacLeod AM, Sinclair RJ, Snyder AZ, Raichle ME ( 1999): Tactile attention tasks enhance activation in somatosensory regions of parietal cortex: A positron emission tomography study. Cereb Cortex 9: 662–674. [DOI] [PubMed] [Google Scholar]
  16. Burton H, Sinclair R, McLaren D ( 2004): Cortical activity to vibrotactile stimulation: A fMRI study in blind and sighted individuals. Hum Brain Mapp 23: 210–228. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Burton H, McLaren D, Sinclair R ( 2006): Reading embossed capital letters: A fMRI study in blind and sighted individuals. Hum Brain Mapp 27: 325–339. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Corbetta M ( 1998): Frontoparietal cortical networks for directing attention and the eye to visual locations: Identical, independent, or overlapping neural systems?. Proc Natl Acad Sci USA 95: 831–838. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Corbetta M, Shulman GL ( 2002): Control of goal‐directed and stimulus‐driven attention in the brain. Nat Rev Neurosci 3: 201–215. [DOI] [PubMed] [Google Scholar]
  20. Corbetta M, Miezin FM, Dobmeyer S, Shulman GL, Petersen SE ( 1991): Selective and divided attention during visual discriminations of shape, color, and speed: Functional anatomy by positron emission tomography. J Neurosci 11: 2383–2402. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Corbetta M, Miezin FM, Shulman GL, Petersen SE ( 1993): A PET study of visuospatial attention. J Neurosci 13: 1202–1226. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Disbrow E, Roberts T, Krubitzer L ( 2000): Somatotopic organization of cortical fields in the lateral sulcus of Homo sapiens: Evidence for SII and PV. J Comp Neurol 418: 1–21. [DOI] [PubMed] [Google Scholar]
  23. Disbrow E, Litinas E, Recanzone GH, Padberg J, Krubitzer L ( 2003): Cortical connections of the second somatosensory area and the parietal ventral area in macaque monkeys. J Comp Neurol 462: 382–399. [DOI] [PubMed] [Google Scholar]
  24. Drury HA, Van Essen DC, Corbetta M, Snyder AZ ( 1999): Surface‐based analyses of the human visual cortex In: Toga AW, editor. Brain Warping. New York: Academic Press; pp 337–363. [Google Scholar]
  25. Eickhoff SB, Amunts K, Mohlberg H, Zilles K ( 2006a): The human parietal operculum. II. Stereotaxic maps and correlation with functional imaging results. Cereb Cortex 16: 268–279. [DOI] [PubMed] [Google Scholar]
  26. Eickhoff SB, Lotze M, Wietek B, Amunts K, Enck P, Zilles K ( 2006b): Segregation of visceral and somatosensory afferents: An fMRI and cytoarchitectonic mapping study. Neuroimage 31: 1004–1014. [DOI] [PubMed] [Google Scholar]
  27. Eickhoff SB, Schleicher A, Zilles K, Amunts K ( 2006c): The human parietal operculum. I. Cytoarchitectonic mapping of subdivisions. Cereb Cortex 16: 254–267. [DOI] [PubMed] [Google Scholar]
  28. Eickhoff SB, Weiss PH, Amunts K, Fink GR, Zilles K ( 2006d): Identifying human parieto‐insular vestibular cortex using fMRI and cytoarchitectonic mapping. Hum Brain Mapp 27: 611–621. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Eidelberg D, Galaburda AM ( 1984): Inferior parietal lobule: Divergent architectonic asymmetries in the human brain. Arch Neurol 41: 843–852. [DOI] [PubMed] [Google Scholar]
  30. Forman SD, Cohen JD, Fitzgerald M, Eddy WF, Mintun MA, Noll DC ( 1995): Improved assessment of significant activation in functional magnetic resonance imaging (fMRI): Use of a cluster‐size threshold. Magn Reson Med 33: 636–647. [DOI] [PubMed] [Google Scholar]
