Nihms402820Bosman Monkey

Attentional Stimulus Selection through SelectiveSynchronization between Monkey Visual Areas

Conrado A. Bosman1,2,*, Jan-Mathijs Schoffelen1,*, Nicolas Brunet1, Robert Oostenveld1,Andre M. Bastos1,3,4, Thilo Womelsdorf1, Birthe Rubehn5, Thomas Stieglitz5, Peter DeWeerd1,6, and Pascal Fries1,3

1Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen, 6525 ENNijmegen, Netherlands 2Cognitive and System Neuroscience Group, Swammerdam Institute forLife Sciences, Center for Neuroscience, University of Amsterdam, 1098 XH, Amsterdam,Netherlands 3Ernst Strngmann Institute (ESI) for Neuroscience in Cooperation with Max PlanckSociety, 60528 Frankfurt, Germany 4Center for Neuroscience and Center for Mind and Brain,University of California, Davis, CA 95618, USA 5Laboratory for Biomedical Microtechnology,Department of Microsystems Engineering (IMTEK) and the Bernstein Center Freiburg, Albert-Ludwigs-Universitt Freiburg, 79110 Freiburg, Germany 6Department of Neurocognition,University of Maastricht, 6229 ER Maastricht, Netherlands

SUMMARYA central motif in neuronal networks is convergence, linking several input neurons to one targetneuron. In visual cortex, convergence renders target neurons responsive to complex stimuli. Yet,convergence typically sends multiple stimuli to a target, and the behaviorally relevant stimulusmust be selected. We used two stimuli, activating separate electrocorticographic V1 sites, and bothactivating an electrocorticographic V4 site equally strongly. When one of those stimuli activatedone V1 site, it gamma-synchronized (6080 Hz) to V4. When the two stimuli activated two V1sites, primarily the relevant one gamma-synchronized to V4. Frequency bands of gamma activitiesshowed substantial overlap containing the band of inter-areal coherence. The relevant V1 site hadits gamma peak frequency 23 Hz higher than the irrelevant V1 site, and 46 Hz higher than V4.Gamma-mediated inter-areal influences were predominantly directed from V1 to V4. We proposethat selective synchronization renders relevant input effective, thereby modulating effectiveconnectivity.

INTRODUCTIONDuring natural vision, many stimuli simultaneously activate our visual system. In primaryvisual cortex, two separate stimuli typically activate two separate groups of neurons. Theseseparate groups of neurons send anatomical connections that converge onto postsynapticneurons in higher visual areas (Fries, 2009). Through this convergence, the postsynapticneurons can respond to either one of the two stimuli. Yet, if one of those stimuli isbehaviorally relevant, it dominates the activity of the postsynaptic neurons at the expense of

2012 Elsevier Inc. All rights reserved.Correspondence to: [email protected] (P.F.), [email protected] (C.A.B.).*These authors contributed equally.Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to ourcustomers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review ofthe resulting proof before it is published in its final citable form. Please note that during the production process errors may bediscovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

NIH Public AccessAuthor ManuscriptNeuron. Author manuscript; available in PMC 2013 September 06.

Published in final edited form as:Neuron. 2012 September 6; 75(5): 875888. doi:10.1016/j.neuron.2012.06.037.

NIH

-PA Author Manuscript

NIH


NIH


the irrelevant stimulus (Moran and Desimone, 1985; Treue and Maunsell, 1996; Reynolds etal., 1999). This effect can be explained as a selective enhancement of synaptic gain of therelevant input (Reynolds et al., 1999). A candidate mechanism for this enhancement needsto fulfill at least the following criteria: (a) It has to be specific for the relevant subset ofsynaptic inputs versus the irrelevant subset, even though the two sets are likely interspersedon a postsynaptic neuron. (b) It has to be flexible to select different subsets of synapses asthe relevant stimulus undergoes changes. (c) It has to be able to switch within a few hundredmilliseconds from strengthening one set of synapses to another set, because switchingattention from one stimulus to another affects the activity of the postsynaptic neurons andbehavior at this time scale (Busse et al., 2008).

To meet these requirements, we and others have proposed that the selective enhancement ofrelevant synaptic input is implemented by the selective rhythmic synchronization of theneuronal target group with the relevant input (Fries, 2005; Brgers and Kopell, 2008; Fries,2009). Rhythmic activity in a target group entails corresponding fluctuations in postsynapticmembrane potentials and postsynaptic shunting, which render input most effective if it isconsistently timed to the peaks of depolarization, i.e. if it is synchronized with the target.This hypothesis has been termed Communication Through Coherence, or CTC (Fries,2005). It has been implemented in mathematical models that demonstrate its plausibility andthe strength with which it can affect neuronal interactions (Brgers and Kopell, 2008;Tiesinga and Sejnowski, 2010; Buehlmann and Deco, 2010; Akam and Kullmann, 2010).There is already experimental support for the mechanistic prediction of the CTC hypothesis:When two groups of neurons are rhythmically active, then the strength of their interactiondepends on the phase relation between their rhythms (Womelsdorf et al., 2007). When threerhythmically active groups are considered, one of them can at the same time be in-phase andtherefore interacting with a second group, while being out-of-phase and thereforenoninteracting with a third group. We aim here to test the cognitive prediction of the CTChypothesis, i.e. that a neuronal target group synchronizes selectively with those inputneurons that provide behaviorally relevant input.

CTC is consistent, yet goes beyond a previous proposal that considered the synchronizationonly among the input neurons, and stated that enhanced synchronization among behaviorallyrelevant input neurons increases their impact onto postsynaptic target neurons throughfeedforward coincidence detection. Tests of this previous proposal obviously confinedthemselves to assessing the synchronization within the input neuron group. These studiesrevealed that neurons activated by an attended as compared to an unattended stimulus showenhanced gamma-band synchronization in monkey area V4 (Fries et al., 2001; Taylor et al.,2005; Bichot et al., 2005; Fries et al., 2008; Buffalo et al., 2011) and area V2 (Buffalo et al.,2011), and either reduced (Chalk et al., 2010), unchanged, or enhanced (Buffalo et al., 2011)gamma-band synchronization in area V1. For area V4, the enhancements of gamma-bandsynchronization have been shown to be functionally relevant: A key behavioral consequenceof attention, an enhanced speed of change detection, is predicted selectively by neuronalsynchronization in the gamma-frequency range, but not by synchronization in otherfrequency ranges or by neuronal firing rates (Womelsdorf et al., 2006; Hoogenboom et al.,2010).

While enhanced gamma-band synchronization among relevant input neurons is fullyconsistent with the CTC hypothesis, CTC crucially entails that those neurons achieve anexclusive or selective synchronization to their postsynaptic target neurons at the expense ofcompeting, behaviorally irrelevant input neurons. Through this, CTC lends a centralmechanistic role to the rhythm of the neuronal target group: It is the synchronization of thisrhythm to the rhythm of the relevant stimulus input, that enhances the gain of this input.

Bosman et al. Page 2

Neuron. Author manuscript; available in PMC 2013 September 06.

