An Event-related Potential Study of Selective …depts.washington.edu/audneuro/cochetal05.pdfAn...

An Event-related Potential Study of Selective AuditoryAttention in Children and Adults

Donna Coch1 Lisa D Sanders2 and Helen J Neville2

Abstract

amp In a dichotic listening paradigm event-related potentials(ERPs) were recorded to linguistic and nonlinguistic probestimuli embedded in 2 different narrative contexts as they wereeither attended or unattended In adults the typical N1attention effect was observed for both types of probes Probessuperimposed on the attended narrative elicited an enhancednegativity compared to the same probes when unattendedOverall this sustained attention effect was greater over medialand left lateral sites but was more posteriorly distributed andof longer duration for linguistic as compared to nonlinguistic

probes In contrast in 6- to 8-year-old children the ERPs weremorphologically dissimilar to those elicited in adults andchildren displayed a greater positivity to both types of probestimuli when embedded in the attended as compared to theunattended narrative Although both adults and childrenshowed attention effects beginning at about 100 msec onlyadults displayed left-lateralized attention effects and a distinctposterior distribution for linguistic probes These resultssuggest that the attentional networks indexed by this taskcontinue to develop beyond the age of 8 years amp

INTRODUCTION

Over the past few decades numerous event-relatedpotential (ERP) studies have investigated the humanability to attend selectively to a single sound source(eg see Luck 1998 Woods 1990 Hillyard amp Picton1987 Naatanen 1982 for reviews) Accumulating evi-dence suggests that this ability may be the result of bothan enhancement of information received from the se-lected source and a suppression of irrelevant competinginformation from other sound sources (Woods 1990)Overall auditory attention appears to be deployed as afinely tuned gradient around the attended source(Teder-Salejarvi Hillyard Roder amp Neville 1999) More-over selective auditory attention can be directed notonly to a specific sound source but to specific featureswithin that sound source such as frequency or pitch orto a conjunction of features (Sanders et al 2001 Woodsamp Alain 1993 Woods Alho amp Algazi 1991) Evidencefurther suggests that the effects of selective attentioncan occur relatively early in the processing of auditoryinformation within primary auditory cortex (Petkovet al 2004 Hugdahl Law et al 2000 Alho et al1999) and within 20 msec of stimulus onset (WoldorffHansen amp Hillyard 1987) and can last for hundreds ofmilliseconds (Hansen amp Hillyard 1980)

Although a number of electrophysiological studieshave investigated the exquisite selectivity and sensitivityof sustained auditory attention in adults very few have

addressed the development of such attentional systemsThe present study was designed to address this gap inknowledge and included adults as well as a group of6- to 8-year-old children In a dichotic listening paradigmERPs were recorded to linguistic and nonlinguistic probestimuli embedded in 2 narrative contexts that were eitherattended or unattended

Event-related Potential Measures of SelectiveAuditory Attention in Adults

In a now classic ERP study of selective auditory atten-tion Hillyard Hink Schwent and Picton (1973) dichoti-cally presented 2 similar series of tone pips and requiredparticipants to attend only to tones played to thedesignated ear comparison of the ERP responses tothe same tones when attended and unattended revealedan enhanced N1 component to attended tones A similarN1 attention effect was reported for syllables and envi-ronmental sounds (Hink Hillyard amp Benson 1978Hink van Voorhis Hillyard amp Smith 1977)

Subsequently researchers began to report not onlyan enhanced N1 for the same stimulus when attendedas compared with when unattended but also a later andbroader negativity termed the processing negativity ornegative difference (Nd) wave (Woldorff amp Hillyard1991 Hillyard amp Picton 1987 Hansen Dickstein Berkaamp Hillyard 1983 Hansen amp Hillyard 1980 Naatanen1979 1982) It was suggested that the Nd reflectedorienting to or further processing of an auditory in-put deemed relevant in preliminary sensory analyses1Dartmouth College 2University of Oregon

D 2005 Massachusetts Institute of Technology Journal of Cognitive Neuroscience 174 pp 605ndash622

(Naatanen 1979) Currently the Nd is thought to consistof both early (Nde) and late (Ndl) components the earlyfronto-central component may reflect rapid initial featur-al analysis of stimuli whereas the late anterior componentreflects further processing of the stimuli and mainte-nance of the attentional trace (eg see Woods 1990Naatanen 1982) The Nde is superimposed on the N1and there is some evidence that the N1 itself is actuallynot modulated by attention as first reported rather theeffects of the slow wave Nde are simply observed in thetypical N1 time window (Woods 1990 Naatanen 1982)Other evidence suggests that the N1M100 (the MEGcounterpart to the N100) itself is modulated by atten-tion (eg Woldorff et al 1993)

A small number of studies have attempted to employ asomewhat more ecologically valid design and haveinvestigated selective spatial attention effects in thecontext of a spoken narrative in most cases the typicalERP attention effects have been observed (Trejo Ryan-Jones amp Kramer 1995 Teder Kujala amp Naatanen 1993Woods Hillyard amp Hansen 1984 Hink amp Hillyard1976) For example irrelevant vowel probes within anattended speech passage elicit more negative ERPs thanthe same probes in the unattended passage in dichoticpresentations (Hink amp Hillyard 1976) In a more complexdesign Woods et al (1984) presented various speech andtone probe stimuli embedded in dichotically presentedprose passages Speech probes spatially and temporallycoincident with the attended passage as compared withthe same probes in the unattended passage elicited anenhanced negativity (presumably the Nd) beginning atabout 50 msec and lasting to 1000 msec In contrast onetype of tone probe elicited an enhanced positivity from200 to 300 msec most likely an attentional modulation ofa different component The authors suggested that thispattern of results indicated that stimulus selection duringlinguistic attention might be specifically tuned to speechsounds (Woods et al 1984)

The possibility of specialized attentional tuning forspeech stimuli in the context of spoken word processingraises a related question concerning specialized neuralprocessing for speech in general In Woods et alrsquos(1984) study the Nd elicited by the CVC speech probewas most prominent over the left hemisphere whereaseffects for other probes evidenced no hemisphericasymmetries Other ERP selective attention studies havealso reported a left-lateralized Ndl for speech-like (pho-nemic change) stimuli as compared with nonspeech(intensity change) stimuli (Szymanski Yund amp Woods1999) In contrast some studies have reported noasymmetries for either speech or tone stimuli witheach eliciting an Nd similar in morphology and distri-bution (although longer in latency for more complexspeech stimuli Hansen et al 1983) DevelopmentalERP studies have suggested that a left hemispherespecialization for speech either precedes the abilityto speak (Molfese Freeman amp Palermo 1975) or be-

gins to develop quite early with spoken language ex-perience (Neville amp Bavelier 2000 Mills Coffey-Corinaamp Neville 1993) Studies using fMRI in 3-month-olds(Dehaene-Lambertz Dehaene amp Hertz-Pannier 2002)and optical imaging in newborns (Pena et al 2003)have reported an early left hemisphere specializationfor speech stimuli Few studies have addressed lateral-ization in school-age children

Development of Selective Auditory AttentionBehavioral Evidence

Behavioral studies have indicated that auditory selectiveattention skills develop throughout childhood at leastuntil adolescence Both the abilities to selectively attendto relevant stimuli and to successfully ignore irrele-vant stimuli improve progressively with increasing ageacross childhood (Lane amp Pearson 1982 Hiscock ampKinsbourne 1980 Geffen amp Wale 1979 Sexton ampGeffen 1979 Geffen amp Sexton 1978 Zukier amp Hagen1978 Doyle 1973 Maccoby amp Konrad 1966) The abilityto shift attention quickly and effectively also developsacross childhood at least until adolescence (Pearson ampLane 1991 Andersson amp Hugdahl 1987 Hiscock ampKinsbourne 1980 Geffen amp Wale 1979) Further thereis some evidence that background noise creates greatermasking effects for younger children as compared withadolescents or adults (Elliott 1979) In a recent studywith children age 6ndash9 years alertness and conflict scoresevidenced change over time whereas orienting scoresdid not (Rueda et al 2004) illustrating that differentaspects of attention have different developmental timecourses

Interestingly there has been some attempt to linksustained selective auditory attention abilities to readingabilities (eg Taub Fine amp Cherry 1994) In partic-ular the ability to make attentional shifts under dich-otic listening conditions appears to be related toreading ability (Asbjoslashrnsen amp Bryden 1998 Hugdahl ampAndersson 1987) Alternately some authors have sug-gested that poorer readers have an attentional capac-ity limitation but not an inability to focus attention(Dickstein amp Tallal 1987)

Development of Selective Auditory AttentionEvent-related Potential Evidence

Although behavioral studies offer evidence for the de-velopment of selective auditory attention in school-agechildren there is very little comparable electrophysio-logical evidence from children in this age range (eg seeRidderinkhof amp van der Stelt 2000) To the best of ourknowledge there is only one published study of a typicalERP dichotic listening attention paradigm using tonesand syllables with young participants (groups with meanage 810 years 1438 years and adults Berman amp Fried-

606 Journal of Cognitive Neuroscience Volume 17 Number 4

man 1995) Stimuli were presented binaurally overheadphones and the expected Nd was observed in allparticipants with Nde amplitude increasing with agemore so for syllables than for tones and Nde distributionmore posterior in adults than in children For all agegroups the Nd attention effect for syllables was smallerand later than that for tones The primary effect of ageappeared to be smaller negative ERPs elicited by stimuliin the unattended channel which the authors suggestedmight reflect a narrowing of attentional focus or greaterfacility in suppressing unattended inputs with age (Ber-man amp Friedman 1995)

In a series of cross-sectional studies with male chil-dren investigating ERP correlates of attention deficithyperactivity disorder Satterfield and colleagues re-ported that attention elicited the typical enhancednegativity in normal control boys aged 6ndash8 years in across-modal auditoryvisual selective attention paradigm(eg Satterfield Schell amp Nicholas 1994 SatterfieldSchell Nicholas amp Backs 1988) However in a longitu-dinal study they reported that the central Nde wasabsent in 6-year-olds but peaked at about 250 msecwhen the same children were tested 2 years later at age8 years the frontal Ndl shortened in latency from age 6to age 8 years (Satterfield Schell Nicholas Satterfield ampFreese 1990) In a more typical selective auditory atten-tion paradigm the expected N1 attention effect wasobserved in adolescents aged 12ndash14 years (LoiselleStamm Maitinsky amp Whipple 1980) Other studies havereported similar attention effects in adolescent boys(Lovrich amp Stamm 1983 Zambelli Stamm Maitinsky ampLoiselle 1977)

