Frontal midline theta rhythms reflect alternative;pdf

Neuroscience Letters 274 (1999) 29±32
www.elsevier.com/locate/neulet
Frontal midline theta rhythms re¯ect alternative activation of
prefrontal cortex and anterior cingulate cortex in humans
Hiroshi Asada a,*, Yutaka Fukuda b, Shigeru Tsunoda a, Masahiko Yamaguchi c,
Mitsuo Tonoike c
a
Department of Earth and Life Sciences, Osaka Prefecture University, Sakai 599±8531, Japan
b
Department of Physiology, Osaka University Medical School, Suita 565±0871, Japan
c
Life Electronics Research Center, MITI, Amagasaki 661±0974, Japan
Received 16 April 1999; received in revised form 12 August 1999; accepted 12 August 1999
Abstract
Frontal midline theta rhythm (Fmu) often appears on electroencephalogram (EEG) during consecutive mental tasks. To
clarify the source of rhythmic activity, magnetoencephalogram (MEG) and EEG were simultaneously measured in six
healthy volunteers during different mental tasks using whole head MEG system. MEG records were averaged every one
cycle of Fmu rhythms using individual positive peaks of Fmu waves in Fz EEG as a trigger. Averaged theta components of
MEG signals were analyzed with a multi-dipole model. Two sources were estimated to the regions both of the prefrontalmedial super®cial cortex and anterior cingulate cortex (ACC). These regions were alternatively activated in about 40 to
1208 phase shift during one Fmu cycle. From above results, we hypothesize that appearance of Fmu during consecutive
mental tasks re¯ects alternative activities of the medial prefrontal cortex and ACC. q 1999 Published by Elsevier Science
Ireland Ltd. All rights reserved.
Keywords: Magnetoencephalography; Electroencephalography; Frontal midline theta rhythm; Prefrontal cortex; Anterior cingulate
cortex; Multi dipole modeling; Attention
It has previously been shown that rhythmic theta activity
often appears over the midfrontal region on the electroencephalogram (EEG) during various mental tasks in normal
subjects [3,8,12,14]. Such rhythmic activity was named
frontal midline theta rhythm (Fmu) [8]. Though previous
studies have suggested that Fmu activity is observed when
a continuous concentration of attention is required to
perform a task [3,14], few studies have attempted to clarify
the generation mechanism. Recently, some source modeling
efforts for Fmu done using magnetoencephalogram (MEG),
have suggested that this signal arises from various parts of
the lateral frontal cortex of both hemispheres [12] or large
area of medial prefrontal cortices including anterior cingulate cortex (ACC) [9]. Although human brain mechanisms
subserving attention have been assigned to prefrontal,
midfrontal, and posterior parietal cortices, as well as to
the anterior cingulate and the thalamus [11], the generation
mechanism of a focal u rhythm on EEG during continuous
* Corresponding author. Tel.: 181-722-549-749; fax: 181-72254-9-749.
E-mail address: [email protected] (H. Asada)
attention has not been clari®ed yet. In this study, we
attempted to extract the magnetic ®eld potentials that corresponded to Fmu rhythms on the EEG recordings and
attempted to clarify the source areas of rhythmic activity
associated with consecutive attention.
The experiments were carried out with six healthy volunteers (subject S1±S6, ®ve males and one female, aged 21±29
years), who had frequently showed Fmu in preliminary EEG
test. MEG and EEG activities were simultaneously recorded
while subjects performed mental tasks. Subjects were
required to perform continuous four mental tasks for about
10 min with 5 min resting periods inbetween. Tasks were:
(1) serial addition of four ®gures presented in front of
subjects; (2) successive subtraction of 7 from 100; (3) recollection of Kanji characters and (4) attention to their breath
with very slow rhythm. The magnetic ®eld outside the head
was measured with a 122-channel neuromagnetometer
covering the whole scalp (Neuromag-122TM) in a magnetically shielded room. EEG signals were collected from Ag/
AgCl electrodes placed at locations Fpz, Fz, Cz, F3, and F4
(10±20 system) and all electrodes were referred to the
shorted right-left earlobes. Both the MEGs and the EEGs
0304-3940/99/$ - see front matter q 1999 Published by Elsevier Science Ireland Ltd. All rights reserved.
