Attention-modulated Alpha-band Oscillations Protect ...people.brandeis.edu/~sekuler/papers/Payne_Alpha... · (Haegens, Luther, & Jensen, 2012). Moreover, the handʼs representation

Attention-modulated Alpha-band Oscillations Protectagainst Intrusion of Irrelevant Information

Lisa Payne, Sylvia Guillory, and Robert Sekuler

Abstract

■ Combining high-density scalp EEG recordings with a sensi-tive analog measure of STMʼs fidelity, we characterized the tem-poral dynamics of intentional ignoring and related thosedynamics to the intrusion of task-irrelevant information. Oneach trial of the task, two study Gabors were briefly presentedin succession. A green or red disc preceding each Gabor signi-fied whether that Gabor should be remembered or ignored, re-spectively. With cue–stimulus intervals of 300, 600, or 900 msecpresented in separate sessions, we found that the onset of pos-terior, prestimulus alpha oscillations varied with the length ofthe interval. Although stimulus onset time was entirely predict-able, the longer the cue–stimulus interval, the earlier the in-crease in prestimulus alpha power. However, the alpha-band

modulation was not simply locked to the cue offset. The tem-poral envelopes of posterior alpha-band modulation were strik-ingly similar for both cued attending and cued ignoring anddiffered only in magnitude. This similarity suggests that cuedattending includes suppression of task-irrelevant, spatial process-ing. Supporting the view that alpha-band oscillations representinhibition, our graded measure of recall revealed that, when thestimulus to be ignored appears second in the sequence, peri-stimulus alpha power predicted the degree to which that irrel-evant stimulus distorted subsequent recall of the stimulus thatwas to be remembered. These results demonstrate that timelydeployment of attention-related alpha-band oscillations can aidSTM by filtering out task-irrelevant information. ■

INTRODUCTION

Cortical oscillations within the alpha band (8–14 Hz) aremarkers of task-related engagement or disengagement ofspecific brain regions. For example, during visual target de-tection tasks, when a cue directs attention to a lateralizedspatial location where an upcoming stimulus may be pre-sented, contralateral brain regions exhibit an anticipatorydecrease in alpha activity relative to precue levels (Huang& Sekuler, 2010a; Thut, Nietzel, Brandt, & Pascual-Leone,2006). Moreover, increases in alpha activity ipsilateral tothe target location have been reported to occur when thereare distracting stimuli within the uncued space (Kelly,Lalor, Reilly, & Foxe, 2006) and also when there were nodistracting stimuli in the unattended space (Rihs, Michel,& Thut, 2007). Similarly, in the somatosensory domain, in-creased alpha activity ipsilateral to a cued location sup-presses detection of stimulation of the uncued hand(Haegens, Luther, & Jensen, 2012). Moreover, the handʼsrepresentation in somatosensory cortex shows decreasedpower after a cue to attend to that hand, but shows in-creased power after a cue to attend to the foot ( Joneset al., 2010). These results suggest that spatial location isitself a distracter, and an important function of directed at-tention is to suppress processes of unattended space re-gardless of whether or not there exists an experimentallydefined, irrelevant stimulus in that space.

Modality-specific changes in alpha activity are also seenduring feature-based selection when participants arecued to attend to input from one particular sensory mod-ality of a compound stimulus. In tasks utilizing audio-visual stimuli, alpha amplitude increases selectively overparietal-occipital regions when participants are cued toattend to the auditory feature, indicating the suppressionof visual processing to attend to the auditory information(Fu et al., 2001; Foxe, Simpson, & Ahlfors, 1998). A double-dissociation between feature dimensions was observedwhen participants were cued to attend either the color orthe motion of a dot array (Snyder & Foxe, 2010). Alphapower over dorsal regions encoding motion increasedwhen color was cued, and alpha power over ventral regionsencoding color increased when motion was cued. There-fore, it is thought that such attentional processes entailnot only a decrease in alpha activity over regions of activeencoding, but also an increase over regions whose poten-tially distracting processing should be suppressed. In thisview, alpha oscillations reflect an active inhibitory mecha-nism, and alpha desynchronization promotes stimulusprocessing (for review, see Foxe & Snyder, 2011; Klimesch,Sauseng, & Hanslmayr, 2007).

Despite these successful manipulations of cued atten-tion, there is still little known about the timing of atten-tion modulation. A number of results have indicated thatsustained, attention-related changes begin approximately400–600 msec after cue onset (Rihs et al., 2007; Thutet al., 2006; Worden, Foxe, Wang, & Simpson, 2000). InBrandeis University

© 2013 Massachusetts Institute of Technology Journal of Cognitive Neuroscience 25:9, pp. 1463–1476doi:10.1162/jocn_a_00395

most of those examples, the onset of differential alphaactivity is also approximately 400–600 msec before thestimulus onset, making it difficult to interpret the timingof attention modulation. Specifically, increased alpha pre-ceding the stimulus that was to be ignored could havebeen adaptive or anticipatory in nature, the product ofparticipantsʼ expectation that the stimulus would occurat a predictable time after the cue; alternatively, the pre-stimulus increase in alpha could simply be time-locked tothe cueʼs offset, reflecting the time required to respondto the cue. To select between these very different accountsof the prestimulus increase in alpha, we manipulated theinterval between cue and stimulus with sessions of either300-, 600-, or 900-msec intervals. Each trial sequence con-sisted of two cue–stimulus pairs. A red cue signaled thatthe following distracting nontarget was to be ignored anda green cue signaled that the subsequent target was to beremembered and reproduced after a brief retention pe-riod. If the first cue of a trial instructed participants toattend, then the second cue would necessarily be theinstruction to ignore, and vice versa. In this way, thisdesign also allowed the comparison of alpha signal onsetbetween conditions in which the cue must be completelyprocessed versus a condition in which the cue would becompletely predictable.

Although compelling evidence suggests that alpha oscil-lations reflect the suppression of irrelevant processing,participants were actually cued to attend toward a spatiallocation or stimulus dimension during these tasks. Evenif participants were explicitly instructed to ignore the un-attended space or feature, the cue indicated the directionof attention, and any ignoring of distracters occurred simul-taneously. Intentional ignoring of a stimulus as a stand-alone trial event was first addressed with a workingmemory task during which a centrally presented stream of16 everyday objects were shown, each following a cue toeither attendor ignore the upcoming stimulus (Freunberger,Fellinger, Sauseng, Gruber, & Klimesch, 2009). After a briefretention period, a probe object was displayed, and parti-cipants indicated whether or not it had been in that trialʼsstream of attended objects. Objects that followed an ignorecue never appeared as the probe to encourage actualignoring of the irrelevant stimuli, thereby reducing thenumber of objects in memory. As hypothesized, a cue toignore an upcoming stimulus elicited higher posterioralpha power than a cue to attend to the stimulus. Althoughnot attending to the irrelevant stimuli would reduce thenumber of objects to be held in memory andmake the taskeasier, there was no way to determine if those stimuli hadbeen successfully ignored.