The Present Study

There is clearly a lack of ERP investigations of thedevelopment of selective auditory attention effectsacross the early school-age years Behavioral evidenceunequivocally indicates that selective attention abilitiesare developing across childhood and into adolescenceIn the present study we employed ERPs and a complexdichotic listening task modeled after that used by Woodset al (1984) to study the development of the neuralsystems important for sustained auditory attention Chil-dren and adults were asked to listen to 1 of 2 concur-rently presented narratives as it periodically switchedbetween a speaker at their right and a speaker at theirleft (as the other narrative played from the oppositespeaker) ERPs were recorded to linguistic (a token ofthe syllable ba) and nonlinguistic (a buzz) probestimuli embedded in the attended and unattendednarratives Comparisons of the ERPs elicited by the sameprobe stimuli when played from the attended andunattended source indexed ERP attention effects Com-parisons of responses to linguistic and nonlinguisticprobes were employed to index the degree of separa-bility of the neural systems processing the different

types of stimuli including any left hemisphere speciali-zation for speech stimuli in the context of the ongoingnarrative Comparisons between children and adultsrevealed the developmental time course of ERP indicesof selective auditory spatial attention Given the evi-dence reviewed above we hypothesized typical N1Ndeffects perhaps differently distributed in children andadults distinct responses for linguistic and nonlinguisticprobes with evidence for left hemisphere specializationfor linguistic processing particularly in adults and evi-dence of development of the ERP waveform betweenadults and 6- to 8-year-old children

RESULTS

Behavioral Testing Results

All scores on the standardized behavioral tests adminis-tered to children were within normal limits PPVT (Dunnamp Dunn 1997) standard scores ranged from 103 to 139with an average of 1188 plusmn 98 corresponding percentilerank scores ranged from 58 to 995 Standard scores onthe Word Identification subtest of the WRMT (Wood-cock 1987) ranged from 94 to 145 with an average of1134 plusmn 143 Standard scores on the Word Attacksubtest of the WRMT ranged from 81 to 145 with anaverage of 1158 plusmn 153

Behavioral Event-related Potential Results

Only participants who answered at least 8 out of 10posttest comprehension questions correctly were in-cluded in analyses Children answered an average of94 plusmn 08 questions correctly whereas adults answeredan average of 98 plusmn 04 questions correctly This differ-ence between groups was significant [t(42) = 25p lt 05]

Event-related Potential Results

Adults

Component analyses (see Table 1) The probesembedded in both the attended and unattendedstories elicited typical early auditory ERP componentsin adults a positivity between 50 and 150 msec (P1)a negativity between 100 and 200 msec (N1) and asubsequent positivity between 200 and 400 msec (P2)Across probe type the P1 was largest at medial andanterior electrode sites [laterality F(121) = 12612p lt 001 anteriorposterior F(5105) = 3627 p lt 001Laterality AnteriorPosterior F(5105) = 3432p lt 001] The P1 peaked slightly earlier for linguisticprobes (mean latency = 98 msec) than for nonlinguisticprobes (mean latency = 103 msec) F(121) = 542p lt 05

Coch Sanders and Neville 607

Analysis of the peak-to-peak amplitude between P1and N1 revealed that the N1 component like the P1 waslargest over medial and anterior electrode sites [later-ality F(121) = 11677 p lt 001 Laterality AnteriorPosterior F(5105) = 2477 p lt 001] Unlike the P1 theN1 peaked significantly earlier for nonlinguistic probes(mean latency = 176 msec) than for linguistic probes(mean latency = 195 msec) F(121) = 2529 p lt 001The N1 was also larger in amplitude for linguistic thannonlinguistic probes (see Figure 1) F(121) = 1969p lt 001 Overall the N1 component was larger toprobes played within the attended story than to probesplayed within the unattended story F(121) = 562p lt 05

The second positivity elicited by the probes P2 waslarger over the right hemisphere F(121) = 590 p lt 05and at medial F(121) = 2102 p lt 001 and anteriorF(5105) = 850 p lt 001 sites The P2 elicited bynonlinguistic probes was larger F(121) = 1131p lt 01 and earlier (mean latency = 279 msec)F(121) = 1168 p lt 01 than the P2 to linguistic probes(mean latency = 304 msec) Across probe type the P2peaked earlier at medial sites for probes played withinthe attended story than those played within the unat-tended story [Attention Laterality F(121) = 471p lt 05] Overall across probe type the amplitude ofP2 was not modulated by attention ( p = 947)

Attention effects Nonlinguistic probes Within the 100-to 175-msec epoch nonlinguistic probes elicited largernegativities when played from the attended speakerthan from the unattended speaker (see Figure 1A)

F(121) = 569 p lt 05 Like the N1 component itselfthis attention effect was largest over medial and anteriorelectrode sites [Attention Laterality F(121) = 782p lt 05 Attention Laterality AnteriorPosteriorF(5105) = 425 p lt 001] Analyses of consecutive25-msec bins between 0 and 300 msec revealed that thisN1 attention effect began within the 100- to 125-msecepoch

Attention effects Linguistic probes Within the 100- to250-msec epoch linguistic probes elicited largernegativities when played from the attended speakerthan from the unattended speaker (see Figure 1B)F(121) = 500 p lt 05 This attention effect waslargest over medial and posterior electrode sites[Attention Laterality F(121) = 1031 p lt 01Attention AnteriorPosterior F(5105) = 357 p lt 05Attention Laterality AnteriorPosterior F(5105) =360 p lt 05] In 25-msec bin analyses this attentioneffect was also shown to begin within the 100- to125-msec epoch

Woods et al (1984) reported that the N1 attentioneffect was sensitive to frequency raising the possibilitythat the N1 attention effect to linguistic probes observedhere may have been influenced by voice-specific acous-tic differences between the probe and the stories readby a man and a woman In a follow-up analysis wemeasured the pitch from 30 random 100-msec intervalsfrom each story recording (mean plusmn SE pitch male150 plusmn 101 Hz female 212 plusmn 82 Hz) and thensubtracted from these measurements the mean pitchof the linguistic probe (158 Hz) An independent sam-ples t test on the absolute value of the pitch differencefor the 2 stories indicated a trend for difference [t(58) =196 p = 06] suggesting that voice-specific differencesbased on the acoustics (in terms of pitch) may haveplayed some role in but were probably not a majorcontributing factor to the observed effects Teder et al(1993) also reported no effect of male or female voiceon the attention effect

Comparison of probe types Overall across probetypes a classic attention effect began around 100 msecsuch that probes played from the attended speakerelicited larger negativities than the same probes playedfrom the unattended speaker (see Figure 1 and Table 2)F(121) = 1114 p lt 01 This attention effect was largestover medial sites [Attention Laterality F(121) =2121 p lt 001] and extended to left lateral sites[Attention Laterality Hemisphere F(121) = 922p lt 01] for both types of probes (Probe Type Attention Laterality Hemisphere p = 971)However for linguistic probes as compared withnonlinguistic probes this attention effect had a moreposterior distribution [Attention Probe Type AnteriorPosterior F(5105) = 670 p lt 01] and

Table 1 Adults Component Analyses

P1(50ndash150 msec)

N1(100ndash200 msec)

P2(200ndash400 msec)

Amplitude(AV)

150 (012) 234 (015)a 127 (011)

Nonlinguisticprobes

161 (014) 213 (015)a 153 (015)

Linguisticprobes

139 (013) 255 (017)a 102 (011)

Latency(msec)

101 (20) 186 (33) 291 (44)

Nonlinguisticprobes

103 (19) 176 (31) 279 (54)

Linguisticprobes

98 (20) 195 (36) 304 (58)

Peak amplitude and latency measured across all sites Standard errorin parentheses

aPeak-to-peak amplitude (see Methods for details)


Figure 1 In adults the typical sequence of P1ndashN1ndashP2 components was observed for both types of probe stimuli with each component largest

at medial anterior sites Each component of interest is identified at site F3 Classic attention effects (probes played from the attended speakerelicited larger negativities than the same probes played from the unattended speaker) were evident for (A) nonlinguistic probes at medial and

anterior sites from 100 to 175 msec and (B) linguistic probes at medial and posterior sites from 100 to 250 msec More anterior sites are toward the

top more posterior sites are toward the bottom left hemisphere sites are on the left right hemisphere sites are on the right medial sites are

toward the middle of the figure lateral sites are toward the edges of the figure and negative is lsquolsquoplotted uprsquorsquo


extended into a later time window [mean amplitudewithin the 175- to 250-msec epoch Attention Probetype F(121) = 1202 p lt 01 for linguistic probesF(121) = 459 p lt 05 for nonlinguistic probesp = 119] Thus in the later time window in adults theincreased negativity with attention was observed onlyfor linguistic probes

Children

Component analyses (see Table 3) The morphologyof the ERP waveforms in children was dissimilar to thatobserved in adults Instead of the distinct positivendashnegativendashpositive sequence seen in adults ERPs inchildren consisted of a single broad sustained posi-tivity that began around 100 msec and continued untilat least 300 msec after stimulus onset In mean am-plitude measurements this positive component waslargest over medial and anterior sites (see Figure 2)[laterality F(121) = 10651 p lt 001 anteriorposteriorF(121) = 5103 p lt 001] Across probe type measuredat the 4 most anterior rows of electrodes the ampli-tude of this component was significantly enhanced withattention [F(121) = 1006 p lt 005]

Attention effects Nonlinguistic probes Within the100- to 250-msec epoch nonlinguistic probes eliciteda greater positivity at anterior electrode sites whenplayed from the attended speaker (see Figure 2A) [At-tention AnteriorPosterior F(5105) = 634 p lt 001attention at anterior 4 rows F(121) = 849 p lt 01] In25-msec bin analyses this attention effect was shown tobegin within the 100- to 125-msec epoch as observedin adults

