involuntary attention and distractibility as evaluated ... et...task unrelated to the mmn-eliciting...

Audiol Neurootol 2000;5:151–166

Involuntary Attention and Distractibilityas Evaluated with Event-Related BrainPotentials

Carles Esceraa Kimmo Alhob,c Erich Schrögerd Istvan Winklerb,e

aNeurodynamics Laboratory, Department of Psychiatry and Clinical Psychobiology, University of Barcelona, Spain;bCognitive Brain Research Unit, Department of Psychology, University of Helsinki, and cDepartment of Psychology,

University of Tampere, Finland; dInstitute for General Psychology, University of Leipzig, Germany; eInstitute for

Psychology, Hungarian Academy of Sciences, Budapest, Hungary

Prof. Carles EsceraNeurodynamics Laboratory, Department of Psychiatry and Clinical PsychobiologyUniversity of Barcelona, P. Vall d’Hebron 171E–08035 Barcelona, Catalonia (Spain)Tel. +34 93 402 1100, ext. 3047, Fax +34 93 403 4424, E-Mail [email protected]

ABCFax + 41 61 306 12 34E-Mail [email protected]

© 2000 S. Karger AG, Basel1420–3030/00/0054–0151$17.50/0

Accessible online at:www.karger.com/journals/aud

Key WordsPassive attention W Orienting response W Change

detector W Sensory memory W Novelty W Mismatch

negativity (MMN) W N1 W P3a W Audition W Event-related

potentials (ERPs)

AbstractThis article reviews recent event-related brain potential

(ERP) studies of involuntary attention and distractibility

in response to novelty and change in the acoustic envi-

ronment. These studies show that the mismatch negativ-

ity, N1 and P3a ERP components elicited by deviant or

novel sounds in an unattended sequence of repetitive

stimuli index different processes along the course to

involuntary attention switch to distracting stimuli. These

studies used new auditory-auditory and auditory-visual

distraction paradigms, which enable one to assess objec-

tively abnormal distractibility in several clinical patient

groups, such as those suffering from closed-head inju-

ries or chronic alcoholism.Copyright © 2000 S. Karger AG, Basel

Our ability to select a subset from the wealth of infor-mation entering the sensory systems is crucial for mostactivities. The human brain does not have a sufficientcapacity to allow the conscious processing of all stimulusinformation that simultaneously impinges on the varioussenses. Therefore, following an initial survey of the senso-ry input, only a part of the incoming information gainsaccess to consciousness. This basic scheme of Broadbent’s[1958] original theory, i.e. the assumption of a large-capacity system performing the initial processing and asubsequent limited-capacity system for task-related eval-uation and decision making, appears also in current theo-ries of human attention [Cowan, 1995; Näätänen, 1992].Versions of the basic model differ in the extent of the ini-tial survey [early vs. late selection theories) [Treisman,1988; Duncan, 1984] as well as in the method of selection[for the classical treatment of the problem, see Broadbent,1958, and Neisser, 1967]. Although other recent theoriesof attention [Allport, 1993] question the existence of aunitary system of selection, it is nevertheless important totrace back the multitude of different forms of selection totheir neural bases.

In principle, the entry of information to the limited-capacity system is controlled by two types of processes:active selection (focussed attention) and breakthrough of

152 Audiol Neurootol 2000;5:151–166 Escera/Alho/Schröger/Winkler

Fig. 1. ERPs elicited by standard and de-viant tones and novel sounds (a), and thecorresponding difference waves obtained bysubtracting the standard tone ERP from thatto deviant tones or novel sounds (b), whichreveal the N1 enhancement, the MMN andthe P3a. Adapted from Escera et al. [1998].

the unattended (passive attention) [James, 1890]. Thefirst is a top-down process, in which channels of informa-tion are selected or rejected under the direction of the cen-tral mechanisms of behavior control. Certain subfunc-tions of the selection processes might be automatic; forexample, auditory streaming is largely automatic [Breg-man, 1990; Sussman et al., 1999], but, at least in ambig-uous cases, it is affected by top-down control [Sussman etal., 1998]. The second is a bottom-up process that enablesthe conscious evaluation of those potentially importantevents that are not currently selected by the first mecha-nism. Without a good balance between the two, one can-not behave adequately in many situations. If the top-down processes dominate one’s attention, one may notreact to vitally important events occurring outside thefocus of attention. On the other hand, if bottom-up pro-cesses can too easily catch one’s attention, then one’sbehavior appears fragmented, making goal-directed ac-tions less effective.

Brain injuries and neurological and psychiatric disor-ders, such as autism, attention-deficit hyperactivity disor-der, dementia or schizophrenia, often affect the precar-ious balance of attentional control. Symptoms of suchstates are usually described in terms of distractibility. Dis-traction denotes the involuntary redirection of one’s at-tention from some goal-oriented behavior to other aspects

of the environment. Lack of distractibility points to thedominance of the top-down control of attention, whereasincreased distractibility suggests an abnormally lowthreshold for the breakthrough of the unattended (in mostcases irrelevant) information. Despite the obvious impor-tance of distractibility in assessing the patients’ neurologi-cal status, no generally accepted index of it has emerged inclinical practice. Currently available tests of attentionfocus on voluntary or controlled attention, such as differ-ent versions of the Continuous Performance Test, theStroop test or the Paced Auditory Serial Addition Test,among others. However, whereas these tests are suitablefor revealing impairment in the control of relevant stimu-li, they fail to provide an indication of any subtle altera-tion in the passive reorienting of attention. The approachto be discussed in the present paper may lead to a set ofobjective measures of involuntary attention and distracti-bility which are applicable across a large variety of clinicalpopulations.

The converging methods reviewed here are based onevent-related brain potential (ERP) components elicitedby short sounds that in some respect discriminably differfrom the ongoing auditory stimulation (fig. 1). These ERPcomponents index different processes along the course ofdistraction. The mismatch negativity (MMN) and N1

mark instances when the preattentive system detected

a b

Objective Evaluation of Distractibility Audiol Neurootol 2000;5:151–166 153

sounds which carry previously unavailable informationthat may require conscious processing. The P3a compo-nent, in turn, is regarded as a reflection of attentionswitching itself. Finally, the reorienting negativity (RON)[Schröger and Wolff, 1998a] may reflect redirecting atten-tion back to the primary task.

The MMN [Näätänen et al., 1978; for recent reviews,see Näätänen and Alho, 1997; Näätänen and Winkler,1999; Schröger, 1997] is elicited by incoming sounds thatviolate some previously invariant characteristic of anauditory stimulus sequence. The simplest paradigm inwhich MMN can be observed is the auditory oddballparadigm: infrequent sounds (deviants) differing in someacoustic feature from the repetitive (standard) stimuluselicit the MMN (independently of the direction of thesubject’s attention). It has been shown that the processgenerating the MMN component is initiated by a discon-cordance (‘mismatch’) between the incoming sound andsome memory record representing the regularities of theimmediate history of auditory stimulation [for a detaileddiscussion, see Näätänen, 1992; Winkler et al., 1996;Winkler and Czigler, 1998]. In accordance with thisaccount, no MMN is elicited by the repetitive stimulusitself, by infrequent sounds presented alone (i.e. in theabsence of a different repetitive sound) [Näätänen et al.,1989] or by a change at the beginning of a sound sequence[Cowan et al., 1993].

One of the most important features of MMN is that itis elicited even when the subject or patient is engaged in atask unrelated to the MMN-eliciting sounds or the audito-ry stimulation in general [Alho et al., 1992]. Although theMMN amplitude is not fully attention independent [Trejoet al., 1995; Woldorff et al., 1991], MMN elicitation is notaffected by task-related top-down processes [Alain andWoods, 1997; Näätänen et al., 1993; Ritter et al., 1999].From these results one can conclude that incomingsounds are extensively analyzed and their regularities (in-cluding abstract ones [Paavilainen et al., 1995]) are de-tected even when these sounds fall outside the focus ofattention. The MMN amplitude was attenuated whensubjects performed a difficult primary auditory discrimi-nation task with the target sequence presented at a veryfast pace in one ear in parallel with the MMN-elicitingsequence presented in the opposite ear [Näätänen et al.,1993; Woldorff et al., 1991]. One interpretation of theseresults is that they reflect a state of reduced susceptibilityto distraction (stimulus-initiated redirection of attention)when subjects under time pressure need to maximallyfocus their attention on the target sound sequence to copewith the primary task. It is possible that such highly

focussed attentional states diminish the amplitude ofthose MMN subcomponents [Alho, 1995] which are, pre-sumably, involved in initiating a call for attention todeviant sounds (see in detail below).