A direct link has not been demonstrated between alphaactivity and the success with which task-irrelevant infor-mation can be kept out of working memory. This poten-tial link is important, as the intrusion of task-irrelevantinformation is a major cause of failures in working mem-ory, particularly in older adults (Hasher & Zacks, 1988).To examine the link between alpha oscillations and intru-

sions into working memory, we used sinusoidal lumi-nance gratings stimuli (Gabors), together with a recalltechnique in which participants adjusted a stimulus tomatch the item held in memory. To enable reproductionof the target from memory, a Gabor patch was displayedwith a slider bar that increased or decreased the spatialfrequency as participants clicked along the bar. In thisway, participantsʼ reproductions of the spatial frequencyof the remembered stimulus yield a sensitive, trial-by-trialcontinuous measure of accuracy. This information-richmeasure of accuracy made it possible not only to gaugewhether there was an error in memory, but more impor-tantly to gauge the degree to which errors reflected theinfluence of the task-irrelevant stimulus (Dube, Zhou, &Sekuler, under review; Huang & Sekuler, 2010b). Manyfactors during each trial may contribute to the total errorincluding pure error of reproduction, but, as explainedbelow, the use of Gabor stimuli and the recall techniquemade it possible to disentangle the influence of otherstimuli seen during the experiment, particularly each trialʼsnontarget stimulus. Our hypothesis is that when a task-irrelevant stimulus is accompanied by higher alpha power,the brain will more effectively filter out that task-irrelevantstimulus so that its influence on memory is curtailed.

METHODS

Participants

Fourteen participants gave written informed consent andcompleted the experiment. Of these, two participantsʼ datahad to be excluded from our analysis because of exces-sive EEG artifacts (epoch rejection rate > 50%). Seven ofthe 12 remaining participants were women. Participantsʼages ranged from 18 to 30 years (mean = 22 years, SD =3.54 years). All were right-handed as determined by theEdinburgh Handedness Inventory (Oldfield, 1971) and hadnormal or corrected-to-normal vision as measured withSnellen targets. All participants denied psychological orneurological disorders. Participants were naive to the pur-pose of the experiment and were paid for participation.

Apparatus and Stimuli

Gabor stimuli were generated and displayed using Matlab 7(Mathworks, Natick, MA), supplemented by extensionsfrom the Psychophysics Toolbox (Brainard, 1997). EachGabor comprised a vertical sinusoidal luminance gratingwindowed by a circular Gaussian carrier with a spaceconstant of 1.14°. Each Gabor subtended 4.65°. The con-trast of a stimulusʼ sinusoidal component was fixed at0.2, a value well above detection threshold. To under-mine the possibility that participants might base theirjudgments on some local correspondence(s) betweenstimuli, the absolute phase of each Gaborʼs sinusoidalcomponent was perturbed on each trial by adding a ran-dom sample from a uniform distribution whose range

1464 Journal of Cognitive Neuroscience Volume 25, Number 9

was 0 to π/2. Stimuli were presented on a 21-in. cathoderay tube monitor with a refresh rate of 99.8 Hz and ascreen resolution of 1280 × 960 pixels. Screen lumi-nance was linearized by means of software adjustment,with the mean luminance of the screen held at 32 cd/m2.The red and green discs used to cue attention wereequiluminant with the displayʼs mean luminance. Duringtesting, participants viewed the computer display binoc-ularly from a distance of 57 cm.

Preliminary Measurements

To take account of individual differences in visual discrim-inability, each participantʼs spatial frequency discrimi-nation threshold was measured for the same type ofGabor that would be later used in testing STM. Each indi-vidualʼs discrimination threshold was then used to scalethe spatial frequencies of stimuli used in the memory test.The discrimination thresholds were used also to normal-ize the recall errors that individual participants made.Spatial frequency discrimination thresholds were esti-mated by two-alternative forced-choice trials controlledby QUEST, an adaptive tracking algorithm (Watson &Pelli, 1983). On each trial, two Gabors were presentedin sequence for 500 msec each. This replicated the tem-poral and spatial conditions that would be subsequentlyused in our study of working memory. A participant usedkeys on a computer keyboard to identify which Gabor, thefirst or second, had the higher spatial frequency. The higher

spatial frequency Gabor was equally likely to occupy thefirst or second positions in the sequence. A computer-generated tone provided feedback about response correct-ness during these preliminary, threshold measurements.

The lower spatial frequency of each trialʼs pair was cho-sen randomly from a uniform distribution that spanned0.5–5 cycles/degree. This covered the range of spatial fre-quencies that was used later to test working memory. TheQUEST algorithm controlled trialwise differences in thetwo Gaborsʼ spatial frequencies. As implemented here,QUEST estimated the difference in spatial frequency thatproduced correct judgments on 80% of attempts. Each par-ticipantʼs discrimination threshold was estimated in threeseparate, successive runs of QUEST. The lowest of thethree resulting threshold estimates was used to representthe participantʼs discrimination threshold. Participantsʼ dis-crimination thresholds ranged from 9% to 20% (mean =13.25%, SD = 3.41%).

Procedure

As Figure 1 illustrates, two study Gabors were presentedin sequence on each trial. A green or red disc immedi-ately before a Gabor signified whether that Gabor wasto be remembered ( ) or ignored ( ). The presentationof the two Gabors was followed by a 1-sec retention pe-riod. Then, the presentation of a comparison Gabor andslidebar signaled the participant to adjust the spatialfrequency to match the remembered spatial frequency

Figure 1. Schematic diagram illustrating a trialʼs event structure. Each trial began with a fixation cross that oriented the participant to the region ofthe computer display within which the trialʼs stimuli would be presented. The fixation point was replaced either by a green disc or a red disc.The green disc cued the participant that the spatial frequency of the ensuing Gabor stimulus should be remembered; a red disc cued the participantthat the next Gaborʼs spatial frequency should be ignored. A cue–stimulus interval of either 300, 600, or 900 msec followed (blocked design),and then the first of two Gabor stimuli was presented. Immediately thereafter, a second cue was presented. This cue was a disc that was green,if the first cue had been red, or red, if the first cue had been green. Then a second cue–stimulus interval followed; this interval was alwaysthe same as the trialʼs initial cue–stimulus interval, that is, either 300, 600, or 900 msec. Next a second Gabor stimulus was presented,which was followed by a 1-sec-long retention interval. Finally, a comparison Gabor appeared whose spatial frequency could be adjusted tomatch the remembered spatial frequency of the Gabor that the participant had been cued to remember. Top row: on half the trials, the targetGabor to be attended appeared first and the nontarget Gabor to be ignored appeared second (hereafter, we refer to this sequence as T1N2).Bottom row: on half of the trials the nontarget Gabor to be ignored appeared first and the target Gabor to be attended appeared second(hereafter, N1T2).

Payne, Guillory, and Sekuler 1465

of the target Gabor. For half of the participants, the initialspatial frequency of the comparison Gabor was at the lowend of the spatial frequency range; for the other half, theinitial spatial frequency of the comparison Gabor was atthe high end.

To quantify the fidelity of working memory, the twoGabors were presented in sequence, as had been thecase during the preliminary threshold determinations.To minimize confusion between the study item thatwas to be remembered and the one that was to be ig-nored, the difference between the spatial frequenciesof the two Gabors was fixed at a large value, four justnoticeable differences. The actual frequencies presentedon each trial were randomized using the method de-scribed by Huang & Sekuler (2010b).