NIH


NIH


NIH


Testing this central prediction requires the simultaneous activation of two competing inputs,and the simultaneous recording of the rhythm in the group of neurons providing input andthe rhythm in their target group. To enable a concrete experimental test of CTC, a strongprediction can be derived about the synchronization among local rhythms in monkey areasV1 and V4 during selective attention to one of two simultaneously presented visual stimuli:If two stimuli activate separate sites in V1, and both activate one V4 site equally strongly,then the V4 site should synchronize selectively to the V1 site driven by the attendedstimulus. Here, we test this prediction, assessing local rhythms through electrocorticographic(ECoG) local field potential (LFP) recordings.

RESULTSTo quantify synchronization between V1 and V4, we used multi-site LFP recordings, whichhave been shown highly effective in assessing long-range, inter-areal synchronization(Roelfsema et al., 1997; von Stein et al., 2000; Tallon-Baudry et al., 2001; Tallon-Baudry etal., 2004). Multi-site LFP recordings are routinely carried out with ECoG grid electrodesimplanted onto the brains of epilepsy patients for pre-surgical focus localization. Theseunique recordings from the human brain have been used for numerous cognitive and/orsystems neuroscience studies (Tallon-Baudry et al., 2001; Canolty et al., 2006), yet theytypically do not include early visual areas. We therefore developed a high-density ECoGgrid of electrodes (1 mm diameter Platinum discs) and implanted it subdurally onto thebrains of two macaque monkeys to obtain simultaneous recordings from 252 electrodesacross large parts of the left hemisphere (Rubehn et al., 2009).

Figure 1A shows the brain of monkey P with the electrode positions superimposed (FigureS1A shows electrode positions for both monkeys). Figure 1B illustrates that a contralateralvisual stimulus induced strong gamma-band activity (Gray et al., 1989), while an ipsilateralstimulus did not. Figure S1B shows respective time-frequency analyses, demonstrating thatstimulus induced gamma was sustained as long as the stimulus was presented. The gamma-band was within the range of frequencies described in previous studies using driftinggratings in human subjects or awake monkeys (Hoogenboom et al., 2006; Fries et al., 2008;Muthukumaraswamy et al., 2009; Swettenham et al., 2009; Vinck et al., 2010; van Pelt etal., 2012). Within that range, the gamma band found here was relatively high, most likelydue to the individual predispositions of the animals and the use of moving stimuli(Swettenham et al., 2009) of high contrast (Ray and Maunsell, 2010). Figure 1C shows forseveral V1 example electrodes (green dots in Fig. 1A) receptive fields (RFs) in the form ofenhanced gamma-band power in response to visual stimulation in specific parts of the lowerright visual field quadrant (see Experimental Procedures and Figure S1C for details). Thewell defined RFs indicate that a given electrode was primarily assessing neuronal activity ina small patch of the underlying visual cortex. Figure 1D shows respective examples forseveral V4 electrodes (red dots in Fig. 1A). In both V1 and V4, the ordered representation ofeccentricity and elevation was as predicted by numerous previous studies (Gattass et al.,2005). Figure S1D shows RF outlines from two recording sessions separated by two months,illustrating the stability of RF positions and thereby suggesting that the electrodes were in astable position on the cortex.

With these recordings at hand, we engaged the monkeys in the selective visual attention taskillustrated in Figure 1E (see Experimental Procedures for details). When the monkeytouched a bar and fixated a central dot, two patches of drifting grating appeared. The twostimuli were always blue and yellow, with the color assigned randomly. After about onesecond, the fixation point assumed the color of one of the stimuli, which was thereby cuedas relevant. In each trial, the relevant grating changed curvature at an unpredictable momentup to 4.5 seconds after the cue, and the monkey was rewarded for bar releases within a short



NIH


NIH


NIH


time window thereafter. Changes in the irrelevant grating were equally probable, butcorresponding bar releases were not rewarded. In monkeys K and P, 92% and 94% of barreleases, respectively, were correct reports of changes in the relevant stimulus. In 10% of thetrials, only one or the other stimulus was shown in isolation (and its changes had to bereported) to assess stimulus selectivity of the recording sites.

For all analyses, we used the period from 0.3 s after cue onset until one of the stimulichanged. Also, for all further analyses, we first calculated local bipolar derivatives, i.e.differences between LFPs from immediately neighboring electrodes. We refer to the bipolarderivatives as sites. Bipolar derivation further enhances spatial specificity of the signaland removes the common recording reference, which is important for the analysis ofsynchronization between sites.

Figure 2 shows the results for a single dataset including a V4 site activated equally by eachof two stimuli, and two V1 sites activated exclusively by either one or the other stimulus.Figures 2AF illustrate the stimulus selectivity of the different sites during isolatedstimulation with stimulus 1 (condition marked red in panel A) or stimulus 2 (conditionmarked blue in panel A): Site V4 was equally driven by both stimuli (Fig. 2B); Site V1aresponded to stimulus 1, but not 2 (Fig. 2C); The opposite was the case for site V1b (Fig.2D). Figures S2AD show the respective raw power spectra. Figures 2E and F demonstratethat V4 showed pronounced inter-areal synchronization in the 6080 Hz band selectivelywith the V1 site that was stimulus driven. In the following, we will refer to the 6080 Hzband as the gamma band. Figure S1E shows the topography across V1 and V4 of stimulus-induced gamma-band power changes and of the gamma-band coherence relative to the V4site. Coherence showed a peak in V1 that coincided with the V1 power-change peak.

Figure 2G illustrates the selective attention conditions with both stimuli presentedsimultaneously, but only one stimulus behaviorally relevant and therefore selected in anygiven trial. In the V4 site, attention to either stimulus gave essentially the same activation(Figure 2H), confirming that the site was equally driven by either stimulus. In both V1 sites,attention to their respective driving stimulus led to a slight but highly consistent increase inthe frequency of the gamma-band activity (Figures 2I and J; p

(Fig. 3G, H; p

The spectra of figures 6A to F are shown again in figures 6G to L, now separately for theconditions attention inside the V1-RF (Figs. 6G, I, K) and attention outside the V1-RF (Figs.6H, J, L), and now comparing directly GC influences in the bottom-up direction (thick lines)versus top-down direction (thin lines). In the gamma band, with attention inside the V1-RF(Fig. 6G), the GC influence in the bottom-up direction was 232% stronger than in the top-down direction (p

local neuronal ensemble activities, and ensemble recordings enable a sensitive investigationof long-range inter-areal communication (Zeitler et al., 2006). The ensemble entrains itsconstituent single neurons (Fries et al., 2001; Womelsdorf et al., 2006) and thereby, theobserved inter-areal LFP coherence likely translates into inter-areal coherence amongneuronal spiking.