Attention effects Linguistic probes Within the 100- to250-msec epoch linguistic probes likewise elicited a

greater positivity at anterior electrode sites when playedfrom the attended speaker (see Figure 2B) [Attention AnteriorPosterior F(5105) = 821 p lt 001 attentionat anterior 4 rows F(121) = 667 p lt 05] As with theattention effect for nonlinguistic probes this positivedifference began within the 100- to 125-msec epoch in25-msec bin analyses

Comparison of probe types

Overall across probe type an attention effect began atabout 100 msec such that probes played from theattended speaker tended to elicit larger positivities thanthe same probes played from the unattended speaker[F(121) = 368 p = 069] Across probe type this effectwas largest at anterior sites [Attention AnteriorPos-terior F(5105) = 1153 p lt 001] There were nosignificant differences in the amplitudes onset latenciesor distributions of the attention effects for the linguisticand nonlinguistic probe stimuli in children

Comparison of Adults and Children

Difference measurements (attendedndashunattended) werenormalized to allow for comparison of the distributionof attention effects across groups Analyses of normal-ized data revealed a significant Group Hemisphere Laterality interaction [F(142) = 1168 p lt 001] con-sistent with the left lateral extension (or relativelygreater left-than-right lateral extension) of attentioneffects observed in adults but not in children (seeFigure 3) Analyses of normalized data also revealed aGroup Probe Type AnteriorPosterior interaction

Table 2 Adults Attention Effects




Nonlinguisticprobes

Attended 050 (014) ndash 071 (013)

Unattended 079 (015) ndash 053 (013)

Linguisticprobes

Attended ndash 011 (017) 043 (020)

Unattended ndash 022 (014) 005 (015)

Mean amplitude measured across all sites Standard error inparentheses

Table 3 Children Component Analyses and Attention Effect

Positivity (100ndash250 msec)

Amplitude (AV) 178 (016)

Nonlinguistic probes 198 (033)

Linguistic probes 160 (030)


Attended 306 (034)

Unattended 210 (034)


Attended 265 (027)


Mean amplitude measured across all sites for overall amplitude meanamplitude measured at 4 most anterior rows for attention effectStandard error in parentheses


Figure 2 In children instead of the typical P1ndashN1ndashP2 series a single sustained broad positivity largest at medial and anterior sitescharacterized the ERPs elicited by both types of probe stimuli An attention effect (probes played from the attended speaker elicited a

larger positivity than the same probes played from the unattended speaker) with similar medial anterior distribution was evident for (A)

nonlinguistic probes beginning at 100ndash125 msec and (B) linguistic probes beginning at 100ndash125 msec All else (although note scale) as in

Figure 1


[F(5210) = 326 p = 054] reflecting the distinctdistributions of attention effects for linguistic (moreposterior) and nonlinguistic (more anterior) probes inadults and the similarity of the distribution of theseeffects in children (see Figure 4) Moreover this inter-action indexes the fact that adults and children showedmore similarly distributed attention effects for nonlin-guistic probes (nonlinguistic probes all ps gt 35)whereas adults evidenced a more posteriorly distrib-uted attention effect for linguistic probes [linguisticprobes Group AnteriorPosterior F(5210) = 924p lt 001]

Correlations between Event-related Potential andBehavioral Measures in Children (Only)

A difference measure of mean amplitude of ERPs elic-ited by attended and unattended probes at sites FC5and FC6 was not correlated with any of our behavioralmeasures of reading skill in the children (all ps gt 45)Further there were no significant correlations between

childrenrsquos age and any measure of reading skill orthis ERP measure of selective auditory attention (allps gt 24)

DISCUSSION

Children and adults attended to one narrative as itperiodically switched from a speaker at their right to aspeaker at their left while another narrative playedfrom the opposite speaker Simultaneously ERPs wererecorded to linguistic and nonlinguistic probe stimulisuperimposed on the narratives at both the attendedand unattended speakers Consistent with previousfindings (Woldorff amp Hillyard 1991 Hillyard amp Picton1987 Woods et al 1984 Hansen Dickstein et al 1983Hansen amp Hillyard 1980 Naatanen 1979 1982 Hinket al 1977 1978 Hillyard et al 1973) in adults the ERPwaveforms evidenced the typical auditory attention ef-fect for both types of probes an enhanced negativity forprobes when attended as compared with those same

Figure 3 In a comparison of the attention effects (attendedndashunattended) in children and adults using normalized difference measures a Group Hemisphere Laterality interaction reflected the relatively greater left-than-right lateral extension of attention effects observed in adults


probes when unattended This effect was longer lastingand more posteriorly distributed for linguistic thannonlinguistic probes In contrast the ERP waveformsof children were morphologically dissimilar to thoseof adults and evidenced a greater broad positivity toboth types of probe stimuli when attended as com-pared with unattended This attention effect had asimilar anterior distribution for the 2 probe typesRemarkably both the typical attention effect involvingan enhanced negativity in adults and the atypical atten-tion effect involving an enhanced positivity in childrenonset at about 100 msec

Adults

Adults evidenced the typical auditory ERP waveformmorphology including P1 N1 and P2 components great-est at medial and anterior sites with both the linguistic

and nonlinguistic probe stimuli eliciting each of thesecomponents Absent attention effects linguistic probeselicited an earlier P1 later and larger N1 and later andsmaller P2 than nonlinguistic probes Differential pro-cessing of the 2 types of probes suggests that these earlycomponents may be sensitive to more than just the fineacoustic characteristics of auditory stimuli The probestimuli were matched in terms of acoustic primitivesincluding spectral frequency but the later and larger N1to linguistic probes in particular suggests that the lin-guistic nature of these probes did affect early processingin the context of the language-based task

Considering the effects of attention directly adultsevidenced the typical N1 or Nd attention effect for bothlinguistic and nonlinguistic probes such that probesplayed from the attended speaker elicited more negativeERPs than the same probes played from the unattendedspeaker This effect was greater over medial sites and

Figure 4 In a comparison of the attention effects (attendedndashunattended) in children and adults using normalized difference measures a Group Probe Type AnteriorPosterior interaction reflected the distinct anteriorposterior distributions of the attention effects to linguistic and

nonlinguistic probes in adults in comparison to the relatively similar distributions in children Further this result highlights the similarity of the

anteriorposterior distribution of the attention effects to nonlinguistic probes in children and adults and the unique posterior distribution of theattention effect for linguistic probes in adults


lateral sites of the left hemisphere across probe typeSome authors have reported a more left-lateralized Ndfor linguistic as compared with nonlinguistic stimuli(Szymanski et al 1999 Woods et al 1984) whereasothers have not (Hansen Dickstein et al 1983) Thepresent pattern of lateral left hemisphere asymmetrysuggests that the context of the linguistic taskmdashlisteningselectively to one narrativemdashmay have influenced thelateralized processing of all stimuli within the lsquolsquospot-lightrsquorsquo of linguistic attention results of MEG PET andfMRI studies have indicated left hemisphere speciali-zation for many aspects of speech processing (egPoldrack et al 2001 Shtyrov Kujala Palva Ilmoniemiamp Naatanen 2000 Hugdahl Broslashnnick et al 1999Zatorre Evans Meyer amp Gjedde 1992) Other aspectsof speech processing such as voicing contrasts havebeen reported to be more dependent on the righthemisphere in both adults and children (Simos Molfeseamp Brenden 1997 Segalowitz amp Cohen 1989 Molfeseamp Molfese 1988)

Although the hemispheric distribution of attentioneffects for the probe types was similar the anteriorposterior distribution was not Linguistic probes clearlyelicited more posteriorly distributed processing thannonlinguistic probes in adults Considering the limitedspatial resolution of the ERP technique it is difficult todraw conclusions regarding this observed pattern How-ever it is interesting to speculate on a possible connec-tion to specialized linguistic processing within posteriorlanguage areas Regardless the differential distributionof effects along the anteriorposterior dimension inadults indicates that linguistic and nonlinguistic probestimuli were distinguished within the spatial linguisticattentional spotlight

The high temporal resolution of the ERP techniquerevealed further evidence for differential processing ofthe probe stimuli in adults The duration of the attentioneffect to linguistic probes extended beyond that fornonlinguistic probes Woods et al (1984) have sug-gested that stimulus selection during linguistic attentionmay be specifically tuned to speech sounds which maybe reflected here in the temporally extended effect onlyfor linguistic probes Although it is possible that thisresult reflects further processing and maintenance ofthe attentional trace in terms of the Ndl (eg seeWoods 1990 Naatanen 1982) the typical Ndl is maxi-mal at anterior sites whereas the present effect (notdistinguishing between early and late Nd) has a poste-rior distribution It is interesting to speculate that withinthe linguistic context of the task only the linguisticprobe stimuli were subject to further analysis preciselybecause of their potentially relevant (ie linguistic andpossible semantic) nature whereas early analysis of thenonlinguistic probes was sufficient to gate them fromfurther processing If this is the case it would suggestthat spatial linguistic attention can be remarkably pre-cise as the linguistic and nonlinguistic stimuli were

spectrally matched This would be consistent withother studies demonstrating the precision of selec-tive auditory attention outside the linguistic domain(Sanders et al 2001 Woods amp Alain 1993 Woodset al 1991)

Children

In contrast to adults children did not evidence typicalauditory ERP waveforms Instead of the common P1ndashN1ndashP2 sequence the ERPs of children were character-ized by a single sustained positivity largest at medial andanterior sites that began at about 100 msec and extend-ed to at least 300 msec Linguistic and nonlinguisticprobe stimuli elicited similarly distributed positivities

