Hearing Musical Streams
Author(s): Stephen McAdams and Albert Bregman
Source: Computer Music Journal, Vol. 3, No. 4 (Dec., 1979), pp. 26-43+60
Published by: The MIT Press
Stable URL: http://www.jstor.org/stable/4617866 .
Accessed: 27/09/2011 13:53
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
The MIT Press is collaborating with JSTOR to digitize, preserve and extend access to Computer Music
Journal.
http://www.jstor.org
Hearing
Musical
Streams
AlbertBregman
Departmentof Psychology
McGillUniversity
Montreal,Quebec,Canada
Stephen McAdams
Hearingand Speech Sciences
Stanford University School of Medicine
Stanford, California 94305
Introduction
Theperceptual
effectsof a soundaredependentupon
the musicalcontextin whichthatsoundis imbedded.Thatis,
a givensound'sperceivedpitch,timbre,andloudnessareinfluencedby the soundsthat precedeit, coincidewith it,
andevenfollowit in time.Thus,thiscontextinfluencesthe
waya listenerwillassociatethe soundwithvariousmelodic,
rhythmic,dynamic,harmonic,and timbralstructureswithin
the musicalsequence.It thus behoovesthe composerand
to understand
the variousperceptualorganizing
interpreter
of musicalcontextfrom
principlesthataffectthe derivation
of
acoustic
We
events.
include
the interpreter
here
sequences
becauseseveralmusicaldimensions,suchas timbre,attack
and decay transients,and tempo, are often not specified
exactlyby the composerandarecontrolledby the performer.
In this articlewe shalldiscussprinciplesthat describe
affectthe perceivedcontinuihowvariousmusicaldimensions
ty of music.LeonvanNoordenhasstatedthat "insequences
wherethe tones follow one anotherin quicksuccession,
effectsareobservedwhichindicatethat the tones arenot
by the perceptionsystem.Onthe one
processedindividually
handwe find varioustypesof mutualinteractionbetween
successivetones, such as forwardand backwardmasking,
anddurationinteractions.
loudnessinteractions
Onthe other
hand,a kindof connectionis foundbetweenthe successive
sonicevents,
perceivedtones."[28, p.1 ]. As for simultaneous
has
that
different
areexsounds
9]
Bregman[5,
suggested
tractedaccordingto variousperceptualand cognitiveorganizationalmechanisms
fromthe superimposed
acousticvibrations.Whilesomeresearchers
wouldlike to attributethese
or earlycentral
to mechanisms
in the peripheral
phenomena
nervoussystem(whicharesurelyinvolvedto someextent,
see [24, 33]), prominentmembersof thismorepsychophysicallyoriented"school"arebeginningto thinkthe organizationis too complexto be so easilyexplained.However,rather
thandelveinto theoreticalexplanations
of thesephenomena,
it will sufficefor the presentpurposeto describethemin
that
generalandto brieflyquantifysomesalientparameters
andAlbertBregman
? 1979 by StephenMcAdams
Page 26
haveintrigueduswithcompositional
possibilities.Thisarticle
thuspresents,in a tutorialfashion,a reviewof research(includingour own) whichhas directimplicationsfor musicians,especiallyfor composers
workingwithcomputermusic.
andin mostotherresearchcited,comInall of ourresearch
wereusedto
PDP-11minicomputers)
puters(predominantly
usedfor
also
were
sounds
the
synthesize
presented.They
of soundstimuli,collectionof responsesfromthe
presentation
listeners,andanalysisof the data.For thoroughsummaries
of thisareaof research,see [2, 5,
andtheoreticaltreatments
9.] .
WhatIs An AuditoryStream?
Auditorystreamformationtheoryis concernedwith
how the auditorysystemdetermineswhethera sequenceof
acousticeventsresultsfromone, or morethanone, "source."
A physical"source"maybe consideredas somesequenceof
acousticeventsemanatingfromone location.A "stream"
such
thatmentallyrepresents
is a psychological
organization
a sequenceanddisplaysa certaininternalconsistency,or continuity,that allowsthat sequenceto be interpretedas a
"whole."By way of example,two possibleperceptualorin
of a repeatingsix-tonesequenceareillustrated
ganizations
axis
and
the
horizontal
is
on
1.