PII: S03 04 - 394 0( 9 9) 00 67 9- 5
織庵■d■■ld，ノ∼■Un＞さ亡J訂Iビ書」〆ねほヱ〃り労針か−ヱ2
Ilぬm油iIllu川畑bilil恒lい1，uJα、dilα11）kIl川・il15卜
0爪hヒrlム≠心r11Iu阜ヾlヾαd山ざ糾）t古仏ヾlヾ凛八〟Pldl．一打
I血山ヾイ舶lいiヽa爪dlh■ヽtri爪ilblヒrl二ゝ≠・dゞi爪l爪山此d血lい
芯師小■− 上長∫
・巾′（：蒜
ゝtヒt；
ー▲！≡
仙（㌫：￡
……皿叫山＝仙山111＝dil’l山ヾilu，爪山huliい恥Ii吋all
血1“佃れM坤肛1Ll阜l，仙Ilαi爪l呼止仙川l川W＝叫血血
Ilい爪l胤hゝu恒α1a爪dヒrDRll止l、叫ヾlヾゝuドJ血叩…血h州
il．
− ▲
ぐいIlヾ川両ヾlheb【恒l山Il、tけ…1Pド∬山川ヒbG
Jピー吋止血ざ、du‡血ざ伯山mヾ爪l■＝加止i爪dlゝl巾iαb．トig．1∧
止．川ゝ加ヾヽ∬l叩IヾlけヒヒG加Jゝ帖Glい八万di哩ゝi爪叫，mヾ
d旭爪J山♭訓山th＝腑“哩旭日川叩両肌k←匹u拍Ii吋加ゝi爪
ゝl巾iαlSl．≠’山鳩ヒヒGlヾ山井di爪ざ、ゝ山川山卜爪I一事u〉爪Iai爪川g
射血血いl−1恒l山Iliヾbul←hl汀くトTllいhd■l桓l山Ilゝ“一山1hヾ
Ⅷll、rI叫l岨叫＝…ヾ血血l胤い恥l＝，IMldi爪l−…lbl
肌い川中血圧帖G心血Il吋、小m準肛1Lトn肌nlト什机．トig．
1b山川ゝ1hih小−uJ仙川l＝汀ゝド血al肌tl満山dヾl汀爪lト几I一事
加Jトm一事hゝl巾iヽ・dSl．「1仙ふトuI仙〉u∫ゝ机ヾlヾ…lヾubh人1
い爪痛いimibl加I爪Vト心匹J、i川ゝ“心l加叫iu｝1山h山
匹扁Iiい爪i爪Ihヾ旭爪ヒゝu恒小J．払い山小l．l九ヾ爪け爪川a爪Jト几I一事
aじ1iIilitゝ叫tTヾ血対日ilu爪III いhdヽヾdi爪血ヾゝa爪V 暮￣lい爪l−
m“lillI哩lい恥．
人血IdトIil址∵山七日ぜ川噂．爪lトm小小川叫匹仙川l、両Ih hiざh
Flg−1− A M G8nd【G ●⊂q■か咽i｛扉l巨訓d竹博iⅦ昭砂山
■mPl仙d●iP●・ぐr■い噂htl血r■ng虚d。旧n■lt■il●n蓼ub鱒Sl・
Upp●r小職●l−∩●i■叩【∈Gr●亡q血n騨■ndth●bⅥ爬rl−n●i■付
和∝川畑∩耶I帽mt和ntalはGd■●亡tqほn●■rlY■如V■【G●● 仁・
trQ血書．Th●urhⅥhl¶加一別i≠●帽■l計卦ut￠．4山．沌l由一亡即旭u帽
qIi印仁一r■l■mPいud●b†Fl†▼r■ng●ほ．ト7．5Hかni脚nt■n●Ou∼
M【GくIdtl一雨【∈Gれghtl 仁1仰W●n軌Ib■dSl抽聯〟dI代‖¶
■lxIY●l．
肌叩Iilulklヾl血dl小h両Iu，山dkヾヽ伯山ヽh｝lα1hヾl心爪Ial
肌い血帖…叩皿l 虹虹G 鮎仙肌lゝ h、凧呵1■」b 加J
ゝゆiαb．トiざ．三川小心加ゎIl虹aId準。爪小爪州即爪申，Ml、
いI’lユニゝ帖Gヾ山肌lゝ加Jlh＝Wl里坤＝血直“卿町仙川hd
IiwLLGヾI肌l、lヾゝ匹Ii、ヾll…I川l頭ヾdinl山J11トbi爪
ゝl巾iαlSl．ト小、如ili闇爪Id捕此Ilば≠1、爪小l山i爪I山川付け−
暮i吋m■l山i爪Ihヾゝl加ialdふ加ibuliい爪い暮￣mh血書封山l血相
帆ヾlぐdiざhi〟dala畑l叩血ざIl叫u瓜いけ・lll）仙．Iilleld
小一illl■b叫IJ一匹ゝ＝｝川．ト又＝レ．P由iliATd山l■恥Vlヾ効ル山
川＝＝拙咤仙加h甲tl闇IdふIal山舶lい圧dい汀−1血∴
hゝll血∫i“血恥Ii汀mati刷り十‘1’＝1井草亜腑い胎ゝ1bG
l
−
−
■
▲
■
＼．． R ノ
、 ● ＋ I − ＝ − レ
ナヽトトーY世ヽ（′しI㌔
IVゝ」L トビ・ゼしセナtr
三こ昌
≡肘↑
㌢；−
● 1 ＼、
；〓
〓．≡
蒜；
鶴茎婁套蔓蔓
†
至芸
訂三
．′′′ サ払
＼鼠．≡−ニ
忘芸賢父
れや㌧・・、−＼、・・＼＼＼、
▲ ● ∵ − 1 ● ・ − J r T ▲ ▼
