central auditory system and central auditory processing disorders

251

Management of Auditory Processing Disorders; Editor in Chief, Catherine V. Palmer, Ph.D.; Guest Editor, Gail D.Chermak, Ph.D. Seminars in Hearing, volume 23, number 4, 2002. Address for correspondence and reprint requests:Dennis P. Phillips, Ph.D., Professor, Hearing Research Laboratory, Department of Psychology, Dalhousie University,Halifax, NS, Canada B3H 4J1. E-mail: [email protected]. 1Hearing Research Laboratory, Department of Psychology,Dalhousie University, Halifax, Nova Scotia, Canada. Copyright © 2002 by Thieme Medical Publishers, Inc., 333 SeventhAvenue, New York, NY 10001, USA. Tel: +1(212) 584-4662. 00734-0451,p;2002,23,04,251,262,ftx,en;sih00219x.

Central Auditory System and CentralAuditory Processing Disorders:Some Conceptual IssuesDennis P. Phillips, Ph.D.1

ABSTRACT

The central auditory system has both parallel and hierarchical af-ferent architectures. In the frequency domain, it is tonotopically con-strained, and in the spatial domain, it is dominated by a representation ofthe contralateral acoustic hemifield. The functions supported by the affer-ent pathways can be somewhat overlapping, and the connectivity amongthe pathways is to some degree plastic. Partial deafferentation (in the formof high-frequency hearing loss) and behavioral experience are capable ofcausing alterations in tonotopic maps in the more rostral auditory system,even in adult animals. Central auditory processing is often frequency-specific. The temporal processes needed for normal auditory function arediverse, which is to be expected given the heterogeneous ways in which au-ditory events are disposed in time and encoded neurally. Central auditorypathologies need not respect structural or functional boundaries in thebrain, and so should be expected to have idiosyncratic presentations. Man-agement strategies based on auditory training may exploit basic neuroplas-ticity, but more evidence is needed to substantiate any hypothesis of theirdifferential efficacy in remediation of central auditory processing disordersor language and reading problems.

KEYWORDS: Auditory neuroscience, neural representation, neuralplasticity, temporal processing, perceptual architecture, auditory training

Learning Outcomes: Upon completion of this article, the reader will (1) understand the basic design principlesof the central auditory pathway, factors that can trigger plastic responses of the central auditory system, andthe forms that those responses take; (2) comprehend the diverse range of functions embraced by the termcentral auditory processing, and the many levels at which specifically temporal processes contribute to hear-ing; and (3) appreciate some of the issues surrounding auditory training paradigms and their effectiveness.

252 SEMINARS IN HEARING/VOLUME 23, NUMBER 4 2002

That information is then disseminated withinthree divisions of the cochlear nuclear com-plex, and from there to a host of caudal audi-tory brainstem nuclei. Almost all of those pro-jections are strictly topographic, that is, theypreserve the general tonotopy (i.e., the spatialarrangement of neurons according to the fre-quency to which they are most sensitive) devel-oped in the cochlea, and connections withinmore central nuclei are usually sufficiently localto preserve (or enhance) the fine frequencytuning of individual neurons.

The divergent projection of the cochlearnerve on the cochlear nucleus begins a patternof parallel processing, in the sense that afferentinformation from the same cochlear sources isreceived and processed, largely independently,by three nuclei at the same time. Those nucleiparticipate in separable and identifiable neuralcircuits. Some of them are involved in binauralspatial hearing. Thus, some nuclei of the su-perior olivary complex receive tonotopicallyconstrained inputs from both ventral cochlearnuclei and, through those routes, can serve ascoincidence detectors for the timing and am-plitudes of stimuli at the two ears. Those inter-aural parameters in turn are cues to soundsource location in the horizontal plane. In thisway, we see the development of hierarchicalprocessing, because the convergence of input re-sults in a serial growth in response complex-ity—from a code for the timing and amplitudeof events at a given ear, to a cross-correlation ofthose encoded events at the two ears.

The cells of the olivary nuclei do not justinherit the properties of their inputs, but exe-cute operations on those inputs so that a newrepresentation emerges. In this instance, the rep-resentation is of the interaural cues for sourcelocation, and because this is done within tono-topically constrained architecture, it followsthat the processing is done on a frequency-by-frequency basis. There is an important behav-ioral outcome of this. As it happens, the magni-tude of the interaural disparities in time andintensity vary not only with the eccentricity ofthe sound source, but also with frequency. Soundsources that are wideband in spectrum thus pro-vide a rich supply of information about sourcelocation and, because the interaural disparityinformation is encoded tonotopically, the fre-

An important key in coming to an ana-lytic understanding of central auditory pro-cessing (CAP) and its disorders (CAPDs) isthe complexity and dynamism of the centralauditory nervous system (CANS) itself. Thebrain, of which the CANS is a relatively smallbut highly integrated part, is even more com-plicated and dynamic. This architectural com-plexity is what supports the diversity of centralauditory function and the brain’s capacity foradaptation and learning. Consider the fol-lowing: the brain must support the detection,discrimination and localization of sound, thesegregation of auditory figure from ground, per-ceptual learning within new or familiar audi-tory dimensions, recognition and identificationof the source, introspection about the perceivedsound, and so on.

The detail of our neurophysiologic de-scriptions of CANS function and behavioraldescriptions of CAP is growing, as is our com-prehension of the links between those two lev-els of description. With these advances arecoming new insights into CAPDs in terms ofboth their genesis and their phenomenologyand, in turn, there is an emergence of new waysof thinking about management and remedia-tion. Because much of this information is new,and because it barely scrapes the surface of thesystem’s structural and functional architecture,much of our understanding is still at a concep-tual level. The purpose of this article is tosketch some of these advances, to illustrate theirstrengths and importance, and to make cau-tionary remarks about some of them. What fol-lows is, of necessity, a somewhat personal viewthat is not intended to cover all aspects of theseissues; it does assume a basic knowledge of theCANS and CAP.

THE CENTRAL AUDITORY NERVOUS SYSTEM

The afferent sensory input to the CANS is thearray of cochlear nerve cell axons from each ofthe two ears. The cochlear transduction pro-cess confers on each afferent fiber narrow fre-quency tuning, and it is the role of each fiber toencode the presence, amplitude and timing ofstimulus energy within its frequency passband.1,2

CAS AND CAPD: CONCEPTUAL ISSUES/PHILLIPS 253

quency-specificity of the information is pre-served in the brain. This provides a rich neuralrepresentation from which to derive a percep-tual judgement, and it is for these reasons thatbehavioral sound localization acuity is far moreacute for wideband sources than for tonal onesthat provide only a single time and/or intensitydisparity at the ears.

The outputs of the olivary nuclei and theprojections of some cell groups within thecochlear nuclei are sent to the nuclei of the lat-eral lemniscus and/or to the inferior colliculus(auditory midbrain). Some of these projectionsalso are parallel in organization. For example,the lateral superior olive projects contralater-ally upon both the dorsal nucleus of the laterallemniscus and the inferior colliculus (with theformer then projecting on to the latter). Thereis also a hierarchical arrangement in this levelof the projections, because some individualneurons of the inferior colliculus receive, forexample, input from both the dorsal cochlearnucleus and the lateral superior olive, and thusinherit the properties of both.3 The inferiorcolliculus ultimately receives input from thecochlear nuclei, olivary nuclei, and nuclei ofthe lateral lemniscus. These inputs are oftenbilateral, but the excitatory or inhibitory natureof the connectivity is of a kind that preservesand/or refines the sensitivity to interaural cuesfor sound location.2 The inferior colliculus pro-jects on the thalamic medial geniculate body,which then supplies the afferent input to theauditory cortex. There is also a well-developeddescending auditory pathway.4 This includesdense cortico-thalamic and cortico-collicularconnections, and other pathways descendingfrom the brainstem ultimately to the cochlearouter hair cells.

There are two further principles of organi-zation to be revealed here. One is that the con-vergence of input onto the inferior colliculususually continues to respect tonotopy and isnot undifferentiated: there are local territoriesdominated (anatomically and physiologically)by afferents from only a subset of those poten-tially available. These patchy territories are pre-served (or renewed) in the auditory thalamusand cortex, with the result that the primaryauditory cortex has been conceptualized as asingle tonotopic array on which is imposed a

patchy mosaic of smaller territories dedicatedto other facets of stimulus representation (lo-cation, amplitude, bandwidth, periodicity, andso on). Second, many of the circuits in the af-ferent pathway have overlapping functions.Thus, binaural convergence that initially occursin the superior olivary nuclei also may occur denovo in the auditory midbrain (and as far ros-tral as the cortex5). These patterns of connec-tivity make it difficult to assign a unique locusto any particular central auditory function. It isprobably because of the multiple layers of bin-aural interaction that lesion of the olivary nu-clei can leave the patterns of binaural input tocortical neurons nearly normal.6 Whether suchinputs can support normal neurophysiologiccoding of interaural disparity cues in the cortexcurrently is unknown.

CENTRAL REPRESENTATIONS,PLASTICITY, AND BEHAVIORALPERFORMANCE

There is some degree of plasticity in the pat-terns of auditory afferent connectivity. Perhapsthe most dramatic expressions of this have comefrom studies in animals with neonatal cochlearablations in whom there can be marked re-arrangement of connections from the survivingear.7 The striking complexity of the brainstempathways offers numerous opportunities for localshuffling of connections following neonatalablations. Thus, in the absence of input fromone ear to the medial superior olive, the intactear may come to innervate both sides of thenucleus. Projections of this sort appear to beable to support physiological responses morerostrally.

The internal organization of tonotopicmaps in the adult CANS also can be influencedby partial (i.e., frequency specific) cochlea dam-age. Most often, this damage has been to basalcochlear sectors, and changes to central tono-topy have taken the form of a loss of neuralspace devoted to the lesioned cochlear sector,and a nearly equivalent expansion of the terri-tory allocated to peri-lesional cochlear places(i.e., cochlear sites adjacent to the damaged sec-tor).8 This form of plasticity is probably differ-ent to that following wholesale neonatal cochlear


ablations, because it is present in adulthood,and because it is expressed within nuclei ratherthan a rearrangement of connectivity betweenpathways. The changes to the affected parts ofthe tonotopic maps have at least two stages.The earlier stage is a loss of sensitivity to stim-ulation of the affected cochlear sector, andwhat residual responsiveness remains is of thekind expected from a pathological endorgan(high thresholds, broad frequency tuning). Thelater stage is the development of near-normalsensitivity to acoustic stimulation, but from theperi-lesional cochlear places (with near-normalfrequency tuning for innervation from thatcochlear site). Note that this re-innervation doesnot support a recovery of behavioral sensitivityto stimulation at the frequencies deprived ofcortical representation (discussed later in thisarticle).

Plasticity in the adult thalamocortical au-ditory system can be driven by peripheral eventsfar less dramatic than ablations. Thus, the be-havioral training on a frequency discriminationtask using tonal stimuli can exert plastic changeson cortical tonotopic maps, notably by expand-ing the territories devoted to the frequenciesinvolved in the training.9 Such changes likelydepend on cholinergic influences from the nu-cleus basalis,10,11 perhaps by influencing activ-ity in the neural loop formed by afferents flow-ing from the midbrain through the thalamus tothe cortex on the one side, and efferents fromthe cortex to the thalamus and colliculus on theother. Through this feedback loop, there maybe a way for the thalamocortical auditory sys-tem to “select” afferent inputs12 of most behav-ioral significance and provide them the mostelaborate representation. In this regard, it istypically an expansion of the cortical territoryallocated to the relevant sector of the cochleathat accompanies behavioral training of func-tion at that cochlear sector and, of the neuro-physiologic (representational) changes that occur,it is the size of the representation that is moststrongly correlated with behavioral acuity.9The further implication of this line of work isthat if the details of cortical (or other) auditoryrepresentations are somewhat experience-de-pendent, then it follows that individual differ-ences in experience will be reflected in individ-

ual differences in the details of auditory repre-sentations and in behavioral performance.