Since the MMN is elicited when the incoming sounddoes not fit the series of the previous stimuli even whenthese stimuli fall outside the focus of the subject’s atten-tion, it marks one of those situations in which unattendedsounds might carry potentially relevant information. Infact, the MMN-generating process can be regarded as aninformation filter: it is activated only when a sound couldnot be (preattentively) ‘predicted’ from the precedingstimulus sequence, i.e. when it carries new information.Therefore, one might suspect that the MMN-generatingprocess could be involved in the passive (bottom-up)directing of attention [Alho et al., 1994; Knight, 1991;Näätänen and Michie, 1979; Schröger, 1996]. Indeed, theautonomic nervous system responses associated with in-voluntary attention switching (heart rate deceleration andthe skin conductance response) tend to be elicited by thesame stimulus events that also elicit the MMN [Lyytinenand Näätänen, 1987; Lyytinen et al., 1992]. Moreover,the MMN is usually followed by the P3a component, anERP sign of attention switching [Squires et al., 1975] (seebelow). On this basis, Näätänen [1990] proposed that theprocess reflected by the MMN can initiate redirecting offocussed attention to infrequent changes in the auditorybackground. As will be reviewed below, converging evi-dence shows that, although the primary generators of theMMN are located in the auditory cortex, there are alsofrontal sources contributing to the observable MMN re-sponse. The existence of frontal MMN generators sup-ports the assumed role of the MMN in calling for atten-tion [Öhman, 1979] because of the involvement of thefrontal lobes in the control of the direction of attention[Fuster, 1989; Stuss and Benson, 1986]. The neural repre-sentations involved in the MMN-generating process [for areview, see Näätänen and Winkler, 1999] quickly adapt tochanges in the auditory regularities. It appears that theMMN-generating process is involved in maintainingthese representations by providing an error signal when-ever the incoming stimulus differs from what could beextrapolated from the model of the auditory environment[Näätänen and Winkler, 1999; Winkler and Czigler, 1998;Winkler et al., 1996]. This function of the MMN is com-patible with its above-described role in passive attention.

A change in a regular sound sequence is not the onlysituation in which the incoming stimulus may signal somepotentially important change in the auditory environ-ment. An unattended sound appearing after a relatively


Fig. 2. The location of the sources determined for the MEG counter-parts of MMN (white circles) and P3a (black circles) activated in thesupratemporal auditory cortex of an individual subject by a frequen-cy change occurring in a repetitive tone while the subject concen-trated on watching a silent movie. The source locations are projectedon a horizontal MRI slice tilted along the Sylvian fissure. Adaptedfrom Alho et al. [1998b].

long silent period, or one having largely different acousticfeatures from those generally encountered in a given envi-ronment (a novel sound), is also likely to intrude to one’sconscious experience. These sounds are also known toelicit a large N1 wave. The auditory N1 is an obligatoryERP response which is sensitive to refractoriness effects[Giard et al., 1994; Hari et al., 1982; Näätänen and Pic-ton, 1987; Näätänen and Winkler, 1999]. A repeatingsound presented at a high rate elicits a small N1, whereasauditory events of salient features elicit N1 waves of a con-siderably higher amplitude. Näätänen’s [1990] model sug-gests that the auditory N1 might reflect a process involvedin redirecting the focus of attention to the onsets of any,especially new, events in the auditory background. Thissuggestion receives support from findings showing thatnovel sounds which elicit a large N1 also elicit a P3a (seebelow), indicating their potential for attention catching[Woods, 1990].

P3a is regarded as an ERP sign of attention switching. Ithas been distinguished from P3b (P300) [for a review, seeDonchin and Coles, 1988] by the different variablesaffecting its elicitation, its somewhat shorter peak latencyand its different scalp topography [Ford et al., 1976;Squires et al., 1975; for reviews, see Knight and Scabini,

1998; Woods, 1990]. Whereas P3b is elicited by relevantinfrequent stimuli in task situations, P3a can be elicited byrare or unusual stimuli even when this stimulus sequenceis not attended. Whereas the P3b is of maximal amplitudeover parietal scalp areas, the scalp distribution of the P3a

has a frontocentral maximum. Findings showing pro-longed reaction times (RTs) to target stimuli immediatelyfollowing an irrelevant P3a-eliciting novel sound supportthe notion that P3a reflects an attention switch to thesesounds [Grillon et al., 1990; Woods, 1992]. As will bereviewed below, several brain structures including theprefrontal, temporal and parietal cortices, as well as thehippocampus and the parahippocampal and the cingulategyri, are involved in generating the P3a.

Finally, the RON, recently discovered by Schröger andWolff [1998a], might index another important stage alongthe full course of brain events associated with distraction:the return of focussed attention to the primary task afterinvoluntary attention switch away from this task. Thiscomponent will also be discussed later in the presentreview.

Cerebral Generators of MMN and P3a

The MMN scalp distribution shows an amplitude max-imum over the frontocentral scalp. This distribution isexplained by bilateral generator sources in the auditorycortices [Scherg et al., 1989; Giard et al., 1995]. Thisexplanation is supported by source modeling of the mag-netoencephalographic (MEG) counterpart of the MMN,the MMNm, which has revealed bilateral MMNm sourcesin the auditory cortex on the superior plane of the tempo-ral lobe (fig. 2) about 1 cm anterior to the source of theN1m (the MEG counterpart of N1), at least for the MMN/MMNm elicited by frequency changes [Alho et al., 1996,1998b; Csépe et al., 1992; Hari et al., 1992; Huotilainen etal., 1993; Levänen et al., 1993, 1996; Levänen and Sams,1997; Sams et al., 1991; Tiitinen et al., 1993; for a review,see Alho, 1995].

Interestingly, MMN/MMNm responses to changes indifferent physical features appear to be generated by atleast partly different neuronal populations in the auditorycortex, as suggested by differences in MMN scalp distri-butions for frequency, intensity and duration changesobserved by Paavilainen et al. [1991] and later confirmedwith dipole source modeling by Giard et al. [1995]. Con-sistently with this, Levänen et al. [1996] have observedMMNm location differences for changes in tone frequen-cy, duration and interstimulus interval. Thus, if we as-


sume that a sensory-memory trace for some auditory attri-bute is located where the MMN is generated by a changein this attribute [Näätänen, 1992], then these results indi-cate that different neuronal populations are involved insensory-memory representations for sound frequency, in-tensity, duration and interstimulus interval. This propo-sal may also be extended to sensory-memory traces forsimple versus more complex sounds, since Alho et al.[1996] observed that MMNm dipoles for frequencychanges occurring in one tone of complex sounds (a 5-tonechord or melody) are located in the auditory cortex about1 cm medial to the MMNm dipole for a similar frequencychange in a simple tone. Moreover, whereas MMN andMMNm responses to changes in pure sinusoidal tones orin more complex nonphonetic auditory stimuli are elicit-ed with larger amplitudes in the right than left hemisphere[Levänen et al., 1996; Paavilainen et al., 1991; Rinne etal., 1999a; Tervaniemi et al., 1999a], MMN and MMNmresponses to phoneme changes usually get a stronger con-tribution from the left than right auditory cortex [Alho etal., 1998a; Näätänen et al., 1997; Rinne et al., 1999a].Furthermore, the MMN source models of Rinne et al.[1999a] suggest that the neuronal population activated byphoneme change is located in the superior temporal cor-tex posteriorly (closer to Wernicke’s area) to the neuronalpopulation activated by changes in nonphonetic sounds.

The generation of MMN in the auditory cortex wasalso indicated by intracranial recordings in animals[Csépe et al., 1987; Javitt et al., 1996; Kraus et al., 1994]and humans [Halgren et al., 1995a; Kropotov et al., 1995;Liasis et al., 1999]. However, intracranial recordings inhumans are performed in neurological, usually epileptic,patients whose brain responses to auditory stimuli maydiffer from those of healthy individuals. Thus, patientsmay show pathologically attenuated or enhanced re-sponses, or even ones with changed source configurations.The generation of MMN in the auditory cortex is also sug-gested by auditory-cortex activity elicited by changes into-be-ignored sound sequences recorded with positronemission tomography (PET) [Tervaniemi et al., 1999b],functional magnetic resonance imaging (fMRI) [Celsis etal., 1999; Opitz et al., 1999] and event-related optical sig-nals [Rinne et al., 1999b]. Moreover, the MMN was atten-uated in patients with temporal-cortex lesions [Alain etal., 1998]. Interestingly, such lesions may affect differ-ently MMNs to different types of sound changes: in twopatients with a lesion of the left posterior-temporal cortexand a speech comprehension deficit, Aaltonen et al.[1993] found no MMN to a phonetic change that elicits anMMN in healthy subjects, whereas the MMN to a change

in tone frequency appeared to be intact. This finding isconsistent with the proposal discussed above that differ-ent neuronal populations generate MMNs to phonemechanges and changes in nonphonetic sounds.