On each trial, the sequence of the cues ( , ) was ran-domized. For terminological convenience, the word targetwill be used for the Gabor that was to be remembered, andnontarget for the Gabor that was to be ignored. On half ofthe trials, the target appeared first, followed by the non-target (T1N2); for the remaining trials, the sequence wasreversed, with the nontarget Gabor appearing first, fol-lowed by the target (N1T2).

Accuracy of recall was tested under three main con-ditions, which differed in the length of the interval sepa-rating cue offset from stimulus onset. Each participantparticipated in three sessions on separate days, eachcomprising a single block of trials with either 300, 600,or 900 msec cue–stimulus timing. The order of the threeconditions was counterbalanced across participants. Foreach condition, the first 16 trials were excluded fromanalysis as practice trials, leaving 160 trials per conditionfor analysis.

Behavioral Analysis

The raw error on each trial is defined by the differencebetween (1) the spatial frequency of that trialʼs targetGabor and (2) the spatial frequency produced by theparticipantʼs adjustment of the comparison Gabor.Following a method described in Huang and Sekuler(2010a, 2010b), we normalized each raw error relativeto the participantʼs Weber fraction for spatial frequency,which we had already ascertained. Let x be the partic-ipantʼs Weber fraction, fT be the spatial frequency oftarget Gabor on some trial, and fR be the comparisonGaborʼs final adjusted frequency. The raw error in repro-duction fR− fT is normalized by the participantʼs Weberfraction, yielding what we call normalized reproductionerror (nRE).

nRE ¼ logð1þxÞfRfT

� �¼ ln fR − ln fT

lnð1þ xÞ ð1Þ

Huang and Sekuler (2010b) showed that each nREcould be thought of as a sum of multiple influences,

including two that arise from stimuli other than the targetstimulus. To start the process of disentangling those in-fluences, the sign of each trialʼs nRE was adjusted relativeto the trialʼs nontarget: a plus sign signified an nRE thatwas displaced toward the nontarget, and a minus signsignified an nRE that was displaced away from the non-target. From these sign-adjusted values of nRE, we usedan algebraic manipulation to extract (1) one componentthat reflects the influence of the spatial frequency of atrialʼs task-irrelevant, nontarget Gabor and (2) a secondcomponent that reflects the influence of the prototypicalspatial frequency defined by the mean frequency of allstimuli that a participant had seen (including both Targetand nontarget) on the preceding trials within the samecondition.Because of planned trial-to-trial variation in the non-

target stimulusʼ spatial frequency, there would be sometrials on which the two putative influences would operatesynergistically, both pulling the reproduction in a com-mon direction, and other trials on which the two influ-ences would work in opposition to one another(Figure 2). Thus, for each participant, we separatedtrials (1) on which the two influences would work inthe same direction (e.g., both values were higher thantargetʼs frequency or both were lower) from trials (2)on which the two influences would oppose one an-other (e.g., one value was higher and the other waslower than targetʼs frequency). As an nRE value witha positive sign signified a bias toward the nontargetstimulus, the difference between the two componentsʼinfluences can be found by subtracting BiasProto fromBiasNontarget.

ErrorSdir ¼ BiasNontarget þ BiasProto ð2Þ

ErrorOdir ¼ BiasNontarget − BiasProto ð3Þ

Equation 4 isolates the contribution of the nontarget,nullifying the contribution from the prototypical stim-ulus, whereas Equation 5 does the opposite, isolatingthe effect of the prototypical stimulus while nullifyingthe contribution of nontarget stimulus.

BiasNontarget ¼ 0:5ðErrorSdir þ ErrorOdirÞ ð4Þ

BiasProto ¼ 0:5ðErrorSdir − ErrorOdirÞ ð5Þ

This detailed decomposition of errors is especially impor-tant as our hypothesis holds that alpha oscillations will beassociated in our task not with errors generally, but withone particular kind of error, namely the nontarget effect.

EEG Recording and Analysis

EEG signals were recorded from the scalp using a high-density, 129-electrode array (Electrical Geodesics, Inc.)


and high-impedance amplifiers. All channels were ad-justed for scalp impedance < 50 kΩ. Sensor signals weresampled at 250 Hz with a 0–125 Hz analogue bandpassfilter and stored for off-line analysis. Bipolar periocularchannels were recorded from above and below eacheye and from a location near the outer canthus of eacheye.EEG signals were preprocessed using the EEGLAB tool-

box (Delorme & Makeig, 2004) for Matlab (Mathworks).The recorded signals were re-referenced to the grand aver-age. A 0.5-Hz Butterworth high-pass filter and a 60-HzParks–McClellan notch filter were applied. Eye blinks wereidentified by visual inspection of independent componentanalysis and eliminated. Epochs containing muscle artifactsor saccades, identified through independent componentanalysis and visual inspection, were rejected. Waveletanalysis and plotting were performed using the FieldTripMatlab toolbox (Oostenveld, Fries, Maris, & Schoffelen,2011). Time–frequency representations were computedusing Morlet wavelets with a width of four cycles per wave-let at center frequencies between 1 and 70 Hz, in 1-Hzsteps. To ensure equivalent resolution at any given timepoint for comparison, wavelets for each of the three cue–stimulus intervals (300, 600, and 900 msec) were made tothe same length.Alpha amplitude was defined by the mean oscillatory

power in the band 8–14 Hz. Wavelet alpha power forall electrodes was calculated for this epoch, extendingfrom cue onset through stimulus offset. Alpha powervalues were log-transformed to approximate a normaldistribution and then to eliminate between-subject dif-ferences in power were converted to standard scores(z scores) for each participant collapsed across the three

cue–stimulus intervals. Topographic plots of stimulus on-set activity (100 msec window from stimulus onset) showstrongest alpha activity in posterior electrodes (Figure 3). Acluster-based, nonparametric, randomization test (Maris &Oostenveld, 2007) between all target and all nontargetconditions was performed to establish a group of EEGelectrodes to be used for all subsequent analysis. Thecluster-based test statistic was calculated by comparing8–14 Hz power for the two conditions at every electrode,leaving the time window open from cue offset to stimu-lus offset. All electrodes for which the t value of the dif-ference between conditions exceeded the thresholdvalue of p < .025 were clustered on the basis of spatialadjacency. The sum of t values from the cluster with themaximum sum was then used as the test statistic, there-by avoiding the problem of multiple comparisons in thesignificance test. A reference distribution of test statisticswas generated by randomly permuting the data acrossthe two conditions 1000 times. A cluster was character-ized as significantly different between target and non-target conditions if the proportion of randomizedvalues larger than the observed test statistic was less thanalpha level, p = .01.