These kinds of observations tend to prompta “bigger is better” view of brain function (i.e.,that a greater volume of neural tissue devotedto a particular function necessarily enhancesperformance of that function). We already haveseen evidence compatible with this view (seethe previous discussion). On the other hand,there are also counter-examples. Monocular de-privation of animals during the critical periodresults in the deprived eye receiving little or nocortical representation (and blindness throughthat eye), while the nondeprived eye presum-ably receives twice the cortical postsynapticspace that it normally would. Despite this, thereis little evidence for behavioral superiority inspatial acuity through the nondeprived eye overthat seen in normal animals. Likewise, we sawabove that in animals, restricted high-frequencyhearing loss can result in an expanded repre-sentation of peri-lesional frequencies. We haveevery reason to believe that this plasticity ex-tends to persons with high-frequency hearingloss. Nevertheless, in the face of any such plas-ticity, while it is no surprise that auditory acuityfor sounds activating the damaged cochlear sec-tor may be poorer than normal, there is littleevidence to support the existence of greater-than-normal frequency acuity for peri-lesionalfrequencies.13

The foregoing observations raise fascinat-ing questions about the nature and extent ofplastic mechanisms in the CANS. The twocases described in the previous paragraph areinstances in which the provision of greater-than-usual neural substrate failed to enhancebehavioral performance above normal. It canbe argued that the naturally-occurring visualspatial acuity in the normal animal is alreadyevolutionarily optimized to be as great as isneurologically supportable, and that throughyears of exposure to speech (and music, etc.)frequency discrimination in man has become ahighly over learned task. The latter may not betrue of animals, and it may be for this reasonthat we see the demonstrable effects of audi-tory frequency discrimination training in ani-mals. From the basic science standpoint then,it would be valuable to ascertain whether train-


ing in a non-over-learned task (e.g., temporalinterval discrimination) in a frequency rangewith an exaggerated central representation (asin peri-lesional frequencies in the case of re-stricted high-frequency hearing loss) would sup-port better acuity than the same training in afrequency range with a normal central repre-sentation. Perhaps a more interesting clinicalquestion concerns the prospects for training inindividuals who at the outset have an impover-ished neurological substrate of some cochlearsector or auditory function. In this general re-gard, developmental dyslexics often have—inaddition to pathological left superior temporalplana—larger-than-normal right plana,14 sug-gesting that an excess of language substrate(whether occurring independently or as a com-pensatory response to the left-sided abnormal-ity) has still left the patients with poor readingperformance.

CENTRAL AUDITORY PROCESSING

As alluded to at the beginning of the article,CAP is an umbrella term for all of the opera-tions executed on peripheral auditory inputs,and which are required for the successful andtimely generation of auditory percepts, theirresolution, differentiation, and identification.Many of the low-level analyses of sound arebased on frequency-specific processing; theseanalyses include sound localization (discussedpreviously; also see Jenkins and Merzenich15),temporal gap detection,16 temporal modulationdetection,17 and loudness perception.18 It alsois clear that more high-level processes (e.g., se-lective auditory attention) can operate locallyin frequency space,19 and that others are con-strained by frequency (e.g., auditory streamsegregation20). The tonotopic architecture ofthe CANS receives behavioral expression infrequency-specificity of much central auditoryprocessing. Sensory information processing isexecuted on a frequency-by-frequency basis(see the previous discussion), and the alloca-tion of sensory signals to perceptual streams(perceptual grouping) has frequency as a prin-cipal guiding force.

Two related features of specifically spatialcoding have emerged in the last two decades.One is that, for centers rostral to the olivarynuclei, neurons that have restricted spatial re-ceptive fields tend to have those receptive fieldslocated in the contralateral auditory hemifield.21

The medial borders of these receptive fields areusually within about 30 degrees of the midline,and are probably determined by binaural inter-actions. Neural response rate for locations withinthe receptive field are shaped more heavily bythe directionality of the pinna within thathemifield. Second, it has become clear that eachside of the CANS is independently capable oflocalizing sound sources in the contralateralauditory hemifield. Each cerebral hemispherecontains the spatial information required forlocalizing sources in the contralateral hemi-field, and unilateral forebrain lesions usuallyresult in deficits in sound localization behavioronly for sources in the auditory hemifield con-tralateral to the lesion.21,22 Psychophysical evi-dence suggests that the human auditory systemprobably has spatial perceptual channels with ahemifield architecture.23 The mechanisms inwhich these hemifield-tuned perceptual chan-nels have their genesis are still being workedout. The revelation of the hemifield architec-ture may offer new insights into a host of audi-tory spatial phenomena, including cocktail partyeffects, spatial masking, and so on. Note thatthe two-channel (left and right) hemifield ar-chitecture of perceptual channels for space isquite different to that for the frequency di-mension—the latter is subject to a multiple-channel (critical band) processing architecture.

The concept of channels is important; itincludes the complete neural representation ofthe stimulus and the perceptual structure sup-ported by that representation.16,24 Auditoryprocessing that is within-channel can often bethought of as low-level or automated and withrelatively little cognitive overhead. Thus, thedetection of temporal gaps between identicalmarkers is done effortlessly and with great acu-ity. In contrast, the detection of temporal gapsbetween dissimilar markers (between-channelgap detection) is more labor-intensive, subjectto a longer learning curve, and has poorer acu-ity.16,23 This difference may exist because the


existence of channel architecture normallyobliges the allocation of attentional resourcesselectively to one channel at a time.16,20,23 Bythe same reasoning, perceptual operations ex-plicitly requiring a comparison of activity indifferent channels (e.g., relative timing opera-tions) have poor acuity.

A vast number of central auditory pro-cesses rely on a specifically temporal analysis ofthe incoming signal. The temporal coincidenceof events that facilitates their perceptual group-ing as an auditory object, the segregation of au-ditory figure and ground by temporal streamingprocesses, the discrimination of phonemes dif-fering only in some temporal aspect (e.g., dura-tion or voice onset time), rhythm perception,temporal order judgments and sequencing, andperiodicity pitch perception are all examples thatmay have their mechanisms rooted deeply incentral temporal representations of sounds. It isimportant to understand that these temporalphenomena are not reflections of some singleprocess. Thus, the fidelity with which the audi-tory forebrain represents the timing of onsettransients and temporal periodicities are deter-mined relatively independently.25 Multiple lev-els of temporal processing might be involved inan apparently single perceptual operation. Ahigh-level process might exploit the output of alow-level one. One example of this is the segre-gation of two sound sources with different peri-odicity pitches on the basis of their differentrhythms.20 A second is the erroneous groupingof two spatially-separated sounds as a singleone on the basis of a contiguous periodicities, asin auditory saltation.26,27 In this paradigm, di-chotic clicks are presented in trains of 6 or 8elements at perfectly regular intervals; the firsthalf of the clicks favor one ear (or side) on thebasis of an interaural time difference, while thelast half of the elements favor the other ear (orside). For inter-click intervals less than about120 ms, the resulting percept is of clicks ema-nating not only from the sites of stimulation,but also from points between them, as if a sin-gle source were moving along a trajectory be-tween the click train’s anchor points. One ac-count of the mechanism is thus that the anchorpoints of the train have locations defined by in-teraural time differences; the auditory system

groups the clicks as a single object on the basisof the perfectly regular inter-click intervals, andhaving made that grouping process, assigns tothe middle clicks of the train spatial locationsthat are intermediate between those of the an-chor point clicks. One level of temporal analy-sis (interaural time disparity coding) perceptu-ally anchors the endpoints of the click train,while a second level of analysis (grouping byinter-click interval) ultimately results in the per-ceptual illusion.

Temporal processes in audition are not re-stricted to the accurate encoding of the stimulustime structure, and the generation of veridicalpercepts from that encoding. The generation ofthe percept itself takes time, and there is proba-bly a host of operations performed on sensoryinputs that take place during that time, and towhich we have little or no direct perceptual ac-cess.26–28 It is the distribution in time of auditorysensory and subsequent linguistic and cognitiveprocesses that makes possible the use of event-related evoked potentials to track the time courseof that processing. Because of the diversity oftemporal processes in audition, identifying aprocess (or a disorder) simply as temporal maynot be helpful.

CENTRAL AUDITORY PROCESSINGDISORDERS AND THEIRMANAGEMENT

CAPDs are probably as idiosyncratic as the in-dividuals they affect. Most of the CAPDs seenin the clinic will not have their origins localiz-able at a structural level. The exceptions areusually rare accidental cases in which precisedelineation of the structural and functional pa-thologies have been possible and tremendouslyinformative. For many of the auditory percep-tual dimensions we can imagine, there have beensingle-case studies revealing isolated functionalloss following focal cerebral lesion. These in-clude deficits in sound localization,29 auditorymotion perception,30 frequency discrimination,31

melody perception,32 and various componentsof auditory, phonological or other levels of analy-sis of spoken language.33


It is probably much more common forCAPDs to be highly idiosyncratic, and there isa host of reasons for this to be true. One is thatcentral nervous system pathology, whether de-velopmental or acquired, need not respect func-tional neurological boundaries. A subtle orga-nizational abnormality in the development ofneural connections, or a transient metabolicproblem during a critical period for the incor-poration of certain membrane channel proteins,could have diverse presentations, precisely be-cause the deficit may be relatively scattered inneural space. Second, even in the case of focallesions, we need to remember that brains arethemselves somewhat individual. In both ani-mals and man, the primary auditory cortex islocated in tissue that varies in fissural patternbetween individuals.34,35 There is some likeli-hood that functional territories will not respectfissural boundaries in any case.34 It is muchmore likely that functional territories will betied to cytoarchitectonic ones. Third, for a givenfunctional territory, there may be individualdifferences in the fine detail of structure andorganization, and different propensities foradaptation and recovery; so similarly locatedabnormalities might have different behavioralsequelae in different individuals. Because somuch of auditory perceptual experience is basedon the time structure of sounds and becausethere are many levels of sound time structureto be processed, some sort of temporal process-ing problem will be common in CAPD pre-sentation.

From a practical standpoint, specificationof a subtle neurological abnormality that un-derlies some processing deficit may not be asimmediately helpful as a detailed mapping ofthe functional disability itself. The tremendousrange of functions ascribed to the CANS, andindividual differences in their relative develop-ment, means that we require probes sensitiveto each function. Probing of auditory functionshould be done in the context of a much moregeneral interrogation of sensory, linguistic, andcognitive capacities. This is for a number of rea-sons. First, more high-level cognitive impair-ments might impair performance on auditoryprocessing probes without necessarily being in-dicative of impairments in the latter. Second,

one needs to know whether any auditory prob-lem is a component of a supramodal processingproblem.36 Third, temporal processing problemsmight be markers for, or causally related to, lan-guage impairments.28 In the classical connec-tionist view of language function, poor auditoryreceptive language function was an unremark-able consequence of an acquired impoverishedinput to the language processor. In a develop-mental scenario (following the general line ofargument provided by Tallal et al37 and Fitch etal38), a poor perceptual processor might affectnot only the acquisition of normal auditory re-ceptive language, but through that mechanism,provide poor targets to guide the developmentof speech production and phonology. This isnot to imply that all developmental languagedisorders are cascade error sequelae to low-levelperceptual problems. One view is that manycases of developmental language delay resultfrom various forms of synergism between manyfactors, of which an auditory processing prob-lem is only one.39

Comprehensive evaluation of persons withsuspected CAPDs is not a small task. Giventhe wealth of auditory functions ascribed tothe brain, it should come as no surprise that awealth of assessment is needed to assay thosefunctions. It also raises the prospect that theevaluation itself might require tools that arenot currently available. In this regard, there areincreasing numbers of reports using auditoryassessments that are not part of the standardaudiologic battery (e.g., various forms of tem-poral gap detection, backward masking,39,40 au-ditory stream segregation,41 auditory saltation42).Currently, this may appear a source of frustra-tion because the nonstandardized testing makesit hard to compare studies, either for the relia-bility with which separate studies have probedthe same mechanism, or because different stud-ies may, by design, not be probing the sametemporal mechanisms. In practice however, thisis often the state of a young science and anystandardization of test materials that is appro-priate will emerge as the scientific issues sortthemselves out over time.