For the most part developmental ERP studies tendto report similar componentry in children and adults(eg Davis Bruce Snyder amp Nelson 2003 CochGrossi Coffey-Corina Holcomb amp Neville 2002 OadesDittmann-Balcar amp Zerbin 1997 Thomas amp Nelson1996) However the N1 is not a unitary event and likelyhas multiple underlying sources (Naatanen amp Picton1987) each of which may have a different developmen-tal time course (eg Takeshita et al 2002) and each ofwhich may be differently affected by task parametersMultiple developmental studies typically using an audi-tory oddball paradigm have reported changes in N1amplitude latency and distribution across childhoodthe general pattern of results indicates a smaller laterand more posteriorly distributed N1 in children as com-pared with adults (Albrecht Suchodoletz amp Uwer 2000Pang amp Taylor 2000 Ponton Eggermont Kwongamp Don 2000 Bruneau Roux Guerin Barthelemy ampLelord 1997 Oades et al 1997 Tonnquist-Uhlen Borgamp Spens 1995 Fuchigami et al 1993 Martin BarajasFernandez amp Torres 1988 Goodin Squires Hendersonamp Starr 1978) Moreover in the waveforms of childrenyounger than age 7ndash9 years it has been reported that theN1 is sometimes lsquolsquounreliablersquorsquo or difficult to identify(Ponton et al 2000 Bruneau et al 1997 Martin et al1988 Daruna amp Rau 1987) perhaps because the gen-erators are not present or are not optimally oriented(Bruneau et al 1997 p 36)

Indeed a number of previous reports in the literaturehave documented that the auditory ERP componentrytypical in adults is not always observed in young chil-dren in fact there is some consistency across studies inreporting an early positivity peaking at about 100 msecfollowed by a negativity peaking at about 250 msec inmismatch negativity auditory refractory period oddballand passive listening paradigms (eg Albrecht et al2000 Ponton et al 2000 Ceponiene Cheour ampNaatanen 1998 Bruneau et al 1997 Sharma KrausMcGee amp Nicol 1997 Paetau Ahonen Salonen amp Sams1995 Korpilahti amp Lang 1994 Courchesne 1990) Thatis in these reports lsquolsquothe polarity of the first wave isoppositersquorsquo in children as compared with adults (Albrecht


et al 2000 p 2271) These early positivities may berelated to the positive component observed here

We speculate that the observed positivity is not areversal in polarity of the N1 in children Rather weinterpret this finding as an absence of an N1 in childrenin the present paradigm (see also Ponton et al 2000Bruneau et al 1997 Martin et al 1988 Daruna amp Rau1987) There is growing evidence to suggest that the N1may be absent in children particularly when stimuli arepresented at short interstimulus intervals (Ceponieneet al 1998 Rojas Walker Sheeder Teale amp Reite 1998Paetau et al 1995) Probe stimuli were played randomlyevery 300ndash800 msec in the present paradigm in additionto the continuous presentation of the 2 stories Thus wehypothesize that the absence of an N1 in the childrenrsquoswaveforms may arise from a saturated auditory systemdue to the simultaneous presentation of the 2 storiesand 2 probe types We speculate that the complexity andexcessive demands of the taskmdashthe amount and contin-uous nature of the auditory inputmdashmay be the criticalfactor in the nonelicitation of the N1 Indeed that12 children and 3 adults failed to answer 8 out of the10 comprehension questions correctly and were exclud-ed from the analyses provides corroborative evidencefor the difficulty and demands of the task

Considering the effects of attention both linguisticand nonlinguistic probes elicited a greater positivitywhen played from the attended sound source thanwhen played from the unattended sound source Thusthe typical effect of an increase in amplitude withattention was observed in childrenmdashbut on the firstclear positivity in the waveform rather than on a nonex-istent first negativity that is attention in both groupsacted to enhance an early component in the waveform(eg Woldorff et al 1993 Woods et al 1984) Both theanterior medial distribution of the attention effect andthe onset at 100- to 125-msec were similar for linguisticand nonlinguistic probes Thus unlike in adults therewas little evidence that the spatial linguistic attentionalspotlight resulted in differential processing of linguisticand nonlinguistic probe stimuli in children This may beanother indication that there is a narrowing of atten-tional focus beyond 8 years old (eg Berman amp Fried-man 1995)

It is possible that the enhanced positive attentioneffects observed here in children might be related tothe enhanced positivity from 200 to 300 msec forattended second formant tone probes reported in aprevious study with adults (Woods et al 1984) Thiseffect showed a symmetrical distribution and was largerat the vertex than at lateral sites the present effect has asimilar medial and bilateral distribution However theprevious effect was attributed in part to the lsquolsquoacousticdeviancersquorsquo of the second formant probes in terms ofhigher pitch (Woods et al 1984 p 774) whereas thepresent probes were matched on spectral frequency andwere not deviant in this or a similar acoustic way

That the effects of attention were observed on a broadpositive wave in children as compared with a well-defined negative peak in adults is consistent with previ-ous ERP research indicating a shift from more broad tomore narrow components over developmental time(eg Thomas amp Nelson 1996) and may be related tobehavioral findings indicating a long time course fordevelopment of attention skills extending into adoles-cence (eg Lane amp Pearson 1982) indeed our ownbehavioral results indicate that adults were better thanchildren at the present task Previous developmentalERP studies of attention effects have reported no Nde in6-year-olds (Satterfield et al 1990) no early Nd effectfor CV stimuli but a typical Nd effect for pure tones inchildren with mean age of 810 years (Berman amp Fried-man 1995) and typical Nd attention effects in adoles-cents (Berman amp Friedman 1995 Lovrich amp Stamm1983 Loiselle et al 1980 Zambelli et al 1977) Acrossstudies these results suggest a long time course ofdevelopment across childhood for typical ERP effectsof sustained endogenous attention To investigate thepossibility that only our youngest participants werecontributing to the observed positive attention effectwe reviewed the individual data from each child andconfirmed that 6- 7- and 8-year-olds all evidenced thepositivity rather than the typical negativities (This isconsistent with the lack of correlation between age andan ERP difference measure in this group) It is possiblethat both the young age of our participants and therelative complexity of our task contributed to our obser-vation of enhanced positivities with attention in childrenFurther studies with younger and older children shouldclarify the course of development of the ERP waveformand sustained attention effects as well as indicating atwhat point in developmental time ERPs elicited within thepresent paradigm appear to be mature

Although other authors have reported a relationshipbetween auditory attention skills and reading ability(eg Asbjoslashrnsen amp Bryden 1998 Taub et al 1994Hugdahl amp Andersson 1987) we found no evidence ofa correlation between our standardized behavioral mea-sures of reading and our ERP measure of attention inchildren consistent with the results of an auditory ERPstudy with adolescents (Lovrich amp Stamm 1983) Thepresent lack of relation may be because of the veryselective nature of the group of children Only thosechildren who scored at least 8 of 10 on the comprehen-sion questions following the ERP task were included inanalyses Behavioral research has shown that the abilityto shift attention effectively as required in the presenttask develops across the childhood years (Pearson ampLane 1991 Andersson amp Hugdahl 1987 Hiscock ampKinsbourne 1980 Geffen amp Wale 1979) By includingonly those children who performed well on our ERPattention task (a necessity to ensure that participantswere indeed attending during the experiment) we mayhave biased our sample to include more developmen-


tally advanced children or better readers indeed scoreson the behavioral tests were mostly above average Inrestricting the range of participants significant correla-tions may have become less likely

Summary and Conclusion Children and Adults

A complex dichotic listening task was successful inindexing the effects of sustained auditory selective spa-tial attention on ERPs in both children and adultsInterestingly these attention effects onset at about thesame time (100 msec) in children and adults Howeverthe differences between groups were prominent Themorphology of the auditory ERP waveforms and thepolarity of the attention effects in children and adultswere markedly different Only adults evidenced left-lateralized effects and only adults showed anteriorposterior distributional differences for linguistic andnonlinguistic probes In particular adults evidencedmore posterior (cf Berman amp Friedman 1995) andlonger duration attention effects for linguistic thannonlinguistic probes The lack of this sort of specializedor distinctive processing in children suggests a longermaturational time course for the processing of linguisticstimuli at least in the context of the present task (seealso Berman amp Friedman 1995) Overall this pattern ofresults suggests further refinement of the attentionalnetworks and skills reflected in ERPs elicited by this taskespecially with regard to the linguistic probes beyondthe age of 8 years

METHODS

Subjects

The final sample included 22 children (range = 78ndash107months mean = 92 months [7 years 9 months] SD = 1014 girls) and 22 adults (range = 228ndash376 months mean =261 months [21 years 9 months] SD = 37 11 women)Participants were right-handed (Oldfield 1971) mono-lingual English speakers with no history of neurological orlanguage disorders All were volunteers paid for theirparticipation The socioeconomic status of childrenrsquosfamilies ranged from lower middle to upper class on theHollingshead Index of Social Position with a middle-classaverage Adults reported normal or corrected-to-normalvision and hearing and all children passed a standardhearing screening (1 2 and 4 kHz under headphones)visual screening for binocular acuity (KindergartenSnellen chart) and oralndashmotor screening

Participants were included in the final analyses onlyif they met behavioral and electrophysiological criteriaOnly participants with a score of at least 8 out of 10on the posttest comprehension questions related tothe attended story (see details below) were includedin the analyses to assure that each participant hadattended to a minimum of 80 of the story segments

In addition only participants with acceptable ERP dataand an adequate number of trials in each condition ofinterest (see details below) were included in analysesNine children were excluded based on failing someaspect of the behavioral screening (eg poor hearingleft-handed) 12 children did not answer at least 8 outof 10 comprehension questions correctly and wereexcluded from the final sample an additional 36children had poor electrophysiological data (eg notenough trials after eye movement rejection noisy) andwere not included in analyses An additional 9 adultswere excluded because of noisy ERP data (primarilycontamination from alpha) and 3 adults were excludedbased on incorrect responses to the comprehensionquestions

Behavioral Testing

As part of a wide battery of testing all children1 weregiven the Word Identification and Word Attack subtestsof the Woodcock Reading Mastery TestsmdashRevised(WRMT Woodcock 1987) and the Peabody PictureVocabulary Test (PPVT Dunn amp Dunn 1997) Behavioraland ERP testing occurred in different sessions separatedby no more than 35 days