Time
represented
Figure
on the verticalaxis.Thedottedlines
frequencyis represented
connectingthe tonesin the figureindicatethe streampercepts.
six tones areheardone afterthe
In the firstconfiguration,
otherin a continuouslyrepeatingcycle (TapedIllustration
la)l; it is easy to followthe entiremelodicpattern.In the
secondpercept,though,onemightheartwo separatethreetone patternswhichappearto havelittle relationshipto
eachother(TapedIllustration1b).It is difficultin thiscase
to followthe originalsix-tonepattern.Notethatin the first
exampleone streamis heard,andin the second,two areheard.
(1) A tapecontainingsoundexamplesis availablefrom
Mr.McAdams.
Descriptionsof the tapedillustrations
arefoundinAppendix1.
Computer Music Journal, Box E, Menlo Park, CA 94025
Volume 3 Number 4
D
D
D
B'
/B
B"B
I
"
II
O
F
-
.
I
II
I
I
%
F
F
Cr
-
ST
%
(a)
%
(b)
Time-
Time -,
Two Streams
One Stream
Figure 1. A repeating6-tone sequencecomposed of interspersedhigh and low tones can result in different percepts. In Figure la,
with high and low tones alternatingat a tempo of 5 tones/sec., one perceptualstream is heard. In Figure1b, at a tempo of
10 tones/sec., the high tones perceptuallysegregatefrom the low tones to form two streams(cf. Taped Illustration1).
D
4
F
t
t
o
0
D
L.
C
A
C
C"=I
U-U
A
,/
u
F
/
Time -+
Time -+
I:JJ
Ji
HighStream
I:JJJ:
Low Stream
J
I
HighStream
l
JU
:I
Low Stream
Figure2. Due to the competition amongstreamorganizations,tone F may be perceivedas belongingto either the higherstream
or the lower stream but not to both. The organizationof the streamschanges, among other things, the perceived rhythmic
structure,as indicatedundereach diagram(cf. Taped Illustration2).
StephenMcAdamsand AlbertBregman:HearingMusicalStreams
Page27
In the rest of this section we will discusssome of the
propertiesexhibited by a stream.For example, it is possible
to focus one's attention on a stream and follow it through
time [6, 28]. In the first taped example one can follow the
six-tone pattern without any trouble. Thus a stream must
exhibit a certaincoherenceover time. However,in the second
example, it is difficult to follow the six-tone pattern,but it is
easy to follow either three-tone pattern. Notice that one can
pay attention to either the higheror lower stream,switching
between them at will, but that it is not possible to attend to
both simultaneously(Taped Illustration ib). Indeed, each
three-tone patternin Figure lb constitutes a separatestream
and maintainsits own temporalcoherence.Whilea listeneris
paying attention to one coherent stream, other acoustic information is perceptually relegated to the background.If
one group of sounds is distinct enough, the foreground-backgroundrelationmay be almost involuntaryand it may require
a greatdeal of attentionaleffort to focus on streamsinitially
relegatedto the background.
The information-processingnatureof the streamsegregation processis suggestedby the observationthat the segregation of a sequenceinto smallerstreamstakes time to occur
[4, 13]. In Figure lb, notice that one can hear a six-tone pattern for the first few cycles before it segregatesinto two
separatestreams(Taped Illustrationlb). It thus appearsthat
the perceptualsystem assumesthings are coming from one
source until it acquiresenough information to suggest an
alternateinterpretation.
When temporalcoherence is lost in a sequence, it becomes more difficult to orderthe events of that sequencein
time [6, 11, 14, 28] . In Figure la, it would be easy to tell the
orderof tones A, B, and C. But as this largerstreambreaks
down into the smallerstreamsof Figureib, i.e., as temporal
coherenceis lost, it becomes more difficult to judge the order
of these tones. In such a case one might notice that tone A
comes before tone C but it would be hard to tell whether
tone B came before A, between A and C, or after C. This
statementshould be qualifiedby noting that in the tone sequencesmentioned all of the tones are of equal loudnessand
timbre. The tone sequences are furthermorecontinually
recyclingand are faded in so that the listenercannot label tone
A as being the first tone in the sequence. Informationin a
musicalcontext, such as the first beat being emphasizedas a
downbeat,may give the listeneran anchorpoint againstwhich
to relatethe temporalpositions of other tones. Thus one can
judge the order of events in time within a given perceptual
streambut not necessarilyacrossstreams.Accompanyingthis
is the observationthat different streamscan appearto overlap
in time even when they do not. Again in Figure lb, notice
that one can hear two three-tone patterns apparentlygoing
on at the same time even though the tones are alternating
(Taped Illustrationib).