There is growing interest in the manage-ment of developmental language delay throughintensive training strategies designed to reme-


diate central auditory processing disorders thatmight underlie the language delay.43,44 Perhapsthe most well-known of these is Fast ForWord.45

The training is based on the premise that brief,closely-spaced acoustic events (of the temporalgrain required to distinguish some phonemes33)are perceptually differentiated relatively poorlyin language-learning impaired children and thatthis results in poor phonology. The hope is,therefore, that provided with a more appropri-ately differentiated (i.e., trained) auditory input,an otherwise intact language processor (or onewhich itself adapts to the new inputs) will havean improved output. Fast ForWord43–45 usesstimuli in which target elements (e.g., stopconsonants in speech stimuli) are selectivelylengthened in time and amplified, and in whichthe sounds to be discriminated are presented atprogrammed inter-stimulus intervals (ISIs). Thetraining is adaptive, in the sense that as thechild learns to perform auditory discriminationswithin stimulus pairs, the extent of the tempo-ral stretching, the amplification, and the ISI allcan be reduced. The training also includes com-ponents above the level of acoustics, notablyexercises that target word recognition, auditorymemory and language comprehension.

Evaluating Fast ForWord or other trainingprograms is necessarily a complex issue.46 First,if a training program targets both auditory andlinguistic levels of analysis, but measures out-come using language performance, then it isnot immediately clear which component(s) ofthe training are responsible for the outcome.Second, if a training program is predicated onan account of language-learning disability thatis true for only a subset of affected listeners, thenparticipant selection in outcome studies is im-portant—because negative outcomes in non-relevant participants have little bearing on thequestion. Third, the differential effectivenessof a training program or other intervention de-pends critically on appropriate control for par-ticipants’ attention, motivation, duration oftreatment, and so on across the independenttreatment variable. The issue of Fast ForWord’seffectiveness has become theoretically impor-tant because of the hypothesized origins of de-velopmental language delay in specifically audi-

tory levels of analysis. The development of FastForWord, and discussion of its effectiveness,thus stands in contrast to other interventionsthat more explicitly acknowledge that they cap-italize on the contributions of language compo-nents of the training.47–49

Initial reports of the effectiveness of FastForWord training on auditory temporal process-ing43 showed great advantage over equivalentlytime-intensive classical language training andnonadaptive auditory intervention. Languageoutcome44 was significantly superior for a groupreceiving adaptive training with the modifiedspeech than for a control group receiving expo-sure to natural stimuli in a nonadaptive fashion.

More recently, independent studies haveproduced mixed findings. One group using aFast ForWord-like program has confirmed thatperceptual training with speech materials canimprove perceptual performance with thosematerials50 and may improve phonologic aware-ness.51 The former study50 did not employ acontrol group, so conclusions about the speci-ficity of any benefit for that training programcannot be drawn. Another study group reportedthat it was unclear whether Fast ForWord train-ing had any effects on language outcome abovethose seen with other, specifically language-based training programs.52 However, the par-ticipant population in that study had been se-lected on the basis of reading difficulties, ratherthan on the basis of the existence of auditoryprocessing difficulties. This is important, be-cause dyslexia itself is something of an um-brella category, and it is not clear that all formsof dyslexia (or language impairments) are nec-essarily associated with CAPD.28,39,53,54

Recently, there have been a series of casereports published on the outcome of Fast For-Word training on language55,56 and auditory per-ceptual57,58 performance. Each study was basedon small numbers of participants and so mustbe regarded as preliminary in its findings.With that caveat, the reports had some poten-tially important conclusions beyond that FastForWord training may improve auditory pro-cessing or language performance in some par-ticipants. One is that language outcome fol-lowing participation in a training program may


be independent of the level of language pro-cessing targeted by the training.56 It followsfrom this that the effectiveness of Fast For-Word training may not necessarily reside in thelink between auditory processing and languageperformance. Second, Marler et al58 demon-strated that the psychoacoustic performance(relation between backward and simultaneousmasking thresholds) of children improved acrosstraining, whether or not those children had nor-mally developing language, and roughly equiv-alently for Fast ForWord and another trainingprogram that did not contain acoustically mod-ified speech. This casts some doubt on the dif-ferential effectiveness of Fast ForWord training.

It is clear that the empirical question ofthe effectiveness of training regimes for CAPDand/or developmental language/reading impair-ments will be difficult to answer—though thisshould not discourage attempts to do so. Be-cause all learning involves plastic changes inthe brain, the newer training strategies are notspecial in this regard. What may prove them tobe special is the effectiveness with which theytarget any impaired process, and the nature ofthe link between resolution of that impairmentand performance benefits in higher-level lan-guage and cognitive domains. That link may beas individual as the listeners themselves.

ACKNOWLEDGMENTSSome of the work described in this article wassupported by grants from NSERC of Canadato the author. Thanks are due to Drs. GailChermak, Guinevere Eden, and Richard Olsonfor their very helpful guidance. Thanks also aredue to Susan E. Boehnke, Susan E. Hall, andDrs. Raymond Klein and Christine Sloan forcountless discussions that shaped the content ofthis article.

ABBREVIATIONS

CANS central auditory nervous systemCAP central auditory processingCAPD central auditory processing disorderISI interstimulus intervals

REFERENCES

1. Ruggero MA. Physiology and coding of sound inthe auditory nerve. In: Popper AN, Fay RR, eds.The Mammalian Auditory Pathway: Neurophysi-ology. New York: Springer-Verlag; 1992:34–93

2. Phillips DP. Introduction to the central auditorynervous system. In: Jahn AF, Santos-Sacchi JR,eds. Physiology of the Ear, 2nd ed. San Diego, CA:Singular; 2001:613–638

3. Semple MN, Aitkin LM. Physiology of pathwayfrom dorsal cochlear nucleus to inferior colliculusrevealed by electrical and auditory stimulation. ExpBrain Res 1980;41:19–28

4. Spangler KM, Warr WB. The descending auditorysystem. In: Altschuler RA, Bobbin RP, CloptonBM, Hoffman DW, eds. Neurobiology of Hearing:The Central Auditory System. New York: RavenPress; 1991:27–46

5. Kitzes LM, Doherty D. Influence of callosal ac-tivity on units in the auditory cortex of ferret(Mustela putorius). J Neurophysiol 1994;71:1740–1751

6. Sally SL, Kelly JB. Effects of superior olivary com-plex lesions on binaural responses in rat inferiorcolliculus. Brain Res 1992;572:5–18

7. Kitzes LM, Kageyama GH, Semple MN, Kil J.Development of ectopic projections from the ven-tral cochlear nucleus to the superior olivary com-plex induced by neonatal ablation of the contralat-eral cochlea. J Comp Neurol 1995;353:341–363

8. Rajan R, Irvine DRF, Wise LZ, Heil P. Effectof unilateral partial cochlear lesions in adult catson the representation of lesioned and unlesionedcochleas in primary auditory cortex. J Comp Neu-rol 1993;338:17–49

9. Recanzone GH, Schreiner CE, Merzenich MM.Plasticity in the frequency representation of primaryauditory cortex following discrimination training inadult owl monkeys. J Neurosci 1993;13:87–103

10. Weinberger NM. Learning-induced receptive fieldplasticity in the primary auditory cortex. SeminNeurosci 1997;9:59–67

11. Kilgard MP, Merzenich MM. Cortical map reor-ganization enabled by nucleus basalis activity. Sci-ence 1998;279:1714–1718

12. Suga N, Yan J, Zhang Y. Cortical maps for hearingand egocentric selection for self-organization.TrendsCogn Sci 1997;1:13–20

13. Irvine DRF, Rajan R, McDermott HJ. Injury-induced reorganization in adult auditory cortex andits perceptual consequences. Hearing Res 2000;147:188–199

14. Galaburda AM, Sherman GF, Rosen GD, ArboitisF, Geschwind N. Developmental dyslexia: four


consecutive patients with cortical anomalies. AnnNeurol 1985;18:222–233

15. Jenkins WM, Merzenich MM. Role of cat primaryauditory cortex for sound localization behavior. JNeurophysiol 1984;52:819–847

16. Phillips DP, Taylor TL, Hall SE, Carr MM,Mossop JE. Detection of silent intervals betweennoises activating different perceptual channels: Someproperties of “central” auditory gap detection. JAcoust Soc Am 1997;101:3694–3705

17. Viemeister NF. Temporal modulation transferfunctions based upon modulation thresholds. JAcoust Soc Am 1979;66:1364–1380

18. Viemeister NF. Auditory intensity discriminationat high frequencies in the presence of noise. Sci-ence 1983;221:1206–1208

19. Scharf B, Quigley S, Aoki C, Peachey N, Reeves A.Focused auditory attention and frequency selectiv-ity. Percept Psychophys 1987;42:215–223

20. Bregman AS. Auditory Scene Analysis: The Per-ceptual Organization of Sound. Cambridge, MA:MIT Press; 1990

21. Phillips DP, Brugge JF. Progress in neurophysiol-ogy of sound localization. Ann Rev Psychol 1985;36:245–274

22. Jenkins WM, Masterton RB. Sound localization:effects of unilateral lesions in central auditorypathways. J Neurophysiol 1982;47:987–1016

23. Boehnke SE, Phillips DP. Azimuthal tuning ofhuman perceptual channels for sound location. JAcoust Soc Am 1999;106:1948–1955

24. Johnson KO. Sensory discrimination: neural pro-cesses preceding discrimination decision. J Neuro-physiol 1980;43:1793–1815

25. Phillips DP, Hall SE. Response timing constraintson the cortical representation of sound time struc-ture. J Acoust Soc Am 1990;88:1403–1411

26. Phillips DP, Hall SE. Spatial and temporal factorsin auditory saltation. J Acoust Soc Am 2001;110:1539–1547

27. Phillips DP, Hall SE, Boehnke SE, RutherfordLED. Spatial stimulus cue information supplyingauditory saltation. Perception 2002;31:875–885

28. Farmer ME, Klein R. The evidence for a temporalprocessing deficit linked to dyslexia: a review.Psych Bull Rev 1995;2:460–493

29. Sanchez-Longo LP, Forster FM. Clinical signifi-cance of impairment of sound localization. Neurol1958;8:119–125

30. Griffiths TD, Rees A, Witton C, et al. Evidencefor a sound movement area in the human cerebralcortex. Nature 1996;383:425–427

31. Tramo MJ, Shah GD, Braida LD. Functionalrole of auditory cortex in frequency processingand pitch perception. J Neurophysiol 2002;87:122–39

32. Peretz I, Kolinsky R, Tramo M, et al. Functionaldissociations following bilateral lesions of auditorycortex. Brain 1994;117:1283–1301

33. Phillips DP. Central auditory processing: a view fromauditory neuroscience. Am J Otol 1995;16:338–352

34. Merzenich MM, Knight PL, Roth GL. Represen-tation of cochlea within primary auditory cortex inthe cat. J Neurophysiol 1975;38:231–249