Event-related Potential Stimuli

Narratives

Two stories were digitally recorded (16 bit 22 kHz)using an ElectroVoice 1750 microphone connected to aMacintosh computer running a sound editing program(SoundEdit 16 Version 2) The beginning portion of MyFatherrsquos Dragon (Gannett 1948) was read by a femalespeaker and lsquolsquoThe Merry Little Breezes Save the GreenMeadowsrsquorsquo and a portion of lsquolsquoThe Stranger in the GreenForestrsquorsquo from Mother West Windrsquos Animal Friends (Bur-gess 1912) were read by a male speaker the latter 2short stories were read consecutively and connectedby the phrase lsquolsquoThe next dayrsquorsquo to form one story Thestories were distinctively different in both style andcontent and each was read at a slow-normal rate Thestories were edited for outside noises speech errorsand pauses Pauses were edited so as not to exceed15 sec to lessen the opportunity for attention switchesto the other channel and to lessen the likelihood thatthe probe stimuli would be heard in the absence of thestory After editing each story was segmented into10 sections of the following lengths 90 30 60 90 3060 60 30 60 and 92 sec (in order) The last segmentwas slightly longer to end each story at a good pointEach segment of each story was pasted consecutivelyinto the alternate channels of a stereo file That is in astereo file segment 1 of story 1 was pasted into the rightchannel and segment 1 of story 2 was pasted into the leftchannel from 0 to 90 sec segment 2 of story 1 was


pasted into the left channel and segment 2 of story 2 waspasted into the right channel from 90 to 120 sec and soon Within each time window the 2 stories were nor-malized to control for amplitude differences Overalleach story switched from one channel to the other at90 120 180 270 300 360 420 450 and 510 sec and theentire experiment lasted about 10 min Stories wereplayed at a comfortable listening level of 65 dB SPL (A-weighted) on average

Probe Stimuli

The linguistic stimulus was a digitized (16 bit 22 kHz)token of the syllable ba spoken in a female voice(different from the female storytellerrsquos) with 100 msecduration The nonlinguistic stimulus was created byediting (CoolEdit) the ba stimulus into 4- to 6-msecsegments between zero crossings Those segments werethen randomly reordered to create a sound that was notrecognizably speech-like some segments were also re-versed as part of the reordering process Thus thenonlinguistic stimulus retained all the primitive acousticcharacteristics of the linguistic stimulus indeed thefrequency spectra of the 2 types of probe stimuli wereidentical The nonlinguistic stimulus sounded like abuzz Each monostimulus could be played from eitherthe left or the right channel thus there were 4 probestimuli ba right ba left buzz right and buzz left Probeswere presented randomly from each speaker but acrossthe experiment an equal number (n = 206) of linguisticand nonlinguistic probes were presented in the at-tended and unattended channels

Visual Fixation Stimuli

Visual stimuli presented on a monitor directly in frontof the participant consisted of 25 25-in coloredimages of various cartoon characters pointing either tothe left or to the right These images served both as afixation point and a reminder of which channel to attend

General Design

A pair of speakers one speaker to either side of theseated participant played the stories and the probestimuli simultaneously2 At the same time a monitor infront of the participant displayed the pointing cartooncharacters (see Figure 5 for the general design) A trialwas as follows At time 0 a cartoon character pointingeither to the left or to the right appeared at the center ofthe screen At time 1 sec the stories began to play fromthe speakers (one story from the left the other from theright speaker simultaneously) At time 4 sec the probestimuli began to play from the speakers and were playedrandomly from the left and right speaker every 300ndash800 msec About 2 sec before the end of each story seg-ment the probe stimuli ceased then the stories stoppedthe cartoon character on the screen was replaced byanother character pointing in the other direction 1 seclater the stories resumed in the opposite channels

General Procedure

The procedures were explained and any questions wereaddressed after which adults and parents signed aconsent form and children signed an assent form After

Figure 5 General design of

the experiment The

participant is seated in a

comfortable chair withspeakers equidistant to the left

and the right and is instructed

to attend to only one narrative

(presented in either a womanrsquosor a manrsquos voice) One

narrative plays from one

speaker at the same time thatthe other narrative plays from

the other speaker the

narratives switch back and

forth between the 2 speakersperiodically throughout the

experiment A monitor directly

in front of the participant

shows a cartoon characterpointing in one direction this

serves as both a fixation point

and a reminder of which sideto attend Simultaneously

ERPs are recorded to linguistic

(ba) and nonlinguistic (buzz)

probe stimuli superimposedover each narrative


fitting of the electrode cap (see below) participants wereseated in a comfortable chair in a sound-attenuatingelectrically shielded booth Small speakers were locatedon either side of the participant (equidistant) with themonitor 57 in in front of the participant A practicesession preceded the actual test session and instructionswere given as part of the practice Attendance to theMother West Wind story (male voice left side start) or theMy Fatherrsquos Dragon story (female voice right side start)was counterbalanced across all participants (children 11left start and 11 right start adults 10 left start and 12right start) The voice that started the practice sessioncorresponded to the attended voice in the actual testsession During the practice each element of the ex-periment was introduced singly (refer to the Appendixfor further details regarding the practice session)

Following the practice session any questions or con-cerns were addressed before the start of the experimentthe participant was again instructed not to move or leanfrom side to side (monitored continuously by a closed-circuit camera and by an experimenter sitting next to thechildren) and to listen carefully to be able to answer thequestions at the end and was reminded which side tolisten to first (by both a verbal label and a manual point)Subsequently the experiment was begun with a cartooncharacter pointing to the indicated side and the start ofthe stories

At the end of the experiment participants were read10 simple yes-or-no questions about salient aspects of thestory to which they had listened (indexing the main ideaof each segment) and 3 simple yes-or-no questions aboutsalient aspects of the story to which they had not beenattending Questions from the 2 stories were presentedintermixed and responses were recorded manually

ElectroencephalogramEvent-related PotentialRecording and Analysis

Electroencephalogram (EEG) was recorded from 29 tinelectrodes mounted in an elastic cap (Electro-Cap Inter-national) These included 3 midline sites (Fz Cz and Pz)and 13 pairs of lateral sites (FP12 F78 FT78 F34 FC56 C34 C56 T34 CT56 P34 T56 TO12 and O12refer to Figure 6) Electrodes were also placed beneaththe lower right eye and at the outer canthi of the left andright eyes to monitor eye movements (electrooculo-gram) in addition recordings from FP12 were used toreject trials that were contaminated by eyeblink artifactsActivity at the left mastoid was recorded during theexperiment but all on-line recordings were referencedto the right mastoid recordings were re-referenced toaveraged mastoids in the final data averaging Eye elec-trode impedances were maintained below 10 k mastoidelectrodes below 2 k and scalp electrodes below 3 k

The EEG was amplified with Grass 7P511 amplifiers(3 dB cutoff bandpass 001 to 100 Hz) and digitizedon-line (sampling rate 4 msec) Off-line separate ERPs to

the 4 types of probe stimuli (linguistic attended linguis-tic unattended nonlinguistic attended nonlinguisticunattended) were averaged for each subject at eachelectrode site over a 500-msec epoch using a 100-msecprestimulus-onset baseline Trials contaminated by eyemovements muscular activity or electrical noise werenot included in analyses Standard artifact rejectionparameters were initially employed including routinesto identify reversals of polarity across the electrodebelow the eye and FP1 (to identify blinks and verticaleye movements) amplitude fluctuations at the horizon-tal eye channels (to identify lateral eye movements)amplifier blocking and high amplitude activity (to iden-tify noise movement) data were subsequently carefullyanalyzed on an individual basis for stricter artifact rejec-tion as necessary A minimum of 30 artifact-free trials ineach of the 4 main conditions was imposed The averagenumber of useable trials in the attended linguisticcondition was 1400 plusmn 517 for adults and 945 plusmn 390for children and the average number of useable trials inthe unattended linguistic condition was 1383 plusmn 531 foradults and 977 plusmn 382 for children The average numberof useable trials in the attended nonlinguistic conditionwas 1382 plusmn 533 for adults and 957 plusmn 360 for childrenand the average number of useable trials in the unat-tended nonlinguistic condition was 1400 plusmn 525 foradults and 956 plusmn 350 for children

Statistical Analyses

To fully characterize the early components of the ERPwaveform in adults local peak amplitude and latencywere measured in 3 time windows positive peak 50ndash150 msec (P1) negative peak 100ndash200 msec (N1 laten-cy only) and positive peak 200ndash400 msec (P2) A peak-to-peak amplitude measurement was taken between P1and N1 to more accurately describe the distribution ofthe N13 Peak measures were subjected to a 6-waymixed-design repeated measures analysis of variance(ANOVA) as described below As data from individualchildren did not evidence peaks clearly defined enoughto permit this type of measurement analyses of child-renrsquos ERPs were limited to the mean amplitude meas-ures described below

To quantify the effect of attention in children andadults the mean amplitude of ERPs to attended andunattended probes was measured within the 100- to250-msec time window in children and within the 100-to 175- and 100- to 250-msec epochs in adults (as wellas within the 175- to 250-msec epoch for comparativepurposes) Six-way mixed-design repeated ANOVAs wereperformed on the mean amplitude data for each agegroup Within-subject factors included attention (2 pos-sible levels attended unattended) probe type (2 pos-sible levels linguistic nonlinguistic) hemisphere (2possible levels left right) laterality (2 possible levelslateral medial) and anteriorposterior (6 possible levels


frontal fronto-temporal temporal central parietal oc-cipital refer to Figure 6) The single between-subjectsfactor was start side (2 possible levels left start rightstart) Because there were no significant effects orinteractions involving start side and attention in omni-bus analyses this factor was removed from subse-quent analyses of mean and peak data as reportedThe GreenhousendashGeisser correction was applied to allwithin-subjects measures with more than 2 levels

In addition within each group analyses by consecu-tive 25-msec epochs from 0 to 300 msec poststimulusonset were conducted with mean amplitude data tomore precisely temporally locate the onset of the atten-tion effects significant results involving attention across3 consecutive epochs were considered an indication ofonset in the earliest epoch

Further to directly compare the distributions of theattention effects observed in children and adults meanamplitude data were normalized as follows For each

group and probe type the grand mean and standarddeviation across electrode site and subject were calcu-lated Then for each subject at each site the grandmean was subtracted from the mean amplitude anddivided by the standard deviation

Finally for the group of children Pearsonrsquos correla-tions were calculated to investigate relationships amongour behavioral and electrophysiological measures spe-cifically between the behavioral measures of readingand the ERP attention effect