獣1il恒仏祖…I川l■山Ii吋加、i爪Judi爪ざ1加u，llヾ…lilヾ
ト爪ItIbuhl、．lゝい1入場1州l＝汀ゝド血al肌tl両Iu血l玩叩川11−
IIUけu、MヒG乱tiIili小机ヾlヾ闇L川心血dIi吋トm帝l訓爪ダいI’5j一
丁．5仙什圧l叫u用lいuJ＝けゝド血aI肌l【満山血IirヒbG
≠ヾlい町献血h肌≠肌・JM山じ山川l血心血転lく，dulll山
ヒヒGゝ）乱丁爪Ialtl山王MbGヾヽa爪l几al皿爪t
爪thゝuu均しt、．
M哩仙elLl⊥暮d、iざ仙I川・ehごdルel準ヾJいl＝i肝ヾI叩川小m、
＿−● −、
＼・→ヰ‡阜ノノノふ
1
｝昌一Eニサ芸ノ
1
ニニケふ芸
言草芸ざ
l I ◆
小一illllヾゝlMl小i山両du■l匹扇山万Pdbい‖Ⅷ川Ⅷ甘“い川
トーヽ■■ ヽ■ん
ト′ヒ上GぬalllざざuP，肌Lb止加“相川伽心㈹門1山Iiい爪1
ヽF
51〉用tlX）小甘lヾ、両Ih、i爪Iil■lduJali）恥i爪トm帝buJゝtゝduJ川ざ
せヽ
高札血t射止．llト）け爪山川Jll■P胱仙川川≠血l九爪Ilh∴肋山一
セナ事■『
坤山血佃…叫M刷1、i爪1三三du爪l山、al、止ddl加此i阜
∼▲｛
uヽi爪ざIl虹mi爪i仙川l−仙吋爪l小暮i爪l血il刑几Idhl止「1’小i血爪1れ
ヽ．．世
ー】 ■■■ ′
“、Mしぃ州山l〉削ぎ山ヾlll…ul山、lざ仙l・．1Ilいlざ心IJl、IIlhu・
＼
1iい恥叫uヾ爪Il止Id両tllい肛叫…aL爪tヾuJT川再伸丸
く虻け●．「Il虹ざlヽ血、ヽ1）ト血＝血い直川l’lhヾ仙ヽ山肌ぬ
u山い血itvh■州・mudl11−11虹m伽uJdIiヾIdl折ぬ
嵐山〉u爪tヾdl．一万bIt加LrD∴1．いIlゝ加ヾ叫〉ulα、いI一肌“哩山
Flg．2一切【G8nd【【Gth■t8用m卿●mi訓●r8騨dI相関1く旧th■t8
1hda、ig几aゝ．bCl大小一助イいul山b＝l闇1−叫山lい鳩lしゎtい
≠■∨●事ln122M【Gi●ni8ほ■ndIiv【【G●−一閃d●i，Fq山一lln●∼IM
血い打Igl仙lゝ咤仙l川lヾぬuⅣdbl■朝一hヾlいI一三くトlユJぬIl肘l、 ●■亡hほGand【【G亡h即川●lihQWtlt・用mPQn●nti⊂■l亡ulddln
IQ山一t■■b．r●i圃W叶
…ulh川一山ia川い爪1al∬どぬk闇血rt九叩l凱鮎■gl山iい爪td＝
H. Asada et al. / Neuroscience Letters 274 (1999) 29±32
31
Fig. 3. (A) Original and estimated mFmu of subject S1. (B) The
upper two graphs show an averaged theta wave in Fz EEG and an
component wave in MEG detector above Fz EEG electrode. The
middle two graphs show dipole moments of sources (1 and 2 in
(D)) in a time-varying two-dipole modeling. The lower graph
shows the g value in a two-dipole modeling during one cycle.
(C) Magnetic ®eld iso-contours (gray for ¯ux exiting the head)
and ECDs at 5, 43 and 160 ms during one Fmu cycle (duration 160
ms) in subject S1 (viewed from front). (D) Source regions superposed in the subject's sagittal MR slice.
under Fpz to Cz in six subjects. The other dipole was estimated in the ACC area in ®ve subjects (S1±S5). In subject