The existence of MMN generators in the left and rightauditory cortices was also implicated by scalp currentdensity (SCD) maps calculated from MMN scalp poten-tial distribution maps [Deouell et al., 1998; Giard et al.,1990; Serra et al., 1998]. However, these SCD maps alsosuggest that the MMN gets an additional contributionfrom the prefrontal cortex (fig. 3). A contribution of pre-frontal neurons to MMN is also suggested by hemody-namic activity caused by stimulus changes and recordedwith PET [Tervaniemi et al., 1999b] and fMRI [Celsis etal., 1999]. This frontal-MMN generator is also supportedby the chaos analysis of the EEG following MMN-elicitingdeviant sounds [Molnar et al., 1995], by MMN attenua-tion in patients with lesions of the dorsolateral prefrontalcortex [Alain et al., 1998; Alho et al., 1994] and by theMEG recordings of Levänen et al. [1996] who reported insome of their subjects an additional frontal source con-tributing to MMNm. The dorsolateral prefrontal cortexhas an important role in the control of the direction ofauditory attention [Alho et al., 1999; Knight et al., 1981].Therefore, the prefrontal MMN subcomponent might beassociated with the initiation of an involuntary attentionswitch towards a change in the acoustic environment[Giard et al., 1990; Näätänen, 1992]. However, an openquestion in the course of involuntary attention switchingis the temporal dynamics of the activation of MMNsources and its relation to behavioral distraction. Accord-ing to the current theoretical background, the supratem-poral MMN subcomponent may be related to the memoryrepresentation of the auditory regularity involved inMMN generation (see above), whereas the frontal MMNsource might generate the neuroelectric signal leading tothe attention-switching response [Giard et al., 1990; Nää-tänen and Michie, 1979]. Consequently, the activation ofthe supratemporal MMN subcomponent should precedein time the activation of the frontal subcomponent,although no concluding evidence has yet been obtained.Very recently, Rinne et al. [submitted] indeed found intheir MMN source current analysis that the right frontalMMN source was activated on average 16 ms later thanthe temporal one, whereas an opposite tendency wasfound by Escera et al. [in preparation], whose SCD analy-sis of the MMN scalp distribution suggested that the rightfrontal MMN source was activated about 20 ms before thesupratemporal one. Escera et al. also observed a positivecorrelation between the activation of the left temporal


3

4


MMN source and behavioral distraction, quantified asthe increased number of errors in a visual discriminationtask when subjects had to respond to visual stimuli follow-ing deviant, MMN-eliciting tones compared with theresponses to stimuli following standard tones (for a de-tailed description of this distraction paradigm, see thenext section). Further studies need to be conducted, how-ever, to clarify the temporal course of the activation of thedifferent MMN sources.

An additional contribution to MMN might originatefrom the parietal lobes, as hemodynamic recordings alsorevealed parietal-cortex activity in response to auditorystimulus changes [Celsis et al., 1999]. Furthermore, Le-vänen et al. [1996] found, in addition to the bilateral audi-tory-cortex MMNm sources, also a right parietal MMNmsource. Perhaps the parietal activity elicited by auditorystimulus changes is associated with the activation of theparietal mechanisms involved in directing spatial audito-ry attention [Alho et al., 1999; Heilman and Valenstein,1972; Mesulam, 1990; Posner et al., 1982].

The positive P3a ERP deflection, with its amplitudemaximum over the central and frontal scalp areas, oftenfollows MMN even to small changes in to-be-ignoredauditory stimulus sequences. Further, a large P3a is elicit-ed by wide changes, e.g. by complex ‘novel’ sounds occur-ring among repetitive tone pips [Woods, 1990]. This P3a

response might indicate the actual orienting of attentionto an MMN-eliciting sound change occurring outside thecurrent focus of attention [Sams et al., 1985; Escera et al.,1998]. Recently, Alho et al. [1998b] have located thesource of the MEG counterpart of the P3a response todeviant tones and novel sounds in the auditory cortex inthe vicinity of the supratemporal MMNm source (fig. 2).

This finding accords with intracranial recordings of P3a

activity in the superior temporal cortex [Halgren et al.,1995a; Kropotov et al., 1995]. Escera et al. [1998] foundthat the early portion of P3a has its maximal amplitudeover the central midline areas, whereas the later P3a por-tion has a more frontal scalp distribution (fig. 4). Judgingfrom its relatively short latency, the auditory-cortex P3a

observed in the MEG study of Alho et al. [1998b] appearsto explain the earlier part of the (auditory) P3a, whereasthe later, more frontal P3a portion might be generated inthe prefrontal cortex, as suggested by P3a recordingsdirectly from the human prefrontal cortex [Baudena et al.,1995], as well as by the P3a attenuation in patients withdorsolateral prefrontal lesions [Knight, 1984]. However,several additional areas also appear to be involved, as P3a

activity has been recorded intracranially also from theparietal cortex, parahippocampal gyrus and anterior cin-gulate gyrus [Alain et al., 1989; Baudena et al., 1995;Halgren et al., 1995a, b; Kropotov et al., 1995], and aslesions of the temporoparietal junction and the posteriorhippocampal region also attenuate the P3a amplitude[Knight, 1996; Knight et al., 1989]. Moreover, sourcemodeling of the scalp-recorded P3a to novel sounds sug-gests the existence of P3a main sources in the frontal andmedial temporal lobes [Mecklinger and Ullsperger,1995].

Distraction Caused by Irrelevant Novel Soundsand Sound Changes

The series of studies to be reviewed in this section pro-vide a wealth of supporting evidence for the participationof the cerebral process involved in the generation of N1,MMN and P3a in the course from involuntary attention tobehavioral distraction. In these studies, behavioral perfor-mance in a primary visual or auditory task was disruptedby unpredictable novel environmental sounds or slightsound changes occurring outside the focus of the subject’sattention.

Distraction during Visual Task PerformanceA recently developed paradigm [Escera et al., 1998]

allows one to study the disruption of visual task perfor-mance caused by irrelevant sound changes and novelsounds. In this paradigm, the subject is instructed to dis-criminate two categories of visual stimuli, e.g. odd andeven numbers, or numbers and letters, presented at a con-stant rate on a computer screen, and to press the corre-sponding response button as fast and accurately as possi-

Fig. 3. Scalp potential (left) and current density (right) maps, show-ing auditory-cortex and prefrontal sources of MMN elicited by fre-quency-deviant tones. Adapted from Escera et al. [in preparation].Fig. 4. Scalp potential distribution of the early (left column) and late(right column) phases of P3a in two different conditions. In the ignorecondition, subjects were instructed to read a book and to ignore theauditory stimulation. In the cueing condition, subjects were in-structed to discriminate visual stimuli presented 300 ms after eachsound and to ignore the auditory stimulation. Note the differentscalp distributions of the two P3a phases. Note also that whereas theearly P3a phase was similar in the two conditions, the late P3a phasewas enhanced in amplitude, particularly over the frontal region in thecueing condition. The central curves show the P3a potential at Fz inthe two conditions; e = Early; l = late; Aa = ignore; AV = cueing.Adapted from Escera et al. [1998].


Fig. 5. An illustration of the auditory-visual distraction paradigm.Subjects are presented with equiprobable odd and even numbers,each preceded by an irrelevant auditory stimulus (interval betweenauditory and visual onsets, 300 ms). Auditory stimuli are randomlyeither standard tones (80%) or distractors (10% deviant tones; 10%novel sounds). Subjects are instructed to press the correspondingresponse button to odd and even numbers and to ignore the auditorystimulation. From Escera et al. [1998].

Fig. 6. The pattern of behavioral results in the auditory-visual dis-traction paradigm. Compared with standard tones (STD), devianttones (DEV) and novel sounds (NOV) prolong the RT to the succes-sive visual targets. Deviant tones also decrease the hit rate in thevisual performance, due to the increase in the number of wrongresponses. When sounds are omitted (V), RTs tend to be slower thanwhen the visual targets are preceded by a standard tone, for thesounds serve as warning stimuli in the auditory-visual condition.Adapted from Escera et al. [1998].

ble. An auditory stimulus is delivered shortly before eachvisual stimulus (for instance, 300 ms from auditory onsetto visual onset; fig. 5). In most cases, the auditory stimu-lus is a repetitive, standard tone. In some (10–20%)unpredictably scheduled cases, however, this stimulus isreplaced by a slightly different (10–15% change in fre-quency), deviant tone or by a natural complex sound(novel sound). The responses to the visual stimuli are clas-sified according to the preceding auditory stimulus. Thepattern of behavioral results obtained in this paradigm isillustrated in figure 6. Compared with the responses tovisual stimuli that follow standard tones, visual stimulifollowing deviant tones and novel sounds show RTswhich are about 5 and 20 ms longer, respectively [Alho etal., 1997; Escera et al., 1998; Jääskeläinen et al., 1996;Yago et al., 1999]. Interestingly, whereas the hit rate issimilar after standard tones and novel sounds, devianttones cause a hit rate decrement of about 2%, resultingfrom an increase in the number of erroneous decisions inthe visual discrimination task [Alho et al., 1997; Escera etal., 1998; Serra et al., 1998; Yago et al., 1999].

The effects described above may not reflect truly invol-untary attention, however, as subjects may have beenattending to the irrelevant auditory stimuli, which oc-curred with a 100% probability and a fixed interval beforethe visual stimuli. In this case, the auditory stimuli mayhave acted as warning cues of the occurrence of the task-relevant stimuli, as indicated by the shorter RT when thevisual stimuli were preceded by an auditory stimuluscompared with when they occurred in a series with nosounds [Escera et al., 1998] (fig. 6).

In a subsequent experiment, attention was more effec-tively withdrawn from the auditory stimuli by presentingsimultaneous visual warning cues informing the subject ofwhether a successive task-relevant stimulus would be pre-sented or not. In this experiment, in which the occurrenceof an auditory stimulus did not predict the occurrence of asubsequent task-relevant visual stimulus, making it un-likely that subjects were monitoring the auditory stream,similar behavioral distraction effects were observed [Alhoet al., 1997]. Moreover, deviant tones preceding visualstimuli caused an attenuation of the occipital N1 ERP tothese visual stimuli. Cued visual stimuli are known to elic-it enhanced N1 responses due to enhanced attention to


these stimuli caused by the cue [Mangun and Hillyard,1991]. Therefore the attenuation of the N1 to a cued task-relevant visual stimulus caused by a preceding devianttone appears to indicate that the involuntary switching ofthe subject’s attention to the deviant tone interfered withthe early, attentive visual processing [Alho et al., 1997].These results strongly support the involuntary nature ofthe attentional mechanisms involved in the observed dis-traction effects.