APPENDIX PRACTICE SESSION

First practice segment attend to male voice conditionMale voice from left speaker saying lsquolsquoIn this experimentyour job is to listen to a story I will be telling the storyso you will be listening to my voice You need to listen tothe story very carefully because we will ask you somequestions about the story at the end While you listen tothe story you must sit as still as you can and lookstraight ahead watching the screen in front of you Onthe screen you can see a cartoon character pointing oneway or the other Right now Woody is pointing to theleft the side you are hearing my voice come from Everyonce in a while a different character will appear on thescreen pointing in the other direction [Peter Pan ap-pears on the screen pointing to the right] and my voicewill change to the other side rsquorsquo Second practice seg-ment male voice from the right speaker lsquolsquo like thisSo now my voice is coming from the right side and PeterPan is pointing to that side It is very important that youlook straight ahead at the screen and watch the cartooncharacters while you are listening to the story becausethey will tell you which side to listen to That way youcan follow my voice as I tell the story from one side tothe other and you wonrsquot miss any of the storyrsquorsquo Thirdpractice segment male voice from the left speakerlsquolsquoDid you see the cartoon character change and pointin the other direction Did you follow my voice to theother side Now my voice is coming from the left andPuumba is pointing in that direction When you can hearonly my voice it is not too hard to pay attention to whatIrsquom saying but during the experiment there will beanother voice telling another story at the same time thatI am telling my story It will sound something like this[add female voice from right speaker] So while I amtalking another person will also be talking at the sametime Remember that your job is to listen to my voicetelling the story You can just ignore the other voice youdonrsquot need to pay any attention to the other voice Youneed to listen to my voice so that you donrsquot miss part ofthe story Remember to ignore the other voice andfollow my voice from side to side as I tell the storyrsquorsquoFinal practice segment female voice from the leftspeaker and male voice from the right speaker sayinglsquolsquoDid you see the cartoon character change and point inthe other direction Are you ignoring the other voice

Figure 6 Schematic representation of the electrode montage and thefactors used in analyses At the top 6 levels of the anteriorposterior

factor are illustrated At the bottom 2 levels of the lateralmedial factor

and 2 levels of the hemisphere factor are indicated


Good you followed my voice back to the other sidewhere Bianca the mouse is pointing and you are payingattention to just my voice Sometimes during the exper-iment there will also be other sounds your job is just tolisten to my voice telling the story The other sounds willsound like this [introduce probe stimuli from left andright speakers] You need to concentrate very hard andpay attention to my voice telling the story even if thereare other sounds at the same time So your job is tolisten to my voice telling the story and follow the story asit changes from one side to the other the cartooncharacters will help you know which side to listen toby pointing in that direction Listen carefully to the storybecause we will ask you some questions about the storyat the end Are you ready to hear the story now Thename of the story is Mother West Windrsquos AnimalFriends and it will start on the other side in just aminutersquorsquo

Acknowledgments

This research was supported by NIH grants DC00481 to HJN(NIDCD) HD08598 to DC (NICHD) and DC0529101 to LDS(NIDCD) We thank the children and parents as well as thecollege students who participated in this study We also thankMark Dow for his excellent storytelling voice his patience andhis technical assistance Ray Vukcevich and Paul Compton fortheir programming and technical expertise and WendySkendzel Libbey White Cheryl Capek Jennifer Woods andCourtney Stevens for their help in running subjects in the ERPparadigm We also thank the students in the CommunicationDisorders and Sciences Department of the University ofOregon including Kim Stahlnecker Rebecca Andes Jill DuthieSuzanne Israels Christine Hinshon Kris Marion Karen BohlinAmy Letts and especially Jessica Fanning who conducted thebehavioral testing under the supervision of Dr MarilynNippold Finally we thank the anonymous reviewers for theirhelpful comments and suggestions

Reprint requests should be sent to Donna Coch Departmentof Education Dartmouth College 203 Raven House HanoverNH 03755 or via e-mail donnajcochdartmouthedu

Notes

1 With the exception of one 7-year-old boy2 Stimuli were presented over speakers because usingheadphones in combination with the electrode cap withchildren is difficult and because using free-field auditory stimuliseemed a more ecologically valid measure3 Because of the nature of the waveforms the mean am-plitude measure of N1 did not accurately represent the dis-tribution of the component (eg refer to Figure 1A whereinN1 appears to be larger at lateral and posterior sites becausethe overall waveform tends to be more negative at these sites)The attention effects are best revealed by the mean amplitudemeasures but the distribution of the component is bestdescribed with the peak-to-peak measure results are reportedaccordingly

REFERENCES

Albrecht R Suchodoletz W v amp Uwer R (2000) Thedevelopment of auditory evoked dipole source activity from

childhood to adulthood Clinical Neurophysiology 1112268ndash2276

Alho K Medvedev S V Pakhomov S V Roudas M STervaniemi M Reinikainen K Zeffiro T amp Naatanen R(1999) Selective tuning of the left and right auditorycortices during spatially directed attention Cognitive BrainResearch 7 335ndash341

Andersson B amp Hugdahl K (1987) Effects of sex age andforced attention on dichotic listening in children Alongitudinal study Developmental Neuropsychology 3191ndash206

Asbjoslashrnsen A E amp Bryden M P (1998) Auditory attentionalshifts in reading-disabled students Quantification ofattentional effectiveness by the Attentional Shift IndexNeuropsychologia 36 143ndash148

Berman S amp Friedman D (1995) The development ofselective auditory attention as reflected by event-relatedbrain potentials Journal of Experimental Child Psychology59 1ndash31

Bruneau N Roux S Guerin P Barthelemy C amp Lelord G(1997) Temporal prominence of auditory evoked potentials(N1 wave) in 4ndash8-year-old children Psychophysiology 3432ndash38

Burgess T W (1912) Mother West Windrsquos animal friendsNew York Grosset amp Dunlap

Ceponiene R Cheour M amp Naatanen R (1998)Interstimulus interval and auditory event-related potentialsin children Evidence for multiple generatorsElectroencephalography and Clinical Neurophysiology108 345ndash354

Coch D Grossi G Coffey-Corina S Holcomb P J ampNeville H J (2002) A developmental investigation ofERP auditory rhyming effects Developmental Science 5467ndash489

Courchesne E (1990) Chronology of postnatal human braindevelopment Event-related potentials positron emissiontomography myelinogenesis and synaptogenesis studiesIn J R Rohrbaugh R Parasuraman amp R Johnson (Eds)Event-related brain potentials Basic issues andapplications (pp 210ndash241) New York OxfordUniversity Press

Daruna J H amp Rau A E (1987) Development of the latecomponents of auditory brain potentials from earlychildhood to adulthood In R Johnson J W Rohrbaugh ampR Parasuraman (Eds) Current trends in event-relatedpotential research EEG Supplement 40 (pp 590ndash595)Amsterdam Elsevier

Davis E P Bruce J Snyder K amp Nelson C A (2003) Thex-trials Neural correlates of an inhibitory control task inchildren and adults Journal of Cognitive Neuroscience 15432ndash443

Dehaene-Lambertz G Dehaene S amp Hertz-Pannier L(2002) Functional neuroimaging of speech perception ininfants Science 298 2013ndash2015

Dickstein P W amp Tallal P (1987) Attentional capabilities ofreading-impaired children during dichotic presentation ofphonetic and complex nonphonetic sounds Cortex 23237ndash249

Doyle A-B (1973) Listening to distraction A developmentalstudy of selective attention Journal of Experimental ChildPsychology 15 100ndash115

Dunn L M amp Dunn L M (1997) Peabody PictureVocabulary Test (3rd ed) Circle Pines MN AmericanGuidance Service

Elliott L L (1979) Performance of children aged 9 to17 years on a test of speech intelligibility in noise usingsentence material with controlled word predictabilityJournal of the Acoustical Society of America 66 651ndash653


Fuchigami T Okubo O Fujita Y Okuni M Noguchi Y ampYamada T (1993) Auditory event-related potentials andreaction time in children Evaluation of cognitivedevelopment Developmental Medicine and ChildNeurology 35 230ndash237

Gannett R S (1948) My fatherrsquos dragon New York RandomHouse

Geffen G amp Sexton M A (1978) The development ofauditory strategies of attention Developmental Psychology14 11ndash17

Geffen G amp Wale J (1979) Development of selectivelistening and hemispheric asymmetry DevelopmentalPsychology 15 138ndash146

Goodin D S Squires K C Henderson B H amp Starr A(1978) Age-related variations in evoked potentials toauditory stimuli in normal human subjectsElectroencephalography and Clinical Neurophysiology 44447ndash458

Hansen J C amp Hillyard S A (1980) Endogenous brainpotentials associated with selective auditory attentionElectroencephalography and Clinical Neurophysiology 49277ndash290

Hansen J C Dickstein P W Berka C amp Hillyard S A(1983) Event-related potentials during selective attention tospeech sounds Biological Psychology 16 211ndash224

Hillyard S A amp Picton T W (1987) Electrophysiology ofcognition Selective attention In F Plum (Ed) Handbook ofphysiology section I The nervous system Higher functionsof the brain Part II (pp 534ndash546) Bethesda MD AmericanPhysiological Society

Hillyard S A Hink R F Schwent V L amp Picton T W(1973) Electrical signs of selective attention in the humanbrain Science 182 177ndash180

Hink R F amp Hillyard S A (1976) Auditory evoked potentialsduring selective listening to dichotic speech messagesPerception and Psychophysics 29 236ndash242

Hink R F Hillyard S A amp Benson P J (1978) Event-relatedbrain potentials and selective auditory attention to acousticand phonetic cues Biological Psychology 6 1ndash16

Hink R F van Voorhis S T Hillyard S A amp Smith T S(1977) The division of attention and the human auditoryevoked potential Neuropsychologia 15 597ndash605

Hiscock M amp Kinsbourne M (1980) Asymmetries of selectivelistening and attention switching in children DevelopmentalPsychology 16 70ndash82

Hugdahl K amp Andersson B (1987) Dichotic listening andreading acquisition in children A one-year follow-upJournal of Clinical and Experimental Neuropsychology 9631ndash649