Inter-areal gamma-band coherence has been shown through intracranical recordings inseveral previous publications (Engel et al., 1991a; Engel et al., 1991b; von Stein et al., 2000;Fell et al., 2001; Buschman and Miller, 2007; Womelsdorf et al., 2007; Gregoriou et al.,2009; Colgin et al., 2009; Popescu et al., 2009; Sigurdsson et al., 2010). For example, vonStein et al. investigated LFPs recorded from visual and associative brain areas of the awakecat and found correlations between gamma-filtered LFPs primarily for novel stimuli (vonStein et al., 2000). Similarly, Buschman et al. investigated coherence between LFPsrecorded in monkey frontal cortex and area LIP and found enhanced gamma-band coherenceduring bottom-up processing (Buschman and Miller, 2007). Gregoriou et al. investigatedspike and LFP recordings from monkey areas V4 and FEF during the same selectiveattention task as used here. Pairs of V4 and FEF sites with overlapping RFs showed gamma-band coherence that was enhanced when attention was inside the joint RF (Gregoriou et al.,2009). Long-range gamma-band coherence has also been studied with non-invasiverecordings in human subjects (Schoffelen et al., 2005; Siegel et al., 2008; Schoffelen et al.,2011; Hipp et al., 2011). For example, Schoffelen et al. showed that cortico-spinal gamma-band coherence indexes a subjects dynamic movement preparation (Schoffelen et al., 2005)selectively among those cortico-spinal neurons involved in the upcoming movement(Schoffelen et al., 2011).

To study inter-areal coherence between monkey areas V1 and V4, we have relied onelectrocorticographic LFP recordings that measure the electrical activity under the electrode.Neither the volume of tissue, nor the way in which it affects the recording are fullyunderstood. Yet, a few statements about ECoG recordings can be made: 1.) ECoG signals donot provide a direct measure of spiking activity, and therefore, our results do not directly testpredictions that might be derived from the CTC hypothesis about spike synchronization. 2.)ECoG recordings from V1 reflect both, V1 neurons with connections to V4, and other V1neurons. Similarly, ECoG recordings from V4 reflect V4 neurons with direct input from V1,and other V4 neurons. Therefore, our results do not directly quantify the coherence betweenV1 output neurons and their postsynaptic target neurons in V4. Such an analysis would haverequired the simultaneous recording of inter-areal pairs of isolated single units, identified tobe monosynaptically coupled to each other. While this would have been technicallyextremely challenging, it would at the same time have rendered the analysis of inter-arealcoherence extremely insensitive. Isolated single neurons reflect with their sparse spikingonly poorly the phase of the underlying rhythm. For two isolated single neurons in V1 andV4, coherence analysis would have been exceedingly insensitive (Zeitler et al., 2006). 3.)ECoG recordings combine spatial resolution in the range of few millimeters (Fig. 1C) withexcellent sensitivity for the rhythms in the respective local neuronal population (Fig. 1B).The core prediction of the CTC hypothesis with regard to selective attention relates to thismesoscopic level: The V4 rhythm is selectively coherent with the V1 rhythm that is drivenby the behaviorally relevant stimulus. To test this prediction, simultaneous multi-area ECoGrecordings are ideal.

Spike recordings in V4 would have allowed testing whether postsynaptic neurons respondedprimarily to the attended stimulus. However, this core result from the attention field (Moranand Desimone, 1985; Reynolds et al., 1999) has been replicated several times andpresumably holds also in our experiment. Thereby, our present results actually also supportthe Binding By Synchronization (BBS) hypothesis. The BBS hypothesis states that



NIH


NIH


NIH


distributed neurons, activated by the same stimulus, are bound together by synchronization(Gray et al., 1989). Most studies testing the BBS hypothesis investigated distributedneuronal activations within a given area (Singer and Gray, 1995). Yet, a stimulus activatesneurons distributed across several brain areas and the BBS hypothesis is meant to apply alsoto such inter-areal neuronal assemblies. As V4 neurons with two stimuli in their RFdynamically represent the attended stimulus, the BBS hypothesis predicts, that they shoulddynamically synchronize to those V1 neurons that represent the same, i.e. the attended,stimulus. This prediction is confirmed by our present results.

Attention affected the gamma rhythm in area V1: While there was no significant attentioneffect on gamma power, there was a very reliable increase in gamma frequency. Theabsence of an attentional effect on gamma power in V1 disagrees with one previous studyusing small static bar stimuli (Chalk et al., 2010), and agrees with another previous studythat used very similar stimuli and task as our paradigm (Buffalo et al., 2011). The attentionalincrease in gamma peak frequency has not been reported before. It is intriguing, becauseattention to a stimulus is similar to an increase in stimulus contrast (Reynolds and Chelazzi,2004), and higher contrast induces higher gamma-band frequencies in monkey area V1 (Fig.S5A) (Ray and Maunsell, 2010). Higher contrast typically results in gamma power toincrease (Henrie and Shapley, 2005; Chalk et al., 2010). Yet, for very high contrast levels,gamma power can saturate or even decrease as is illustrated in figure S5B, which explainswhy attention to our full-contrast stimuli did not lead to further gamma powerenhancements.

Figure 5 shows that the local gamma peaks had a certain width, overlapping for their largerparts. While the gamma peak frequency at the relevant V1 site was 23 Hz higher than at theirrelevant V1 site, it was 46 Hz higher than in V4. If one considered these slightly differentgamma peak frequencies without the coherence results, then the simplest interpretationwould be the following: The rhythms at the attended V1 site, the unattended V1 site and theV4 site reflect three independent sine wave oscillators with slightly, but distinctly differentfrequencies; The width of the respective frequency bands is due to moment-to-momentdeviations from perfect sine waves of the respective frequencies; Those deviations areirrelevant noise. This interpretation entails that the three oscillators constantly precessrelative to each other, because their peak frequencies differ. For example, in monkey P, theV1-attended gamma peak frequency was 65.3 Hz and the V4 gamma peak frequency was59.5 Hz (Fig. 5), i.e. the peak frequencies differed by roughly 6 Hz. In the abovementionedinterpretation, this would result in the relative gamma phase to precess over full 360 degreecycles roughly 6 times per second. Such a precession would result in fully inconsistentgamma phase relations and a complete absence of gamma coherence. Gamma coherencewould actually be destroyed by any of the observed frequency differences: A six Hertzfrequency difference would lead to complete precession and loss of coherence six times persecond, and a two Hertz difference would lead to complete precession and loss of coherencetwo times per second. An absence of coherence would be inconsistent with CTC. However,we found clear V1-V4 gamma coherence. The presence of gamma coherence demonstratesthat gamma phases did not freely precess against each other, but rather that gamma rhythmshad a consistent phase-relation. Thereby, the observation of coherence rules out theabovementioned simple interpretation of the slight frequency differences.

Rather, the synopsis of our findings suggests one of the following scenarios or acombination of them: 1.) The frequencies of the synchronized rhythms at the V4 site and therelevant V1 site are always identical on a moment-by-moment basis, yet the commonfrequency fluctuates and the local circuits resonate at different frequencies, giving rise todifferent peak frequencies in the time-averaged power spectra. 2.) Our ECoG recordings inV4 reflect a mixture of (at least) two gamma rhythms in V4, one entrained by the attended



NIH


NIH


NIH


V1 gamma, and a second at a slightly lower frequency. 3.) In the third scenario, the differentgamma frequencies play mechanistic roles in bringing about the selective inter-arealsynchronization. There is one crucial additional ingredient to this scenario, namely a theta-rhythmic gamma-phase reset across V1 and V4, which we have described previously(Bosman et al., 2009). After the reset, the attended-V1 gamma and the unattended-V1gamma partly precess relative to the slightly slower V4 gamma. The attended-V1 gamma isof higher frequency than the unattended-V1 gamma, and therefore precesses faster.Correspondingly, in each gamma cycle, the attended-V1 input enters V4 before theunattended-V1 input. The earlier entry together with feedforward inhibition makes theattended-V1 input entrain V4 at the expense of the unattended-V1 input (Fries et al., 2007;Vinck et al., 2010). The selective entrainment of V4 by the attended-V1 gamma rhythmfurther enhances the gain of the attended-V1 input and reduces the gain of the unattended-V1 input (Fries, 2005; Brgers and Kopell, 2008). The theta-rhythmic reset of inter-arealgamma-band synchronization is supported by our data (Fig. 8 and S4). It likely correspondsto a reset of attentional selection, and under natural viewing conditions, might subserve thetheta-rhythmic sampling of multiple objects in a scene, either overtly (Otero-Millan et al.,2008) or covertly (Fries, 2009; Landau and Fries, 2012).