35. Musiek FE, Reeves AG. Asymmetries of the audi-tory areas of the cerebrum. J Am Acad Audiol1990;1:240–245

36. Cacace AT, McFarland DJ. Central auditory pro-cessing disorder in school-aged children: a criticalreview. J Speech Lang Hear Res 1998;41:355–373

37. Tallal P, Miller S, Fitch RH. Neurobiological basisof speech: a case for the preeminence of temporalprocessing. Ann NY Acad Sci 1993;682:27–47

38. Fitch RH, Read HL, Benasich AA. Neurophysiol-ogy of speech perception in normal and impairedsystems. In: Jahn AF, Santos-Sacchi JR, eds. Phys-iology of the Ear, 2nd ed. San Diego, CA: Singular;2001:651–672

39. Bishop DVM, Carlyon RP, Deeks JM, Bishop SJ.Auditory temporal processing impairment: neithernecessary nor sufficient for causing language im-pairment in children. J Speech Lang Hear Res1999;42:1295–1310

40. Wright BA, Lombardino LJ, King WM, et al.Deficits in auditory temporal and spectral resolu-tion in language-impaired children. Nature 1997;387:176–178

41. Helenius P, Uutela K, Hari R. Auditory stream seg-regation in dyslexic adults. Brain 1999;122:907–913

42. Hari R, Kiesilä P. Deficit of temporal auditory pro-cessing in dyslexic adults. Neurosci Lett 1996;205:138–140

43. Merzenich MM, Jenkins WM, Johnston P, et al.Temporal processing deficits of language-learningimpaired children ameliorated by training. Science1996;271:77–81

44. Tallal P, Miller SL, Bedi G, et al. Language com-prehension in language-learning impaired childrenimproved with acoustically modified speech. Sci-ence 1996;271:81–84

45. Scientific Learning Corporation. Fast ForWord™[Computer Software]. Berkeley, CA: Author; 1996

46. Gillam RB, Frome Loeb D, Friel-Patti S. Lookingback: a summary of five exploratory studies of FastForWord. Am J Speech-Lang Pathol 2001;10:269–273

47. Chermak GD. Managing central auditory process-ing disorders. Metalinguistic and metacognitiveapproaches. Semin Hear 1998;19:379–392

48. Sloan C. Management of auditory processing diffi-culties. A perspective from speech-language pa-thology. Semin Hear 1998;19:367–378


49. Musiek FE, Schochat E. Auditory training andcentral auditory processing disorders. A case study.Semin Hear 1998;19:357–366

50. Louis M, Espesser R, Rey V, et al. Intensive train-ing of phonological skills in progressive aphasia: amodel of brain plasticity in neurodegenerative dis-ease. Brain Cogn 2001;46:197–201

51. Habib M, Espesser R, Rey V, et al. Training dyslex-ics with acoustically modified speech: evidence ofimproved phonological performance. Brain Cogn1999;40:143–146

52. Hook PE, Macaruso P, Jones S. Efficacy of Fast For-Word training on facilitating acquisition of readingskills by children with reading difficulties—a longi-tudinal study. Ann Dyslexia 2001;51:75–96

53. Tallal P, Stark RE. Perceptual/motor profiles ofreading impaired children with or without concom-itant oral language deficits. Ann Dyslexia 1982;32:163–176

54. Van der Lely HKJ, Rosen S, McClelland A. Evi-dence for a grammar-specific deficit in children.Curr Biol 1998;8:1253–1258

55. Friel-Patti S, DesBarres K, Thibodeau L. Case stud-ies of children using Fast ForWord. Am J SpeechLang Pathol 2001;10:203–215

56. Gillam RB, Crofford JA, Gale MA, Hoffman LM.Language change following computer-assisted lan-guage instruction with Fast ForWord or LaureateLearning Systems software. Am J Speech LangPathol 2001;10:231–247

57. Marler JA, Champlin CA, Gillam RB. Backwardand simultaneous masking measured in childrenwith language-learning impairments who receivedintervention with Fast ForWord or Laureate Learn-ing Systems software. Am J Speech Lang Pathol2001;10:258–268

58. Thibodeau LM, Friel-Patti S, Britt, L. Psycho-acoustic performance in children completing FastForWord training. Am J Speech Lang Pathol 2001;10:248–257

263

Management of Auditory Processing Disorders; Editor in Chief, Catherine V. Palmer, Ph.D.; Guest Editor, Gail D.Chermak, Ph.D. Seminars in Hearing, volume 23, number 4, 2002. Address for correspondence and reprint requests:Frank E. Musiek, Ph.D., Department of Communication Sciences and Otolaryngology, School of Medicine, Universityof Connecticut, Storrs, CT 06269. E-mail: [email protected]. 1Department of Communication Sciences andOtolaryngology, School of Medicine, University of Connecticut, Storrs, Connecticut; 2Dartmouth-Hitchcock MedicalCenter, Lebanon, New Hampshire. Copyright © 2002 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, NewYork, NY 10001, USA. Tel: +1(212) 584-4662. 00734-0451,p;2002,23,04,263,276,ftx,en;sih00220x.

Plasticity, Auditory Training, and AuditoryProcessing DisordersFrank E. Musiek, Ph.D.,1 Jennifer Shinn, M.S.,2 and Christine Hare, M.A.1

ABSTRACT

Auditory training (AT) for the treatment of auditory processing dis-orders (APD) has generated considerable interest recently. There is emergingevidence that well conceived AT programs can improve higher auditory func-tion. The plasticity of the brain underlies the success of AT. This article re-views brain plasticity and the role of plasticity in AT outcomes, and highlightskey studies that provide insight into the clinical use of AT for APD.

KEYWORDS: Auditory processing disorder, auditory training, plasticity

Learning Outcomes: Upon completion of this article, the reader will be able to (1) understand the rationalebehind auditory plasticity as it applies to APD, (2) understand new therapy techniques for APD, and (3) formu-late therapy approaches using plasticity.

auditory nervous system (CANS). Improve-ment in higher auditory function is related tothe capacity of the central nervous system tochange. This change is tightly linked to, and isa result of, plasticity. This article reviews plas-ticity and AT within the framework of APD.

AUDITORY PLASTICITY DEFINED

There is a significant body of evidence fromboth animals and humans that supports im-provement in auditory tasks through the use ofAT. As we begin to gain new insight into the

Recently, there has been great interest inthe use of auditory training (AT) for the treat-ment of auditory processing disorders (APD).The use of AT for peripheral auditory prob-lems is not a new concept, in that it has an ex-tensive history.1 The use of AT for treatmentof APD is new and its application is differentfrom the classic use of AT. Most important tothis difference is that AT applied to APD istargeting the brain as the main site of media-tion, and the brain, unlike the auditory periph-ery, is plastic. Though peripheral sensorineuralloss cannot be improved upon with AT, thismay not be the case for deficits of the central


fundamental mechanisms involved in improvedperformance on specific auditory tasks, we musttake into consideration the concept of auditoryplasticity. The scientific literature on the sub-ject of plasticity is immense. For the purposesof our discussion, we will concentrate specifi-cally on auditory plasticity and overview its re-lationship to AT.

Neural plasticity is defined by Lund2 as“neural form and connections that take on apredictable pattern (probably genetically deter-mined) often referred to as neural specificity;exceptions to this predictable pattern, in cer-tain circumstances, represent neural plasticity.”Auditory (specific) plasticity can be defined asthe alteration of nerve cells to better conformto immediate environmental influences, withthis alteration often associated with behavioralchange.1 It is necessary for both clinicians andresearchers to have a strong foundation in thefundamentals of plasticity and its relationshipto AT, as it is key to management of patientswith APD.

ANIMAL RESEARCH

Perhaps the greatest body of evidence that sug-gests plasticity in the auditory system relatedto training lies in the extensive number of ani-mal studies. Though the scope of this articledoes not allow extensive coverage of this topic,we will highlight a few studies that hopefullyprovide some principles of interest regardingplasticity. Some of the early research on plas-ticity and training via animal models focusedon sub-cortical levels and the brainstem; how-ever, the auditory cortex has recently gainedconsiderable attention. The evidence of corti-cal plasticity is shown impressively by func-tional reorganization, that has attracted muchattention from scientists and clinicians in a va-riety of disciplines.

Studies on rodents, monkeys, cats, and theavian species have shown reorganization withinthe auditory cortex.3–7 There are two main con-ditions that can result in reorganization of theauditory cortex, sub cortex, and brainstem. Oneof these is deprivation that, experimentally, isinduced by creating hearing loss in some man-

ner. This can be achieved by creating a conduc-tive loss, which is often done by plugging theear and/or removing or damaging middle earstructures. Sensorineural hearing loss is experi-mentally created by exposing the ear to ototoxicdrugs, high intensity noise, or mechanicallydamaging the cochlea. These hearing losses willdeprive the higher auditory system of stimula-tion and will result in a reorganization of theauditory cortex.3,8–10 Raising animals in acous-tically attenuated chambers is another methodof creating deprivation effects. High stimulationor tasking of the auditory system is the othercondition that can create plastic changes in theCANS.4,11

The barn owl is one of the most studiedanimal models for demonstrating plasticity.One of the reasons for studying the barn owl isthat its optic tectum (or midbrain), which issimilar in nature to the superior colliculus (SC)of the mammal, contains neural substrate withrespect to both auditory and visual maps. TheSC is one of the most highly studied corticalstructures because of its ability to reorganize.The SC is a useful structure to investigate, as itis particularly sensitive to experience-dependentplasticity and alteration of auditory maps. Thebarn owl provides an extensive amount of mod-eling information to base on the derivation ofits auditory map.12

Evidence of auditory plasticity has beengained through both auditory deprivation andauditory training with the barn owl. Previousdeprivation studies, in which barn owls wereraised with monaural occlusion, provided phys-iologic evidence of inter-aural level differencesoccurring in the posterior division of the ven-tral nucleus of the lateral lemniscus within thebrainstem.13 Through the use of acoustic filter-ing devices, Gold and Knudsen14 investigatedthe effect of altered auditory experience on inter-aural timing differences of barn owls. Relativeto owls raised under normal auditory conditions,owls raised with unilateral auditory deprivationshowed neuronal changes of the inferior col-liculus (in the optic tectum) and changes in be-havioral responses, demonstrating that plasticitywithin the inferior colliculus plays a significantrole in adjustments in the inter-aural timingdifference and frequency tuning. These plas-

AUDITORY TRAINING PROGRAMS FOR APD/MUSIEK ET AL 265

tic changes in auditory experience as a resultof deprivation and accommodation are notedat both the subcortical12 and higher-level sys-tems.15

A substantial body of research suggeststhat reorganization takes place in all mammals.These findings have strong implications forauditory plasticity in humans. The general find-ings show that sensory maps have training de-pendent reorganization capabilities and sug-gest that a majority, if not all of the brain, maybe plastic.10 There also are indications that inmammals plasticity is lateralized. That is, cor-tical reorganization seems to take place in thehemisphere contralateral to the ear deprived ofacoustic input or for which there is peripheraldamage, which reduces peripheral input.10

MECHANISMS UNDERLYINGCORTICAL PLASTICITY ANDREORGANIZATION

Before auditory plasticity and its relationshipto AT can be fully appreciated, there must be acomprehensive understanding of the mecha-nisms that underlie this critical process. Nu-merous theories have been proposed and themore widely accepted will be briefly discussedfor the purpose of this review.