RESULTS

Our findings are presented in three sections. The firstsection includes our principal behavioral results; the sec-ond section presents the analysis of EEG recordings. Thefinal section describes the relationships between alphaoscillations and behavioral measures of memory fidelity.Throughout, results of analyses of variance are reported

Figure 2. Bar chartsshowing absolute valuesof nRE (left set of bars),the effect of the nontargetstimulus (middle set ofbars), and the effect ofthe prototypical stimulus(right set of bars) for the300-, 600-, and 900-mseccue–stimulus intervalconditions. In each pair ofbars, results for T1N2 andN1T2 trials are shown separately.Data are meansover participants. Error barsare ±1 within-subjectstandard errors of the mean.A significant differencebetween two test conditionsat p < .05 is indicatedby the asterisk (*).


as Greenhouse–Geisser-corrected values. Calculations ofwithin-subject errors are based on the method presentedby Cousineau (2005).

Behavioral Results

There was no difference in reproduction error (nRE) be-tween the three experimental conditions regardless ofthe session order in which the cue–stimulus intervalswere presented, F(2, 22) = 0.069, p = .929. For session1, mean = 1.536 and SD = 0.331; for session 2, mean =1.5 and SD = 0.33; for session 3, mean = 1.52 and SD =0.335. This outcome suggests that the order of testingwith the three conditions is not consequential and neednot be included as a factor in our subsequent analyses.

A repeated-measures multivariate analysis assessedthe influence of main factors trial Sequence (T1N2, N1T2)and Cue–Stimulus Interval (300, 600, and 900 msec) onthree dependent measures: nRE, the nontarget effect,and prototype effect. As shown in Figure 2, the sequenceof stimulus presentation, that is, T1N2 versus N1T2, signifi-cantly affected nRE, with T1N2 trials producing larger meanreproduction errors than N1T2 trials, F(1, 11) = 135.38, p<.001. Follow-up t tests reveal that this difference was sig-nificant between T1N2 and N1T2 across each of the threecue–stimulus intervals [300 msec (t(11) = 5.178, p <.001), 600 msec (t(11) = 4.786, p = .001), and 900 msec(t(11) = 4.075, p = .002)]. Stimulus sequence influencedneither the prototype nor the nontarget effect [F(1, 11) =0.733, p = .41), F(1, 11) = 2.378, p = .151, respectively].Interval had no significant effect on any of the three depen-

dent measures in the multivariate analysis: nRE, F(1, 11) =1.034, p= .364, the prototype effect, F(1, 11) = 0.604, p=.551, or the nontarget effect, F(1, 11) = 0.834, p = .422.Thus, the behavioral impact of cued attention was constantdespite a threefold variation in the interval separating cueand stimulus. No interactions were statistically significant.

Alpha Oscillations

Alpha Cluster

Results of the cluster-based, permutation test reveal aposterior cluster of 26 electrodes that showed higheralpha power during a peristimulus epoch of −100 msecto +150 msec surrounding onset for nontarget versustarget stimuli. Correlation analyses between alpha powerat this posterior cluster and the intrusion of the non-target into STM were performed separately for 100 msecpre- and 100 msec post-onset epochs in an effort to dis-tinguish effects due to pre- and post-stimulus alpha activ-ity. As described below, post-onset alpha activity drovethe relationship between alpha activity and the intrusionof the nontarget into the subsequent Gabor reproduc-tion. Therefore, alpha power during the 100-msec onsetepoch across this posterior cluster of electrodes (Figure 4)will be used for all subsequent analyses.

Alpha Power

A three-way, repeated-measures ANOVA with factorsCue–Stimulus Interval (300, 600, and 900 msec), Cue

Figure 3. Grand-averaged topographic maps of alpha power for the first Attend and Ignore stimuli (left) and the second Attend and Ignorestimuli (right). (Top) Results with 300 msec between cueʼs end and onset of stimulus; (middle) results with 600 msec between cueʼs end andonset of stimulus; (bottom) results with 900 msec between cueʼs end and onset of stimulus. Results are averaged over the 0- to 100-msec timewindow from the stimulus onset. Rounded ears are to the right and left sides and a triangle nose is at the top of each map.


(Ignore, Attend), and Position of the stimulus within thesequence (first Gabor, second Gabor) was calculated.There was a significant main effect of Cue in thatIgnore-related alpha was stronger than Attend-related al-pha, F(1, 11) = 28.495, p < .001. There was also a signif-icant main effect of Position, with alpha related to thesecond stimulus in a trial being reduced compared withalpha power related to the first stimulus in a trial, F(1,

11) = 16.57, p = .002. Despite the difference betweenthe mean posterior alpha values between interval in Fig-ure 3, there was no significant main effect of interval fol-lowing z-score normalization, F(1, 11) = .373, p = .554.None of the interactions were significant. Figure 5 showsthe result of a wavelet analysis on signals taken from theposterior cluster. Results shown are for oscillations re-corded over the interval (i) from onset of the first ofthe trialʼs color cues until (ii) the offset of the trialʼs firststudy Gabor.

Onset Latency and Resolution

The onset of increase in alpha power was defined as thefirst time point at which the power reached 50% of peakpower (Luck et al., 2009; Kiesel, Miller, Jolicoeur, &Brisson, 2008; Luck et al., 2006). To identify the 50% peakmoment, the average power at each individual frequencywithin the 8–14 Hz band from −500 to 0 msec precuewas subtracted from the epoch between cue onset tostimulus offset, thus bringing power values at the cueepoch onset close to zero. On each trial, participantswere presented with one cue–target pair and one cue–nontarget pair. Although the sequence of attend firstversus ignore first was randomized and counterbalanced,participants knew that if the first Gabor was to be attendedthen the second would be ignored, and vice versa. Thus,the response to the second cue–stimulus pair was influ-enced by prior knowledge and existing task demands.The influence of these factors can be seen in Figure 6

Figure 5. Grand-averaged, time–frequency wavelets averaged across the cluster of 26 posterior electrodes for the first Attend and Ignore stimuli(left) and the second Attend and Ignore stimuli (right). The timescale is the same for each wavelet (500 msec is the same length on all wavelets).The Attend and Ignore columns are lined up on the stimulus onset for ease of comparing the timing of the increases in alpha power. (Top)Results with 300 msec between cueʼs end and onset of stimulus; (middle) results with 600 msec between cueʼs end and onset of stimulus;(bottom) results with 900 msec between cueʼs end and onset of stimulus.

Figure 4. Topographic display of ignore-related alpha power(nontarget > target cluster; p < .01). Electrodes within the clustershowing a significant difference between the two test conditions atp < .025 are indicated by the asterisk (*).


(right) in the variability of alpha power and temporalresolution. For several participants in each of the cue–stimulus interval conditions, there was no time point atwhich the power reached 50% peak without being at50% during the previous time point. For this reason, toidentify the onset latency of the prestimulus alpha signal,only a trialʼs first stimulus was included in this analysis.

