Pii: s0304-3940(00)01429-4

Suppression of transient 40-Hz auditory response by haloperidol suggests modulation of human selective attention Jyrki Ahveninena,b,c,*, Seppo KaÈhkoÈnena,b,d, Hannu Tiitinena,b, Eero Pekkonena,b,c, Juha Huttunenb,c, Seppo Kaakkolac, Risto J. Ilmoniemib, Iiro P. JaÈaÈskelaÈinena,b,e aCognitive Brain Research Unit, Department of Psychology, PO Box 13, FIN-00014 University of Helsinki, Helsinki, Finland bBioMag Laboratory, Medical Engineering Centre, Helsinki University Central Hospital, Hospital District of Helsinki and Uusimaa, cDepartment of Neurology, University of Helsinki, Helsinki, Finland dDepartment of Psychiatry, Helsinki University Central Hospital, Hospital District of Helsinki and Uusimaa, Helsinki, Finland eMassachusetts General Hospital-NMR Center, Harvard Medical School, Charlestown, Massachusetts, USA Received 14 June 2000; received in revised form 7 August 2000; accepted 8 August 2000 Cognitive processes including selective attention may depend on synchronous activity of neurons at the gamma-band (around 40 Hz). To determine the effect of neuroleptic challenge on transient auditory evoked 40-Hz response, simulta- neous measurement of 122-channel magnetoencephalogram (MEG) and 64-channel electroencephalogram (EEG) was used. Either 2 mg of dopamine D2-receptor antagonist haloperidol or a placebo was administered orally to 11 healthy subjects in a double-blind randomized crossover design in two separate sessions. The subjects attended to tones presented to one ear and ignored those presented to the other ear. Haloperidol signi®cantly suppressed the transient 40-Hz electric response to the attended stimuli, while no signi®cant effect was observed in the electric responses to the unattended tones or in the magnetic responses. The present result suggests that dopamine D2 receptors modulate selective attention. q 2000 Published by Elsevier Science Ireland Ltd.
Keywords: 40-Hz response; Auditory; D2 receptors; Dopamine; Event-related potentials; Magnetoencephalography; Electroencephalo- Selective attention refers to the ability to pick relevant the electrically and magnetically observed responses have stimuli embedded in a vast amount of information. The been shown to be synchronized at the gamma band [13].
neurochemical basis of this fundamental function remains EEG studies indicate that the transient 40-Hz response is to be elucidated. Previous studies have indicated the dopa- enhanced by selective attention [18] and gradually attenu- mine system may be associated with higher-order atten- ated after long-term stimulation due to lessened vigilance tional processes such as mental set-shifting or working [10]. It is produced by a separate source than the 40-Hz memory [4]. In addition, selective-attention de®cits have steady-state response [13], postulated to represent summa- been found in disorders of dopamine system, including tion of overlapping middle-latency auditory evoked responses [6]. Thus, it might represent an essential compo- Higher-order cognitive processes, such as selective atten- nent of auditory attention and perceptual processing.
tion, may be based on synchronization of neural discharges The transient auditory 40-Hz response is attenuated by in the gamma-band (around 40 Hz) [5,18]. In the human g-aminobutyric acidA (GABAA)-agonist temazepine [8], auditory system, a transient 40-Hz response can be detected and modulated by cholinergic muscarine receptors [2].
with electroencephalography (EEG) and magnetoencepha- The dopaminergic modulation of transient 40-Hz auditory lography (MEG) [13]. After 20±130 ms from stimulus onset, responses has not yet been studied. However, other event- related potential (ERP) components related to selective * Corresponding author. Tel: 1358-9-191-23761; fax 1358-9- attention have been shown to be affected by dopamine D2- receptor antagonists, such as droperidol [15]. Therefore, we E-mail address: jyrki.ahveninen@helsinki.® (J. Ahveninen).
0304-3940/00/$ - see front matter q 2000 Published by Elsevier Science Ireland Ltd.
J. Ahveninen et al. / Neuroscience Letters 292 (2000) 29±32 studied the effects of the dopamine D2-receptor antagonist band-pass ®ltered off-line at 32±48 Hz. The data were haloperidol on the transient auditory 40-Hz response with a simultaneous 122-channel MEG and 64-channel EEG Each two-channel MEG sensor unit measures two inde- pendent magnetic-®eld gradient components, dBz/dx and Two milligrams of haloperidol (2 £ Serenase (1 mg, dBz/dy, the z-axis being normal to the local helmet surface.