Importantly, also the third scenario, with partial precession, leads to selective coherence, asobserved here. The presence of coherence is not only necessary, but also sufficient for CTC.CTC requires two rhythms with a phase relation that is (partly) consistent across time (ormultiple observation epochs). The consistency of phase relations is precisely what isquantified by coherence. Crucially, coherence entails that the phase estimates of the twosignals do not reflect noise, because with a pure noise signal on either one of the sides, phaserelations would be random and there would be no coherence. Thereby, coherence in itselfdemonstrates 1.) the presence of two meaningful rhythms on the two sides, and 2.) thepresence of synchronization. As exemplified in the above scenarios, coherence does notrequire that two sites show rhythms with the same or similar peak frequency. And we notealso that rhythms with the same or similar peak frequency are not sufficient for coherence. Ife.g. the two visual hemispheres are separated by cutting the corpus callosum, then thegamma rhythms in the two hemispheres of a given animal are essentially identical, but thereis no coherence (Engel et al., 1991a).

We found that Granger-causal influences in the gamma band were substantially stronger inthe bottom-up V1-to-V4 direction than vice versa. Granger analyses alone can ultimately notprove or disprove one particular network organization. Yet, the strong bottom-updirectedness of the V1V4 gamma GC influence combines with two additional pieces ofevidence: 1.) Both in V1 and V4, neuronal spiking is gamma synchronized almostexclusively in the superficial layers, while neuronal spiking in infragranular layers lacksgamma synchronization (Buffalo et al., 2011). 2.) V1 neurons projecting to V4 are locatedalmost exclusively in supragranular layers, while V4 neurons projecting to V1 are locatedalmost exclusively in infragranular layers (Barone et al., 2000). These three pieces ofevidence together suggest that: 1.) In V1, gamma synchronization emerges in supragranularlayers, and the behaviorally relevant V1 gamma influences V4 through feedforwardprojections with their respective delay; 2.) In V4, gamma synchronization also emerges insupragranular layers and primarily influences areas further downstream of V4; 3.) The top-down influence from V4 to V1 originates from deep V4 layers and is therefore mediated to amuch lesser extent through the gamma band. A direct test of these predictions will requirelaminar recordings in both areas simultaneously.

Most importantly, we demonstrate strong inter-areal gamma-band synchronization that linksV4 dynamically to the relevant part of V1, precisely as predicted by the CTC hypothesis.The CTC hypothesis states that a local neuronal rhythm modulates input gain rhythmically,



NIH


NIH


NIH


that input is therefore most effective if it is consistently timed to moments of maximal gain,and that thereby the synchronization between input and target modulates effectiveconnectivity (Fries, 2005; Schoffelen et al., 2005; Womelsdorf et al., 2007; Fries, 2009; vanElswijk et al., 2010; Schoffelen et al., 2011). The context dependent modulation of effectiveconnectivity is at the heart of cognition. A paradigmatic case for this is selective attention, inwhich relevant stimulus input is routed preferentially, and the result of this selective routingcan be read directly from the activity of the target neurons (Moran and Desimone, 1985;Treue and Maunsell, 1996; Reynolds et al., 1999). Our current results strongly suggest thatthe selective routing of attended input is implemented by selective gamma-bandsynchronization between the target and the attended input, according to the CTCmechanism.

EXPERIMENTAL PROCEDURESVisual stimulation and attention paradigm

All procedures were approved by the ethics committee of the Radboud University,Nijmegen, NL. Stimuli and behavior were controlled by the software CORTEX(http://www.cortex.salk.edu). Stimuli were presented on a CRT monitor at 120 Hz non-interlaced. When the monkey touched a bar, a gray fixation point appeared at the center ofthe screen. When the monkey brought its gaze into a fixation window around the fixationpoint (0.85 degree radius in monkey K; 1 deg radius in monkey P), a pre-stimulus baselineof 0.8 s started. If the monkeys gaze left the fixation window at any time, the trial wasterminated. The measured eye positions during correct trials used for analysis differed onlyby an average of 0.03 deg of visual angle between the two attention conditions. After thebaseline period, two physically isoluminant patches of drifting sinusoidal grating appeared(diameter: 1.2 degree; spatial frequency: 0.40.8 cycles/degree; drift velocity: 0.6 degree/s;resulting temporal frequency: 0.240.48 cycles/s; contrast: 100%). The two grating patcheschosen for a given recording session always had equal eccentricity, size, contrast, spatialfrequency and drift velocity. The two gratings always had orientations that were orthogonalto each other, and they had drift directions that were incompatible with a Chevron patternmoving behind two apertures, to avoid pre-attentive binding. Positions and sizes of the twostimuli were aimed to achieve the following: a) There should be one or more sites in area V4that were activated by the two stimuli to an equal amount. b) There should be one or moresites in area V1 that were activated by only one of the two stimuli.

In any given trial, one grating was tinted yellow, the other blue, with the color assignedrandomly across trials. The yellow and blue colors were physically equiluminant. After 11.5 s (0.81.3 s in monkey P), the fixation point changed color to match the color of one ofthe two gratings, thereby indicating this grating as the relevant stimulus and the other asirrelevant. For each trial, two independent change times for the two stimuli were determinedrandomly between stimulus onset and 4.5 s after cue onset, according to a slowly risinghazard rate. If the relevant stimulus changed (before or after the irrelevant stimulus changed)and the monkey released the bar within 0.150.5 s thereafter, the trial was terminated and areward was given. If the monkey released the bar at any other time, the trial was terminatedwithout reward. The stimulus changes were small changes in the grating pattern, with thestripes undergoing a gentle bend (Fig. 1E). During the bend, the outer ends of the gratingstripes lagged increasingly behind the center of the stripes, until the lag reached 0.1 degreeat 75 ms after the start of the bend. Over the course of another 75 ms, the stripesstraightened again. We used this shape change detection task, because previous studies ongamma-band activity in monkey area V4 had found larger attention effects for a shapetracking task (Taylor et al., 2005) than a color change detection task (Fries et al., 2001; Frieset al., 2008).



NIH


NIH


NIH


On 10 percent of the trials, only one of the two stimuli was presented, randomly at one orthe other position and tinted yellow or blue. In these trials, the fixation point alwaysassumed the color of this one grating, the change time was determined according to the samehazard rate and if the monkey released within 0.150.5 s thereafter, a reward was given.