S6, dipoles were estimated only in the prefrontal super®cial
cortex. These actual source locations were within ^ 10 mm
from the displayed slice. The two ECDs estimated in each
subject, alternatively ®red at about 40±1208 phase shift
during each Fmu cycle. The g values in the two-dipole
modeling were typically over 80% in four subjects.
The localized topographical distribution of the isocontours of spectral amplitude for theta activity in MEG
and EEG suggests that the frontal mFmu and Fmu rhythms
most likely arises from a compact, relatively medial and
super®cial source. For this reason, it is convincing that
ECDs for Fmu were localized to both regions of the medial
ACC and the prefrontal-medial super®cial cortex in almost
all subjects. The prefrontal cortex and the limbic association
cortex including ACC are known to be mutually connected
and to make a neural network [10]. From above ®ndings we
propose a hypothesis that the appearance of Fmu rhythms
during mental tasks is associated with interactions of both
regions of the ACC and the prefrontal-medial super®cial
cortex and re¯ects a composite of potentials from them.
If Fmu rhythm generates from the interactions of alternate
activities that were represented by those two dipoles, the
location of averaged activity of the two dipoles may be
shown in the middle area between the ACC and the prefrontal-medial super®cial cortex. In fact, Ishii et al. [9] demonstrated the large area of medial prefrontal cortex between
superior frontal gyrus and anterior cingulate gyrus, as the
source of the Fmu using Synthetic Aperture Magnetometry
components among four tasks. In six subjects, spatial distribution of these theta components showed similar pattern but
phase lag was found between each subject.