Orion-Pharma, Espoo, Finland), suf®cient to affect cogni- The position of the subject's head relative to the MEG tive performance without side effects (e.g. dysphoria, instrument was determined by measuring magnetic ®elds akathisia) [9], or a matching placebo was administered produced by three marker coils attached to the scalp. Before orally to 11 right-handed healthy paid volunteers (aged the measurement, the location of the marker coils in relation 20±28 years; six females) in a double-blind, randomized to cardinal points of the head (nasion, left, and right pre- crossover design, 3±4 h before the recording (to reach auricular points) were determined using an Isotrak 3D-digi- peak effects [12]). The subjects were instructed to avoid tizer (Polhemus, Colchester, VT, USA).
alcohol for at least 48 h, and caffeine and tobacco for The magnetic 40-Hz response amplitudes for both 12 h, before MEG and EEG recordings. The subjects attended and unattended stimuli were quanti®ed as the reported having had no neurological or psychiatric disorders mean amplitude during the latency range of 30±130 ms or using no drugs for 2 weeks before the study. None had from the channel pair showing the highest response been exposed to any class of neuroleptics. The study was amplitude over the left and right temporal areas approved by the National Agency for Medicines, Finland, …uau ˆ ‰…dBz=dx†2 1 …dBz=dy†2Š1=2†. Four MEG variables and Ethics Committee of the Department of Clinical were calculated for each subject in each drug condition as Neurosciences, Helsinki University Central Hospital. Writ- follows: responses ipsilateral to the ear stimulated with (1) ten informed consent was obtained after the procedures had attended or (2) unattended stimuli and responses contralat- been fully explained to the subjects. All experimental eral to the ear stimulated with (3) attended or (4) unattended sessions were carried out between 08:00 and 12:00 h, and stimuli. The dipole source of the gamma response was the sessions were separated by 1 week.
modeled using a spherical head model with a subset of 34 During the recordings, the subject sat in a comfortable channels above each hemisphere. The dipole locations were chair in a magnetically shielded room (Euroshield, Eura, ®tted at consecutive peaks of the magnetic gamma oscilla- Finland) with the head inside a helmet-shaped 122-channel tions (only for responses contralateral to the ear stimulated) whole-head MEG instrument [1] (Neuromag, Helsinki, during 30±130 ms post-stimulus, during which the magnetic Finland). The subjects were presented, dichotically through and electric 40-Hz responses have been observed to be ear pieces and plastic tubes, with standard (700 Hz to the synchronous [13]. The source loci were estimated by calcu- left ear, 1100 Hz to the right ear, P ˆ 0:44 for each), and lating the mean of the dipole-®t x-, y-, and z-coordinates of deviant (770 Hz to the left ear, and 1210 Hz to the right ear, P ˆ 0:06 for each) 50-ms pure tones (with 5-ms rise and fall The amplitudes of the electric 40-Hz response for both times) at 60 dB above the individually-determined subjec- attended and unattended stimuli were quanti®ed as the mean tive hearing threshold. The stimuli were presented asyn- amplitude during the latency range of 30±130 ms analyzed chronously to the left and right ears. The inter-stimulus using a signal-space projection (SSP) [17] calculated from interval (ISI) was random at 210±410 ms. The subjects signals obtained at 12 sites around frontal and vertex elec- were to attend to tones presented to one ear and to silently trodes (F1, Fz, F2, FC1, FCz, FC2, C1, Cz, C2, TP1, CPz, TP2).
count the number of deviants, while ignoring tones Because of insuf®cient signal-to-noise ratio (fewer summa- presented to the other ear. The ear to be attended was coun- tions), the electromagnetic signals elicited by the deviant tones were not analyzed. The EEG data of one subject The magnetic 40-Hz response was measured with a 122- was rejected because of excessive noise.
channel planar gradiometer and the simultaneous mastoid- The effect of haloperidol on amplitude of the 40-Hz referenced electric 40-Hz response with a 64-channel EEG magnetic response over the contralateral and ipsilateral using an electrode cap with 64 Ag/AgCl electrodes. Verti- hemispheres to the attended ear was tested by a 4-way cal and horizontal eye movements were recorded with an (attended ear by drug by attention by hemisphere) repeated electrooculogram (EOG). The recording pass-band was measures ANOVA with contrasts. The effect of haloperidol 0.03±100 Hz for EEG and MEG data and 0.5±30 Hz for on SSP of electric 40-Hz response was analyzed using a 3- the EOG (sampling rate 394 Hz for each). The analysis way (attended ear by drug by attention) ANOVA with period was 800 ms (including a 100 ms pre-stimulus base- line). Epochs containing over 150 mV peak-to-peak Corroborating previous observations [18], the transient changes in the EOG and EEG or over 3-pT/cm MEG 40-Hz electric response was signi®cantly enhanced by changes, as well as the responses to the ®rst few stimuli attention (F…1; 8† ˆ 13:2, P , 0:01) in the placebo condi- of the sequence, were automatically rejected. At least 500 tion. Fig. 1 shows the electric 40-Hz auditory response artefact-free responses for the attended, and 500 for the elicited by the attended tones during the haloperidol and unattended tones, were recorded, averaged and digitally placebo conditions in a representative subject. A signi®cant J. Ahveninen et al. / Neuroscience Letters 292 (2000) 29±32 Fig. 1. Transient electric 40-Hz auditory response elicited by the attended tones presented to the right ear in one subject.