Several sessions (either separate or after attention-task sessions) were devoted to themapping of receptive fields (RFs), using 60 patches of moving grating as illustrated in figureS1C. Receptive field positions were stable across recording sessions (Fig. S1D).

Neurophysiological recording techniques and signal preprocessingNeuronal recordings were made from two left hemispheres in two monkeys through amicromachined 252-channel electrocorticogram-electrode array (ECoG) implantedsubdurally (Rubehn et al., 2009). Briefly, a 6.54 cm craniotomy over the left hemisphere ineach monkey was performed under aseptic conditions with isoflurane anesthesia. The durawas opened and the ECoG was placed directly onto the brain under visual control. Severalhigh resolution photos were taken before and after placement of the ECoG for latercoregistration of ECoG signals with brain regions. After ECoG implantation, both the boneand the dural flap were placed back and secured in place. After a recovery period ofapproximately three weeks, we started with neuronal recordings.

Signals obtained from the 252-electrode grid were amplified 20 times by eight Plexonheadstage amplifiers (Plexon, USA), then low-pass filtered at 8 kHz and digitized at 32 kHzby a Neuralynx Digital Lynx system (Neuralynx, USA). LFP signals were obtained by low-pass filtering at 200 Hz and downsampling to 1 kHz. Powerline artifacts were removed bydigital notch filtering. The actual spectral data analysis included spectral smoothing thatrendered the original notch invisible.

Data analysis generalAll analyses were done in Matlab (The MathWorks, MA) and using FieldTrip (Oostenveldet al., 2011) (http://fieldtrip.fcdonders.nl).

For the analysis of the data recorded during the attention task, we used the time period from0.3 s after cue onset (the change in the fixation point color) until the first change in one ofthe stimuli. For each trial, this period was cut into non-overlapping 0.5 s data epochs,discarding remaining time at the end of the period that was less than 0.5 s long.

We calculated local bipolar derivatives, i.e. differences (sample-by-sample in the timedomain) between LFPs from immediately neighboring electrodes. We refer to the bipolarderivatives as sites. Bipolar derivation further enhances spatial specificity of the signaland removes the common recording reference, which is important when analyzingsynchronization between sites. Subsequently, per site and individual epoch, the mean wassubtracted, and then, per site and session, the signal was normalized by its standarddeviation. These normalized signals were pooled across sessions with identical stimulus andtask, unless indicated otherwise.

Spectral AnalysisSpectral power, coherence and GC influences were estimated by applying a fast Fouriertransform (FFT) after multitapering (Mitra and Pesaran, 1999) with seven tapers. Givenepoch lengths of 0.5 s, this resulted in a spectral smoothing of 7 Hz. The resulting spectraare shown from 8 Hz to 140 Hz. We performed a separate analysis of the lower frequencies(4 Hz to 28 Hz), in which the same 0.5 s data epochs were Hanning tapered. This did notreveal any consistent attentional effect.



NIH


NIH


NIH


For the analysis of GC influences, we applied non-parametric spectral matrix factorization tothe cross-spectral density (Dhamala et al., 2008). We performed this factorization separatelyfor each pair of sites. GC influence spectra were first estimated with the same spectralconcentration parameters as all spectra, and then smoothed with a two-frequency-bin boxcarwindow. If in a site pair, one site has a higher signal-to-noise ratio (SNR), then the analysisof GC influences has a bias towards estimating a stronger influence from the high-SNR siteto the low-SNR site (Nalatore et al., 2007). To control for this, we stratified for SNR. Wedefined SNR as the absolute power of the bipolar derived, de-meaned and standard-deviation normalized signal in the frequency band for which the stratification was intended.There were two types of comparisons related to the Granger analysis and two correspondingtypes of stratification:

1. We compared bottom-up with top-down GC influences. In this case, we stratifiedSNR per site-pair across the two areas.

2. We compared GC influences in a given direction between two attention conditions.In this case, we stratified SNR per site-pair across the two attention conditions.

In both cases, per site-pair, trials were discarded until the mean SNR was essentiallyidentical (and the SNR distribution across trials was as similar as possible) across sites (case1.) or across attention conditions (case 2.). If for a given site pair, this left fewer than 100trials, the site pair was discarded from the stratified analysis.

Statistical TestingStatistical testing included two steps: We first tested across all frequencies for significancesat a p

distributions (Nichols and Holmes, 2001; Maris et al., 2007). Frequencies with significantcoherence differences are marked with a gray bar in all figures.

This procedure, as explained for the example coherence spectrum, was applied similarly forthe average across the entire sample of coherence spectra. The only difference was that, foreach randomization, the random condition assignment was done per coherence spectrumcontributing to the average, rather than per epoch.

The same approach as used for coherence spectra was also applied to GC influence spectra.

Similar randomization procedures were also used for the average across the 6080 Hz bandand for gamma peak frequencies or amplitudes: The observed difference was compared tothe randomization distribution of differences. A correction for multiple comparisons was notnecessary in this case.

Statistical testing for the cross-frequency interaction is described below.

Site selectionThe main requirement for V1V4 site pairs to be included in the analysis was that the V4site had to be driven roughly equally by the two stimuli, while the V1 site had to be drivenprimarily by one of the two stimuli. In order to employ an objective selection, sites and sitepairs had to satisfy a number of quantitative criteria. The results that we report were robustagainst moderate changes in those criteria. Obviously, coherence and correspondingattention effects got weaker when e.g. sites were included that were not properly stimulusdriven, or pairs of sites whose receptive fields did not overlap well. The selection wasperformed according to the following steps:

1. For each site, we normalized power spectra to make the values more directlyinterpretable: We calculated the gamma-band power (P; 40100 Hz) averagedacross all pre-stimulus baseline periods (Pb) and during stimulation (Ps). Wecalculated normalized power spectra during stimulation (nPs): nPs=(PsPb)/Pb.

2. We selected those sites that were driven by the stimuli: In V1 and V4, we selectedall sites for which nPs during simultaneous presentation of both stimuli, i.e. duringthe two conditions of the attention task pooled, was 0.2 or larger.

3. For each site, we calculated a response map (RM) as shown in figure 1C and D,using the data from the RF mapping (sub-) sessions: We created a spatial map ofnPs for the 60 different mapping stimuli (Fig. S1C), smoothed this map with aGaussian kernel and normalized it to values between zero and one.

4. From the driven sites (step 2), we selected sites with RFs that overlapped well withone of the stimuli used in the attention paradigm. To calculate the overlap, wemultiplied the sites response map with the respective stimulus map, i.e. a map withthe value one where the stimulus fell and zero everywhere else, and subsequentlysummed across space. We selected sites whose overlap with one of the stimuliexceeded 30% of the distribution across the driven sites.

5. We selected V1V4 site pairs that had some chance of being anatomicallyconnected, based on a quantification of overlap between their response maps: Wemultiplied the V1-RM with the V4-RM and summed across space. We selectedpairs whose overlap exceeded 30% of the distribution across the driven sites. RFoverlap was positively related to V1V4 gamma-band coherence (Fig. S1F).

6. In V1, we selected those sites that had a preference for one of the two stimuli: Wecalculated a stimulus selectivity index (SI), by relating the gamma-band power



NIH


NIH


NIH


induced by the two stimuli, i.e. Ps1 and Ps2, respectively: SI=(Ps1Ps2)/(Ps1+Ps2).We selected those V1 sites for which the absolute value of the selectivity index was0.4 or larger. This created one group (Gs1) of sites with preference for stimulus 1and one group (Gs2) with preference for stimulus 2.