In the auditory-visual distraction paradigm, the ERPselicited by the auditory-visual stimulus pairs in which thetone is deviant show an MMN followed by a small P3a

[Alho et al., 1997; Escera et al., 1998; Yago et al., 1999]. Inpairs in which the auditory stimulus is a novel sound, inturn, the ERPs show an enhanced N1, compared with thatelicited in pairs containing the standard tone, and a large,biphasic P3a potential [Escera et al., 1998; Yago et al.,1999] (fig. 7). These ERP results, in combination with thebehavioral results discussed above, led the authors to pro-pose that two different attention-switching mechanismsare involved in the distraction observed during visual per-formance in the auditory-visual distraction paradigm.Novel sounds elicited an enhanced N1 wave, probablycaused by an enhanced supratemporal N1 [Näätänen andPicton, 1987] or some other N1 component such as thefrontal N1 [Giard et al., 1994], and MMN [Alho et al.,1998b] (fig. 7). This suggests that the attention-switchingsignal is probably triggered by a combined response of thetransient-detector mechanism reflected by the N1 [Nää-tänen, 1990, 1992; Näätänen and Picton, 1987] and thestimulus change detector mechanism reflected by theMMN [Näätänen, 1990, 1992], resulting in an effectiveattention switch as indicated by the subsequent large P3a

wave and the clearly delayed RT to the following visualstimulus. For deviant tones, in turn, a distinct MMN, fol-lowed by a small P3a wave, was observed (fig. 7), suggest-ing an attention switch initiated by the stimulus changedetector mechanism generating the MMN [Schröger,1996]. This results in a less effective attention switch, asindicated by the smaller P3a and the more modest behav-ioral effect compared with that caused by the novelsounds.

As already mentioned, the large, biphasic P3a potentialto novel sounds of Escera et al. [1998] had an early cen-trally dominant scalp distribution with polarity reversalin posterior and inferior electrodes, whereas the late P3a

displayed a frontal scalp maximum (fig. 4). The scalp dis-tribution of these two P3a phases is in agreement withstudies suggesting a critical role of the temporal-parietaljunction in the generation of the P3a to auditory [Halgren

Fig. 7. The component structure of ERPs to deviant tones and novelsounds, as revealed by the ERP difference waves (deviant-standard,novel-standard) at Fz and the right mastoid (RM) referred to an elec-trode at the nose. Deviant tones elicit MMN and a small P3a, whereasnovel sounds elicit an overlapping N1 enhancement and MMN [Alhoet al., 1998b], and a large biphasic P3a potential. MMN and the earlypart of the P3a invert in polarity at the right mastoid below the audi-tory cortex, suggesting an auditory-cortex contribution to theseresponses; e = Early; l = late. Adapted from Escera et al. [1998].

et al., 1995a,b; Knight et al., 1989], somatosensory [Ya-maguchi and Knight, 1991, 1992] and visual novel stimu-li [Knight, 1991, 1997], and also with results suggesting aprefrontal contribution to the P3a [Baudena et al., 1995;Friedman and Simpson, 1994; Friedman et al., 1993;Knight, 1984; Mecklinger and Ullsperger, 1995]. It is pos-sible that the two phases of P3a reflect two different pro-cesses in the course of involuntary attention switching, asthey have different cerebral sources and react differential-ly to attentional manipulations. Interestingly, Escera et al.[1998] found that the early P3a subcomponent was of sim-ilar amplitude irrespective of whether their subjects werereading a book or performing a visual discrimination task,whereas the late P3a subcomponent was enhanced inamplitude when subjects were monitoring the auditorystimuli during the visual performance (fig. 4). As dis-cussed above, the task-irrelevant auditory stimuli mayhave been to some extent attended as they cued the occur-rence of a successive visual target (fig. 6). A similar atten-


Fig. 8. An illustration of the auditory-audi-tory distraction paradigm. Subjects are pre-sented with equiprobable short- and long-duration tones and instructed to discrimi-nate them. Infrequent task-irrelevant fre-quency deviations in these tones serve as dis-tractors. Adapted from Schröger and Wolff[1998b].

tional modulation of the P3a elicited by novel soundshas already been reported [Holdstock and Rugg, 1995;Woods, 1992]. Thus, judging from its neural generatorsand insensitiveness to attentional manipulation, the earlyP3a might reflect violation of a multimodal model of theexternal world maintained in the temporal-parietal asso-ciation cortex [Yamaguchi and Knight, 1991], whereasthe late P3a may be more closely related to the actualorienting of attention, as indicated by its attentionaldependence and prefrontal generators.

Distraction during Auditory Task PerformanceThe processing of task-relevant auditory information

can be distracted by irregularities in task-irrelevant as-pects of the acoustic input. In one type of these auditory-auditory distraction paradigms, subjects have to performa same-different judgement on two tones (S1 and S2) sep-arated by an interval of several seconds (delayed match-to-sample task). The interval between S1 and S2 is silentor filled with different types of intervening stimuli suchas sinusoidal tones or environmental sounds. The inter-ference in immediate memory (e.g. for tonal pitch)caused by intervening sounds has been studied as a func-tion of several variables such as the type of interveningstimuli, particular brain lesions or age [Chao and Knight,1995, 1997; Deutsch, 1970; Pechmann and Mohr, 1992;Semal et al., 1996]. This paradigm is suited for studyingeffects of task-irrelevant sounds on short-term and work-ing memory. In another type of paradigm, subjects whopresented with dichotic sounds are instructed to attendto the input of one ear and to respond to particular targetsounds occurring in this ear (dichotic selective-attentiontask). The discrimination of targets preceded by irregularor novel sounds in the attended and even in the unat-

tended ear may be deteriorated [Schröger, 1996; Woods,1992; Woods et al., 1993]. This paradigm is suited forstudying distractibility of selective attention. In a thirdtype of paradigm, the distracting and task-relevant as-pects of stimulation appear in the same acoustic event. Inthis paradigm, subjects are to discriminate the durationof tones that are equiprobably of short (200 ms) or long(400 ms) duration. These tones are of the standard fre-quency (e.g. 600 Hz) with a high probability, or of thedeviant frequency (e.g. 650 Hz) with a low probability(fig. 8), this frequency variation having no task relevance.This auditory-auditory distraction paradigm recently de-veloped by Schröger and Wolff [1998a, b] is suited forstudying distraction mediated by a sensory-memory-relat-ed mechanism. We will examine this paradigm in moredetail.

The RT in the duration discrimination task is pro-longed for frequency-deviant tones compared with stan-dard frequency tones [Schröger and Wolff, 1998a, b].Theoretically, this RT effect could be due to ‘costs’ in pro-cessing duration information in deviant trials, but also to‘benefits’ in processing duration in standard trials. Intheir control condition, Schröger and Wolff [1998a]showed that the RT prolongation in deviant trials wasmainly due to distraction caused by task-irrelevant fre-quency deviations, that is, there are genuine costs in theprocessing of duration information in deviant trials. Inthe ERPs, the present distraction effects are reflected bythe MMN and P3a components (fig. 9), which differ fromeach other with respect to their dependence on the alloca-tion of attention to the sounds. When sounds are notattended, the moderate frequency deviations that wereused in this paradigm only elicit an MMN, indicating thepreattentive registration of the deviant sound, but no P3a.


When sounds are in turn attended, a distinct P3a is elicitedby frequency deviations, although the MMN is not aug-mented relative to the MMN in the ignore condition[Schröger et al., submitted; Schröger and Wolff, 1998b].One may assume that an attention switch indicated by P3a

is elicited in this case because task-irrelevant sound infor-mation cannot be easily disregarded when the sound car-ries task-relevant information also (see below). This hy-pothesis is supported by the subjects’ self-report statingthat they often felt distracted from the duration discrimi-nation task by frequency deviations when sounds wereattended, whereas they did not feel distracted by these fre-quency deviations when ignoring the sounds.

Notably, in the present paradigm, P3a is followed bya frontocentrally distributed negativity at the 400- to600-ms range. Since this negativity was confined to condi-tions in which subjects discriminated long sounds fromshort ones but did not occur when frequency deviationswere task relevant nor when sounds were ignored, it wassuggested to reflect the reorienting of attention towardstask-relevant aspects of stimulation following distraction[Schröger and Wolff, 1998b]; therefore, this negativitywas termed RON. In a recent study comparing the presentauditory-auditory distraction paradigm with its visualanalog, a RON was also observed, indicating that thiscomponent is not specific to the auditory modality [Bertiand Schröger, in press]. SCD maps for RON reveal bilat-eral frontal sinks around FC1 and FC2 electrodes, sug-gesting frontal generators, and complex current fields oflower amplitudes over the centroparietal regions [Schrö-ger et al., submitted].