drug main effect (F…1; 8† ˆ 6:6, P , 0:05) and a signi®cant A recent PET study indicated that haloperidol signi®- drug by attention interaction (F…1; 8† ˆ 13:8, P , 0:01) cantly reduces glucose utilization in the frontal and anterior indicated that that the transient 40-Hz auditory electric cingulate cortex [3]. This was interpreted to re¯ect binding response was attenuated by haloperidol (Table 1). The a of haloperidol in the striatum and other D2-rich areas, which priori contrasts indicated that this attenuation by haloperi- results in an increased GABAergic inhibition of the thala- dol was signi®cant for the 40-Hz electric response elicited mus, and ®nally reduces neocortical activity [3]. The by the attended stimuli (F…1; 8† ˆ 21:8, P , 0:01), but no suppression of 40-Hz response to attended stimuli could, signi®cant effect was observed for the response to unat- therefore, re¯ect haloperidol-induced hypoactivity in the tended stimuli (Table 1). According to the ANOVA, the frontal and anterior cingulate cortex, brain regions that are effects caused by haloperidol were very similar in the known to be crucial for selective attention.
subjects attending to the left- and right-ear stimuli.
Despite the EEG results, no signi®cant haloperidol effects The differences in the amplitude of the transient 40-Hz were found in MEG. MEG detects only the activity of audi- magnetic response by haloperidol were not statistically tory-cortical sources oriented tangentially to the temporal signi®cant (Table 2). The dipole-source locations of the tran- scalp surface [11]. In EEG, these sources project to the sient 40-Hz magnetic response were not affected by the drug fronto-central electrodes (used in the present analysis).
or the level of attention. Supporting previous ®ndings [13] However, radial activity within, for instance, the frontal the dipole sources of transient 40-Hz magnetic responses brain regions might additionally contribute to the signal were, on the average, located bilaterally 5 cm lateral to the recorded at the fronto-central EEG leads. As already sagittal and 1 cm anterior to the coronal (determined with left mentioned, haloperidol signi®cantly reduced the activity and right pre-auricular points) mid-line of the head.
in the frontal lobes [3], unfortunately, the contribution of The amplitude of the transient 40-Hz auditory electric these regions on the transient auditory 40-Hz responses is response for the selectively attended tones was signi®cantly not well known. The absence of the dopamine D2-receptors reduced by administration of the dopamine D2-receptor from the human primary auditory cortex [7], the suggested antagonist haloperidol, but no such effect was observed origin of the magnetic 40-Hz response [13], however, might for the response to the unattended stimuli. During the partially explain the absence of signi®cant haloperidol placebo condition, the previously reported [18] enhance- effects in MEG. It has to be also noted that at the cellular ment of 40-Hz response by selective attention was clearly signi®cant. The present results thus suggest that the block- age of dopamine D2 receptors speci®cally abolished the Amplitudes of the 40-Hz electric response (mV) attentional modulation of sources affecting the transient auditory gamma oscillations. This basically supports previous result [15] that the processing negativity (PN), elicited by selective attention, was reduced by the dopamine D2-receptor antagonist droperidol (with a slightly different J. Ahveninen et al. / Neuroscience Letters 292 (2000) 29±32 Amplitudes of the 40-Hz magnetic response (fT/cm) level, the generators of gamma oscillations might be rather [6] Galambos, R., Makeig, S. and Talmachoff, P.J., A 40-Hz complex, and that some of their aspects (even the tangential) auditory potential recorded from the human scalp, Proc.
generating clear EEG signal might produce rather weak Natl. Acad. Sci. USA, 78 (1981) 2643±2647.
[7] Goldsmith, S.K. and Joyce, J.N., Dopamine D magnetic ®eld on the scalp surface [16]. One might spec- organized in bands in normal human temporal cortex, ulate that these generator differences, together with the tentative radial sources, might account for the present [8] JaÈaÈskelaÈinen, I.P., Hirvonen, J., Saher, M., Pekkonen, E., discrepancies in the EEG and MEG results. However, Sillanaukee, P., NaÈaÈtaÈnen, R. and Tiitinen, H., Benzodiaze- further studies on differences between magnetic and electric pine temazepam suppresses the transient auditory 40-Hz response amplitude in humans, Neurosci. Lett., 268 (1999) Finally, haloperidol is known to bind with high af®nity to [9] King, D.J., Psychomotor impairment and cognitive distur- sigma receptors, and with lower af®nity to adrenergic, sero- bances induced by neuroleptics, Acta Psychiatr. Scand., tonergic, and GABAergic sites [14,20]. Thus, the present results might also re¯ect the direct binding of haloperidol [10] May, P., Tiitinen, H., Sinkkonen, J. and NaÈaÈtaÈnen, R., Long- term stimulation attenuates the transient 40-Hz response, at non-dopaminergic sites in D2-poor areas of cerebral [11] NaÈaÈtaÈnen, R., Ilmoniemi, R.J. and Alho, K., Magnetoence- In conclusion, the suppression of transient auditory- phalography in studies of human cognitive brain function, evoked 40-Hz response by haloperidol suggests the modu- Trends Neurosci., 17 (1994) 389±395.