If an event is potentially a memberof more than one
"competing"stream,one may perceiveit as belongingto one
streamor anotherbut not to both simultaneously[3, 10, 11] .
This is not to say that a musiciancannot hear severalsimultaneous lines. The point is that it is impossibleto use several
parsing schemes at the same time. Figure 2 illustrates the
effect on our second example of moving the higher triplet
into closer frequencyproximity to the lower triplet. We can
find an intermediateposition where the lowest tone, tone F,
Page 28
can group with either the higher or lower stream (Taped
Illustration2). Notice that this regroupingresultsin a rhythmic transformationas shown under each configuration.
Some non-musicalexamplesof this phenomenonare the facevase illusion and the reversibleNecker cubes; Escherand
Vasarelyhave produced art works based on such principles
of perceptualorganization.
Finally, this bringsup the relationshipbetween "source"
and "stream."A streamis perceivedas emanatingfrom a single
source. So, in the first example, the pattern was fairly continuous and was easily recognizedas coming from one source;
but in Figure lb, the large frequency distance between the
two three-note groups introduced a sort of discontinuity
that caused the perceptualsystem to interpretthe sequence
as resultingfrom two sources. Since at any givenmoment the
composite pressurevariationsstimulatingthe ear result from
severalsources,the auditorysystem needs a battery of heuristics to parse, or segregate,the information into separate
streams.It thus needs to build a descriptionof the acoustic
environmentfrom separatedescriptionsof the variousstreams
and the relationshipsbetween them [2]. Factorswhich the
perceptualsystem uses to build descriptionsof streams,and
subsequentlysources, are frequency, rate of occurrenceof
events (or tempo), intensity, timbre, and attack/decay transients. In the rest of the paperthese will be discussedin some
detail. Of course it is obvious that sounds are assignedperceptually to different sources when the physical sources are at
different spatial positions. In this case, intensity, spectral,
and temporal cues are all utilized to parse the sound into
separate sources. However, we will primarilyconfine our
discussion to the illusion of many sources which occurs
due to the organizationswithin a single emanationof sound.
Frequencyand Tempo
Considerthat a repetitivecycle of tones spreadover a
certain frequency rangemay be temporally coherent, or integrated,at a particulartempo. It is possible to graduallyincrease the tempo until certain tones group together into
separatestreamson the basis of frequency,as discussedabove.
The faster the tempo, the greaterthe degreeof breakdownor
decomposition into narrowerstreams until ultimately every
given frequency might be beating along in its own stream.
This last possibility is dependent upon a number of other
factorswhich will be discussedlater.
Figure 3 illustratesthe possible stages of perceptual
decompositionfor a recyclingsix-tone patternas one gradually increases the tempo (Taped Illustration 3). Note that
streamsper se are not trackedbeyond a certainpoint, but a
texture or timbreis perceivedsince the ability to temporally
resolvethe individualtones degeneratesaltogether.Of course,
one could hold the tempo constant and graduallyexpand the
frequency relationshipsto achieve a similarstreamingeffect
[3, 6, 14, 17], but the musicalconsequenceswould be vastly
different as one can hear in TapedIllustration4. Dowling [17]
used simple melodies to illustrate this frequency-based
streamingprinciple.He interleavedtwo melodies in the same
frequencyrangethereby makingit very difficult, without prior
knowledge of the melodies, to separatethem perceptually.
But as they were pulled apartin frequency,i.e. when all the
tones of one of the melodies were transposedupward,each
melody becameapparent(Taped Illustration5).
Computer Music Journal, Box E, Menlo Park, CA 94025
Volume 3 Number 4
f
0
-
I ?
r
I
I
i
I
!
II
CL
.2-.
ccL
0)0
r-
I
•
,
,
=.
A
A
I
0
04
cm
%
Q/)
A&B
Absorbed
I
I
0.
" .
A&B
Isolated
r
IE
S
je.-"-x,0.1
Y
S
O"
c>
B
B
'O
a)
0
oo
(D
O
CD
E
------- CC
u,
L--
X
-------QI
-----,
--o,)=----------
x
Time
Time-
----------------------?
