Auditory Illusions and Confusions
These failures of perception are studied because they isolate and
clarify sonle fundamental processes that normally lead to accurac.
of perception and appropriate Interpretation of ambiguous sound.
For more than a century visu al illu-sions have been of particular inter-est to students of perception. AI-
though they are in effect misjudgmentsof the l'eal world, they apparently reBectthe operation of fundamental perceptualmechanisms, and they serve to isolateand clarify visual processes that are nor-mally inaccessible to investigation. Audi-tory illusions, on the other hand, havereceived little scientific attention. Until
recently the Beeting nature of auditorystimuli made it difficult to create, control
and reproduce sound patterns as readilyas visu al ones. The tape recorder made iteasy to manipulate sounds, and yet fora time there was little examination of
auditory illusions, perhaps because therewas no historical tradition to build on-
no puzzles inherited from the experi-mental psychologists of the past cen-tury, as there were in the case of opticalillusions. Some new investigations, how-ever, have led to the discovery of illu-sions in hearing that help to explain thehuman ability to extract informationfrom Beeting patterns of sound. Theseinvestigations have also led to the iden-
tification of confusions in hearing thathelp to explain some limitations of thatability.
Consider for a moment that you are ata convention banquet. \Vhile you
are still finishing your dinner the after-dinner speeches begin. The clatter ofdishes masks some of the speech sounds,as do occasion al coughs from your neigh-bors and your own munching. Nonethe-less, you may be able to understand
what the speaker is saying by utilizingthe information that reaches you duringintervals that are relatively free of theseinterfering noises. ln order to understandhow speech perception functions in thepresence of transient noises, we and
Charles J.Obusek did some experiments30
by Richard M. Warren and Roslyn P. Warrcn
last year in our laboratory at the Univer-sity of 'Wisconsin at Milwaukee. First werecorded the sentence "The state gover-nors met with their respective legisla-tures convening in the capital city."Then we carefully cut out of the tape re-cording of the sentence one phoneme, orspeech :;ound: the first "s" in "Iegisla-tures." We also cut out enough of thepreceding and following phonemes toremove any transition al eues to the iden-tity of the missing speech sound. Final-Iy, we spliced the recorded sound of acough of the same duration into the tapeto replace the deleted segment.
\Vhen this doctored sentence was
played to listeners, we found that wehad created an extremely compelling il-lusion: the missing speech sound washeard as clearly as were any of the pho-nemes that were physically present. Wecalled this phenomenon "phonemic res-toration." Even on hearing the sentenceagain, after having been told that asound was missing, our subjects couldnot distinguish the ilIusory sound fromthe real one. One might expect that themissing phoneme could be identified bylocating the position of the cough, butthis strategy was of no help. The coughhad no clear location in the sentence; it
seemed to coexist with other speechsounds without interfering with theirintelligibility. Phonemic restoration alsooccurred with other sounds, such as abuzz or tone, when these sounds were asloud as or louder than the loudest sound
in the sentence. Moreover, phonemicrestorations were not limited to singlespeech sounds. The en tire syllable "gis"in "Iegislatures" was heard clearly whenit was replaced by an extraneous soundof the same duration.
We did find a condition in which the
missing sound was not restored. \Vhena silent gap replaced the "s" in "legisla-tures," the gap could be located within
the sentence and the missing sound idtified. ln visual terrns, it was as iferasure of a letter in a printed text coud'~be detected, whereas an opaque blot']over the same symbol would ;esult iJi~
iIlusory perception of the obliterated let-'ter, with the blot appearing as a trans-parent smear over another portion of the!text [see top illustration on pages32 and!33]. Of course, in vision a blot can 001
localized readily, and even the mor~j1 . " f d ' .11 . " i!e USlve proo rea el' 5 1 USIOns can u
eliminated when the reader is told in ad]vance just where the error in the text J1'curs. With phonemic restorations, holever, knowledge of the nature of tIjextraneous sound and of the identitybl,~.,the missing phoneme does not preventclear perception of the missing sound!:even when the stimulus is played to thc~;Jlistener as many times as he whes. :t:
The inability to localize an extraneo~jsound in a sentence was first reported2J1960 by the
British workers Peter L
foged and Donald E. Broadbent. S'they employed brief intrusive 50(clicks and short hisses) and took
that no phoneme was obliterated, Pnemic restorations did not arisc. Sn .