  31. Fox MD, Snyder AZ, Vincent JL, Corbetta M, Van Essen DC, Raichle ME ( 2005): The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc Natl Acad Sci USA 102: 9673–9678. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Fujiwara N, Imai M, Nagamine T, Mima T, Oga T, Takeshita K, Toma K, Shibasaki H ( 2002): Second somatosensory area (SII) plays a significant role in selective somatosensory attention. Brain Res Cogn Brain Res 14: 389–397. [DOI] [PubMed] [Google Scholar]
  33. Garner W ( 1974): The Processing of Information and Structure. Potomac, MD: Erlbaum. [Google Scholar]
  34. Hamada Y, Okita H, Suzuki R ( 2003): Effect of interstimulus interval on attentional modulation of cortical activities in human somatosensory areas. Clin Neurophysiol 114: 548–555. [DOI] [PubMed] [Google Scholar]
  35. Hämäläinen H, Hiltunen J, Titievskaja I ( 2002): Activation of somatosensory cortical areas varies with attentional state: An fMRI study. Behav Brain Res 135: 159–165. [DOI] [PubMed] [Google Scholar]
  36. Harada T, Saito DN, Kashikura KI, Sato T, Yonekura Y, Honda M, Sadato N ( 2004): Asymmetrical neural substrates of tactile discrimination in humans: A functional magnetic resonance imaging study. J Neurosci 24: 7524–7530. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Hsiao SS, O'Shaughnesy DM, Johnson KO ( 1993): Effects of selective attention on spatial form processing in monkey primary and secondary somatosensory cortex. J Neurophysiol 70: 444–447. [DOI] [PubMed] [Google Scholar]
  38. Hyvärinen J ( 1982): Posterior parietal lobe of the primate brain. Physiol Rev 62: 1060–1129. [DOI] [PubMed] [Google Scholar]
  39. Johansen‐Berg H, Christensen V, Woolrich M, Matthews PM ( 2000): Attention to touch modulates activity in both primary and secondary somatosensory areas. Neuroreport 11: 1237–1241. [DOI] [PubMed] [Google Scholar]
  40. Jonides J, Mack R ( 1984): On the cost and benefit of cost and benefit. Psychol Bull 96: 29–44. [Google Scholar]
  41. Kahneman D ( 1973): Attention and Effort. Englewood Cliffs, NJ: Prentice‐Hall; 246 p. [Google Scholar]
  42. Katz D ( 1989): The World of Touch. Krueger LE, translator. Hillsdale, NJ: Lawrence Erlbaum; 260 p. [Google Scholar]
  43. Kurth R, Villringer K, Curio G, Wolf KJ, Krause T, Repenthin J, Schwiemann J, Deuchert M, Villringer A ( 2000): fMRI shows multiple somatotopic digit representations in human primary somatosensory cortex. Neuroreport 11: 1487–1491. [PubMed] [Google Scholar]
  44. Ledberg A, O'Sullivan BT, Kinomura S, Roland PE ( 1995): Somatosensory activations of the parietal operculum of man. A PET study. Eur J Neurosci 7: 1934–1941. [DOI] [PubMed] [Google Scholar]
  45. Macaluso E, Driver J ( 2005): Multisensory spatial interactions: A window onto functional integration in the human brain. Trends Neurosci 28: 264–271. [DOI] [PubMed] [Google Scholar]
  46. Macaluso E, Frith CD, Driver J ( 2002): Directing attention to locations and to sensory modalities: Multiple levels of selective processing revealed with PET. Cereb Cortex 12: 357–368. [DOI] [PubMed] [Google Scholar]
  47. Maldjian JA, Gottschalk A, Patel RS, Detre JA, Alsop DC ( 1999): The sensory somatotopic map of the human hand demonstrated at 4 Tesla. Neuroimage 10: 55–62. [DOI] [PubMed] [Google Scholar]
  48. Meador KJ, Allison JD, Loring DW, Lavin TB, Pillai JJ ( 2002): Topography of somatosensory processing: Cerebral lateralization and focused attention. J Int Neuropsychol Soc 8: 349–359. [DOI] [PubMed] [Google Scholar]