Hugdahl K Broslashnnick K Kyllingsbaeligk S Law I Gade Aamp Paulson O B (1999) Brain activation during dichoticpresentations of consonantndashvowel and musicalinstrument stimuli A 15O-PET study Neuropsychologia37 431ndash440

Hugdahl K Law I Kyllingsbaeligk S Broslashnnick K Gade A ampPaulson O B (2000) Effects of attention on dichoticlistening An 15O-PET study Human Brain Mapping 1087ndash97

Korpilahti P amp Lang H A (1994) Auditory ERP componentsand mismatch negativity in dysphasic childrenElectroencephalography and Clinical Neurophysiology 91256ndash264

Lane D M amp Pearson D A (1982) The development ofselective attention Merrill-Palmer Quarterly 28 317ndash337

Loiselle D L Stamm J S Maitinsky S amp Whipple S C(1980) Evoked potential and behavioral signs of attentivedysfunctions in hyperactive boys Psychophysiology 17193ndash201

Lovrich D amp Stamm J S (1983) Event-related potential andbehavioral correlates of attention in reading retardationJournal of Clinical Neuropsychology 5 13ndash37

Luck S J (1998) Neurophysiology of selective attention In HPashler (Ed) Attention (pp 257ndash295) East Sussex UKPsychology Press

Maccoby E E amp Konrad K W (1966) Age trends in selectivelistening Journal of Experimental Child Psychology 3113ndash122

Martin L Barajas J J Fernandez R amp Torres E (1988)Auditory event-related potentials in well-characterizedgroups of children Electroencephalography and ClinicalNeurophysiology 71 375ndash381

Mills D L Coffey-Corina S A amp Neville H J (1993)Language acquisition and cerebral specialization in20-month-old infants Journal of Cognitive Neuroscience 5317ndash334

Molfese D L amp Molfese V J (1988) Right-hemisphereresponses from preschool children to temporal cues tospeech and nonspeech materials Electrophysiologicalcorrelates Brain and Language 33 245ndash259

Molfese D L Freeman R B amp Palermo D S (1975) Theontogeny of brain lateralization for speech and nonspeechstimuli Brain and Language 2 356ndash368

Naatanen R (1979) Early selective attention effects on theevoked potential A critical review and reinterpretationBiological Psychology 8 81ndash136

Naatanen R (1982) Processing negativity An evoked-potentialreflection of selective attention Psychological Bulletin 92605ndash640

Naatanen R amp Picton T (1987) The N1 wave of the humanelectric and magnetic response to sound A review and ananalysis of the component structure Psychophysiology 24375ndash425

Neville H J amp Bavelier D (2000) Specificity and plasticityin neurocognitive development in humans InM S Gazzaniga (Ed) The new cognitive neurosciences(2nd ed pp 83ndash98) Cambridge MIT Press

Oades R D Dittmann-Balcar A amp Zerbin D (1997)Development and topography of auditory event-relatedpotentials (ERPs) Mismatch and processing negativity inindividuals 8ndash22 years of age Psychophysiology 34677ndash693

Oldfield R C (1971) The assessment and analysis ofhandedness The Edinburgh Inventory Neuropsychologia9 97ndash113

Paetau R Ahonen A Salonen O amp Sams M (1995)Auditory evoked magnetic fields to tones and pseudowordsin healthy children and adults Journal of ClinicalNeurophysiology 12 177ndash185

Pang E W amp Taylor M J (2000) Tracking the developmentof the N1 from age 3 to adulthood An examination ofspeech and non-speech stimuli Clinical Neurophysiology111 388ndash397

Pearson D A amp Lane D M (1991) Auditory attentionswitching A developmental study Journal of ExperimentalChild Psychology 51 320ndash334

Pena M Maki A Kovacic D Dehaene-Lambertz GKoizumi H Bouquet F amp Mehler J (2003) Sounds andsilence An optical topography study of language recognitionat birth Proceedings of the National Academy of SciencesUSA 100 11702ndash11705

Petkov C I Kang X Alho K Bertrand O Yund E W ampWoods D L (2004) Attentional modulation of humanauditory cortex Nature Neuroscience 7 658ndash663

Poldrack R A Temple E Protopapas A Nagarajan STallal P Merzenich M amp Gabrieli J D E (2001) Relationsbetween the neural bases of dynamic auditory processing


and phonological processing Evidence from fMRI Journalof Cognitive Neuroscience 13 687ndash697

Ponton C W Eggermont J J Kwong B amp Don M (2000)Maturation of human central auditory system activityEvidence from multi-channel evoked potentials ClinicalNeurophsyiology 111 22ndash236

Ridderinkhof K R amp van der Stelt O (2000) Attention andselection in the growing child Views derived fromdevelopmental psychophysiology Biological Psychology 5455ndash106

Rojas D C Walker J R Sheeder J L Teale P D amp ReiteM L (1998) Developmental changes in refractoriness ofthe neuromagnetic M100 in children NeuroReport 91543ndash1547

Rueda M R Fan J McCandliss B D Halparin J D GruberD B Lercari L P amp Posner M I (2004) Development ofattentional networks in childhood Neuropsychologia 421029ndash1040

Sanders L D Spezio M L Dow M Weaver JBaufman D J Woods D L amp Neville H J (2001)Attention to auditory location and pitch A behavioralERP and fMRI study [abstract] Annual Meeting of theCognitive Neuroscience Society New York CognitiveNeuroscience Society

Satterfield J H Schell A M amp Nicholas T (1994)Preferential neural processing of attended stimuli inattention-deficit hyperactivity disorder and normal boysPsychophysiology 31 1ndash10

Satterfield J H Schell A M Nicholas T amp Backs R W(1988) Topographic study of auditory event-relatedpotentials in normal boys and boys with attention deficitdisorder with hyperactivity Psychophysiology 25591ndash606

Satterfield J H Schell A M Nicholas T W Satterfield B Tamp Freese T E (1990) Ontogeny of selective attentioneffects on event-related potentials in attention-deficithyperactivity disorder and normal boys BiologicalPsychiatry 28 879ndash903

Segalowitz S J amp Cohen H (1989) Right hemisphereEEG sensitivity to speech Brain and Language 37220ndash231

Sexton M A amp Geffen G (1979) Development of threestrategies of attention in dichotic monitoringDevelopmental Psychology 15 299ndash310

Sharma A Kraus N McGee T J amp Nicol T G (1997)Developmental changes in P1 and N1 central auditoryresponses elicited by consonantndashvowel syllablesElectroencephalography and Clinical Neurophysiology104 540ndash545

Shtyrov Y Kujala T Palva S Ilmoniemi R J amp Naatanen R(2000) Discrimination of speech and complex nonspeechsounds of different temporal structure in the left and rightcerebral hemispheres Neuroimage 12 657ndash663

Simos P G Molfese D amp Brenden R A (1997) Behavioraland electrophysiological indices of voicingndashcuediscrimination Laterality patterns and development Brainand Language 57 122ndash150

Szymanski M D Yund E W amp Woods D L (1999) Humanbrain specialization for phonetic attention NeuroReport 101605ndash1608

Takeshita K Nagamine T Thuy D H D Satow TMatsuhashi M Yamamoto J Takayama M Fujiwara Namp Shibasaki H (2002) Maturational change of parallelauditory processing in school-aged children revealedby simultaneous recording of magnetic and electriccortical responses Clinical Neurophysiology 1131470ndash1484

Taub C F Fine E amp Cherry R S (1994) Finding a linkbetween selective auditory attention and reading problemsin young children A preliminary investigation Perceptualand Motor Skills 78 1153ndash1154

Teder W Kujala T amp Naatanen R (1993) Selection of speechmessages in free-field listening NeuroReport 5 307ndash309

Teder-Salejarvi W A Hillyard S A Roder B amp Neville H J(1999) Spatial attention to central and peripheral auditorystimuli as indexed by event-related potentials CognitiveBrain Research 8 213ndash227

Thomas K M amp Nelson C A (1996) Age-related changes inthe electrophysiological response to visual stimulus noveltyA topographical approach Electroencephalography andClinical Neurophysiology 98 294ndash308

Tonnquist-Uhlen I Borg E amp Spens K E (1995)Topography of auditory evoked long-latency potentials innormal children with particular reference to the N1component Electroencephalography and ClinicalNeurophysiology 95 34ndash41

Trejo L Ryan-Jones D L amp Kramer A F (1995) Attentionalmodulation of the mismatch negativity elicited by frequencydifferences between binaurally presented tone burstsPsychophysiology 32 319ndash328

Woldorff M Hansen J C amp Hillyard S A (1987)Evidence for effects of selective attention in the mid-latencyrange of the human auditory event-related potentialIn R Johnson J W Rohrbaugh amp R Parasuraman (Eds)Current trends in event-related potential research EEGSupplement 40 (pp 146ndash154) Amsterdam Elsevier

Woldorff M G Gallen C C Hampson S A Hillyard S APantev C Sobel D amp Bloom F E (1993) Modulation ofearly sensory processing in human auditory cortex duringauditory selective attention Proceedings of the NationalAcademy of Sciences USA 90 8722ndash8726

Woldorff M G amp Hillyard S A (1991) Modulation of earlyauditory processing during selective listening to rapidlypresented tones Electroencephalography and ClinicalNeurophysiology 79 170ndash191

Woodcock R W (1987) The Woodcock Reading MasteryTestsmdashRevised Circle Pines MN American GuidanceService

Woods D L (1990) The physiological basis of selectiveattention Implications of event-related potential studies InJ W Rohrbaugh R Parasuraman amp J R Johnson (Eds)Event-related potentials Basic issues and applications(pp 178ndash209) New York Oxford University Press

Woods D L amp Alain C (1993) Feature processing duringhigh-rate auditory selective attention Perception andPsychophysics 53 391ndash402

Woods D L Alho K amp Algazi A (1991) Brain potential signsof feature processing during auditory selective attentionNeuroReport 2 189ndash192

Woods D L Hillyard S A amp Hansen J C (1984)Event-related brain potentials reveal similar attentionalmechanisms during selective listening and shadowingJournal of Experimental Psychology Human Perceptionand Performance 10 761ndash777