7. In V4, we selected those sites that were roughly equally driven by the two stimuli.We calculated the same selectivity index as in V1, but selected sites with aselectivity index (absolute value) of 0.1 or smaller.

8. We excluded those V1 sites that, despite their relative preference for one stimulus,still picked up activity induced by the other stimulus. To this end, we did thefollowing: For each site in Gsx (x being 1 or 2), we calculated the coherencespectrum to all sites in Gsy (y being 2 if x was 1 and vice versa). We averaged allthose coherence spectra. If they showed coherence, this was a sign that the Gsx sitepicked up activity induced by stimulus 2. Thus, we eliminated a Gsx site if itsaverage coherence spectrum to all Gsy sites was not flat. In order to quantifywhether a coherence spectrum was flat, we compared it to its bias estimate. Weestimated the coherence bias by shuffling all trials before calculating coherence.Those bias estimates were subtracted for each Gsx-Gsy pair before averaging thecoherence spectra between a given Gsx site and all Gsy sites. The average biassubtracted coherence spectra were then rectified and summed across allfrequencies. The resulting value had to be smaller than 0.06.

Cross-frequency analysisWe analyzed the effect of the theta frequency phase in V4 on the high frequencysynchronization between V1 and V4 as follows. The phase of the V4 theta oscillation wasdetermined from a set of average referenced sites overlying V4. Signals obtained from thesesites were band-pass filtered between 3 and 5 Hz, and the time points of the peaks of the lowfrequency oscillation were determined using the Hilbert transform, after averaging acrosssites. Subsequently, we computed the time frequency representation of V1V4 coherence,time-locked to the peak of the low frequency V4 theta oscillation. We only included thosetrials for a given V1V4 pair when the stimulus encoded by the V1 site was the attendedstimulus. Coherence was computed using a frequency dependent sliding window of 10cycles, between 40 and 100 Hz, in steps of 2 Hz. The resulting time-frequencyrepresentations showed high coherence in the gamma-band in slightly different bands forboth monkeys (monkey K: 7080 Hz, monkey P: 6070 Hz). The magnitude of coherenceseemed to systematically change as a function of the low frequency phase. We evaluated thisstatistically by performing a non-parametric randomization test and repeated the followingprocedure 1000 times: We randomly permuted the time axis of the individual peak-lockedsnippets, prior to re-computing the coherence. This shuffling essentially destroyed thetemporal profile of the phase of the theta oscillation, and served to construct a null-distribution of the amplitude of a cosine function (with a frequency of 4 Hz.) fitted to thetemporal profile of V1V4 coherence in a predefined frequency band. The estimatedamplitude of the cosine function from the unshuffled data was tested against thisdistribution, to obtain a p-value.

Supplementary MaterialRefer to Web version on PubMed Central for supplementary material.

AcknowledgmentsWe thank Mark Roberts and Eric Lowet for support, Edward Chang for help with implanting monkey P, MingzhouDing for providing the code for spectral matrix factorization, and Karl Friston and Wolf Singer for helpful



NIH


NIH


NIH


comments on earlier versions of this manuscript. This work was supported by the European Young Investigatorprogram of the European Science Foundation (PF), by the European Unions seventh framework program (PF), theNational Science Foundation Graduate Student Fellowship Program (AB), and a Fulbright grant from the U.S.Department of State (AB). RO and JMS gratefully acknowledge the support of the BrainGain Smart MixProgramme of the Netherlands Ministry of Economic Affairs and the Netherlands Ministry of Education, Cultureand Science.

Reference ListAkam T, Kullmann DM. Oscillations and filtering networks support flexible routing of information.

Neuron. 2010; 67:308320. [PubMed: 20670837]Barone P, Batardiere A, Knoblauch K, Kennedy H. Laminar distribution of neurons in extrastriate

areas projecting to visual areas V1 and V4 correlates with the hierarchical rank and indicates theoperation of a distance rule. J Neurosci. 2000; 20:32633281. [PubMed: 10777791]

Bichot NP, Rossi AF, Desimone R. Parallel and serial neural mechanisms for visual search in macaquearea V4. Science. 2005; 308:529534. [PubMed: 15845848]

Brgers C, Kopell NJ. Gamma oscillations and stimulus selection. Neural Comput. 2008; 20:383414.[PubMed: 18047409]

Bosman CA, Womelsdorf T, Desimone R, Fries P. A microsaccadic rhythm modulates gamma-bandsynchronization and behavior. J Neurosci. 2009; 29:94719480. [PubMed: 19641110]

Bragin A, Jando G, Nadasdy Z, Hetke J, Wise K, Buzski G. Gamma (40100 Hz) oscillation in thehippocampus of the behaving rat. J Neurosci. 1995; 15:4760. [PubMed: 7823151]

Buehlmann A, Deco G. Optimal information transfer in the cortex through synchronization. PLoSComput Biol. 2010; 6

Buffalo EA, Fries P, Landman R, Buschman TJ, Desimone R. Laminar differences in gamma andalpha coherence in the ventral stream. Proc Natl Acad Sci U S A. 2011; 108:1126211267.[PubMed: 21690410]

Buschman TJ, Miller EK. Top-down versus bottom-up control of attention in the prefrontal andposterior parietal cortices. Science. 2007; 315:18601862. [PubMed: 17395832]

Busse L, Katzner S, Treue S. Temporal dynamics of neuronal modulation during exogenous andendogenous shifts of visual attention in macaque area MT. Proc Natl Acad Sci U S A. 2008;105:1638016385. [PubMed: 18922778]

Canolty RT, Edwards E, Dalal SS, Soltani M, Nagarajan SS, Kirsch HE, Berger MS, Barbaro NM,Knight RT. High gamma power is phase-locked to theta oscillations in human neocortex. Science.2006; 313:16261628. [PubMed: 16973878]

Chalk M, Herrero JL, Gieselmann MA, Delicato LS, Gotthardt S, Thiele A. Attention reducesstimulus-driven gamma frequency oscillations and spike field coherence in V1. Neuron. 2010;66:114125. [PubMed: 20399733]

Colgin LL, Denninger T, Fyhn M, Hafting T, Bonnevie T, Jensen O, Moser MB, Moser EI. Frequencyof gamma oscillations routes flow of information in the hippocampus. Nature. 2009; 462:353357.[PubMed: 19924214]

Dhamala M, Rangarajan G, Ding M. Analyzing information flow in brain networks withnonparametric Granger causality. Neuroimage. 2008; 41:354362. [PubMed: 18394927]

Engel AK, Knig P, Kreiter AK, Singer W. Interhemispheric synchronization of oscillatory neuronalresponses in cat visual cortex. Science. 1991a; 252:11771179.