short, nonmasking extraneouS 50were later used by a group at the,sachusetts Institute of Technology
included Jerry A. Fodor, Merrill F. -rett and Thomas Bever. They have:
ported that systematic errors in localthe clicks are caused by various fealof sentence structure, and they have';the errors to explore those featureS.
perceptual synthesis of the ?h
is accomplished on the basI.s ~fbal context. ln the case of the rnlSSIll
in "legislatures" the context pri~rabsent sound suffices for identifi,vVhat about a sentence so const
that the context necessary to idenobliterated sound does not co
aTIJe governors metstate
b~ ~ ~--
with legislatures conveningtheir respective in the capital---------- -------------
The governorsm~ with ~~ir_- - respect~~~~~I-atu~~~-~~~~~~~~-g___~-~h~_- ~i~~'
PHONEMIC RESTORATION is an illu,ion that ,how, the im-
portance of context in determining what sound is heard. A sentcneewas recorded on tape (a). Then the first "s" in "Iegiolatures" was
]ater? \Vith the symbo] " representing aloue! cough that replaces a speech sound,consider a spoken sentence bcginning,"It \Vas found that the" ee] was on the
_." The context provided by the lastword in the sentence shou]d resolve the
ambiguity and de termine the appropri-ate phonemic restoration, Among the\\'ords that eould complete the sentenceare "axJe," "shoe," "orange" and "tab]e."Each implies a e!ifj'erent speech soundfor the preceding \\"OH] fragment, re-spectivelv "wheel," "hec]," "pecJ" am]"mea1." PrcJiminary studies by Cary
H!GH TONE EUZZ
one 7 th re
excised and a cough of the same duration (black rectangle) wasspliced in its pla,'e (b L Wh en the altered sentence was played tsuhject" the mi",ing "s" was heard clearly Ic) and localization 07
Sherman in our laboratorv have indi-
cated that the listener e!oes experieneethe appropriate phonemie restoration,apparent]y by storing the ineomp]ete in-formation until the neccssary context issuppliee! so that the required phonemel'an he synthesizee!. \Ve are still im'esti-gating the influence of such factors asthe e!uration of extraneous sounds in re-
lation to the e!uration of the missingphoneme and the maximum temporalseparation bel\veen the ambiguous \\'ore!fragment and the resolving context thatwill still permit phonemic restoration.
LOIN TONE HISS
The use of subsequent context for Cor-reding errors hae! been suggested On]ogiea] groune!s by George A. l\JilIer ofRoekefeJJer University. He reasoned thatunless some sueh strategy were avaj].able, a mistake once made while Hsten-ing to spoken diseourse wauld cause
errors in interpreting the foJJowing par.tions of the message ta pile up, until the".
entire system eventuaJJy staJJed. The,long delays in museu]ar acth'jty that'have been observee! in the skilled tran-
scription of an ineoming message a]sosuggst that storage of ineaming ]an-:
HIGH TONE BU
7eight 7two 70ne
TEMPORAL CONFUSION was o]!'cncd "hen a high tone, a huzz,a low tone and a hiss (repre,ented here ,('hcmati('ally " eaeh la'ting
200 milli,c('on(b, werc l>re,ented repeatcdly (top). Subjects rouI
not report the sequcnec of the sounds "roperly whetber they tri
e e e e e e e e e e e 100 0 0 0 0 0 0 0 0 01i i i i i i i i i i i i i i i i i i ii i i il U'U U U U U U U U U u le e e e e e e
( 200 >MILLISECONDS
i iiiiiiiiiiiiiiii UUUUUUUU", ,~ ~ "
e e e e e e e'e
FOUR VOWEL SOUNDS were used in another experim"nt ontemporal confusion. When the vowel sound, of "beet," "boot," "bit"
and "hut" wcre presented at a sustained level for 200 JO.thcir scquen('e ('ould not hl' detcrmined (top), De]e