It has been shown in several visual and auditory stud-ies that distractors are more effective if task-relevant anddistracting information is carried by the same object orperceptual group than if distracting information appearsin a different object [Alain and Woods, 1993, 1997; Baylisand Driver, 1992; Bregman and Rudnicky, 1975; Driverand Baylis, 1991; Jones and Macken, 1993; Kramer andJacobsen, 1991]. That is, large distraction effects are to beexpected if a deviation occurs in a task-irrelevant dimen-sion of a task-relevant sound, e.g. in the frequency of thesound of a duration discrimination task. Indeed, theabove-discussed auditory-auditory distraction paradigm,in which distracting and task-relevant events occur in thesame to-be-attended stimuli, yields larger behavioral dis-traction effects than the auditory-visual [Escera et al.,1998] or the other types of auditory-auditory distractionparadigms [Schröger, 1996] in which distractors occurwithin the to-be-ignored stimulus sequence. Importantly,each subject studied in the paradigm with distracting and

Fig. 9. Typical ERP difference waves (calculated by subtractingERPs to tones of standard frequency from those to tones of deviantfrequency) obtained with the auditory-auditory distraction para-digm. During the duration discrimination task, MMN, P3a and RONare elicited. When subjects are instructed to read a book of their ownchoice and to ignore the auditory stimuli, only MMN is elicited.Adapted from Schröger et al. [1999].

target events occurring in the same stimulus showed abehavioral effect of distraction [e.g. Jääskeläinen et al.,1999; Schröger et al., submitted; Schröger and Wolff,1998a,b; Berti and Schröger, 1999]. Moreover, in thisparadigm, the test-retest replicability of the RT prolonga-tion and of the MMN, P3a and RON is rather high, withproduct-moment correlations for those parameters mea-sured in two sessions separated by 25 days being 0.77 forMMN, 0.88 for P3a, 0.81 for RON and 0.90 for the RTprolongation [Schröger et al., submitted].

Due to the large and consistent distraction effects whendistracting stimulus changes occur in a task-irrelevantattribute of the target sounds, considerably smaller stimu-lus changes can be used as distractors in this paradigmthan in the traditional distraction paradigms. In theabove-mentioned dichotic selective-attention type of par-adigm, widely deviant distractors such as dog-barking ortelephone-ringing are required in order to produce clearbehavioral distraction effects when the distractors are pre-sented in the to-be-ignored ear. With widely deviant dis-tractors, it is likely that distraction is triggered by a tran-sient-detector mechanism activated by firing patterns ofneural populations specifically responding to the featuresof the infrequently presented distracting sound, as theseneurons are less refractory than those specifically re-


sponding to the repetitive (standard) acoustic input [Nää-tänen and Picton, 1987]. However, as already reviewed,distraction can also be mediated by the memory-basedmismatch mechanism which can detect irregular eventswithin regular stimulation by comparing the current stim-ulus with the representation of the invariances inherent inthe recent stimulation [Näätänen, 1992; Öhman, 1979;Schröger, 1997; Sokolov, 1975]. In order to study distrac-tion mediated exclusively by this change detector mecha-nism, distractors should then be physically close to non-distractors.

Use of ERPs in Clinical Research of InvoluntaryAttention and Distractibility

The distraction paradigms reviewed in the precedingsection offer valuable objective tools for investigating theneural mechanisms of involuntary attention and their dis-orders. The main advantage of these paradigms is thatthey provide, in a reliable way, behavioral measures ofdistraction and electrophysiological indices of the un-derlying cerebral processes. A limitation, however, is thatpatients are required to perform a highly demanding maintask, precluding the examination of widely impaired sub-jects. In this case, an alternative may consist of instructingthe patients to ignore the auditory stimuli and to concen-trate on reading a book or watching a silent video movie,excluding the behavioral measures. This would yieldMMN and P3a components, although the results may bedifferent in comparison with the active condition, at leastfor P3a (see the previous section). Future research in thisdirection will clarify to what extent a passive version ofthe distraction paradigms reviewed here provides reliableindices of involuntary attention.

In a study of patients with closed-head injuries, Kaipioet al. [1999] reported MMNs to deviant tones similar tothose in control subjects but an enhanced late subcompo-nent of P3a to novel sounds, suggesting a ‘normal’ activa-tion of the change detector mechanism associated withMMN but a stronger attention-switching response to largechanges in the unattended acoustic environment. Consis-tently with this, Polo et al. [1998] reported similar MMNsbut enhanced P3a responses to deviant tones in a group ofchronic abstinent alcoholics. Both studies indicate a func-tional dissociation between the neural processes involvedin MMN and P3a generation: the MMN generator processmay be related to triggering a call for attention [Öhman,1979], and this call may result in a more or less effectiveattention switch reflected by the P3a component. These

results also indicate the utility of recording ERPs (MMNand P3a) to deviant tones and novel sounds in an attemptto identify particular attention deficits in different clinicalpopulations.

In a preliminary report, Polo et al. [1999] found anenhanced P3a to novel sounds in chronic alcoholics, whichwas due to an enhanced left frontal amplitude of the laterP3a phase. This finding might provide the physiologicalsubstrate of the abnormally strong reactivity to irrelevantstimuli as a common problem in alcoholic patients [DeSo-to et al., 1985; Fein et al., 1990; Hansen, 1980] and sug-gests the involvement of certain left prefrontal areas, inaddition to the right prefrontal cortex, in P3a generation.Several functional neuroimaging studies have indicatedthe involvement of the left prefrontal cortex in memoryencoding of novel stimuli [Berns et al., 1997; Tulving etal., 1994]. Thus, in the Polo et al. [1999] study, theobserved enhanced P3a resulting from a left prefrontalcontribution may indicate enhanced memory encoding ofnovel sounds in chronic alcoholics, as suggested by the(small) parietal P3b, a scalp signature of memory updating[Donchin and Coles, 1988], elicited in the patients afterthe P3a.

Increased involuntary attention switch in chronic alco-holics, as indicated by the enhanced P3a to deviant tonesand novel sounds, suggests increased distractibility inalcoholism. Thus, in addition to clinical symptoms, ex-perimental behavioral evidence of increased distractibili-ty in chronic alcoholics should be expected. Ahveninen etal. [submitted], using the auditory-auditory distractionparadigm described above, found prolonged RTs afterdeviant stimuli in abstinent chronic alcoholics. This RTprolongation was larger in alcoholics than in age-, sex- andeducation-matched controls, supporting the hypothesis ofincreased distractibility in alcoholics. Furthermore, theRT prolongation correlated positively with the MMNamplitude (r = 0.7), indicating that the abnormal atten-tional reactivity of chronic alcoholics to irrelevant acous-tic changes was accompanied by an increased response ofthe change detector mechanism reflected by the MMN.

There is currently a lack of instruments capable ofrevealing deficits in the brain’s ability to move attentionfrom a main task to unexpected potentially relevant stim-uli occurring outside the focus of attention. Such aninstrument appears to be provided by the above-discussedparadigms, which may become also a useful tool to prove‘reduced’ attention-switching capabilities. Jääskeläinen etal. [1999], using occasional distracting frequency changesin target tones of a duration discrimination task, foundthat a moderate dose of alcohol (0.3 g/kg, corresponding


to about 0.04% blood alcohol concentration) reduced theRT prolongation observed after the distracting events rel-ative to a placebo condition. Furthermore, similar results,with the auditory-visual distraction paradigm, were ob-tained by Jääskeläinen et al. [1996] who found that duringa mild ethanol intoxication (0.05% blood alcohol concen-tration), the hit rate reduction observed in distractingtrials was significantly smaller than during a placebo con-dition. Taken together, these results suggest that acutealcohol challenge, even at a very small dose, suppressesthe attention-capturing effect of deviant sounds, indicat-ing a detrimental influence of alcohol on the neural mech-anisms of involuntary attention. These detrimental ef-fects may result in damage and subsequent adaptivechanges caused by long-term alcohol consumption, whichmay, in turn, explain the abnormal attention reactivity toirrelevant stimuli observed in chronic alcoholics. It is,however, an open question how these short-term effects ofalcohol intoxication relate to the abnormal attention reac-tivity to irrelevant stimuli observed in chronic alcoholics[Ahveninen et al., submitted; Polo et al., 1998, 1999; seeAhveninen et al., in press].

Another area in which the paradigms reviewed mayprovide insightful information is the study of childrenand adolescents with attention, language or learning dis-abilities. There is a relatively extensive ERP literature inthe area of attention in these children, but mainly fo-cussed on selective and voluntary attention. In childrenwith attention-deficit hyperactivity disorders, some au-thors have suggested a deficit in preferential processing ofattended stimuli as opposed to an inability to block irrele-vant stimuli [Satterfield et al., 1990, 1994]. However,some recent results showing an attenuation of the MMNelicited by deviant tones suggest the dampening of mecha-nisms of involuntary attention [Winsberg et al., 1993;Kemner et al., 1996]. Also in adolescents, Lähteenmäki etal. [in press] found similar MMNs but reduced P3a todeviant tones in survivors to childhood leukemia, suggest-ing that the P3a amplitude might be used to evaluate thedeteriorating effects of prophylactic CNS radiotherapy inchild cancer patients on attention functions, possiblyexplaining their learning disabilities. These results strong-ly encourage the use of the new distraction paradigms infuture research in infants, children and adolescents.

Concluding Remarks

The studies reviewed in the present article shed somelight on the brain mechanisms of involuntary attention

and distractibility, indicating that the MMN, N1, P3a andRON components of the ERP mark different cerebralprocesses in the course of distraction. The MMN and N1

mark those instances when the preattentive system de-tects sounds which may carry new, potentially relevantinformation. A small P3a elicited by slightly deviantsounds and a larger P3a elicited by complex novel soundsprobably reflect an actual switching of attention to unat-tended stimuli. The RON, in turn, might reflect the returnof attention to the primary task after an involuntary atten-tion switch. The above-reviewed studies also suggest acomplex cerebral circuitry, including the supratemporalauditory cortex and the temporoparietal and prefrontalcortices as well as some other brain areas, such as the cin-gulate cortex and the hippocampus, underlying involun-tary attention.