-I
I>
Time
Timbre/Texture
Figure 3. This figure illustrates the decomposition of an
acoustic sequence into smallerand smallerperceptualstreams
as the frequencyseparationbetween the tones or the tempo of
the sequence increases.In the latter case, a point is ultimately
reachedwhere one can no longer perceive individualtonal
events;a texture or timbre is heardinstead(cf. Taped Illustrations 3 and 4).
Thefrequencyrangewithinwhichthe perceptual
system
groupstones on the basisof frequencyproximityis not
constant;the groupingcan varywith the particular
pattern
of frequenciespresented.For example(see Figure4), two
tones, A and B, are arrangedwith givenfrequencyand
temporalseparationssuch that they will alwaysstream
togetherwhenno othertones arepresent.Wecancreatea
similarorganization
withtonesX andY in anotherfrequency
range.Thesetwo groupswill each forma streamas can be
heardin TapedIllustration
6a. By bringingthe pairsinto the
samefrequencyrange,new streamscan be formed,suchas
A-X and B-Y, on the basisof an alternativeproximity
organization
6b). Thus,the particular
[3] (TapedIllustration
betweenfrequencies
in a tonalpattern,andnot
relationships
just the frequencyseparationbetweenadjacenttones, plays
a vitalrolein the formationof streams.
It appearsfromourexamplesthatthereis anessentially
inverseand strictly interdependentrelationshipbetween
amongindividualtones
tempoand frequencyrelationships
the
faster
the
tones
follow
one another,the smaller
28]:
[26,
the frequencyseparation
at whichthey segregateinto separate
perceptualstreams.Conversely,the greaterthe frequency
the slowerthe tempoat whichsegregation
occurs.
separation,
Thisillustratesyet anotheraspectof musicwhich,with the
aidof the computer,cancomeunderthe composer'scontrol.
Supposewe makea graph,as in Figure5, whichrelates
of two alternating
toneson one axisto
frequencyseparation
StephenMcAdamsand AlbertBregman:HearingMusicalStreams
Figure4. This figure illustratesthe effect of frequencycontext
as opposed to frequencyseparationon streamformation. In
the first part,tones A and B form one perceptualstreamwhich
is unaffected by the simultaneousstreamcomposed of tones X
and Y. Whentones X and Y are moved into the proximity of
tones A and B, which remain unchanged, an alternate
perceptualinterpretationresultswhereby tones A and B are
assignedto separatestreamson the basisof a new frequency
context (cf. Taped Illustration6).
of the toneson the otheraxis.Note
the rateof alternation
that the horizontalaxisindicatesincreasingtone repetition
to decreasing
tempo.Onecandraw
time,whichcorresponds
boundarieson this graphindicatingthe frequency-tempo
regionsin whichthe tonescohereas a singlestreamandthose
streamsof
in whichthey segregateinto two simultaneous
differentfrequencies.Thereare two suchboundaries.Leon
VanNoorden[28] hastermedthesethefissionboundaryand
temporalcoherenceboundaryandnotedthatthey areslightly
differentfor eachperson.In Figure5 the uppercurveis the
temporalcoherenceboundary.Abovethis boundaryit is
tones into one
impossibleto integratethe two alternating
stream.Belowthis boundarylies the temporalcoherence
region,whereit is possibleto integrateeventsinto a single
perceptualstream.The lower boundaryis the fission
boundary.Belowthisit is impossibleto hearmorethanone
stream.Note that the regionsof fissionand coherencealso
overlap,creatingan ambiguousregionwhereeitherpercept
maybe heard.
As the tempo decreasestemporalcoherencecan be
maintainedat greaterfrequencyseparations.