Engel AK, Kreiter AK, Knig P, Singer W. Synchronization of oscillatory neuronal responses betweenstriate and extrastriate visual cortical areas of the cat. Proc Natl Acad Sci U S A. 1991b; 88:60486052. [PubMed: 2068083]

Fell J, Klaver P, Lehnertz K, Grunwald T, Schaller C, Elger CE, Fernndez G. Human memoryformation is accompanied by rhinal-hippocampal coupling and decoupling. Nat Neurosci. 2001;4:12591264. [PubMed: 11694886]

Fries P. A mechanism for cognitive dynamics: neuronal communication through neuronal coherence.Trends Cogn Sci. 2005; 9:474480. [PubMed: 16150631]

Fries P. Neuronal gamma-band synchronization as a fundamental process in cortical computation.Annu Rev Neurosci. 2009; 32:209224. [PubMed: 19400723]



NIH


NIH


NIH


Fries P, Nikolic D, Singer W. The gamma cycle. Trends Neurosci. 2007; 30:309316. [PubMed:17555828]

Fries P, Reynolds JH, Rorie AE, Desimone R. Modulation of oscillatory neuronal synchronization byselective visual attention. Science. 2001; 291:15601563. [PubMed: 11222864]

Fries P, Womelsdorf T, Oostenveld R, Desimone R. The effects of visual stimulation and selectivevisual attention on rhythmic neuronal synchronization in macaque area V4. J Neurosci. 2008;28:48234835. [PubMed: 18448659]

Gattass R, Nascimento-Silva S, Soares JG, Lima B, Jansen AK, Diogo AC, Farias MF, Botelho MM,Mariani OS, Azzi J, Fiorani M. Cortical visual areas in monkeys: location, topography,connections, columns, plasticity and cortical dynamics. Philos Trans R Soc Lond B Biol Sci. 2005;360:709731. [PubMed: 15937009]

Gray CM, Knig P, Engel AK, Singer W. Oscillatory responses in cat visual cortex exhibit inter-columnar synchronization which reflects global stimulus properties. Nature. 1989; 338:334337.[PubMed: 2922061]

Gregoriou GG, Gotts SJ, Zhou H, Desimone R. High-frequency, long-range coupling betweenprefrontal and visual cortex during attention. Science. 2009; 324:12071210. [PubMed: 19478185]

Henrie JA, Shapley R. LFP power spectra in V1 cortex: the graded effect of stimulus contrast. JNeurophysiol. 2005; 94:479490. [PubMed: 15703230]

Hipp JF, Engel AK, Siegel M. Oscillatory synchronization in large-scale cortical networks predictsperception. Neuron. 2011; 69:387396. [PubMed: 21262474]

Hoogenboom N, Schoffelen JM, Oostenveld R, Fries P. Visually induced gamma-band activitypredicts speed of change detection in humans. Neuroimage. 2010; 51:11621167. [PubMed:20307670]

Hoogenboom N, Schoffelen JM, Oostenveld R, Parkes LM, Fries P. Localizing human visual gamma-band activity in frequency, time and space. Neuroimage. 2006; 29:764773. [PubMed: 16216533]

Kaminski M, Ding M, Truccolo WA, Bressler SL. Evaluating causal relations in neural systems:granger causality, directed transfer function and statistical assessment of significance. BiolCybern. 2001; 85:145157. [PubMed: 11508777]

Landau AN, Fries P. Attention samples stimuli rhythmically. Curr Biol. 2012; 22:10001004.[PubMed: 22633805]

Lee H, Simpson GV, Logothetis NK, Rainer G. Phase locking of single neuron activity to thetaoscillations during working memory in monkey extrastriate visual cortex. Neuron. 2005; 45:147156. [PubMed: 15629709]

Maris E, Schoffelen JM, Fries P. Nonparametric statistical testing of coherence differences. J NeurosciMethods. 2007; 163:161175. [PubMed: 17395267]

Mitra PP, Pesaran B. Analysis of dynamic brain imaging data. Biophys J. 1999; 76:691708.[PubMed: 9929474]

Moran J, Desimone R. Selective attention gates visual processing in the extrastriate cortex. Science.1985; 229:782784. [PubMed: 4023713]

Muthukumaraswamy SD, Edden RA, Jones DK, Swettenham JB, Singh KD. Resting GABAconcentration predicts peak gamma frequency and fMRI amplitude in response to visualstimulation in humans. Proc Natl Acad Sci U S A. 2009; 106:83568361. [PubMed: 19416820]

Nalatore H, Ding M, Rangarajan G. Mitigating the effects of measurement noise on Granger causality.Phys Rev E Stat Nonlin Soft Matter Phys. 2007; 75:031123. [PubMed: 17500684]

Nichols TE, Holmes AP. Nonparametric permutation tests for functional neuroimaging: a primer withexamples. Hum Brain Mapp. 2001; 15:125. [PubMed: 11747097]

Oostenveld R, Fries P, Maris E, Schoffelen JM. FieldTrip: Open source software for advanced analysisof MEG, EEG, and invasive electrophysiological data. Comput Intell Neurosci. 2011;2011:156869. [PubMed: 21253357]

Otero-Millan J, Troncoso XG, Macknik SL, Serrano-Pedraza I, Martinez-Conde S. Saccades andmicrosaccades during visual fixation, exploration, and search: foundations for a common saccadicgenerator. J Vis. 2008; 8:2118. [PubMed: 19146322]

Popescu AT, Popa D, Par D. Coherent gamma oscillations couple the amygdala and striatum duringlearning. Nat Neurosci. 2009; 12:801807. [PubMed: 19430471]



NIH


NIH


NIH


Ray S, Maunsell JH. Differences in gamma frequencies across visual cortex restrict their possible usein computation. Neuron. 2010; 67:885896. [PubMed: 20826318]

Reynolds JH, Chelazzi L. Attentional modulation of visual processing. Annu Rev Neurosci. 2004;27:611647. [PubMed: 15217345]

Reynolds JH, Chelazzi L, Desimone R. Competitive mechanisms subserve attention in macaque areasV2 and V4. J Neurosci. 1999; 19:17361753. [PubMed: 10024360]

Roelfsema PR, Engel AK, Knig P, Singer W. Visuomotor integration is associated with zero time-lagsynchronization among cortical areas. Nature. 1997; 385:157161. [PubMed: 8990118]

Rubehn B, Bosman C, Oostenveld R, Fries P, Stieglitz T. A MEMS-based flexible multichannelECoG-electrode array. J Neural Eng. 2009; 6:036003. [PubMed: 19436080]

Schoffelen JM, Oostenveld R, Fries P. Neuronal coherence as a mechanism of effective corticospinalinteraction. Science. 2005; 308:111113. [PubMed: 15802603]

Schoffelen JM, Poort J, Oostenveld R, Fries P. Selective movement preparation is subserved byselective increases in corticomuscular gamma-band coherence. J Neurosci. 2011; 31:67506758.[PubMed: 21543604]

Siegel M, Donner TH, Oostenveld R, Fries P, Engel AK. Neuronal synchronization along the dorsalvisual pathway reflects the focus of spatial attention. Neuron. 2008; 60:709719. [PubMed:19038226]

Sigurdsson T, Stark KL, Karayiorgou M, Gogos JA, Gordon JA. Impaired hippocampal-prefrontalsynchrony in a genetic mouse model of schizophrenia. Nature. 2010; 464:763767. [PubMed:20360742]

Singer W, Gray CM. Visual feature integration and the temporal correlation hypothesis. Annu RevNeurosci. 1995; 18:555586. [PubMed: 7605074]