Zambelli A J Stamm J S Maitinsky S amp Loiselle D L(1977) Auditory evoked potentials and selective auditoryattention in formerly hyperactive adolescent boys AmericanJournal of Psychiatry 134 742ndash747

Zatorre R J Evans A C Meyer E amp Gjedde A (1992)Lateralization of phonetic and pitch discrimination in speechprocessing Science 256 846ndash849

Zukier H amp Hagen J W (1978) The development ofselective attention under distracting conditions ChildDevelopment 49 870ndash873


(Naatanen 1979) Currently the Nd is thought to consistof both early (Nde) and late (Ndl) components the earlyfronto-central component may reflect rapid initial featur-al analysis of stimuli whereas the late anterior componentreflects further processing of the stimuli and mainte-nance of the attentional trace (eg see Woods 1990Naatanen 1982) The Nde is superimposed on the N1and there is some evidence that the N1 itself is actuallynot modulated by attention as first reported rather theeffects of the slow wave Nde are simply observed in thetypical N1 time window (Woods 1990 Naatanen 1982)Other evidence suggests that the N1M100 (the MEGcounterpart to the N100) itself is modulated by atten-tion (eg Woldorff et al 1993)

A small number of studies have attempted to employ asomewhat more ecologically valid design and haveinvestigated selective spatial attention effects in thecontext of a spoken narrative in most cases the typicalERP attention effects have been observed (Trejo Ryan-Jones amp Kramer 1995 Teder Kujala amp Naatanen 1993Woods Hillyard amp Hansen 1984 Hink amp Hillyard1976) For example irrelevant vowel probes within anattended speech passage elicit more negative ERPs thanthe same probes in the unattended passage in dichoticpresentations (Hink amp Hillyard 1976) In a more complexdesign Woods et al (1984) presented various speech andtone probe stimuli embedded in dichotically presentedprose passages Speech probes spatially and temporallycoincident with the attended passage as compared withthe same probes in the unattended passage elicited anenhanced negativity (presumably the Nd) beginning atabout 50 msec and lasting to 1000 msec In contrast onetype of tone probe elicited an enhanced positivity from200 to 300 msec most likely an attentional modulation ofa different component The authors suggested that thispattern of results indicated that stimulus selection duringlinguistic attention might be specifically tuned to speechsounds (Woods et al 1984)

The possibility of specialized attentional tuning forspeech stimuli in the context of spoken word processingraises a related question concerning specialized neuralprocessing for speech in general In Woods et alrsquos(1984) study the Nd elicited by the CVC speech probewas most prominent over the left hemisphere whereaseffects for other probes evidenced no hemisphericasymmetries Other ERP selective attention studies havealso reported a left-lateralized Ndl for speech-like (pho-nemic change) stimuli as compared with nonspeech(intensity change) stimuli (Szymanski Yund amp Woods1999) In contrast some studies have reported noasymmetries for either speech or tone stimuli witheach eliciting an Nd similar in morphology and distri-bution (although longer in latency for more complexspeech stimuli Hansen et al 1983) DevelopmentalERP studies have suggested that a left hemispherespecialization for speech either precedes the abilityto speak (Molfese Freeman amp Palermo 1975) or be-

gins to develop quite early with spoken language ex-perience (Neville amp Bavelier 2000 Mills Coffey-Corinaamp Neville 1993) Studies using fMRI in 3-month-olds(Dehaene-Lambertz Dehaene amp Hertz-Pannier 2002)and optical imaging in newborns (Pena et al 2003)have reported an early left hemisphere specializationfor speech stimuli Few studies have addressed lateral-ization in school-age children

Development of Selective Auditory AttentionBehavioral Evidence

Behavioral studies have indicated that auditory selectiveattention skills develop throughout childhood at leastuntil adolescence Both the abilities to selectively attendto relevant stimuli and to successfully ignore irrele-vant stimuli improve progressively with increasing ageacross childhood (Lane amp Pearson 1982 Hiscock ampKinsbourne 1980 Geffen amp Wale 1979 Sexton ampGeffen 1979 Geffen amp Sexton 1978 Zukier amp Hagen1978 Doyle 1973 Maccoby amp Konrad 1966) The abilityto shift attention quickly and effectively also developsacross childhood at least until adolescence (Pearson ampLane 1991 Andersson amp Hugdahl 1987 Hiscock ampKinsbourne 1980 Geffen amp Wale 1979) Further thereis some evidence that background noise creates greatermasking effects for younger children as compared withadolescents or adults (Elliott 1979) In a recent studywith children age 6ndash9 years alertness and conflict scoresevidenced change over time whereas orienting scoresdid not (Rueda et al 2004) illustrating that differentaspects of attention have different developmental timecourses

Interestingly there has been some attempt to linksustained selective auditory attention abilities to readingabilities (eg Taub Fine amp Cherry 1994) In partic-ular the ability to make attentional shifts under dich-otic listening conditions appears to be related toreading ability (Asbjoslashrnsen amp Bryden 1998 Hugdahl ampAndersson 1987) Alternately some authors have sug-gested that poorer readers have an attentional capac-ity limitation but not an inability to focus attention(Dickstein amp Tallal 1987)

Development of Selective Auditory AttentionEvent-related Potential Evidence

Although behavioral studies offer evidence for the de-velopment of selective auditory attention in school-agechildren there is very little comparable electrophysio-logical evidence from children in this age range (eg seeRidderinkhof amp van der Stelt 2000) To the best of ourknowledge there is only one published study of a typicalERP dichotic listening attention paradigm using tonesand syllables with young participants (groups with meanage 810 years 1438 years and adults Berman amp Fried-




The Present Study



RESULTS






Adults











P1(50ndash150 msec)



Amplitude(AV)

150 (012) 234 (015)a 127 (011)

Nonlinguisticprobes

161 (014) 213 (015)a 153 (015)

Linguisticprobes

139 (013) 255 (017)a 102 (011)

Latency(msec)

101 (20) 186 (33) 291 (44)

Nonlinguisticprobes

103 (19) 176 (31) 279 (54)

Linguisticprobes

98 (20) 195 (36) 304 (58)











Children













Nonlinguisticprobes

Attended 050 (014) ndash 071 (013)


Linguisticprobes

Attended ndash 011 (017) 043 (020)









Attended 306 (034)



Attended 265 (027)







Figure 1






DISCUSSION





Adults












Children















METHODS

Subjects




Behavioral Testing



Narratives




Probe Stimuli




General Design


General Procedure



the experiment The











































Acknowledgments



Notes


REFERENCES































































































The Present Study



RESULTS






Adults











P1(50ndash150 msec)



Amplitude(AV)

150 (012) 234 (015)a 127 (011)

Nonlinguisticprobes

161 (014) 213 (015)a 153 (015)

Linguisticprobes

139 (013) 255 (017)a 102 (011)

Latency(msec)

101 (20) 186 (33) 291 (44)

Nonlinguisticprobes

103 (19) 176 (31) 279 (54)

Linguisticprobes

98 (20) 195 (36) 304 (58)











Children













Nonlinguisticprobes

Attended 050 (014) ndash 071 (013)


Linguisticprobes

Attended ndash 011 (017) 043 (020)









Attended 306 (034)



Attended 265 (027)







Figure 1






DISCUSSION





Adults












Children















METHODS

Subjects




Behavioral Testing



Narratives




Probe Stimuli




General Design


General Procedure



the experiment The











































Acknowledgments



Notes


REFERENCES





































































































P1(50ndash150 msec)



Amplitude(AV)

150 (012) 234 (015)a 127 (011)

Nonlinguisticprobes

161 (014) 213 (015)a 153 (015)

Linguisticprobes

139 (013) 255 (017)a 102 (011)

Latency(msec)

101 (20) 186 (33) 291 (44)

Nonlinguisticprobes

103 (19) 176 (31) 279 (54)

Linguisticprobes

98 (20) 195 (36) 304 (58)











Children













Nonlinguisticprobes

Attended 050 (014) ndash 071 (013)


Linguisticprobes

Attended ndash 011 (017) 043 (020)









Attended 306 (034)



Attended 265 (027)







Figure 1






DISCUSSION





Adults












Children















METHODS

Subjects




Behavioral Testing



Narratives




Probe Stimuli




General Design


General Procedure



the experiment The











































Acknowledgments



Notes


REFERENCES




































































































Children













Nonlinguisticprobes

Attended 050 (014) ndash 071 (013)


Linguisticprobes

Attended ndash 011 (017) 043 (020)









Attended 306 (034)



Attended 265 (027)







Figure 1






DISCUSSION





Adults












Children















METHODS

Subjects




Behavioral Testing



Narratives




Probe Stimuli




General Design


General Procedure



the experiment The











































Acknowledgments



Notes


REFERENCES
































































































Figure 1






DISCUSSION





Adults












Children















METHODS

Subjects




Behavioral Testing



Narratives




Probe Stimuli




General Design


General Procedure



the experiment The











































Acknowledgments



Notes


REFERENCES

































































































DISCUSSION





Adults












Children















METHODS

Subjects




Behavioral Testing



Narratives




Probe Stimuli




General Design


General Procedure



the experiment The











































Acknowledgments



Notes


REFERENCES






























































































Adults












Children















METHODS

Subjects




Behavioral Testing



Narratives




Probe Stimuli




General Design


General Procedure



the experiment The











































Acknowledgments



Notes


REFERENCES

































































































Children















METHODS

Subjects




Behavioral Testing



Narratives




Probe Stimuli




General Design


General Procedure



the experiment The











































Acknowledgments



Notes


REFERENCES







































































































METHODS

Subjects




Behavioral Testing



Narratives




Probe Stimuli




General Design


General Procedure



the experiment The











































Acknowledgments



Notes


REFERENCES






























































































Probe Stimuli




General Design


General Procedure



the experiment The











































Acknowledgments



Notes


REFERENCES




















































































































Acknowledgments



Notes


REFERENCES





























































































An Event-related Potential Study of Selective …depts.washington.edu/audneuro/cochetal05.pdfAn...

Documents

Transcript of An Event-related Potential Study of Selective …depts.washington.edu/audneuro/cochetal05.pdfAn...