lation of auditory selective attention by dopamine D [12] NordstoÈm, A.L., Farde, L. and Halldin, C., Time course of D2- dopamine receptor occupancy examined by PET after tors. The present ®nding thus enhances our understanding of single oral doses of haloperidol, Psychopharmacology, the role of neurotransmitter systems in human cognition.
[13] Pantev, C., Elbert, T., Makeig, S., Hampson, S., Eulitz, C. and This work was supported by the Finnish Cultural Founda- Hoke, M., Relationship of transient and steady-state audi- tion, Helsinki University Central Hospital Research Funds, tory evoked ®elds, Electroencephalogr. clin. Neurophysiol., the Ella and Georg Ehrnrooth's Foundation, and the Acad- [14] Richelson, E. and Nelson, A., Antagonism by neuroleptics emy of Finland. We thank Ms Suvi HeikkilaÈ and Mr Teemu of neurotransmitter receptors of normal human brain in Peltonen for their assistance in carrying out the experiments.
vitro, Eur. J. Pharmacol., 103 (1984) 197±204.
[15] Shelley, A.M., Catts, S.V., Ward, P.B., Andrews, S., Mitch- ell, P., Michie, P. and McConaghy, N., The effect of [1] Ahonen, A., HaÈmaÈlaÈinen, M., Kajola, M., Knuutila, J., Laine, decreased catecholamine transmission on ERP indices of P., Lounasmaa, O., Parkkonen, L., Simola, J. and Tesche, C., selective attention, Neuropsychopharmacology, 16 (1997) 122-Channel SQUID instrument for investigating the magnetic signals from the human brain, Physica Scripta, [16] Tallon-Baudry, C., Bertrand,O. and Pernier,J., A ring-shaped distribution of dipoles as a source model of induced gamma- [2] Ahveninen, J., Tiitinen, H., Hirvonen, J., Pekkonen, E., band activity, Clin. Neurophysiol., 110 (1999) 660±665.
Huttunen, J., Kaakkola, S. and JaÈaÈskelaÈinen, I.P., Scopola- [17] Tesche, C.D., Uusitalo, M.A., Ilmoniemi, R.J., Huotilainen, mine augments transient auditory 40-Hz magnetic M., Kajola, M. and Salonen, O., Signal-space projections of response in humans, Neurosci. Lett., 277 (1999) 115±118.
MEG data characterize both distributed and well-localized [3] Bartlett, E.J., Brodie, J.D., Simkowitz, P., Dewey, S.L., Rusi- neuronal sources, Electroenceph. clin. Neurophysiol., 95 nek, H., Wolf, A.P., Fowler, J.S., Volkow, N.D., Smith, G.
Wolkin, A. and Cancro, R., Effects of haloperidol challenge [18] Tiitinen, H., Sinkkonen, J., Reinikainen, K., Alho, K., Lavikai- on regional cerebral glucose utilization in normal human nen, J. and NaÈaÈtaÈnen, R., Selective attention enhances the subjects, Am. J. Psychiatry, 151 (1994) 681±686.
auditory 40-Hz transient response in humans, Nature, 364 [4] Coull, J.T., Neural correlates of attention and arousal: insights from electrophysiology, functional neuroimaging [19] Vieregge, P., Verleger, R., Wascher, E., Stuven, F. and and psychopharmacology, Prog. Neurobiol., 55 (1998) 343± Kompf, D., Auditory selective attention is impaired in Parkinson's disease±event-related evidence from EEG [5] Engel, A.K., Konig, P., Kreiter, A.K., Schillen, T.B. and potentials, Brain Res. Cogn. Brain Res., 2 (1994) 117±129.
Singer, W., Temporal coding in the visual cortex: new vistas [20] Weissman, A.D., Su, T.P., Hedreen, J.C. and London, E.D., on integration in the nervous system, Trends Neurosci., 15 Sigma receptors in post-mortem human brains, J. Pharma-

Source: http://becs.tkk.fi/bml/pdf/ahveninen_et_al_2000d.pdf