It is well established that plastic changes area result of neuronal responses to both externalstimuli (i.e., sounds in our environment) andinternal stimuli (e.g., thinking about certainsounds when we listen to music or read). Wetend to think of our brain as a highly stableorgan, when in fact it is its instability that we relyon most for development and auditory learning.There are generally three types of plasticity inthe auditory domain: (1) developmental plastic-ity, (2) compensatory plasticity resulting from alesion occurring somewhere within the auditorysystem, and (3)learning-related plasticity.6

Simple auditory tasks, such as detecting thepresence of a pure tone, require relatively littleneural substrate in the CANS. In contrast, com-plex tasks—such as discriminating small differ-ences among various acoustic stimuli in noise—require considerable neural activity. The simpleauditory task may not be improved signifi-

cantly with AT; however, complex tasks aremore likely to benefit as they necessitate morebrain function (and neural substrate) and, there-fore, are more likely to benefit from brain plas-ticity.1

Generally there are two ways in which thebrain is thought to reorganize. Although bothrequire neuronal changes, their mechanismsare quite different. Reorganization may involvethe activation of neurons and neural connec-tions that were previously in a state of rest.Thesereserve neurons replace the nonfunctional neu-rons. That is, the reserve neurons would re-place connections that are no longer active dueto damage or lack of stimulation.The other pro-posed mechanism underlying brain reorganiza-tion requires that new connections be formed.Both mechanisms involve changes that may beslow and require an extensive amount of timeand training, or the changes can be extremelyrapidly occurring through development andmaturation without intervention.10

It should be noted that neural reorganiza-tion might not be considerable and obvious.According to Kaas,10 these structural changesmay be rather subtle, resulting from modifica-tions in the extent of synaptic connections thatbecome more effective, or from changes inneural location, resulting in a more effectivetargeting or spreading of axonal and dendriticarbors, that in turn results in not only more butbetter located synaptic connections. If newdendrites are either being recruited or devel-oped, their connections must be appropriate toensure that they evolve meaningful function.

CLINICAL CORRELATES OF AUDITORY PLASTICITY IN HUMANS

Clinical experience with hearing aids andcochlear implants provide important insightand evidence of the plasticity phenomenon andraise several key questions as well. As mentionedpreviously, auditory reorganization can resultfrom lack of input to central systems, usuallyafter damage to the periphery. It is known thatthe auditory cortex will reorganize tonotopi-cally if input is markedly reduced at certain


frequencies secondary to damage to the corre-sponding part of the cochlea. The cortical re-gion that previously responded to the frequen-cies that have been lost will shift to adjacentfrequencies that are active and viable.5 Thispermits the formerly deprived region of theauditory cortex to become physiologically ac-tive and retain its functional status. This neuralreorganization is a result of the dynamic andactive process of plasticity rather than passiveconsequences of the lesion at the cochlear level.16

The observations provided by clinical investi-gators of patients with bilateral sensorineuralhearing loss who wear hearing aids in only oneear provide information potentially related tocortical reorganzation.17 These patients, overa period of time, will demonstrate decreasedspeech recognition in the ear not amplified,while the amplified ear’s speech recognition re-mained stable or, in some cases, improved. Onemight theorize that these clinical results indi-cate possible reorganization in the cortex.

The success of patients with cochlear im-plants also provides some insights into neuralplasticity. Cochlear implant patients have pro-found hearing loss that in turn results in a formof auditory deprivation to the CANS, whichcould result in reorganization of the auditorycortex. The implant allows stimulation of thecentral system, such that it may reconstitutethe more normal organization and activity ofauditory cortical neurons that have assumed adifferent role because of lack of auditory input.Manrique et al18 demonstrated that a group ofpre-lingually deafened children implanted priorto 3 years of age surpassed the speech percep-tion performance of post-lingually implantedpeers as early as the third year of follow-up.They suggested that the critical period of de-velopmental plasticity extended up to 6 yearsof age; beyond 6 years of age, peripheral stimu-lation was not sufficient to overcome the de-privation sustained by the CANS. Their find-ings reaffirm the expectation that the youngerthe brain, the greater the capacity for plasticity.

A number of studies have focused ondemonstrating that the younger the age at thetime of implantation, and the sooner the im-plantation following deafness, the better the

speech perception abilities.19 Although thesefindings often were used as reasons against theimplantation of deafened individuals outsidethis ideal group, evidence continues to mountfor the brain’s ability to adapt in response toAT later in life and following sustained periodsof deprivation. Labadie and coworkers20 foundsignificant increases in post-operative perfor-mance on word and sentence identification fortwo groups of implant patients with mean agesof 46.9 years and 71.5 years. No significant dif-ferences were reported relative to the medicalprocedure itself, and it was concluded thatneural plasticity continues to exist throughoutlife and that age alone should not be a criterionfor implantation. Geier et al21 demonstratedthat those implanted at a younger age and witha shorter term of deafness (defined as less than60% of life with deafness) showed the bestspeech recognition abilities 3 months post-im-plantation; however, even adults who had beendeaf for over 60% of their lives demonstratedsignificant speech recognition improvements.Although the rate of improvement was slowerfor this latter group, speech perception abilitiescontinued to improve with implant experiencefollowing re-assessment at 6 months after im-plantation. Moreover, deaf subjects who usedsign language to communicate demonstratedsuccessful cochlear implantation.22 While si-multaneously listening and watching sign lan-guage, positron emission topography (PET)scans showed a shift from no activity in the au-ditory cortex after short-term experience withthe implant to much higher levels of activityfollowing long-term experience.22 If the audi-tory areas of the brain previously deprived ofauditory input are able to be stimulated, thereis a potential for success, and the degree ofsuccess will largely depend on the length andquality of follow-up and the habilitation pro-cess. Imaging techniques, such as functionalmagnetic resonance imaging (fMRI), positronemission topography (PET) scans, and corticalevoked potentials are gaining in popularity fortheir potential role in demonstrating neuralplasticity of previously deprived systems in re-sponse to stimulation via cochlear implants andthrough AT.


AUDITORY TRAINING WITH HUMANS

Diagnostic central test procedures can guidethe clinician to the types of AT that are re-quired by the patient. A key component of anyAT program is to measure AT training effectsusing various test procedures. Traditionally, be-havioral speech perception tests have been usedas measures of auditory processing abilities,and as measures of treatment outcomes and ef-ficacy. Although these tests are crucial in theauditory processing (AP) assessment battery,performance on these measures is affected byfactors such as attention, motivation, and learn-ing. Recently, attention has focused on audi-tory evoked potentials as noninvasive and moreobjective tools to measure auditory processingand plasticity.23–31 Functional imaging tech-niques also are being used to examine physio-logic changes of the CANS subsequent to AT.These tools provide an objective means to dif-ferentiate subjects with known APD fromnormal subjects. In addition, these objectivemeasures are increasingly demonstrating phys-iological changes as a result of behavioral ATin normal subjects and those with particulardeficits.

Auditory evoked potentials are used to as-sess auditory function from the periphery tothe cortex through the use of electrophysio-logic responses. There are primarily two typesof cortical evoked potentials—obligatory anddiscriminative.24 Obligatory potentials are typ-ically elicited by clicks, tone bursts, and speechphonemes that may vary in duration. Thesecortical potentials (e.g., auditory brainstem re-sponse [ABR] and middle latency response[MLR]) are used clinically in auditory process-ing assessment. Discriminative evoked poten-tials use an oddball paradigm and are elicitedpassively (e.g., mismatch negativity [MMN]),or are elicited with an active paradigm (e.g.,P300).

Using the MMN, Kraus et al23 showedthat physiologic changes in central auditoryfunction and auditory discrimination could beimproved with training. Tremblay et al28 dem-onstrated that AT generalizes to listening situ-

ations beyond those used in the training para-digm, both behaviorally and physiologically (asrecorded by the MMN). Tremblay et al30 con-cluded that behavioral changes are likely to fol-low neurophysiologic changes resulting fromAT and that these neurophysiological mea-sures would serve to determine the efficacy ofAT. Jirsa32 showed an increase in the amplitudeof the P300 and decrease in its latency in chil-dren who underwent a general AT program.These changes in the P300 were not noted in acontrol group of children. Based on the MMN,musicians were found to have superior audi-tory processing relative to nonmusicians whilelistening to perfect versus impure chords33 andto have a more precise and longer temporalwindow of integration.34 These studies provideevidence that improvement in higher auditoryfunction can result from AT, this improvementcan be measured with current assessment tools,and evoked potentials can be used to differ-entiate a range of auditory abilities in normallisteners.

FORMAL AUDITORY TRAINING

Typically, the process of diagnosing and man-aging APD begins with parental or schoolconcerns that result in a child being referred toan audiologist for AP testing. Often the con-cerns involve language, learning, reading, andauditory memory rather than, or in addition to,APD. Due to the complexity and interdepen-dence of these processes, it is essential thatchildren referred for AP testing also be evalu-ated by a speech-language pathologist and ed-ucational psychologist, and that the informa-tion from these professionals be incorporatedin the management of these children. A recentconsensus report provides a good reference foridentification and differential diagnosis of APD,and recommends a minimal test battery for thispurpose.35

Formal AT requires the use of special in-strumentation to manipulate auditory stimuliand would typically be conducted by an audiol-ogist in the clinic or laboratory setting.36 For-mal AT usually incorporates an intensive train-


ing schedule and targets specific stimuli usingadaptive techniques to foster improved process-ing in a graded manner.The use of commerciallyavailable computer software can complementformal training at school or at home when reg-ular visits to the lab or clinic are not feasible.

Because plasticity in response to AT ishighly correlated with behavioral contexts suchas motivation,37 it is particularly important withchildren to use feedback and reinforcement tomaintain a high level of motivation during for-mal AT. These considerations have been incor-porated in commercial software supportingAT, but they need to be addressed in any ATprogram, especially in training in the clinic orlab. Modifying stimuli to maintain a fair de-gree of success while challenging the listeneralso needs to be an integral part of formal AT.Various types of training have shown that ifthe task is too easy or too difficult, optimalchanges will not result.1 It is important, there-fore, that task difficulty be carefully selectedif adaptive approaches are not employed (andsometimes even when adaptive approaches areused!).

A variety of formal AT techniques areoutlined below. The reader is referred to Cher-mak and Musiek (see pages 297–308, this issue)for an additional discussion of formal and in-formal AT tasks, as well as general principlesunderlying AT.

Temporal Processing

There is a significant body of evidence thatsuggests that temporal processing abilities arethe foundation of auditory processing, specifi-cally with respect to speech perception. Muchdebate, however, concerns the role of temporalauditory processing in children’s language andreading problems.38,39 Considerable researchhas concluded that poor temporal processingskills, measured behaviorally, are found in manycases of dyslexia.38,39 Unresolved were ques-tions regarding the role of temporal processingfor speech versus nonspeech signals and theprimacy of auditory versus visual processingdeficits for language and reading problems.Recently, studies incorporating brain imaging

and electrophysiologic tests have concludedthat temporal processing deficits underlie poorreading skills.40 Using fMRI, Temple et al41

demonstrated a lack of differential activity inthe left prefrontal region of the cortex forrapidly and slowly changing acoustic stimuli inadults with dyslexia. Improvement in differen-tial sensitivity was identified in this area bothfor normal adult subjects and dyslexic subjectsfollowing intensive training. Also using fMRI,Temple et al42 demonstrated physiologic deficitsfor both phonologic and orthographic processesin children with dyslexia.

New training techniques have developedin parallel with the increased study of electro-physiologic measures sensitive to temporal pro-cessing deficits. Merzenich et al43 demonstratedthe efficacy of computer games in training tem-poral processing thresholds of both speech andnonspeech stimuli for children 5 to 10 years ofage. This study included two-tone ordering andsequencing of consonant-vowel syllables (CVs)modified by duration and intensity of conso-nants and duration of interstimulus intervals.Training was intensive (20 days), tasks incorpo-rated various forms of feedback, and reinforce-ment was used as the basis for the developmentof currently available commercial software (i.e.,Fast ForWord [FFW]; Scientific Learning Cor-poration, 1997). Tallal et al44 reported that for-mal AT using these specially designed computergames incorporating modified temporal speechcues and enhanced speech transitions improvedspeech discrimination and language comprehen-sion among language-impaired children follow-ing an intensive 4-week therapy program.