The auditory-visual and auditory-auditory distractionparadigms discussed above have been developed onlyrecently. In these paradigms, the performance in an audi-tory or visual task is disrupted by the occurrence of anirrelevant auditory (distracting) event, i.e. a deviant ornovel sound, preceding task-relevant stimuli. These testsprovide objective measures of distraction, e.g. an increasein RT or a decrease in the hit rate, for task-relevant stimu-li immediately following the distracting events. In addi-tion, the simultaneous recording of ERPs during perfor-mance in these distraction paradigms allows one to testthe functional integrity of the underlying cerebral pro-cesses. Although the distracting effects yielded by theauditory-auditory distraction paradigm are larger thanthose obtained with the auditory-visual distraction para-digm, both tests provide reliable measures of distractibili-ty. These measures can now be, as reviewed above, usedto reveal attentional dysfunctions at the group level inseveral clinical populations and might provide, in thenear future, a new tool for the evaluation of attentionalintegrity in individual patients.

Acknowledgements

This work was supported by BIOMED-2 contract BMH4-CT96-0189-COBRAIN, the Generalitat de Catalunya (ACES98-65/8), theSpanish Ministry of Science and Culture (SAF98-1822-E), the Acad-emy of Finland, the Deutsche Forschungsgemeinschaft (Schr 375/8-1) and the Hungarian National Research Fund (OTKA 022681).This review was written while Carles Escera was in the CognitiveBrain Research Unit, University of Helsinki, supported by a MarieCurie grant (contract BMH4-CT98-5118). The authors would like tothank Heikki Lang and two anonymous referees for their construc-tive comments.


References

Aaltonen O, Tuomainen J, Laine M, Niemi P: Cor-tical differences in tonal versus vowel process-ing as revealed by an ERP component calledmismatch negativity (MMN). Brain Lang1993;44:139–152.

Ahveninen J, Escera C, Polo MD, Grau C, Jääske-läinen I: Acute and chronic effects of alcohol inpreattentive auditory processing as reflected bymismatch negativity. Audiol Neurootol, inpress.

Ahveninen J, Jääskeläinen IP, Pekkonen E, Hall-berg A, Hietanen M, Näätänen R, Schröger E,Sillanaukee P: Increased distractibility by un-expected auditory stimulus changes in alcohol-ics. Submitted.

Alain C, Richer F, Achim A, Saint-Hilaire JM:Human intracerebral potentials associatedwith target, novel and omitted auditory stimu-li. Brain Topogr 1989;1:237–245.

Alain C, Woods DL: Distractor clustering enhancesdetection speed and accuracy during selectivelistening. Percept Psychophys 1993;54:509–514.

Alain C, Woods DL: Attention modulates auditorypattern memory as indexed by event-relatedbrain potentials. Psychophysiology 1997;34:534–546.

Alain C, Woods DL, Knight RT: A distributed cor-tical network for auditory sensory memory inhumans. Brain Res 1998;812:23–37.

Alho K: Cerebral generators of mismatch negativi-ty (MMN) and its magnetic counterpart(MMNm) elicited by sound changes. Ear Hear1995;16:38–51.

Alho K, Connolly JF, Cheour M, Lehtokoski A,Huotilainen M, Virtanen J, Aulanko R, Ilmo-niemi RJ: Hemispheric lateralization in preat-tentive processing of speech sounds. NeurosciLett 1998a;258:9–12.

Alho K, Escera C, Dıaz R, Yago E, Serra JM:Effects of involuntary auditory attention onvisual task performance and brain activity.Neuroreport 1997;8:3233–3237.

Alho K, Medvedev SV, Pakhomov SV, RoudasMS, Tervaniemi M, Reinikainen K, Zeffiro T,Näätänen R: Selective tuning of the left andright auditory cortices during spatially directedattention. Cogn Brain Res 1999;7:335–341.

Alho K, Tervaniemi M, Huotilainen M, Lavikai-nen J, Tiitinen H, Ilmoniemi RJ, Knuutila J,Näätänen R: Processing of complex sounds inthe human auditory cortex revealed by magnet-ic brain responses. Psychophysiology 1996;33:369–375.

Alho K, Winkler I, Escera C, Huotilainen M, Vir-tanen J, Jääskeläinen IP, Pekkonen E, Ilmo-niemi RJ: Processing of novel sounds and fre-quency changes in the human auditory cortex:Magnetoencephalographic recordings. Psycho-physiology 1998b;35:211–224.

Alho K, Woods DL, Algazi A, Knight RT, Näätä-nen R: Lesions of frontal cortex diminish theauditory mismatch negativity. Electroencepha-logr Clin Neurophysiol 1994;91:353–362.

Alho K, Woods DL, Algazi A, Näätänen R: Inter-modal selective attention. II. Effects of atten-tional load on processing of auditory and visualstimuli in central space. ElectroencephalogrClin Neurophysiol 1992;82:356–368.

Allport A: Attention and control: Have we beenasking the wrong questions? A critical review oftwenty-five years; in Meyer DE, Kornblum S(eds): Attention and Performance 14: Synergiesin Experimental Psychology, Artificial Intelli-gence, and Cognitive Neuroscience. Cam-bridge, MIT Press, 1993, pp 183–218.

Baudena P, Halgren E, Heit G, Clarke JM: Intrace-rebral potentials to rare target and distractorauditory and visual stimuli. III. Frontal cortex.Electroencephalogr Clin Neurophysiol 1995;94:251–264.

Baylis GC, Driver J: Visual parsing and responsecompetition: The effect of grouping factors.Percept Psychophys 1992;51:145–162.

Berns GS, Cohen JD, Mintum MA: Brain regionsresponsive to novelty in the absence of aware-ness. Science 1997;276:1272–1275.

Berti S, Schröger E: A comparison of auditory andvisual distraction. J Psychophysiol, in press.

Bregman AS: Auditory Scene Analysis. The percep-tual Organization of Sound. Cambridge, MITPress, 1990.

Bregman AS, Rudnicky AI: Auditory segregation:Stream or streams? J Exp Psychol Hum Percept1975;1:263–267.

Broadbent DE: Perception and Communication.New York, Pergamon Press, 1958.

Celsis P, Boulanouar K, Doyon B, Ranjeva JP, Ber-ry I, Nespoulous JL, Chollet F: DifferentialfMRI responses in the left posterior superiortemporal gyrus and left supramarginal gyrus tohabituation and change detection in syllablesand tones. Neuroimage 1999;9:135–144.

Chao LL, Knight RT: Human prefrontal lesionsincrease distractibility to irrelevant sensory in-puts. Neuroreport 1995;6:1605–1610.

Chao LL, Knight RT: Prefrontal deficits in atten-tion and inhibitory control with aging. CerebCortex 1997;7:63–69.

Cowan N: Attention and Memory: An IntegratedFramework. Oxford, Oxford University Press,1995.

Cowan N, Winkler I, Teder W, Näätänen R: Memoryprerequisites of the mismatch negativity in theauditory event-related potential (ERP). J ExpPsychol Learn Mem Cogn 1993;19:909–921.

Csépe V, Karmos G, Molnar M: Evoked potentialcorrelates of stimulus deviance during wakeful-ness and sleep in cat – Animal model of mis-match negativity. Electroencephalogr ClinNeurophysiol 1987;66:571–578.

Csépe V, Pantev C, Hoke M, Hampson S, Ross B:Evoked magnetic responses of the human audi-tory cortex to minor pitch changes: Localiza-tion of the mismatch field. ElectroencephalogrClin Neurophysiol 1992;84:538–548.

Deoull L, Bentin S, Giard MH: Mismatch negativi-ty in dichotic listening: Evidence for interhemi-spheric differences and multiple generators.Psychophysiology 1998;35:355–365.

DeSoto CB, O’Donnell WE, Allred LJ, Lopes CE:Symptomatology in alcoholics at various stagesof abstinence. Alcoholism. Clin Exp Res 1985;9:505–512.

Deutsch D: Tones and numbers: Specificity ofinterference in immediate memory. Science1970;168:1604–1605.

Donchin E, Coles MGH: Is the P300 component amanifestation of context updating? BehavBrain Sci 1988;11:357–374.

Driver J, Baylis GC: Target-distractor separationand feature integration in visual attention toletters. Acta Psychol 1991;76:101–119.

Duncan J: Selective attention and the organizationof visual information. J Exp Psychol Gen 1984;113:501–517.

Escera C, Alho K, Winkler I, Näätänen R: Neuralmechanisms of involuntary attention switchingto novelty and change in the acoustic environ-ment. J Cogn Neurosci 1998;10:590–604.

Escera C, Serra JM, Giard MH, Yago E, Alho K: Aneural change detector mechanism in the audi-tory system as a driver of involuntary atten-tion. In preparation.

Fein G, Bachman L, Fisher S, Davenport L: Cogni-tive impairments in abstinent alcoholics. WestJ Med 1990;152:531–537.

Ford JM, Roth WT, Kopell BS: Auditory evokedpotentials to unpredictable shifts in pitch. Psy-chophysiology 1976;13:32–39.

Friedman D, Simpson G: ERP amplitude and scalpdistribution to target and novel events: Effectsof temporal order in young, middle-aged andolder adults. Cogn Brain Res 1994;2:49–63.

Friedman D, Simpson G, Hamberger M: Age-relat-ed changes in scalp topography to novel andtarget stimuli. Psychophysiology 1993;30:383–396.