A two-way repeated-measures ANOVA showed a maineffect of Cue–Stimulus Interval (300, 600, 900 msec), F(2,22) = 81.93, p < .001. Neither Cue type (Ignore, Attend)nor any interaction was significant. Follow-up t tests re-vealed that, for every pairwise comparison for 300, 600,and 900 msec, alpha latency onset was significantly shorterfor the longer of the two intervals being compared[earlier for 900 vs. 600 msec (t(11) = −5.194, p <.001), for 900 vs. 300 msec (t(11) = −9.976, p <

.001), and for 600 vs. 300 msec (t(11) = −16.74, p < .001)].Expressed relative to stimulus onset, mean onset laten-cies were 300 msec = −0.185, 600 msec = −0.472,and 900 msec =−0.699. So despite the fact that the tim-ing of stimulus onset was constant and entirely predict-able trial after trial in a session, the longer the cuepreceded the stimulus, the earlier the onset of thealpha-band response.Visual inspection of the time–frequency wavelets (Fig-

ure 5) and time–power traces (Figure 6) indicated thatthe enhanced peristimulus alpha power declined well be-fore the offset of the stimulus. In other words, althoughthe study item remained on the computer screen, alphapower appeared to have returned to preonset levels.With this intriguing phenomenon in mind, we carriedout an additional post hoc analysis, in which the end of

Figure 6. Grand Averaged alpha power over time for attend (green) and ignore (red) for the first stimulus (left) and the second stimulus(right). (Top) 300-msec condition, (middle) 600-msec condition, and (bottom) 900-msec condition. Ribbons indicate ±1 within-subject standarderrors of the mean. The black rectangles along the time axis mark the duration between the onset and offset of 50% peak power.


the signal was defined as the first time point after thestimulus onset at which the power had fallen back to50% of peak power. A two-way ANOVA with factorsCue–Stimulus Interval (300, 600, and 900 msec) andCue type (Ignore, Attend) showed no difference in offsettimes with Cue–Stimulus Interval, Cue type, or theirinteraction. The mean offset latencies were 300 msec =0.220, 600 msec = 0.204, and 900 msec = 0.184, confirm-ing that the offset of alpha power began at approximately200 msec postcue, a time several hundred millisecondsbefore the stimulus disappeared.

Alpha and Memory Fidelity

To test the prediction that alpha power would be negativelycorrelated with the degree to which the nontarget stimulusintruded into memory, trials across participants and acrosscue–stimulus interval conditions were aggregated for theT1N2 trials in which the Gabor to be ignored appeared sec-ond in the sequence and the N1T2 trials in which the Gaborto be ignored appeared first in the sequence.

For the T1N2 trials, each participant contributed onaverage 204 artifact-cleaned trials (SD = 17 trials). Alpha

Figure 7. Alpha power predicts the nontarget effect (intrusion of the nontarget information) when the nontarget stimulus follows the targetstimulus. The relationship shows that as alpha power increases, the nontarget effect decreases. The header depicts the trial type with a boxaround the nontarget stimulus of interest. (A) Ignore–first trials (N1T2) and (B) Ignore–second trials (T1N2) were sorted into eight equally populousbins according to the ongoing alpha power across the cluster of posterior electrodes. Bin 1 comprises trials with the lowest alpha amplitude,and bin 8 includes trials on which alpha amplitude was highest. (Top) Prestimulus 100 msec time window; (bottom) 100 msec time window followingnontarget onset.


power over a −100 to +100 msec peristimulus epochwas log-transformed and converted to z scores for eachparticipant across all three cue–stimulus interval con-ditions. Additionally, nRE values for these trials wereconverted to standardized scores (z scores) for each par-ticipant across the three conditions to mute any impactof between-subject differences in overall fidelity of recall.The resulting 2445 total trials were sorted in order ofincreasing alpha power and divided into eight bins of306 trials each, except for bin 8, which held 303 trials.Ordering the bins according to increasing alpha power,bin 1 comprises trials across participants with the lowestalpha amplitude, and bin 8 includes trials on which alphaamplitude was the highest. The nontarget effect asso-ciated with trials in each of bin was then calculated.The strength of the relationship over the eight orderedbins between the nontarget effect and alpha power wasthen evaluated with linear regression.

The same analysis was carried out for the N1T2 trials. Eachparticipant contributed on average 196 artifact-cleaned trials(SD = 20 trials). The 2,352 total trials resulting from N1T2trials were sorted into eight bins of 294 trials each.

As shown in Figure 7B, in the T1N2 trials, that is, whenthe nontarget stimulus appeared second, alpha power isa predictor of the nontarget effect during both the pre-stimulus epoch and the stimulus onset epoch. As alphapower increases, so does the ability to ignore distractinginformation and so the intrusion of the nontarget de-creases. For the first 100 msec following stimulus onset,this linear trend accounts for 86% of the variance in thenontarget effect across the eight alpha-defined bins, F(1,11) = 35.63, p = .001. Alpha power during the 100 msecleading up to the stimulus onset was a significant butweaker predictor of the intrusion of the distracting stimu-lus, F(1, 11) = 6.02, p = .050. This analysis performed fornontarget stimuli in the N1T2 trials produced no compa-rable relationship for prestimulus, F(1, 11) = 0.438, p =.533, or stimulus onset, F(1, 11) = 0.059, p = .817, as canbe seen in Figure 7A.

DISCUSSION

When a stimulus is intentionally ignored rather thanattended, the ignored stimulus is preceded and accom-panied by a relative increase in posterior alpha power (Fig-ure 5). Importantly, this increase in alpha power is directlylinked to the ability to prevent task-irrelevant stimulus frominfluencing recall of an accompanying task-relevant stimu-lus (Figure 7). The onset latency, resolution, and modula-tion of the cued-increase in alpha power provide valuableinsights into the nature of the process that serves to pro-tect memory.

Alpha Oscillations

The increase in alpha power we observed over posteriorbrain regions when participants attempted to ignore a stim-

ulus (Figure 3) is consistent with previous reports of atten-tional modulations of alpha during visual (Foxe & Snyder,2011; Klimesch, Fellinger, & Freunberger, 2011) and audi-tory (Banerjee, Snyder, Molholm, & Foxe, 2011) stimulusencoding. Parietal regions are believed to be a part of afrontoparietal network of attentional control that modu-lates activity in sensory cortices (Greenberg et al., 2012;Bollimunta, Mo, Schroeder, & Ding, 2011; Corbetta &Shulman, 2002). In line with the view that posterior alphapower reflects an inhibitory mechanism, evidence showsthat attentional biasing in visual cortex can suppress com-peting information (McMains & Kastner, 2011; Kastner &Ungerleider, 2000) and task-irrelevant information (Payne& Allen, 2011). A similar, task-related posterior locus ofalpha-band activity is also observed during STM retention(Michels, Moazami-Goudarzi, Jeanmonod, & Sarnthein,2008; Klimesch et al., 2007; Jensen, Gelfand, Kounios, &Lisman, 2002) and again seems to reflect the suppressionof distracting information to protect memory (Bonnefond& Jensen, 2012; Freunberger, Werkle-Bergner, Griesmayr,Lindenberger, & Klimesch, 2011; Payne & Kounios, 2009;Jokisch & Jensen, 2007).The reduced alpha power for the second stimulus of

a sequence compared with the first was an unexpectedfinding (Figure 3). We are uncertain if this reductionoccurred because of a relative lack of attentional re-sources late in the trialʼs sequence or because the stimu-lus immediately before the retention period was moreefficiently modulated. The latter seems more likely, asthe dissociation between attend and ignore signals beganimmediately upon cue processing and the differencebetween the two is quite exaggerated compared withtheir first-stimulus counterparts. This early dissociationbetween attend and ignore for the second cue–stimuluspair likely manifests from participantʼs prior knowledgegiven that if the first cue had been to attend, then thesecond would always be to ignore, and vice versa. Indi-vidual differences in the use of this prior knowledge werereflected in the variability in alpha onset during thissecond prestimulus interval (Figure 6). Regardless of thisdifference in overall strength of alpha power between thefirst and second cue–stimulus pairs, alpha power wasalways greater for ignore versus attend across posteriorelectrodes (Figure 3) and across the same approximatetemporal envelope (Figures 3 and 6).