Several recent studies have examined theefficacy of FFW.45–48 In general, these studiesdemonstrate some advantages to the FFW pro-gram, but overall, they question its efficacy.There is concern about the types of problemsand types of populations that may be best suitedfor FFW. It is interesting that the more recentstudies do not demonstrate the same overallvalue of FFW that was shown in the originalstudies.44 However, earlier and more recentFFW efficacy studies did not use the same eval-uative procedures. It also is curious that none ofthe studies employed commonly used centralauditory tests to measure the effects of FFW.


Kujala et al49 demonstrated—through bothbehavioral and physiologic (MMN) measures—that a more general temporal processing deficit,rather than a primarily linguistic deficit, is seenin dyslexic adults. Dyslexic subjects had diffi-culty discriminating temporal sound featuresof tonal stimuli.49 Kujala et al50 also publishedthe results of a unique, auditory-visual discrim-ination training procedure for first-graders iden-tified as having dyslexia. A computer game wasdeveloped to graphically represent the frequency,duration, and intensity of sound as representedby the relative height, length, and thickness ofrectangles. One task was to identify which ofthe two displayed visual patterns matched thesubsequent sound pattern. The second task wasto initially look at a visual pattern while listen-ing to the sound pattern and identify when thelast element was heard. Correct responses wererewarded and incorrect responses were fol-lowed by a repeated presentation of the pat-tern. Rectangles changed colors as they werepresented in order to help the child track thepattern. Following auditory-visual discrimina-tion training, children showed improved read-ing skills, as evidenced through correct identi-fication of words and reading speed, as well asphysiologic changes (as measured by the MMN).They also reported a high correlation betweenthe behavioral and physiologic measures. It isimportant to point out that the Kujala et al50

study used sound and visual patterns for train-ing that required temporal sequencing and cross-modality correlations.

Several studies of commercially availableprograms suggest that through the use of intenseAT, temporal integration abilities improve sig-nificantly.51 Several informal AT techniques—asthose reviewed in Musiek et al36 and Musiek25—also are beneficial for strengthening temporalprocessing skills. Identifying or mimicking pat-terns with differing durations or frequencies, andattention to stress and prosody while reading orlistening to poetry are examples of informal ATtasks that can be used to strengthen temporalprocessing. Similarly, sequencing tones with theuse of the commercial game Simon™ strength-ens frequency discrimination as well as sequenc-ing. The Simon game also provides a visual cor-relate to the acoustic patterns used in the game.

In light of the work by Kujala et al,50 the bi-modal element of the Simon game could proveto be quite beneficial.

Auditory Discrimination

Auditory discrimination is one of the most fun-damental auditory processes. Discrimination offrequency, intensity, and duration of tonal stim-uli can be critical to the discrimination of morecomplex acoustic stimuli such as speech seg-ments and or phonemes. Studies of adult owlmonkeys showed improved frequency discrimi-nation after systematic training, and physiologicand anatomic changes that correlated to the fre-quencies trained, versus those that were not.11

More specifically, this study showed an increasein neural substrate (expanding into adjacentneural areas) related to the frequencies involvedin the AT. Improved performance on both be-havioral tasks and increased amplitude of bothN1 and MMN have been reported53 after train-ing on auditory discrimination tasks. Tremblayet al30 demonstrated that as voice-onset-timediscrimination improved, the amplitude of theN1-P2 late potential waveform also increased.They attributed these changes to an increasednumber of neural connections and neural syn-chrony as a result of AT. Naatanen et al54

demonstrated a gradual sharpening or improveddiscrimination following AT in normal adultsubjects that involved simply repeated presen-tations for comparisons of various complexsounds. The comparison of complex sounds wasset up in an oddball type paradigm to generatean MMN response. Although not initially pres-ent, the MMN was elicited later in the training.Kraus et al23 demonstrated significant physio-logical changes, measured by MMN, in normaladult subjects following speech discriminationtraining to differentiate the CV /da/ with alter-ation of the onset frequencies of the second andthird formant transitions.

Formal auditory discrimination trainingstrategies include improving difference limens(DLs) for nonspeech stimuli. This type of ATbecomes most important when abnormally largeDLs are noted for frequency, intensity, or dura-tion in children with APD. Sloan55 provides


practical ideas for training the discriminationof speech stimuli necessary for good vowel andconsonant identification, particularly in youngchildren. It is well accepted that vowel and con-sonant identification and discrimination are crit-ical to reading and language skills.

Earobics (available from Cognitive Con-cepts, Inc.) is a relatively inexpensive computerprogram that has versions available for homeand school situations, and is easily integratedinto most AT programs. This program trainsphonemic segmentation and auditory discrimi-nation of vowels, consonants, and blends. It alsoaddresses such skills as auditory memory, at-tention, and listening in low redundancy situa-tions. Like FFW, Earobics uses adaptive tech-niques, but unlike FFW, Earobics does notemphasize temporal processing skills. In an in-depth study of the effects of 4 weeks of Earobicstraining for children with APD, Wheadon56

reported a significant training effect on the fre-quency pattern test and the Test of AuditoryComprehension of Language (TACL); therewere no changes seen on dichotic digits, dura-tion patterns, or MLR. Despite the absence ofimprovement on some central auditory tests,the author concluded that Earobics trainingwas of benefit to children with APD enrolledin this study.

Auditory Closure

Children identified with poor auditory closureskills, with language and learning disabilities,and those with poor speech perception can po-tentially benefit from a vocabulary buildingtechnique based on Miller and Gildea’s researchof how children learn words.57 In this proce-dure, the child is exposed to unknown wordsthrough reading, listening, and pronouncingthe new words. The new words are then placedin context and the child is required to use thiscontext to deduce the meaning of the word(i.e., contextual derivation). It is important thatthe context has a sufficient number of contex-tual cues so the child can successfully derivethe meaning of the unknown word. Improveduse of contextual cues and a larger vocabularybase can serve as a useful compensatory strat-

egy when confronting an unfamiliar or misper-ceived word. As a pre-learning technique, tar-get words can be chosen by school personnel orparents who can identify new words for up-coming lessons. This allows the child to focusattention on the lesson itself and may fosterself-confidence, which may improve the likeli-hood of class participation. This procedure in-corporates multiple processes, such as auditory,visual, and cognitive skills for word learning,rather than mere memorization of vocabularylists.

A recent article outlines specific parametersfor conducting this contextual approach to vo-cabulary building and includes results of a smallsurvey of school personnel following use of thecontextual word derivation approach.37 Elevenof 12 respondents favored the use of this vocab-ulary building program and noted improvementin several broad areas, such as word knowledge,reading, and academics in general.37

Binaural Integration and

Binaural Separation

Binaural integration and binaural separationtasks are warranted when deficits are identifiedduring dichotic evaluations. A common find-ing in children with APD is a left ear deficit ondichotic speech tasks. Musiek and Shochatdemonstrated significant improvement in bin-aural listening when dichotic training taskswere incorporated in AT.58 Before training, thelistener demonstrated a unilateral deficit on di-chotic digits tests and moderate, bilateral deficitson compressed speech with reverberation tests.Training involved directing the stimuli to thestronger ear at a reduced level, while maintain-ing the higher intensity level to the weaker ear.This paradigm maintains good performancelevels in the weaker ear as the intensity level ofthe stronger ear is gradually raised over a pe-riod of time. This procedure can also be modi-fied by using temporal offsets that lag in thepoorer ear, which improves the poorer ear’s per-formance. By using adaptive techniques, theoffset differentials are reduced over multiplepractice sessions. This allows the improved per-


formance of the good ear to stabilize back tonormal and maintain the improvement of theweak ear at a higher level of performance.

Dichotic listening training can supportboth binaural integration and binaural separa-tion. The integration task requires the patientto respond to the stimuli in both ears. In theseparation task, the patient is usually asked toignore one ear and respond to stimuli presentedin the other. Dichotic training should employ awide variety of dichotic stimuli such as words,CVs, numbers, phrases, sentences, and com-plex acoustic stimuli. Other modifications ofbinaural integration and separation procedurescan enhance training.59 In a study on the train-ing of dichotic listening similar to that justmentioned,60 improvements were shown in di-chotic listening abilities for a group of childrenwith learning disabilities and dichotic deficitscompared to a control group.

Auditory Vigilance

Auditory vigilance is essentially the ability fora listener to remain attentive to auditory stim-ulation over a sustained period of time. Audi-tory vigilance underlies listening and learningin the classroom, influences APD diagnostictest outcomes, and indeed influences AT itself,as all require the listener to actively focus onauditory stimuli for extended periods of time.It is hypothesized that sustained, increasedawareness to acoustic stimuli will improve au-ditory vigilance. Informal AT techniques in-clude the use of complex or multi-step audi-tory directives requiring that the child firstlisten to instructions in their entirety beforeperforming a task.36 Providing target words,sounds, categories or ideas randomly scatteredthroughout increasingly longer passages thatare read to the child, provides a simple meansof training this ability.

Auditory Memory Enhancement (AME)

As indicated by frequent concerns from par-ents, school personnel, and educational psy-chologists, children with APD often display

poor auditory memory. Auditory memory isinvolved in a number of tasks employed inAP testing, including frequency and durationpatterns tests, dichotic digits, and competingword tests (e.g., the Staggered Spondaic WordTest [SSW]).62 For this reason, AME has beenrecommended when deficits are found acrossthe AP test battery, across multiple academicsubjects, or whenever auditory memory is aconcern.

Based on psychologic research for enhanc-ing recall, AME has been modified for usewith the APD population.52 The procedure in-volves having the child read aloud or havingthe adult read to the child, a paragraph or longerpassage that constitutes a segment. A segmentis defined as the context that conveys a mainidea. The child’s tasks following the reading ofa segment are to distill the segment into themain idea and to sketch the main idea on paper.The process is repeated for each segment untilthe end of the passage. When the child hascompleted all the sketches for each of the de-fined segments, he/she is required to review allthe sketches and verbally relate the conceptsunderlying the segments. The goal of AME isto develop the child’s ability to listen for gen-eral concepts rather than trying to hear, pro-cess, and recall large amounts of detail. A timelimit of one minute for each sketch is recom-mended. This forces the child to reduce infor-mation by transferring it for Gestalt process-ing. This helps the child formulate conceptsand readies the information for easy memoryintegration. Sketching assists memory throughvisual, spatial and motor (multisensory) repre-sentations, that also increase the number ofneural circuits available to memory areas.52

Though based on well-founded concepts re-lated to memory, the AME still awaits furtherevaluation.

SUMMARY AND CLOSINGCOMMENTS

The link between successful AT and plasticityis a critical one. AT for individuals with APDwill benefit from increased understanding ofneural plasticity. As discussed in this article,


there is evidence that the brain can reorganizerelatively quickly when it is deprived of acous-tic input. This reorganization is an importantform of neural plasticity. Questions remain asto whether this plasticity evolves so that theneural substrate can remain physiologically vi-able or to accommodate changing demands onthe system. Neural plasticity is demonstratedwhen the auditory system is trained and im-proves in its function. Whether plasticity evolvesbecause there are neurons held in reserve forjust these purposes or because synaptic endingsare growing and making new connections alsoremains to be clarified. These are some criticalquestions that science is on the verge of an-swering. Clinicians involved in treating andmanaging APD should continue to follow ad-vances in brain plasticity in order to best selectand modify training programs for their patients.