Results of dipole modeling using the extracted mFmu
components are shown in Figs. 3 and 4. Fig. 3 shows a
time-varying two-dipole modeling applied to one cycle of
Fmu in subject S1. A clear dipolar pattern over the midfrontal area, example in Fig. 3C, was observed in isocontour maps at the peak of Fmu wave in all subjects. In
subject S1, two obvious sources were estimated in the
prefrontal super®cial lobe underneath Fz and in the ACC
(Fig. 3D). Note that these two sources ®red with reverse
direction at a constant 908 phase shift during one Fmu
cycle (middle ®gures in Fig. 3B). When the two ECDs
were introduced into a two-dipole model, the measured
MEG signals in 32 detectors over the medial frontal area
were explained to better than 80% up in g value during one
cycle (bottom ®gure in Fig. 3B). The original and estimated
signals practically coincided (Fig. 3A). Similar results were
obtained in other three subjects.
Fig. 4 shows source locations projected on sagittal MR
slices of six subjects. Of the two-dipole sources during one
Fmu cycle calculated in each subject, one was estimated in
various medial parts of the prefrontal super®cial cortex
Fig. 4. Source locations projected on sagittal MR slices of six
subjects.
32
H. Asada et al. / Neuroscience Letters 274 (1999) 29±32
(SAM) analysis in MEG. Their result based upon statistical
parametric imaging by SAM does not con¯ict with our
hypothesis.
Precedent studies using MEG estimated the ECDs for
Fmu to prefrontal area [7] or various parts of the dorsolateral
frontal cortex in both hemispheres [12]. One of the reasons
for the differences between their and our results, may be that
their data analysis were done for raw MEG data through
band-pass ®lter. In this study, we extracted MEG components (mFmu) that were related to only the Fmu rhythm by
using EEG peak potential as the trigger. The use of averaging for the background rhythmic activity eliminated the
effects of variability of the other activities unrelated to Fmu,
facilitating the identi®cation of Fmu-dependent activity
generated by Fmu sources. Though the MEG signals from
the deeper cortex were very week, the averaging and twodipole model method enabled us to estimate the second
dipole ACC at other time points except at the time point
of maximum amplitude for mFmu or Fmu wave (Fig. 3).
The estimation to the ACC region as one of the sources of
Fmu is also supported by the study of Gevins et al. [3]. By
using EEG and MRI system, they reported that the average
dipoles computed over 10 Fmu bursts were found to cluster
around the ACC, similar to the regions that we found in the
present study. It has also been reported in many positron
emission tomography (PET) studies that attentional tasks
activate both the prefrontal cortex and ACC [1,2]. Furthermore, in PET studies medial prefrontal region including
both the prefrontal cortex and ACC is reported to play an
important role in selective or continuous performance of
attention [6,13]. Above PET data also support our present
result.
Waveform and distribution of extracted mFmu components were insensitive to the type of information being
processed. This insensitivity to the type of mental tasks
suggests that the extracted theta components in MEG
were responsible for continuous focused attention that is
required in common to perform the task. This sustained
attention might be accomplished via the reciprocal connections known to exist between the ACC and widespread
regions of frontal association cortex [4]. Though alternative
®re of the two regions may be operated by subcortical structures as thalamus, it is likely to be closely related to the
ongoing cortico-ACC feedback presumably required for
the recruitment of neuronal subpopulations into a coherent
distributional network. The control of cortico-ACC activities may contribute to suppress unnecessary regions
performing the engaging task and to activate more necessary
regions performing it. This brain activity for `positive
suppression' control probably results in an increasing
blood ¯ow in the regions of prefrontal medial cortex including ACC in the PET studies [1,2] and makes the state like as
concentration of attention in the subject.