Swettenham JB, Muthukumaraswamy SD, Singh KD. Spectral properties of induced and evokedgamma oscillations in human early visual cortex to moving and stationary stimuli. J Neurophysiol.2009; 102:12411253. [PubMed: 19515947]

Tallon-Baudry C, Bertrand O, Fischer C. Oscillatory synchrony between human extrastriate areasduring visual short-term memory maintenance. J Neurosci. 2001; 21:RC177. [PubMed: 11588207]

Tallon-Baudry C, Mandon S, Freiwald WA, Kreiter AK. Oscillatory synchrony in the monkeytemporal lobe correlates with performance in a visual short-term memory task. Cereb Cortex.2004; 14:713720. [PubMed: 15054050]

Taylor K, Mandon S, Freiwald WA, Kreiter AK. Coherent oscillatory activity in monkey area v4predicts successful allocation of attention. Cereb Cortex. 2005; 15:14241437. [PubMed:15659657]

Tiesinga PH, Sejnowski TJ. Mechanisms for Phase Shifting in Cortical Networks and their Role inCommunication through Coherence. Front Hum Neurosci. 2010; 4:196. [PubMed: 21103013]

Treue S, Maunsell JH. Attentional modulation of visual motion processing in cortical areas MT andMST. Nature. 1996; 382:539541. [PubMed: 8700227]

van Elswijk G, Maij F, Schoffelen JM, Overeem S, Stegeman DF, Fries P. Corticospinal beta-bandsynchronization entails rhythmic gain modulation. J Neurosci. 2010; 30:44814488. [PubMed:20335484]

van Pelt S, Boomsma DI, Fries P. Magnetoencephalography in twins reveals a strong geneticdetermination of the peak frequency of visually induced gamma-band synchronization. J Neurosci.2012; 32:33883392. [PubMed: 22399760]

Vinck M, Lima B, Womelsdorf T, Oostenveld R, Singer W, Neuenschwander S, Fries P. Gamma-phase shifting in awake monkey visual cortex. J Neurosci. 2010; 30:12501257. [PubMed:20107053]

von Stein A, Chiang C, Knig P. Top-down processing mediated by interareal synchronization. ProcNatl Acad Sci U S A. 2000; 97:1474814753. [PubMed: 11121074]

Womelsdorf T, Fries P, Mitra PP, Desimone R. Gamma-band synchronization in visual cortex predictsspeed of change detection. Nature. 2006; 439:733736. [PubMed: 16372022]

Womelsdorf T, Schoffelen JM, Oostenveld R, Singer W, Desimone R, Engel AK, Fries P. Modulationof neuronal interactions through neuronal synchronization. Science. 2007; 316:16091612.[PubMed: 17569862]



NIH


NIH


NIH


Zeitler M, Fries P, Gielen S. Assessing neuronal coherence with single-unit, multi-unit, and local fieldpotentials. Neural Comput. 2006; 18:22562281. [PubMed: 16846392]



NIH


NIH


NIH


HIGHLIGHTS

There is pronounced gamma-band synchronization between monkey areas V1and V4.

Area V4 gamma-synchronizes selectively to the behaviorally relevant portion ofV1.

Relevant V1-gamma peaks 23 Hz above irrelevant V1, and 46 Hz above V4-gamma.

In the gamma band, V1 influences V4 much stronger than vice versa.



NIH


NIH


NIH


Fig. 1.High density monkey ECoG grid and attention paradigm. (A) Rendering of the brain ofmonkey P. Lines indicate the covered area with the major sulci. Dots indicate the 252subdural electrodes. (B) Power change relative to baseline in a V1 electrode for contra- andipsilateral stimulation. (C) Receptive fields of the 35 V1 example electrodes shown ingreen in panel (A). (D) Receptive fields of the 33 V4 example electrodes shown in red inpanel (A). Color bar applies to (C) and (D). (E) Selective attention paradigm. SeeExperimental Procedures for details. See also Figure S1.



NIH


NIH


NIH


Fig. 2.Example triplet recording of one V4 and two V1 sites in monkey P. (A) Illustration of thetwo single-stimulus conditions, corresponding to the red/blue lines in (B) through (F). Bothconditions activated V4, but only the condition labeled red (blue) activated site V1a (V1b).The double arrow illustrates the likely pattern of interaction between neuronal groups. (B)through (D): Spectral power changes (relative to pre-stimulus baseline) for the indicatedsites. (E), (F) Coherence spectra for the indicated site pairs. Gray bars indicate frequencieswith a significant effect (p

Fig. 3.Analysis of GC influences for the example triplet. The spectra are GC influence spectrabetween the indicated sites and in the indicated direction. (B) through (E) show the resultsfor the single-stimulus conditions illustrated in (A). (G) through (J) show the results for thetwo-stimulus conditions with selective attention to either stimulus, as illustrated in (F). Graybars indicate frequencies with a significant effect (p

Fig. 4.Average results from the attention paradigm. (A) Average spectral power change relative topre-stimulus baseline in V4. (B) Average spectral power change relative to pre-stimulusbaseline in V1, when the stimulus activating the respective site was behaviorally relevant(red) or irrelevant (blue). (C) Average V1V4 coherence spectrum, when the stimulusactivating the respective V1 site was relevant (red) or irrelevant (blue). Gray bar indicatesfrequencies with a significant effect (p

corresponding to a V1V4 site pair and a recording session, comparing attention outside theV1-RF (x-axis) to attention inside the V1-RF (y-axis).



NIH


NIH


NIH


Fig. 5.The gamma peaks of the two monkeys. All panels show the individual gamma peaks of theindicated monkeys. Area (combinations) and attention conditions are color coded asindicated at the top. The values of visually induced power and of coherence were divided bytheir respective maximum values in order to scale all peaks to unit peak height and therebyease direct comparison of peak frequencies. Scaled spectra are shown as dots. Gaussianswere fitted to the top third of each peak and are shown as solid lines (all r-square valueswere above 0.98). The panels on the right show the peaks in detail. The Gausians means,i.e. the gamma-band peaks, are shown as vertical lines (shaded regions correspond to 95%confidence bounds) and as numbers. See also Figure S5.



NIH


NIH


NIH


Fig. 6.Average GC influences. Each panel shows the average GC influences for the indicatedmonkeys, attention conditions and directions. Gray bars indicate frequencies with asignificant effect (p

Fig. 7.Coherence spectra before and after cue presentation. (A) Coherence spectra between V4 andthe indicated V1a site (same sites as Fig. 2). The pink spectrum is from the pre-cue epoch.The red (blue) spectrum is from the post-cue epoch with attention inside (outside) the V1-RF. Colored bar(s) indicate significant differences between the pre-cue spectrum and thespectrum with the same color as the significance bar (p

Fig. 8.Cross-frequency analysis of inter-areal synchronization as a function of time around thetheta peak. (A) V1V4 coherence as a function of time relative to peaks in the 4 Hz rhythm.(B) Coherence between V1 and V4 in a 7080 Hz band, as a function of time relative topeaks in the 4 Hz rhythm. Data are from monkey K. See figure S4 for data from monkey P.



NIH


NIH


NIH


Documents

Nihms402820Bosman Monkey