Fuster J: The Prefrontal Cortex: Anatomy, Physiol-ogy, and Neuropsychology of the Frontal Lobe.New York, Raven Press, 1989.

Giard MH, Lavikainen J, Reinikainen K, Perrin F,Bertrand O, Thévenet M, Pernier J, NäätänenR: Separate representation of stimulus frequen-cy, intensity, and duration in auditory sensorymemory. J Cogn Neurosci 1995;7:133–143.

Giard MH, Perrin F, Echallier JF; Thévenet M,Froment JC, Pernier J: Dissociation of tempo-ral and frontal components in the auditory N1wave: A scalp current density and dipole modelanalysis. Electroencephalogr Clin Neurophys-iol 1994;92:238–252.

Giard MH, Perrin F, Pernier J, Bouchet P: Braingenerators implicated in processing of auditorystimulus deviance: A topographic event-relatedpotential study. Psychophysiology 1990;27:627–640.

Grillon C, Courchesne E, Ameli R, Geyer M, BraffDL: Increased distractibility in schizophrenicpatients. Arch Gen Psychiatry 1990;47:171–179.

Halgren E, Baudena P, Clarke JM, Heit G, LiegeoisC, Chauvel P, Musolino A: Intracerebral poten-tials to rare target and distractor auditory andvisual stimuli. I. Superior temporal plane andparietal lobe. Electroencephalogr Clin Neuro-physiol 1995a;94:191–220.


Halgren E, Baudena P, Clarke JM, Heit G, Marin-kovic K, Devaux B, Vignal JP, Biraben A:Intracerebral potentials to rare target and dis-tractor auditory and visual stimuli. II. Medial,lateral, and posterior temporal lobe. Electroen-cephalogr Clin Neurophysiol 1995b;94:229–250.

Hansen L: Treatment of reduced intellectual func-tioning of alcoholics. J Stud Alcohol 1980;41:156–158.

Hari R, Kaila K, Katila T, Tuomisto T, Varpula T:Interstimulus interval dependence of the audi-tory vertex response and its magnetic counter-part: Implications for their neural generation.Electroencephalogr Clin Neurophysiol 1982;54:561–569.

Hari R, Rif J, Tiihonen J, Sams M: Neuromagneticmismatch fields to single and paired tones.Electroencephalogr Clin Neurophysiol 1992;82:152–154.

Heilman KM, Valenstein E: Auditory neglect inman. Arch Neurol 1972;26:32–35.

Holdstock JS, Rugg MD: The effect of attention onthe P300 deflection elicited by novel sounds. JPsychophysiol 1995;9:18–31.

Huotilainen M, Ilmoniemi RJ, Lavikainen J, Tii-tinen H, Alho K, Sinkkonen J, Knuutila J, Nää-tänen R: Interaction between representationsof different features of auditory sensory memo-ry. Neuroreport 1993;4:1279–1281.

Jääskeläinen IP, Alho K, Escera C, Winkler I, Silla-naukee P, Näätänen R: Effects of ethanol andauditory distraction on forced choice reactiontime. Alcohol 1996;13:153–156.

Jääskeläinen IP, Schröger E, Näätänen R: Effects ofethanol on auditory distraction: An ERP andbehavioral study. Psychopharmacology 1999;141:16–21.

James W: The Principles of Psychology, New York,Holt, 1890.

Javitt DC, Steinschneider M, Schroeder CE, Arez-zo JC: Role of cortical N-methyl-D-aspartatereceptors in auditory sensory memory and mis-match negativity generation: Implications forschizophrenia. Proc Natl Acad Sci USA 1996;93:11962–11967.

Jones DM, Macken WJ: Irrelevant tones producean irrelevant speech effect: Implications forphonological coding in working memory. J ExpPsychol Learn Mem Cogn 1993;19:369–381.

Kaipio ML, Alho K, Winkler I, Escera C, Surma-aho O, Näätänen R: Event-related potentialsreveal covert distractibility in closed head in-jured patients. Neuroreport 1999;10:2125–2129.

Kemner C, Verbaten MN, Koelega HS, BuitelaarJK, van der Gaag RJ, Camfferman G, vanEngeland H: Event-related brain potentials inchildren with attention-deficit and hyperactivi-ty disorder: Effects of stimulus deviance andtask relevance in the visual and auditory mo-dality. Biol Psychiatry 1996;40:522–534.

Knight RT: Decreased response to novel stimuliafter prefrontal lesion in man. Electroencepha-logr Clin Neurophysiol 1984;59:9–20.

Knight RT: Evoked potential studies of attentioncapacity in human frontal lobe lesions; in LevinH, Eisenberg H, Benton F (eds): Frontal LobeFunction and Dysfunction. Oxford, OxfordUniversity Press, 1991, pp 139–153.

Knight RT: Contribution of human hippocampalregion to novelty detection. Nature 1996;383:256–259.

Knight RT: Distributed cortical network for visualattention. J Cogn Neurosci 1997;9:75–91.

Knight RT, Hillyard SA, Woods DL, Neville HJ:The effects of frontal cortex lesions on event-related potentials during auditory selective at-tention. Electroencephalogr Clin Neurophysiol1981;52:571–582.

Knight RT, Scabini D: Anatomic bases of event-related potentials and their relationship to nov-elty detection in humans. J Clin Neurophysiol1998;15:3–13.

Knight RT, Scabini D, Woods DL, Clayworth C:Contributions of temporal-parietal junction tothe human auditory P3. Brain Res 1989;502:109–116.

Kramer A, Jacobsen A: Perceptual organizationand focussed attention: The role of objects andproximity in visual processing. Percept Psycho-phys 1991;50:267–284.

Kraus N, McGee T, Carrell T, King C, Littman T,Nicol T: Discrimination of speech-like con-trasts in the auditory thalamus and cortex. JAcoust Soc Am 1994;96:2758–2768.

Kropotov JD, Näätänen R, Sevostianov AV, AlhoK, Reinikainen K, Kropotova OV: Mismatchnegativity to auditory stimulus change record-ed directly from the human temporal cortex.Psychophysiology 1995;32:418–422.

Lähteenmäki P, Holopainen I, Krause C, HeleniusH, Salmi TT, Lang AH: The cognitive func-tions of adolescent childhood cancer survivorsassessed by event-related potentials. Clin Neu-rophysiol, in press.

Levänen S, Ahonen A, Hari R, McEvoy L, Sams M:Deviant auditory stimuli activate human leftand right auditory cortex differently. CerebCortex 1996;6:288–296.

Levänen S, Hari R, McEvoy L, Sams M: Responsesof the human auditory cortex to changes in oneversus two stimulus features. Exp Brain Res1993;97:177–183.

Levänen S, Sams M: Disrupting auditory changedetection: Chopin is superior to white noise.Psychophysiology 1997;34:259–265.

Liasis A, Towell A, Boyd S: Intracranial auditorydetection and discrimination potentials as sub-strates of echoic memory in children. CognBrain Res 1999;7:503–506.

Lyytinen H, Blomberg AP, Näätänen R: Event-related potentials and autonomic responses to achange in unattended auditory stimuli. Psycho-physiology 1992;29:523–534.

Lyytinen H, Näätänen R: Autonomic and ERPresponses to deviant stimuli: Analysis of covar-iation; in Johnson R Jr, Parasuraman R, Rohr-baugh JW (eds): Current Trends in Event-Related Potential Research. EEG Supplement40. Amsterdam, Elsevier, 1987, pp 108–117.

Mangun GR, Hillyard SA: Modulations of sensory-evoked brain potentials indicate changes inperceptual processing during visual-spatialpriming. J Exp Psychol Hum Percept Perform1991;17:1057–1074.

Mecklinger A, Ullsperger P: The P300 to novel andtarget events: A spatio-temporal dipole modelanalysis. Neuroreport 1995;7:241–245.

Mesulam MM: Large-scale neurocognitive net-works and distributed processing for attention,language, and memory. Ann Neurol 1990;28:597–613.

Molnar M, Skinner JE, Csépe V, Winkler I, Kar-mos G: Correlation dimension changes accom-panying the occurrence of the mismatch nega-tivity and the P3 event-related potential com-ponent. Electroencephalogr Clin Neurophysiol1995;95:118–126.

Näätänen R: The role of attention in auditory infor-mation processing as revealed by event-relatedpotentials and other brain measures of cogni-tive function. Behav Brain Sci 1990;13:201–288.

Näätänen R: Attention and Brain Function. Hills-dale, Erlbaum, 1992.

Näätänen R, Alho K: Mismatch negativity (MMN)– The measure for central sound representationaccuracy. Audiol Neurootol 1997;2:341–353.

Näätänen R, Gaillard AWK, Mäntysalo S: Earlyselective attention effect on evoked potentialreinterpreted. Acta Psychol 1978;42:313–329.

Näätänen R, Lehtokoski A, Lennes M, Cheour M,Huotilainen M, Iivonen A, Vainio M, Alku P,Ilmoniemi RJ, Luuk A, Allik J, Sinkkonen J,Alho K: Language-specific phoneme represen-tations revealed by electric and magnetic brainresponses. Nature 1997;385:432–434.

Näätänen R, Michie PT: Early selective attentioneffects on the evoked potential: A critical re-view and reinterpretation. Biol Psychol 1979;8:81–136.