Onset Latency

As Figure 6 showed, the onset latency of increased pre-stimulus alpha power tracked the length of the intervalseparating the cue from the ensuing stimulus. As thecue–stimulus interval grew from 300 to 900 msec, theonset of increased alpha activity led stimulus onset byan increasing amount. Within a session of isochronictrials, the time at which the stimulus would appear wascompletely predictable, so it is unsurprising that onsetlatency of increased alpha power would be linked to


predictable timing within the subsecond range used in ourstudy. After all, a wide range of behaviors including targetinterception and collision avoidance depend upon beingable to predict timing within this range (Zarco, Merchant,Prado, & Mendez, 2009). Moreover, the Weber fractionfor intervals within the range covered by our manipula-tion, 300–900 msec, is ∼5% and 15%, which should makeeach of our three intervals highly distinctive (Merchant,Zarco, Bartolo, & Prado, 2008; Thompson, Schiffman, &Bobko, 1976). Despite this predictability of stimulus on-set time, the latency of the prestimulus increase in alphawas not, as one might expect, tightly linked to stimulusonset. Instead, the increase in alpha power precededstimulus onset by as much as several hundred milli-second, a value many times the Weber fraction for timeinterval discrimination. When the cue–stimulus intervalwas varied between sessions, the latency of alpha in-crease did not occur at a fixed interval before the stimu-lus. Rather, the temporal relationship between cue andstimulus strongly influenced the timing of anticipatoryalpha. Given only scant time to decide whether to attendor ignore, as in our 300-msec condition, modulation ofalpha began almost immediately upon cue offset; however,given additional warning time, as in our 900-msec condi-tion, alpha modulation was delayed until several hundredmilliseconds after cue offset.

Signal Resolution

Regardless of the cue–stimulus interval within a block oftrials, the increase in alpha power anticipated stimulusonset and then resolved by ∼200 msec following stimulusonset. Note that the resolution is essentially completewell before the stimulus would disappear from view.Interestingly, Freunberger et al. (2009) found highly simi-lar timing of alpha signal onset and offset, despite the useof a longer, 1000-msec stimulus (see their Figure 3).Within approximately 200 msec of a visual stimulusʼ onset,EEG recordings differentiate between characteristics suchas spatial location, category meaning, and task relevance(VanRullen & Thorpe, 2001). Reliable cue-related in-creases in prestimulus alpha power and its resolutionbefore stimulus offset demonstrates the importance ofsuppressing the onset and early processing of irrelevantinformation. The sudden onset of a stimulus has beenfound to be particularly hard to ignore (Ludwig &Gilchrist, 2002; Yantis, 1993), but directed attention canattenuate the response to the onset of task-intrusive stim-uli (Fukuda & Vogel, 2009; Ludwig & Gilchrist, 2003;Yantis & Jonides, 1990).Although the response to nontarget onset may be atten-

uated, there is no evidence that the alpha suppressionthat always occurs in response to visual stimulus onsetcan be completely prevented during an eyes-open task.Regardless of cued warning to ignore an upcomingstimulus, alpha returns to precue levels within a couplehundred milliseconds of a stimulus onset. Successful

attentional modulations of neural activity associated withdistracting information are consistently observed duringthis early epoch of visual processing (Hillyard, Vogel, &Luck, 1998). This concept was illustrated clearly by Zantoand Gazzaley (2009), who demonstrated that poormemory performance was associated with the unsuc-cessful filtering of distracting information during stim-ulus encoding, as evidenced by early ERP componentsN1 and P1. In agreement with these reports, successfulemployment of goal-directed, attentional control withinthe first 100 msec of the stimulus presentation limitedthe interference from task-irrelevant stimuli (Figure 7).Our findings also agree with a number of empirical resultsdemonstrating a relationship between alpha oscillationsand early event-related signatures of visual processing(Klimesch et al., 2011; Rajagovindan & Ding, 2011).

Attentional Modulation

Although there were significant differences in alphapower for attended and ignored stimuli, the temporalenvelopes of increased alpha power during a trial arestrikingly similar for the two conditions (see Figure 6).Although some reports demonstrate that cueing atten-tion is associated with a desynchronization of alpha activ-ity in advance of a stimulusʼ presentation (Huang &Sekuler, 2010a; Thut et al., 2006; Sauseng et al., 2005),that result may not be universal (Rihs et al., 2007; Wordenet al., 2000). Additionally, as our own results show, alphasynchronization may increase during the interval betweenthe cue offset and the onset of the stimulus (Freunbergeret al., 2009). It is tempting to ascribe some general execu-tive function to alpha synchronization, perhaps a functionrelated to timing (Gooch, Wiener, Hamilton, & Coslett,2011). For example, Min et al. (2008) proposed that pos-terior alpha power observed during a cue–stimulus inter-val represented temporal expectancy. However, temporalexpectancy would not differ between nontarget and targetstimuli, as the timing is the same for both. Also, temporalexpectancy did not differ between studies that report desyn-chronization rather than synchronization.

Instead, we propose that increased EEG alpha powerduring cued attention to some stimulus reflects the sup-pression of processing across the rest of the visual fieldwhen it is necessitated by centrally presented stimuli orirrelevant information within the same hemifield. The re-tinotopic organization of attentional modulation of alphaactivity recorded over posterior cortex was demonstratedacross the four visual field quadrants (Worden et al.,2000) and across eight equally spaced locations, placedequidistant around fixation (Rihs et al., 2007). Recently,retinotopically organized connections between intra-parietal sulcus and the visual cortex, correlated with atten-tional biasing, were illuminated in an elegant series ofexperiments using diffusion spectrum imaging in humanparticipants (Greenberg et al., 2012). These studies revealselective attentionʼs spatial specificity. When attention is


directed to the location of a stimulus, alpha desynchro-nizes in cortical areas involved in processing that stimulus.When a target stimulus is localized to one visual hemifield,alpha desynchronizes over the contralateral occipital cor-tex (Thut et al., 2006), provided the attended stimulus isspatially separated from distracting stimuli. However, alphasynchronization will be prevalent if a stimulus that shouldbe ignored can potentially occupy the same location asan attended stimulus. That is the case for the centrally-presented stimuli in our study and for Freunberger et al.(2009), for multiple locations within the same hemifield(Rihs et al., 2007), and for spatially-overlapping attributes(Worden et al., 2000).