  49. Miezin FM, Maccotta L, Ollinger JM, Petersen SE, Buckner RL ( 2000): Characterizing the hemodynamic response: Effects of presentation rate, sampling procedure, and the possibility of ordering brain activity based on relative timing. Neuroimage 11: 735–759. [DOI] [PubMed] [Google Scholar]
  50. Mima T, Nagamine T, Nakamura K, Shibasaki H ( 1998): Attention modulates both primary and second somatosensory cortical activities in humans: A magnetoencephalographic study. J Neurophysiol 80: 2215–2221. [DOI] [PubMed] [Google Scholar]
  51. Mountcastle VB ( 2005): The Sensory Hand: Neural Mechanisms of Somatic Sensation. Cambridge, MA: Harvard University Press; 616 p. [Google Scholar]
  52. Mugler JPD, Brookeman JR ( 1990): Three‐dimensional magnetization‐prepared rapid gradient‐echo imaging (3D MP RAGE). Magn Reson Med 15: 152–157. [DOI] [PubMed] [Google Scholar]
  53. Nelson AJ, Staines WR, Graham SJ, McIlroy WE ( 2004): Activation in SI and SII; the influence of vibrotactile amplitude during passive and task‐relevant stimulation. Brain Res Cogn Brain Res 19: 174–184. [DOI] [PubMed] [Google Scholar]
  54. Ollinger JM, Corbetta M, Shulman GL ( 2001a): Separating processes within a trial in event‐related functional MRI. I. The method. Neuroimage 13: 218–229. [DOI] [PubMed] [Google Scholar]
  55. Ollinger JM, Shulman GL, Corbetta M ( 2001b): Separating processes within a trial in event‐related functional MRI. II. Analysis. Neuroimage 13: 210–217. [DOI] [PubMed] [Google Scholar]
  56. Picard N, Strick PL ( 1996): Motor areas of the medial wall: A review of their location and functional activation. Cereb Cortex 6: 342–353. [DOI] [PubMed] [Google Scholar]
  57. Posner MI ( 1986): Chronometric Explorations of Mind. New York: Oxford University Press; 271 [Google Scholar]
  58. Prather SC, Votaw JR, Sathian K ( 2004): Task‐specific recruitment of dorsal and ventral visual areas during tactile perception. Neuropsychologia 42: 1079–1087. [DOI] [PubMed] [Google Scholar]
  59. Raczkowski D, Kalat JW, Nebes R ( 1974): Reliability and validity of some handedness questionnaire items. Neuropsychologia 12: 43–47. [DOI] [PubMed] [Google Scholar]
  60. Reed CL, Shoham S, Halgren E ( 2004): Neural substrates of tactile object recognition: An fMRI study. Hum Brain Mapp 21: 236–246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Reed CL, Klatzky RL, Halgren E ( 2005): What vs. where in touch: An fMRI study. Neuroimage 25: 718–726. [DOI] [PubMed] [Google Scholar]
  62. Rizzolatti G, Fogassi L, Gallese V ( 2002): Motor and cognitive functions of the ventral premotor cortex. Curr Opin Neurobiol 12: 149–154. [DOI] [PubMed] [Google Scholar]
  63. Robinson CJ, Burton H ( 1980a): Organization of somatosensory receptive fields in cortical areas 7b, retroinsula, postauditory and granular insula of M. fascicularis . J Comp Neurol 192: 69–92. [DOI] [PubMed] [Google Scholar]
  64. Robinson CJ, Burton H ( 1980b): Somatic submodality distribution within the second somatosensory (SII), 7b, retroinsular, postauditory, and granular insular cortical areas of M. fascicularis . J Comp Neurol 192: 93–108. [DOI] [PubMed] [Google Scholar]
  65. Roland PE, O'Sullivan B, Kawashima R ( 1998): Shape and roughness activate different somatosensory areas in the human brain. Proc Natl Acad Sci USA 95: 3295–3300. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Saad ZS, Reynolds RC, Argall B, Japee S, Cox RW ( 2004): SUMA: An Interface for Surface‐Based Intra‐ and Inter‐Subject Analysis with AFNI. pp 1510–1513.