Näätänen R, Paavilainen P, Alho K, ReinikainenK, Sams M: Do event-related potentials revealthe mechanism of the auditory sensory memo-ry in the human brain? Neurosci Lett 1989;98:217–221.

Näätänen R, Paavilainen P, Tiitinen H, Jiang D,Alho K: Attention and mismatch negativity.Psychophysiology 1993;30:436–450.

Näätänen R, Picton TW: The N1 wave of thehuman electric and magnetic response tosound: A review and an analysis of the compo-nent structure. Psychophysiology 1987;24:375–425.

Näätänen R, Winkler I: The concept of auditorystimulus representation in cognitive neurosci-ence. Psychol Bull 1999;125:826–859.

Neisser U: Cognitive Psychology. New York, Ap-pleton-Century-Crofts, 1967.

Öhman A: The orienting response, attention, andlearning: An information-processing perspec-tive; in Kimmel HD, van Olst EH, Orlebeke JF(eds): The Orienting Response in Humans.Hillsdale, Erlbaum, 1979, pp 443–471.


Opitz B, Mecklinger A, Von Cramon DY, KruggelF: Combined electrophysiological and hemo-dynamic measures of the auditory oddball.Psychophysiology 1999;36:142–147.

Paavilainen P, Alho K, Reinikainen K, Sams M,Näätänen R: Right-hemisphere dominance ofdifferent mismatch negativities. Electroen-cephalogr Clin Neurophysiol 1991;78:466–479.

Paavilainen P, Saarinen J, Tervaniemi M, Näätä-nen R: Mismatch negativity to changes in ab-stract sound features during dichotic listening.J Psychophysiol 1995;9:243–249.

Pechmann T, Mohr G: Interference in memory fortonal pitch: Implications for a working-memo-ry model. Mem Cogn 1992;20:314–320.

Polo MD, Yago E, Grau C, Alho K, Gual A, EsceraC: Involuntary attentional processing inchronic alcoholism; in Tervaniemi M, Escera C(eds): Abstracts of the First InternationalWorkshop on Mismatch Negativity and ItsClinical Applications. Helsinki, MultiprintUniversity of Helsinki, 1998, p 88.

Polo MD, Yago E, Gual A, Grau C, Alho K, EsceraC: Abnormal activation of cerebral networks oforienting to novelty in chronic alcoholics. Psy-chophysiology 1999;36:S90.

Posner MI, Cohen Y, Rafal RD: Neural systemscontrol over spatial orienting. Philos Trans RSoc Lond B 1982;298:187–198.

Rinne T, Alho K, Alku P, Holi M, Sinkkonen J,Virtanen J, Bertrand O, Näätänen R: Analysisof speech sounds is left-hemisphere predomi-nant at 100–150 ms after sound onset. Neuro-report 1999a;10:1113–1117.

Rinne T, Alho K, Ilmoniemi RJ, Virtanen J, Nää-tänen R: Separate time behaviors of the tempo-ral and frontal MMN sources. Submitted.

Rinne T, Gratton G, Fabiani M, Cowan M, MaclinE, Stinard A, Sinkkonen J, Alho K, NäätänenR: Scalp-recorded optical signals make soundprocessing in the auditory cortex visible? Neu-roimage 1999b;10:620–624.

Ritter W, Sussman E, Deacon D, Cowan N,Vaughan HG Jr: Two cognitive systems simul-taneously prepared for opposite events. Psy-chophysiology 1999;36:835–838.

Sams M, Kaukoranta E, Hämäläinen M, NäätänenR: Cortical activity elicited by changes in audi-tory stimuli: Different sources for the magneticN100m and mismatch responses. Psychophysiol-ogy 1991;28:21–29.

Sams M, Paavilainen P, Alho K, Näätänen R:Auditory frequency discrimination and event-related potentials. Electroencephalogr ClinNeurophys 1985;62:437–448.

Satterfield JH, Schell AM, Nicholas TW: Preferen-tial neural processing of attended stimuli inattention-deficit hyperactivity disorder andnormal boys. Psychophysiology 1994;31:1–10.

Satterfield JH, Schell AM, Nicholas TW, Satter-field BT, Freese T: Ontogeny of selective atten-tion effects on event-related potentials in atten-tion-deficit hyperactivity disorder and normalboys. Biol Psychiatr 1990;28:879–903.

Scherg M, Vajsar J, Picton T: A source analysis ofthe human auditory evoked potentials. J CognNeurosci 1989;1:336–355.

Schröger E: A neural mechanism for involuntaryattention shifts to changes in auditory stimula-tion. J Cogn Neurosci 1996;8:527–539.

Schröger E: On the detection of auditory deviants:A pre-attentive activation model. Psychophysi-ology 1997;34:245–257.

Schröger E, Giard MH, Wolff Ch: Auditory distrac-tion: Event-related potential and behavioral in-dices of auditory distraction. Submitted.

Schröger E, Wolff Ch: Attentional orienting andreorienting is indicated by human event-relat-ed brain potentials. Neuroreport 1998a;9:3355–3358.

Schröger E, Wolff Ch: Behavioral and electrophysi-ological effects of task-irrelevant sound change:A new distraction paradigm. Cogn Brain Res1998b;7:71–87.

Semal C, Demany L, Ueda K, Halle PA: Speechversus nonspeech in pitch memory. J AcoustSoc Am 1996;100:1132–1140.

Serra JM, Giard MH, Yago E, Alho K, Escera C:Bilateral contribution from frontal lobes toMMN. Int J Psychophysiol 1998;30:236–237.

Sokolov EN: The neuronal mechanisms of theorienting reflex; in Sokolov EN, VinogradovaOS (eds): Neuronal Mechanisms of the Orient-ing Reflex. Hillsdale, Erlbaum, 1975, pp 217–338.

Squires KC, Squires NK, Hillyard SA: Decision-related cortical potentials during an auditorysignal detection task with cued observation in-tervals. J Exp Psychol Hum Percept Perform1975;1:268–279.

Stuss DT, Benson DF: The Frontal Lobes. NewYork, Raven Press, 1986.

Sussman E, Ritter W, Vaughan HG Jr: Attentionaffects the organization of auditory input asso-ciated with the mismatch negativity system.Brain Res 1998;789:130–138.

Sussman E, Ritter W, Vaughan HG Jr: An investi-gation of the auditory streaming effect usingevent-related brain potentials. Psychophysiolo-gy 1999;36:22–34.

Tervaniemi M, Kujala A, Alho K, Virtanen J,Ilmoniemi RJ, Näätänen R: Functional spe-cialization of the human auditory cortex in pro-cessing phonetic and musical sounds: A magne-toencephalographic (MEG) study. Neuroimage1999a;9:330–336.

Tiitinen H, Alho K, Huotilainen M, Ilmoniemi RJ,Simola J, Näätänen R: Tonotopic auditory cor-tex and the magnetoencephalographic (MEG)equivalent of the mismatch negativity. Psycho-physiology 1993;30:537–540.

Treisman AM: Features and objects: The four-teenth Bartlett memorial lecture. Q J Exp Psy-chol 1988;40A:201–237.

Trejo LJ, Ryan-Jones DL, Kramer AF: Attentionalmodulation of the mismatch negativity elicitedby frequency differences between binaurallypresented tone burst. Psychophysiology 1995;32:319–328.

Tulving E, Markowitsch HJ, Kapur S, Habib R,Houle S: Novelty encoding networks in thehuman brain: Positron emission tomographydata. Neuroreport 1994;5:2525–2528.

Winkler I, Czigler I: Mismatch negativity: De-viance detector or the maintenance of the ‘stan-dard’. Neuroreport 1998;9:3809–3813.

Winkler I, Karmos G, Näätänen R: Adaptive mod-eling of the unattended acoustic environmentreflected in the mismatch negativity event-related potential. Brain Res 1996;742:239–252.

Winsberg BG, Javitt DC, Silipo GS, Doneshka P:Mismatch negativity in hyperactive children:Effects of methylphenidate. PsychopharmacolBull 1993;29:229–233.

Woldorff MG, Hackley SA, Hillyard SA: The ef-fects of channel-selective attention on the mis-match negativity wave elicited by devianttones. Psychophysiology 1991;28:30–42.

Woods DL: The physiological basis of selectiveattention: Implications of event-related poten-tial studies; in Rohrbaugh JW, Parasuraman R,Johnson R Jr (eds): Event-Related Potentials:Basic Issues and Applications. New York, Ox-ford University Press, 1990, pp 178–209.

Woods DL: Auditory selective attention in middle-aged and elderly subjects: An event-relatedbrain potential study. Electroencephalogr ClinNeurophysiol 1992;84:456–468.

Woods DL, Knight RT, Scabini D: Anatomicalsubstrates of auditory selective attention: Be-havioral and electrophysiological effects of pos-terior association cortex lesions. Cogn BrainRes 1993;1:227–240.

Yamaguchi S, Knight RT: Anterior and posteriorassociation cortex contributions to the somato-sensory P300. J Neurosci 1991;11:2039–2054.

Yamaguchi S, Knight RT: Effects of temporal-pari-etal lesions on the somatosensory P3 to lowerlimb stimulation. Electroencephalogr ClinNeurophysiol 1992;84:139–148.

Yago E, Alho K, Escera C: Electrophysiological andbehavioral evidence of attentional reorienta-tion to auditory stimulus change and novelty.Psychophysiology 1999;36:S127.

involuntary attention and distractibility as evaluated ... et...task unrelated to the mmn-eliciting...

Documents