For the understanding of the mechanisms of Fmu that is
responsible for attention in human brain, further information
on spontaneous activity of neurons in deeper brain parts will
be needed.
[1] Barch, D.M., Braver, T.S., Nystrom, L.E., Forman, S.D., Noll,
D.C. and Cohen, J.D., Dissociating working memory from
task dif®culty in human prefrontal cortex. Neuropsychologia, 35 (1997) 1373±1380.
[2] Davis, K.D., Taylor, S.J., Crawley, A.R., Wood, M.L. and
Mikulis, D.J., Functional MRI of pain- and attention-related
activations in the human cingulate cortex. J. Neurophysiol.,
77 (6) (1997) 3370±3380.
[3] Gevins, A., Smith, M.E., McEvoy, L. and Yu, D., High-resolution EEG mapping of cortical activation related to working
memory: effects of task dif®culty, type of processing, and
practice. Cereb. Cortex, 7 (1997) 374±385.
[4] Goldman-Rakic, P., Circuitry of primate prefrontal cortex
and regulation of behavior by representational memory.
In F. Blum (Ed.), Handbook of Physiology, the Nervous
System-Higher Functions of the Brain, Vol. 1, American
Physiology Association, 1987, pp. 373±417.
[5] HaÈmaÈlaÈinen, M.S., Functional localization based on
measurements with a whole-head magnetometer system.
Brain Topogr., 7 (1995) 283±289.
[6] Hazneder, M.M., Buchsbaum, M.S., Luu, C., Hazlett, E.A.,
Siegel, B.V., Lohr, J., Wu, J., Haier, R.J. and Bunneu Jr.,
W.E., Decreased anterior cingulate gyrus metabolic rate in
schizophrenia. Am. J. Psychiatry, 154 (1997) 682±684.
[7] Iramina, K., Ueno, S. and Matsuoka, S., MEG and EEG topography of frontal midline theta rhythm and source localization. Brain Topogr., 8 (1996) 329±331.
[8] Ishihara, T. and Yoshii, N., Multivariate analytic study of
EEG and mental activity in juvenile delinquents. Electroenceph. clin. Neurophysiol., 33 (1972) 71±80.
[9] Ishii, R., Shinosaki, K., Ukai, S., Inoue, T., Ishihara, T., Yoshimine, T., Hirabuki, N., Asada, H., Kihara, T., Robinson, S.E.
and Takeda, M., Medial prefrontal cortex generates frontal
midline theta rhythm. NeuroReport, 10 (1999) 675±679.
[10] Pandya, D.N. and Seltzer, B., Association areas of cerebral
cortex. Trends Neurosci., 5 (November) (1982) 386±390.
[11] Posner, M. and Rothbart, M.K., Attentional mechanisms
and conscious experience. In A.D. Milner and M.D. Rugg
(Eds.), The Neuropsychology of Consciousness, American
Press, San Diego, CA, 1992, pp. 91±111.
[12] Sasaki, K., Nambu, A., Tsujimoto, T., Matsuzaki, R., Kyuhou,
S. and Gemba, H., Studies on integrative functions of the
human frontal association cortex with MEG. Cogn. Brain
Res., 5 (1996) 165±174.
[13] Siegel Jr., B.V., Buchsbaum, M.S., Bunney Jr, W.E.,
Gottschalk, L.A., Hair, R.J., Lohr, J.B., Lottenberg, S., Naja®,
A., Nuechterlein, K.H., Potkin, S.G. and Wu, J.C., Corticalstriatal-thalamic circuits and brain glucose metabolic activity in 70 unmedicated male schizophrenic. Am. J. Psychiatry, 150 (1993) 1325±1336.
[14] Yamaguchi, Y., Frontal midline theta activity. In N. Yamaguchi and K. Fujisawa (Eds.), Recent Advances in EEG and
EMG Data Processing, Elsevier, Amsterdam, 1981, pp. 391±
396.

Download Report