Alpha and the Nontarget Effect

Our novel finding is that as alpha power increases, the in-trusion of the distracting stimulus decreases (Figure 7). Inother words, as alpha power increases, so does the abilityto filter out task-irrelevant information. Interestingly, thisrelationship was true only for the T1N2 trials in which thenontarget stimulus appeared second in the sequence. Thetotal error was greater for these T1N2 than for the N1T2trials (Figure 2); however, the nontarget effect was notsignificantly different between the two sequences, despiteT1N2 trials also yielding a greater average nontarget effect.The combination of Gabor stimuli and the target repro-duction technique used in this experiment enables thedissociation of errors toward stimuli other than the target.There would also be, on any given trial, errors arising frommany other sources such as internal or external distrac-tion, or poor reproduction. Our finding indicates thaterrors not systematically in the direction of the other stim-uli drove the significant difference in total error betweenthe trial types. It seems likely that a major source of errorin the T1N2 trials that would not bias reproduction towardthe nontarget would be the disruption of target mainte-nance because of the sudden onset of the second cueand the subsequent nontarget stimulus.

The finding of a positive relationship between alphapower and the nontarget effect only for T1N2 trials im-plies that although the nontarget effect did not differ sig-nificantly between the two types of trial sequences (T1N2,N1T2) these two populations have two distinct distribu-tions affecting the correlation with alpha power. It is clearthat the mean nontarget effect for the N1T2 trials is quitesmall, near zero, which would hinder efforts to correlatewith such a small range of values. Previous work withcued attention to Gabor stimuli located in separate hemi-fields failed to produce a nontarget effect (Huang &Sekuler, 2010a), indicating that characteristics of non-target stimuli that appear in a separate visual hemifieldor first in a sequence can be more successfully filteredout of memory. Nontargets that appear in the same visualfield as a target already being maintained have a decreas-ing influence on target reproduction with increasingalpha power. When there are even stronger alpha oscilla-

tions, as for the first cue–stimulus interval compared withthe second cue–stimulus interval (Figures 3, 4, and 6),there is no such linear relationship.Thus, the ability to ignore distraction is fundamental to

perception and memory, and it is correlated with in-creased alpha oscillations. In combination, our findingsdemonstrate the significance of the timely deploymentof alpha-related attentional processes during cued, inten-tional ignoring.

Acknowledgments

The authors thank Abigail Noyce and Stefan Berteau for theirhelp with this project. Supported in part by CELEST, a NationalScience Foundation Science of Learning Center (NSF SMA-0835976), NIH grants MH068404 and T32-NS07292.

Reprint requests should be sent to Lisa Payne, Department ofPsychology, Brandeis University, Room MS013, 415 South St.,Waltham, MA 02453, or via e-mail: [email protected].

REFERENCES

Banerjee, S., Snyder, A. C., Molholm, S., & Foxe, J. J. (2011).Oscillatory alpha-band mechanisms and the deployment ofspatial attention to anticipated auditory and visual targetlocations: Supramodal or sensory-specific controlmechanisms? Journal of Neuroscience, 31, 9923–9932.

Bollimunta, A., Mo, J., Schroeder, C. E., & Ding, M. (2011).Neuronal mechanisms and attentional modulation ofcorticothalamic alpha oscillations. Journal of Neuroscience,31, 4935–4943.

Bonnefond, M., & Jensen, O. (2012). Alpha oscillations serve toprotect working memory maintenance against anticipateddistracters. Current Biology, 22, 1–6.

Brainard, D. H. (1997). The psychophysics toolbox. SpatialVision, 10, 433–436.

Corbetta, M., & Shulman, G. L. (2002). Control of goal-directedand stimulus-driven attention in the brain. Nature ReviewsNeuroscience, 3, 201–215.

Cousineau, D. (2005). Confidence intervals in within-subjectdesigns: A simpler solution to Loftus and Massonʼs method.Tutorials in Quantitative Methods for Psychology, 1,42–45.

Delorme, A., & Makeig, S. (2004). EEGLAB: An open sourcetoolbox for analysis of single-trial EEG dynamics includingindependent component analysis. Journal of NeuroscienceMethods, 134, 9–21.

Dube, C., Zhou, F., & Sekuler, R. (under review). Similarity-based distortion of short-term visual memory is perceptual.

Foxe, J. J., Simpson, G. V., & Ahlfors, S. P. (1998). Parieto-occipital approximately 10 Hz activity reflects anticipatorystate of visual attention mechanisms. NeuroReport, 9,3929–3933.

Foxe, J. J., & Snyder, A. C. (2011). The role of alpha-band brainoscillations as a sensory suppression mechanism duringselective attention. Frontiers in Psychology, 2, 154.

Freunberger, R., Fellinger, R., Sauseng, P., Gruber, W., &Klimesch, W. (2009). Dissociation between phase-lockedand nonphase-locked alpha oscillations in a working memorytask. Human Brain Mapping, 30, 3417–3425.

Freunberger, R., Werkle-Bergner, M., Griesmayr, B.,Lindenberger, U., & Klimesch, W. (2011). Brain oscillatorycorrelates of working memory constraints. Brain Research,1375, 93–102.


Fu, K. M., Foxe, J. J., Murray, M. M., Higgins, B. A., Javitt, D. C.,& Schroeder, C. E. (2001). Attention-dependent suppressionof distracter visual input can be cross-modally cued asindexed by anticipatory parieto-occcipital alpha-bandoscillations. Brain Research, 12, 145–152.

Fukuda, K., & Vogel, E. K. (2009). Human variation inoverriding attentional capture. Journal of Neuroscience,29, 8726–8733.

Gooch, C. M., Wiener, M., Hamilton, A. C., & Coslett, H. B.(2011). Temporal discrimination of sub- and suprasecondtime intervals: A voxel-based lesion mapping analysis.Frontiers in Integrative Neuroscience, 5, 59.

Greenberg, A. S., Verstynen, T., Chiu, Y.-C., Yantis, S., Schneider,W., & Behrmann, M. (2012). Visuotopic cortical connectivityunderlying attention revealed with white-matter tractography.Journal of Neuroscience, 32, 2773–2782.

Haegens, S., Luther, L., & Jensen, O. (2012). Somatosensoryanticipatory alpha activity increases to suppress distractinginput. Journal of Cognitive Neuroscience, 24, 677–685.

Hasher, L., & Zacks, R. T. (1988).Workingmemory, comprehension,and aging: A review and a new view. The psychology of learningand memory. In G. H. Bower (Ed.), (Vol. 22, pp. 193–225).New York: Academic Press.

Hillyard, S. A., Vogel, E. K., & Luck, S. J. (1998). Sensory gaincontrol (amplification) as a mechanism of selective attention:Electrophysiological and neuroimaging evidence.Philosophical Transactions of the Royal Society of London,Series B, Biological Sciences, 353, 1257–1270.

Huang, J., & Sekuler, R. (2010a). Attention protects the fidelityof visual memory: Behavioral and electrophysiologicalevidence. Journal of Neuroscience, 30, 13461–13471.

Huang, J., & Sekuler, R. (2010b). Distortions in recall fromvisual memory: Two classes of attractors at work. Journal ofVision, 10, 24.1–24.27.

Jensen, O., Gelfand, J., Kounios, J., & Lisman, J. E. (2002).Oscillations in the alpha band (9-12 Hz) increase withmemory load during retention in a short-term memory task.Cerebral Cortex, 12, 877–882.