A number of AT techniques are emerging,especially in the area of temporal processing.Probably the best known is the FFW program.Reports of the efficacy of FFW are mixed. Theinitial research report by Tallal et al44 was im-pressive; however, more recent reports haveraised questions about the outcomes.45–48 Thediscrepancies in outcomes may be related tothe selection of patients. The current patientbase may be broader than the base in earlierstudies. Some patients may be better suitedthan others to use and benefit from FFW. Ad-ditional research is needed to determine whichauditory processing deficits are best trainedthrough this procedure. Clinicians also shouldconsider the variety of temporal processingtraining procedures that can be employed in amore informal or formal manner to treat par-ticular deficits. In addition, clinicians shouldconsider the use of other commercial programsthat focus on auditory-language (e.g., phone-mic awareness training) (e.g., Earobics™, Cog-nitive Concepts, Inc., Evanston, IL). Theseprograms may be more applicable to a broaderrange of auditory-language deficits. Thoughsome studies indicate Earobics to have thera-peutic value,56 more research needs to be done.

One of the therapies that is rather intrigu-ing and for which some important mechanismsare known is dichotic listening therapy. Thisadaptive technique of training binaural listen-

ing attacks a common dichotic listening deficitobserved in children with APD—asymmetricalear performance. The auditory memory en-hancement procedure (AME) is a simple ther-apy that incorporates some complex processesthat should benefit memory and, therefore, sup-port auditory processing. The AME procedureinvolves multi-modality representations, orga-nizational perspective and concept formulation;therefore, it also should have a positive influ-ence on academic performance.

In closing, we would like to mention twoaspects of AT that have only been alluded to inthis article but are critical to success—attentionand motivation. Though attention and motiva-tion are not considered part of mainstream au-ditory processing or AT, they are factors thatinfluence outcomes and must, therefore, be ad-dressed by clinicians. Therapies will only workif the patient is willing to put forth effort. Ef-fort requires motivation and attention, and ef-fort is key to triggering plasticity. Though manycomputer programs have attractive animationsto keep the child captivated and working, noth-ing will replace the many advantages of a car-ing, supportive and innovative therapist in cre-ating high motivation and sustained attentionduring therapy. In the days when new instru-mentation and computer software are rapidlybecoming a dominant part of our habilitativeapproach, we must understand that the thera-pist remains the critical link to success.

ABBREVIATIONS

ABR auditory brainstem responseAME auditory memory enhancementAP auditory processingAPD auditory processing disordersAT auditory trainingCANS central auditory nervous systemCVs consonant-vowel syllablesDL difference limenFFW Fast ForWordfMRI functional magnetic resonance imagingMLR middle latency responseMMN mismatch negativity potentialPET positron emission topographySC superior colliculus


SSW Staggered Spondaic Word TestTACL Test of Auditory Comprehension of

Language

REFERENCES

1. Musiek FE, Berge B. A neuroscience view of audi-tory training/stimulation and central auditory pro-cessing disorders. In: Katz J, Masters MG, SteckerN, eds. Central Auditory Processing Disorders:Mostly Management. Needham Heights, MA:Allyn & Bacon; 1998:15–32

2. Lund R. Development and Plasticity of the Brain:An Introduction. New York: Oxford UniversityPress; 1978

3. Robertson D, Irvine DRF. Plasticity of frequencyorganization auditory cortex of guinea pigs withpartial unilateral deafness. J Comp Neurol 1989;282:456–471

4. Hassamannova J, Myslivecek J, Novakova V. Ef-fects of early auditory stimulation on cortical areas.In: Syka J, Aitkin K, eds. Neuronal Mechanisms ofHearing. New York: Plenum Press; 1981

5. Schwaber M, Garraghty P, Morel A, Kaas J. Neu-roplasticity of the adult primate auditory cortexfollowing cochlear hearing loss. Am J Otol 1993;14:252–258

6. Scheich H. Auditory cortex: comparative aspects ofmaps and plasticity. Current Opinion in Neurobi-ology 1991;1:236–247

7. Chermak G, Musiek F. Central Auditory Disorders:New Perspectives. San Diego, CA: Singular; 1997

8. Willot J. Effects of aging, hearing loss, and ana-tomical location on thresholds of inferior colliculusneurons in C57BL/6 and CBA mice. J Neurophys-iol 1986;56:391–408

9. Rajan R, Irvine D, Wise L, Heil P. Effect of unilateralpartial cochlear lesions in adult cats on the representa-tion of lesioned and unlesioned cochleas in primaryauditory cortex. J Comp Neurol 1993;338:17–49

10. Kaas J. Plasticity of sensory representations in theauditory and other systems of adult mammals. In:Salvi R, Henderson D, Fiorino F, Colletti V, eds.Auditory System Plasticity and Regeneration. NewYork: Thieme; 1996:213–223

11. Recanzone G, Schreiner C, Merzenich M. Plas-ticity in the frequency representation of primaryauditory cortex following discrimination train-ing in adult owl monkeys. J Neurosci 1993;13:97–103

12. Brainard M, Knudsen E. Experience-dependentplasticity in the inferior colliculus: a site for visualcalibration in the neural representation of auditoryspace in the barn owl. J Neurosci 1993;13:4589–4608

13. Miller G, Knudsen E. Early auditory experienceinduces frequency specific adaptive plasticity in theforebrain gaze field of the barn owl. J Neurophysiol2001;85:2154–2194

14. Gold J, Knudsen E. Abnormal auditory experienceinduces frequency-specific adjustments in unit tun-ing for binaural localization cues in the optic tectumof juvenile owls. J Neurosci 2000;20:862–77

15. Korte D, Rauschecker J. Auditory spatial tuning ofcortical neurons is sharpened in cats with earlyblindness. J Neurophysiol 1993;70:1717–1721

16. Irvine D, Rajan R, Brown M. Injury- and use-related plasticity in adult auditory cortex. Audiol-ogy & Neuro-Otology 2001;6:192–195

17. Silman S, Silverman C, Emmer A, Gelfand S.Adult onset auditory deprivation. J Am Acad Au-diol 1992;3:390–396

18. Manrique M, Cervera-Paz F, Huarte A, et al.Cerebral auditory plasticity and cochlear implants.Int J Pediatr Otorhinolarygol 1999;49(Suppl 1):S193–S197

19. Allum-Mecklenburg D, Babighian G. Cochlearimplant performance as an indicator of auditoryplasticity in humans. In: Salvi R, Henderson D,eds. Auditory System Plasticity and Regeneration.New York: Thieme;1996:395–404

20. Labadie R, Carrasco V, Gilmer C, Pillsbury, H.Cochlear implant performance in senior citizens.Otolaryngol Head Neck Surg 2000;123:419–424

21. Geier L, Barker M, Fisher L, Opie J. The effect oflong-term deafness on speech recognition in post-lingually deafened adult CLARION cochlearimplant users. Ann Otol Rhinol Laryngol 1999;177(Suppl):80–83

22. Nishimura H, Hashikawa K, Doi K. Sign lan-guage ‘heard’ in the auditory cortex. Nature 1999;397:116

23. Kraus N, McGee T, Carrell T, et al. Central auditorysystem plasticity associated with speech discrimina-tion training. J Cogn Neurosci 1195;7:25–32

24. Kraus N. Speech sound perception, neurophysiol-ogy, and plasticity. Int J Pediatric Otorhinolaryngol1999; 47:123–129

25. Ponton C. Possible application of functional imag-ing of the human auditory system in the study of ac-climatization and late onset deprivation. Ear Hear1996;17(Suppl):78S–86S

26. Ponton C, Moore J, Eggermont J. Prolonged deaf-ness limits auditory system developmental plastic-ity: evidence from an evoked potentials study inchildren with cochlear implants. Scand Audiol1999;51(Suppl):13–22

27. Ponton C, Eggermont J, Kwong B, Don M. Matu-ration of human central auditory system activity:evidence from multi-channel evoked potentials.Clin Neurophysiol 2000;111: 220–236


28. Tremblay K, Kraus N, Carrell T, McGee T. Centralauditory system plasticity: generalization to novelstimuli following listening training. J Acoust SocAm 1997;102:3762–3773

29. Tremblay K, Kraus N, McGee T. The time courseof auditory perceptual learning: neurophysiologicalchanges during speech-sound training. Neurore-port 1998;9:3557–3560

30. Tremblay K, Kraus N, Mcgee T, Ponton C, Otis B.Central auditory plasticity: changes in the N1-P2complex after speech sound-training. Ear Hear2001;22:79–90

31. Purdy S, Kelly A, Thorne P. Auditory evoked po-tentials as measures of plasticity in humans. Audi-ology & Neuro-Otology 2001;6:211–215

32. Jirsa R. The clinical utility of the P3 AERP in chil-dren with auditory processing disorders. J SpeechHear Res 1993;35:903–912

33. Koelsch S, Shroger E, Tervaniemi M. Superiorpre-attentive auditory processing in musicians.Neuroreport 1999;10:1309–1313

34. Russeler J, Altenmuller E, Nager W, Kohlmetz C,Munte T. Event-related brain potentials to soundomissions differ in musicians and non-musicians.Neurosci Lett 2001;308:33–36

35. Jerger J, Musiek FE. Report of the consensus con-ference on the diagnosis of auditory processing dis-orders in school-aged children. J Am Acad Audiol2000;11:467–474

36. Musiek FE, Baran J, Schochat E. Selected man-agement approaches to central auditory process-ing disorders. Scand Audiol 1999;28(Suppl 51):63–76

37. Ahissar E, Vaadia E, Ahissar M, et al. Dependenceof cortical plasticity on correlated activity of singleneurons and on behavioral context. Science 1992;257:1412–1415

38. Farmer M, Klein R. The evidence for a temporalprocessing deficit linked to dyslexia: a review. PschonBull & Rev 1995;2:460–493

39. Bornstein S, Musiek F. Implication of temporal pro-cessing for children with learning and languageproblems. In: Beasley D, ed. Contemporary Issues inAudition. San Diego, CA: College Hill; 1984:25–54

40. Nagarajan S, Mahncke H, Salz T, et al. Cortical au-ditory signal processing in poor readers. Proc NatlAcad Sci USA 1999;96:6483–6488

41. Temple E, Poldrack R, Propopapas A, et al. Dis-ruption of the neural response to rapid acousticstimuli in dyslexia: evidence from functional MRI.Proc Natl Acad Sci USA 2000;97:13907–13912

42. Temple E, Poldrack R, Salidis J, et al. Disruptedneural responses to phonological and orthographi-cal processing in dyslexic children: an fMRI study.Neuroreport 2001;12:299–307

43. Merzenich M, Jenkins W, Johnston P, et al. Tem-poral processing deficits of language-learning im-

paired children ameliorated by training. Science1996;271:77–81

44. Tallal P, Miller S, Bedi G, et al. Language compre-hension in language-learning impaired childrenimproved with acoustically modified speech. Sci-ence 1996;271:81–84

45. Friel-Patti S, Frome-Loeb D, Gillam R. Lookingahead: an introduction to five exploratory studiesof Fast ForWord. Am J Speech Lang Pathol 2001;10:195–2002

46. Frome-Loeb D, Stoke C, Fey M. Languagechanges associated with Fast ForWord language:evidence from case studies. Am J Speech LangPathol 2001;10:216–230

47. Thibodeau L, Friel-Patti S, Britt L. Psychoacousticperformance in children completing Fast ForWordtraining. Am J Speech Lang Pathol 2001;10:248–257

48. Gillam R, Frome-Loeb D, Friel-Patti S. Lookingback: a summary of five exploratory studies of FastForWord. Am J Speech Lang Pathol 2001;10:269–273

49. Kujala T, Myllyviita K, Tervaniemi M, et al. Basicauditory dysfunction in dyslexia as demonstratedby brain activity measurements. Psychophysiology2000;37:262–266