  67. Sankoh A, Huque M, Dubey S ( 1997): Some comments on frequently used multiple endpoint adjustments in clinical trials. Sat Med 16: 2529–2542. [DOI] [PubMed] [Google Scholar]
  68. Sapir A, d'Avossa G, McAvoy M, Shulman GL, Corbetta M ( 2005): Brain signals for spatial attention predict performance in a motion discrimination task. Proc Natl Acad Sci USA 102: 17810–17815. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Shulman GL, d'Avossa G, Tansy AP, Corbetta M ( 2002): Two attentional processes in the parietal lobe. Cereb Cortex 12: 1124–1131. [DOI] [PubMed] [Google Scholar]
  70. Shulman GL, McAvoy MP, Cowan MC, Astafiev SV, Tansy AP, d'Avossa G, Corbetta M ( 2003): Quantitative analysis of attention and detection signals during visual search. J Neurophysiol 90: 3384–3397. [DOI] [PubMed] [Google Scholar]
  71. Sinclair R, Kuo J, Burton H ( 2000): Effects on discrimination performance of selective attention to tactile features. Somatosens Mot Res 17: 145–157. [DOI] [PubMed] [Google Scholar]
  72. Stoeckel MC, Weder B, Binkofski F, Buccino G, Shah NJ, Seitz RJ ( 2003): A fronto‐parietal circuit for tactile object discrimination: An event‐related fMRI study. Neuroimage 19: 1103–1114. [DOI] [PubMed] [Google Scholar]
  73. Stoeckel MC, Weder B, Binkofski F, Choi HJ, Amunts K, Pieperhoff P, Shah NJ, Seitz RJ ( 2004): Left and right superior parietal lobule in tactile object discrimination. Eur J Neurosci 19: 1067–1072. [DOI] [PubMed] [Google Scholar]
  74. Talairach J, Tournoux P ( 1988): Coplanar Stereotaxic Atlas of the Human Brain. New York: Thieme Medical; 122 p. [Google Scholar]
  75. Van Boven RW, Ingeholm JE, Beauchamp MS, Bikle PC, Ungerleider LG ( 2005): Tactile form and location processing in the human brain. Proc Natl Acad Sci USA 102: 12601–12605. [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Van de Winckel A, Sunaert S, Wenderoth N, Peeters R, Van Hecke P, Feys H, Horemans E, Marchal G, Swinnen SP, Perfetti C, De Weerdt W ( 2005): Passive somatosensory discrimination tasks in healthy volunteers: Differential networks involved in familiar versus unfamiliar shape and length discrimination. Neuroimage 26: 441–453. [DOI] [PubMed] [Google Scholar]
  77. Van Essen DC ( 2005): A population‐average landmark‐ and surface‐based (PALS) atlas of human cerebral cortex. NeuroImage 28: 635–662. [DOI] [PubMed] [Google Scholar]
  78. Ward B ( 2006): Deconvolution analysis of fMRI time series data In: Cox RW, editor. MCW AFNI. Milwaukee, WI: Medical College of Wisconsin; pp 1–100. [Google Scholar]
  79. Young JP, Herath P, Eickhoff S, Choi J, Grefkes C, Zilles K, Roland PE ( 2004): Somatotopy and attentional modulation of the human parietal and opercular regions. J Neurosci 24: 5391–5399. [DOI] [PMC free article] [PubMed] [Google Scholar]
  80. Zhang MM, Mariola E, Stilla R, Stoesz M, Mao H, Hu XP, Sathian K ( 2005): Tactile discrimination of grating orientation: fMRI activation patterns. Hum Brain Mapp 25: 370‐377. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

This article contains supplementary material available via the Internet at http://www.interscience.wiley.com/jpages/1065-9471/suppmat

Articles from Human Brain Mapping are provided here courtesy of Wiley

RESOURCES