Jokisch, D., & Jensen, O. (2007). Modulation of gamma and alphaactivity during a working memory task engaging the dorsalor ventral stream. Journal of Neuroscience, 27, 3244–3251.

Jones, S. R., Kerr, C. E., Wan, Q., Pritchett, D. L., Hamalainen,M., & Moore, C. I. (2010). Cued spatial attention drivesfunctionally relevant modulation of the mu rhythm in primarysomatosensory cortex. Journal of Neuroscience, 30,13760–13765.

Kastner, S., & Ungerleider, L. G. (2000). Mechanisms of visualattention in the human cortex. Annual Review of Neuroscience,23, 315–341.

Kelly, S. P., Lalor, E. C., Reilly, R. B., & Foxe, J. J. (2006).Increases in alpha oscillatory power reflect an activeretinotopic mechanism for distracter suppression duringsustained visuospatial attention. Journal of Neurophysiology,95, 3844–3851.

Kiesel, A., Miller, J., Jolicoeur, P., & Brisson, B. (2008).Measurement of ERP latency differences: A comparison ofsingle-participant and jackknife-based scoring methods.Psychophysiology, 45, 250–274.

Klimesch, W., Fellinger, R., & Freunberger, R. (2011). Alphaoscillations and early stages of visual encoding. Frontiers inPsychology, 2, 118.

Klimesch, W., Sauseng, P., & Hanslmayr, S. (2007). EEG alphaoscillations: The inhibition-timing hypothesis. Brain ResearchReviews, 53, 63–88.

Luck, S. J., Fuller, R. L., Braun, E. L., Robinson, B., Summerfelt,A., & Gold, J. M. (2006). The speed of visual attention inschizophrenia: Electrophysiological and behavioral evidence.Schizophrenia Research, 85, 174–195.

Luck, S. J., Kappenman, E. S., Fuller, R. L., Robinson, B.,Summerfelt, A., & Gold, J. M. (2009). Impaired responseselection in schizophrenia: Evidence from the P3 wave andthe lateralized readiness potential. Psychophysiology, 46,776–786.

Ludwig, C. J. H., & Gilchrist, I. D. (2002). Stimulus-drivenand goal-driven control over visual selection. Journal ofExperimental Psychology: Human Perception andPerformance, 28, 902–912.

Ludwig, C. J. H., & Gilchrist, I. D. (2003). Goal-driven modulationof oculomotor capture. Perception & Psychophysics, 65,1243–1251.

Maris, E., & Oostenveld, R. (2007). Nonparametric statisticaltesting of EEG- and MEG-data. Journal of NeuroscienceMethods, 164, 177–190.

McMains, S., & Kastner, S. (2011). Interactions of top–down andbottom-up mechanisms in human visual cortex. Journal ofNeuroscience, 31, 587–597.

Merchant, H., Zarco, W., Bartolo, R., & Prado, L. (2008).The context of temporal processing is represented in themultidimensional relationships between timing tasks.PLoS One, 3, e3169.

Michels, L., Moazami-Goudarzi, M., Jeanmonod, D., & Sarnthein,J. (2008). EEG alpha distinguishes between cuneal andprecuneal activation in working memory. Neuroimage,40, 1296–1310.

Min, B. K., Park, J. Y., Kim, E. J., Kim, J. I., Kim, J. J., & Park, H. J.(2008). Prestimulus EEG alpha activity reflects temporalexpectancy. Neuroscience Letters, 438, 270–274.

Oldfield, R. C. (1971). The assessment and analysis ofhandedness: The Edinburgh inventory. Neuropsychologia,9, 97–113.

Oostenveld, R., Fries, P., Maris, E., & Schoffelen, J.-M. (2011).FieldTrip: Open source software for advanced analysisof MEG, EEG, and invasive electrophysiological data.Computational Intelligence and Neuroscience, 2011,156869.

Payne, H. E., & Allen, H. A. (2011). Active ignoring in earlyvisual cortex. Journal of Cognitive Neuroscience, 23,2046–2058.

Payne, L., & Kounios, J. (2009). Coherent oscillatory networkssupporting short-term memory retention. Brain Research,1247, 126–132.

Rajagovindan, R., & Ding, M. (2011). From prestimulus alphaoscillation to visual-evoked response: An inverted-U functionand its attentional modulation. Journal of CognitiveNeuroscience, 23, 1379–1394.

Rihs, T. A., Michel, C. M., & Thut, G. (2007). Mechanisms ofselective inhibition in visual spatial attention are indexed byalpha-band EEG synchronization. European Journal ofNeuroscience, 25, 603–610.

Sauseng, P., Klimesch, W., Stadler, W., Schabus, M.,Doppelmayr, M., Hanslmayr, S., et al. (2005). A shift of visualspatial attention is selectively associated with human EEGalpha activity. European Journal of Neuroscience, 22,2917–2926.

Snyder, A. C., & Foxe, J. J. (2010). Anticipatory attentionalsuppression of visual features indexed by oscillatoryalpha-band power increases: A high-density electricalmapping study. Journal of Neuroscience, 30, 4024–4032.

Thompson, J. G., Schiffman, H. R., & Bobko, D. J. (1976). Thediscrimination of brief temporal intervals. Acta Psychologica,40, 489–493.

Thut, G., Nietzel, A., Brandt, S. A., & Pascual-Leone, A. (2006).Alpha-band electroencephalographic activity over occipitalcortex indexes visuospatial attention bias and predictsvisual target detection. Journal of Neuroscience, 26,9494–9502.


VanRullen, R., & Thorpe, S. J. (2001). The time course of visualprocessing: From early perception to decision-making.Journal of Cognitive Neuroscience, 13, 454–461.

Watson, A. B., & Pelli, D. G. (1983). QUEST: A Bayesian adaptivepsychometric method. Perception & Psychophysics, 33, 113–120.

Worden, M. S., Foxe, J. J., Wang, N., & Simpson, G. V. (2000).Anticipatory biasing of visuospatial attention indexed byretino-topically specific alpha-band electroencephalographyincreases over occipital cortex. The Journal of Neuroscience,20, 1–6.

Yantis, S. (1993). Stimulus-driven attentional capture. CurrentDirections in Psychological Science, 2, 156–161.

Yantis, S., & Jonides, J. (1990). Abrupt visual onsets andselective attention: Voluntary versus automatic allocation.Journal of Experimental Psychology: Human Perceptionand Performance, 16, 121–134.

Zanto, T. P., & Gazzaley, A. (2009). Neural suppressionof irrelevant information underlies optimal workingmemory performance. Journal Neuroscience, 29,3059–3066.

Zarco, W., Merchant, H., Prado, L., & Mendez, J. C. (2009).Subsecond timing in primates: Comparison of intervalproduction between human subjects and rhesus monkeys.Journal of Neurophysiology, 102, 3191–3202.


Documents

Attention-modulated Alpha-band Oscillations Protect ...people.brandeis.edu/~sekuler/papers/Payne_Alpha... · (Haegens, Luther, & Jensen, 2012). Moreover, the handʼs representation