50. Kujala T, Karma K, Ceponiene R, et al. Plasticneural changes and reading improvement causedby audiovisual training in reading-impaired chil-dren. PNAS 2001;98:10509–10514

51. Tallal P, Merzenich M, Miller S, Jenkins W. Lan-guage learning impairments: integrating basic sci-ence, technology, and remediation. Exp Brain Res1998;123:210–219

52. Musiek FE. Habilitation and management of audi-tory processing disorders: overview of selected pro-cedures. J Am Acad Audiol 1999;10:329–342

53. Menning H, Roberts LE, Pantev C. Plastic changesin the auditory cortex induced by intensive fre-quency discrimination training. Neuroreport 2000;11:817–822

54. Naatanen R, Schroger E, Karakas S, TervaniemeiM, Paavilainen P. Development of a memory tracefor a complex sound in the human brain. Neurore-port 1993;4:503–506

55. Sloan C. Treating Auditory Processing Difficultiesin Children. San Diego, CA: College Hill; 1986

56. Wheadon L. The influence of auditory training onbehavioral and electrophysiologic test measures inchildren with central auditory processing disorders(Dissertation). Auckland, New Zealand: Universityof Auckland; 1999

57. Miller G, Gildea P. How children learn words. Sci-entific American 1987;257:94–99

58. Musiek FE, Shochat E. Auditory training and cen-tral auditory processing disorders: a case study.Semin Hear 1998;19:357–366


59. Musiek FE, Pinheiro M. Dichotic speech tests in thedetection of central auditory dysfunction. In: Pin-heiro M, Musiek FE, eds. Assessment of CentralAuditory Dysfunction: Foundations in Clinical Cor-relates. Baltimore, MD: Williams & Wilkins; 1985

60. Katz J, Chertoff M, Sawusch R. Dichotic training,J Audit Res 1984;24:251–264

61. Zattore R, Halpern A, Perry D, Meyer E, Evans A.Hearing in the mind’s ear: a PET investigation of

musical imagery and perception. J Cognitive Neu-rosci 1996;8:29–46

62. Bornstein S, Musiek FE. Implication of temporalprocessing for children with learning and languageproblems. In: Beasley D, ed. Audition in Child-hood: Methods of Study. San Diego, CA: CollegeHill Press; 1984

263

Management of Auditory Processing Disorders; Editor in Chief, Catherine V. Palmer, Ph.D.; Guest Editor, Gail D.Chermak, Ph.D. Seminars in Hearing, volume 23, number 4, 2002. Address for correspondence and reprint requests:Frank E. Musiek, Ph.D., Department of Communication Sciences and Otolaryngology, School of Medicine, Universityof Connecticut, Storrs, CT 06269. E-mail: [email protected]. 1Department of Communication Sciences andOtolaryngology, School of Medicine, University of Connecticut, Storrs, Connecticut; 2Dartmouth-Hitchcock MedicalCenter, Lebanon, New Hampshire. Copyright © 2002 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, NewYork, NY 10001, USA. Tel: +1(212) 584-4662. 00734-0451,p;2002,23,04,263,276,ftx,en;sih00220x.

Plasticity, Auditory Training, and AuditoryProcessing DisordersFrank E. Musiek, Ph.D.,1 Jennifer Shinn, M.S.,2 and Christine Hare, M.A.1

ABSTRACT

Auditory training (AT) for the treatment of auditory processing dis-orders (APD) has generated considerable interest recently. There is emergingevidence that well conceived AT programs can improve higher auditory func-tion. The plasticity of the brain underlies the success of AT. This article re-views brain plasticity and the role of plasticity in AT outcomes, and highlightskey studies that provide insight into the clinical use of AT for APD.

KEYWORDS: Auditory processing disorder, auditory training, plasticity

Learning Outcomes: Upon completion of this article, the reader will be able to (1) understand the rationalebehind auditory plasticity as it applies to APD, (2) understand new therapy techniques for APD, and (3) formu-late therapy approaches using plasticity.

auditory nervous system (CANS). Improve-ment in higher auditory function is related tothe capacity of the central nervous system tochange. This change is tightly linked to, and isa result of, plasticity. This article reviews plas-ticity and AT within the framework of APD.

AUDITORY PLASTICITY DEFINED

There is a significant body of evidence fromboth animals and humans that supports im-provement in auditory tasks through the use ofAT. As we begin to gain new insight into the

Recently, there has been great interest inthe use of auditory training (AT) for the treat-ment of auditory processing disorders (APD).The use of AT for peripheral auditory prob-lems is not a new concept, in that it has an ex-tensive history.1 The use of AT for treatmentof APD is new and its application is differentfrom the classic use of AT. Most important tothis difference is that AT applied to APD istargeting the brain as the main site of media-tion, and the brain, unlike the auditory periph-ery, is plastic. Though peripheral sensorineuralloss cannot be improved upon with AT, thismay not be the case for deficits of the central


fundamental mechanisms involved in improvedperformance on specific auditory tasks, we musttake into consideration the concept of auditoryplasticity. The scientific literature on the sub-ject of plasticity is immense. For the purposesof our discussion, we will concentrate specifi-cally on auditory plasticity and overview its re-lationship to AT.

Neural plasticity is defined by Lund2 as“neural form and connections that take on apredictable pattern (probably genetically deter-mined) often referred to as neural specificity;exceptions to this predictable pattern, in cer-tain circumstances, represent neural plasticity.”Auditory (specific) plasticity can be defined asthe alteration of nerve cells to better conformto immediate environmental influences, withthis alteration often associated with behavioralchange.1 It is necessary for both clinicians andresearchers to have a strong foundation in thefundamentals of plasticity and its relationshipto AT, as it is key to management of patientswith APD.

ANIMAL RESEARCH

Perhaps the greatest body of evidence that sug-gests plasticity in the auditory system relatedto training lies in the extensive number of ani-mal studies. Though the scope of this articledoes not allow extensive coverage of this topic,we will highlight a few studies that hopefullyprovide some principles of interest regardingplasticity. Some of the early research on plas-ticity and training via animal models focusedon sub-cortical levels and the brainstem; how-ever, the auditory cortex has recently gainedconsiderable attention. The evidence of corti-cal plasticity is shown impressively by func-tional reorganization, that has attracted muchattention from scientists and clinicians in a va-riety of disciplines.

Studies on rodents, monkeys, cats, and theavian species have shown reorganization withinthe auditory cortex.3–7 There are two main con-ditions that can result in reorganization of theauditory cortex, sub cortex, and brainstem. Oneof these is deprivation that, experimentally, isinduced by creating hearing loss in some man-

ner. This can be achieved by creating a conduc-tive loss, which is often done by plugging theear and/or removing or damaging middle earstructures. Sensorineural hearing loss is experi-mentally created by exposing the ear to ototoxicdrugs, high intensity noise, or mechanicallydamaging the cochlea. These hearing losses willdeprive the higher auditory system of stimula-tion and will result in a reorganization of theauditory cortex.3,8–10 Raising animals in acous-tically attenuated chambers is another methodof creating deprivation effects. High stimulationor tasking of the auditory system is the othercondition that can create plastic changes in theCANS.4,11

The barn owl is one of the most studiedanimal models for demonstrating plasticity.One of the reasons for studying the barn owl isthat its optic tectum (or midbrain), which issimilar in nature to the superior colliculus (SC)of the mammal, contains neural substrate withrespect to both auditory and visual maps. TheSC is one of the most highly studied corticalstructures because of its ability to reorganize.The SC is a useful structure to investigate, as itis particularly sensitive to experience-dependentplasticity and alteration of auditory maps. Thebarn owl provides an extensive amount of mod-eling information to base on the derivation ofits auditory map.12

Evidence of auditory plasticity has beengained through both auditory deprivation andauditory training with the barn owl. Previousdeprivation studies, in which barn owls wereraised with monaural occlusion, provided phys-iologic evidence of inter-aural level differencesoccurring in the posterior division of the ven-tral nucleus of the lateral lemniscus within thebrainstem.13 Through the use of acoustic filter-ing devices, Gold and Knudsen14 investigatedthe effect of altered auditory experience on inter-aural timing differences of barn owls. Relativeto owls raised under normal auditory conditions,owls raised with unilateral auditory deprivationshowed neuronal changes of the inferior col-liculus (in the optic tectum) and changes in be-havioral responses, demonstrating that plasticitywithin the inferior colliculus plays a significantrole in adjustments in the inter-aural timingdifference and frequency tuning. These plas-


tic changes in auditory experience as a resultof deprivation and accommodation are notedat both the subcortical12 and higher-level sys-tems.15

A substantial body of research suggeststhat reorganization takes place in all mammals.These findings have strong implications forauditory plasticity in humans. The general find-ings show that sensory maps have training de-pendent reorganization capabilities and sug-gest that a majority, if not all of the brain, maybe plastic.10 There also are indications that inmammals plasticity is lateralized. That is, cor-tical reorganization seems to take place in thehemisphere contralateral to the ear deprived ofacoustic input or for which there is peripheraldamage, which reduces peripheral input.10

MECHANISMS UNDERLYINGCORTICAL PLASTICITY ANDREORGANIZATION

Before auditory plasticity and its relationshipto AT can be fully appreciated, there must be acomprehensive understanding of the mecha-nisms that underlie this critical process. Nu-merous theories have been proposed and themore widely accepted will be briefly discussedfor the purpose of this review.

It is well established that plastic changes area result of neuronal responses to both externalstimuli (i.e., sounds in our environment) andinternal stimuli (e.g., thinking about certainsounds when we listen to music or read). Wetend to think of our brain as a highly stableorgan, when in fact it is its instability that we relyon most for development and auditory learning.There are generally three types of plasticity inthe auditory domain: (1) developmental plastic-ity, (2) compensatory plasticity resulting from alesion occurring somewhere within the auditorysystem, and (3)learning-related plasticity.6

Simple auditory tasks, such as detecting thepresence of a pure tone, require relatively littleneural substrate in the CANS. In contrast, com-plex tasks—such as discriminating small differ-ences among various acoustic stimuli in noise—require considerable neural activity. The simpleauditory task may not be improved signifi-

cantly with AT; however, complex tasks aremore likely to benefit as they necessitate morebrain function (and neural substrate) and, there-fore, are more likely to benefit from brain plas-ticity.1

Generally there are two ways in which thebrain is thought to reorganize. Although bothrequire neuronal changes, their mechanismsare quite different. Reorganization may involvethe activation of neurons and neural connec-tions that were previously in a state of rest.Thesereserve neurons replace the nonfunctional neu-rons. That is, the reserve neurons would re-place connections that are no longer active dueto damage or lack of stimulation.The other pro-posed mechanism underlying brain reorganiza-tion requires that new connections be formed.Both mechanisms involve changes that may beslow and require an extensive amount of timeand training, or the changes can be extremelyrapidly occurring through development andmaturation without intervention.10

It should be noted that neural reorganiza-tion might not be considerable and obvious.According to Kaas,10 these structural changesmay be rather subtle, resulting from modifica-tions in the extent of synaptic connections thatbecome more effective, or from changes inneural location, resulting in a more effectivetargeting or spreading of axonal and dendriticarbors, that in turn results in not only more butbetter located synaptic connections. If newdendrites are either being recruited or devel-oped, their connections must be appropriate toensure that they evolve meaningful function.

CLINICAL CORRELATES OF AUDITORY PLASTICITY IN HUMANS

Clinical experience with hearing aids andcochlear implants provide important insightand evidence of the plasticity phenomenon andraise several key questions as well. As mentionedpreviously, auditory reorganization can resultfrom lack of input to central systems, usuallyafter damage to the periphery. It is known thatthe auditory cortex will reorganize tonotopi-cally if input is markedly reduced at certain