However,the
frequencyseparationrequiredfor fission remainsfairly
constantwithdecreasing
tempo.Thetemporalcoherenceand
fission boundariesabove and below a given tone are
with respectto pitch.Thisindicatesthat the
symmetrical
phenomenaof temporalcoherenceand fissionmay occur
dependingon the absolutemusicalintervalbetweenthe tones
of the directionof pitchchange.Whilethere
but irrespective
are substantialquantitative
differencesin theseboundaries
Page29
FrequencyTrajectories
20
o
Another frequency-based effect involves frequency
trajectories.These are important on two levels. The first
involves trajectoriesbetween tones (see Figure 6). Bregman
and Dannenbring[7] have found that tones that are connected by glissandiare much less likely to segregateunder
given conditions of tempo and frequency separationthan
those which make abruptfrequency transitions.Intermediate
situations beget intermediateresults. Yet even when using a
sine wave for frequencymodulation of a sine tone, it is possible to discerna sort of streamingeffect of the higherand
3
15
10
Al
Alwaysays
Segregated,
/
O
c/
Tempo (Tones/Sec)
5
10
5
I
_I
a
Tone Repetition TiAm biguous
=
I
D
IRegion
I
I
I
Always
I
Fission BoundaryCoen
Ramped Semi- I
I
0
50
100
150
200
250
300
350
I
400
r-
Tone Repetition Time (msec)
I
a.,
Sta
a
Figure 5. Boundaries of temporal coherence (upper curve) and
fission (lower curve) define three perceptual regions in the
stream relationship between two alternating tones each lasting
40 msec. This relationship is a function of both the tempo of
alternation and the frequency separation between the tones
(after [28] ).
I
Discrete
between listeners, the qualitativetrends tend to be similar
[28]. Note that the difference between boundaries is
increasinglysubstantialfor tempi below about 10 tones/sec.
(tone repetition time = 100 msec.). This means that at very
slow tempi of, say, five tones/sec., a separationof more than
a minor 10th is necessaryto induce streaming.Below this it
becomes virtually impossible to induce streaming. In the
limiting case, if the temporal distance between tones is too
great, they do not seem to be connected at all but sound
as isolated events.
A musicallyrelevantaspect of these boundariesshould
be mentioned. The regionbetween the two boundariesmay be
consideredto be an ambiguousregionsince either a segregated
or an integrated percept may be heard. The primary
determiningfactor in this regionis attention. In other words,
it is possible to shift one's attention back and forth between
the two perceptswhen the sequence falls in this region. For
example, one may focus either on a whole streampercept or
on smallerindividualstreams. It appearsthat the closer the
sequencelies to one of the boundaries,the easierit is to focus
on the percept which is predominantbeyond that boundary.
Conversely,it is very difficult in this situation to shift one's
attention back to the other percept. Once the physicalvalues
go beyond either of these boundaries,attaining the complementarypercept may be consideredto be impossible.The role
of attention will be discussedin more detail later.
Page 30
1
a
Tas
rapeady-
SteadyState
__
_
__
_
aaTime-+
Transition
SteadyState
Figure 6. These are the 3 types of frequency transitions
between tones used by Bregman and Dannenbring (1973). In
the top section the tones are completely connected by a
frequency glissando. In the middle section an interrupted
glissando is directed towards the succeeding tone. No
frequency glide occurs in the bottom section; the first tone
ends on one frequency and the next tone begins on another
(after [71 ). The unconnected tones segregate more readily
than the others.
lower peaks of the modulation at certainmodulation frequencies (Taped Illustration7). At lower modulating frequencies
one can track the modulation, but higher modulatingfrequenciesresultin the effect of a texture or timbre. This continuum is found for modulation involving discrete changes
in frequencyas well.
The second type of trajectorymight be called a melodic
trajectory. The basic rule goes: large jumps and sudden
changes in direction of a melody produce discontinuity in
that melody. In terms of streamformation,one or two tones
in a melody that are removedfrom the melodic continuity of
the rest could be perceivedas coming from a differentsource
Computer Music Journal, Box E, Menlo Park, CA 94025
Volume 3 Number 4
and would not be integratedinto the melody as a whole,
perhapsleavinga rhythmic gap in the phrasedependingon
the natureof the main sequence. It has been reported,however, that the excluded tones are sometimes noticed in the
background,with their absencehavinglittle effect on the main
melody line [23]. Implied polyphony in a solo compound
melody line, as in the suites for solo instrumentsby Bach, is a
compositionaluse of this principle.
Schouten [30] reported that if an ascendingand descendingmajorscale fragmentplayed with sine tones is continually repeated, temporal coherence is maintainedup to a
tempo of about 20 tones/sec. However,if these same tones
are arrangedat random,the maximumtempo at which coherence still occursis reducedto about 5-10 tones/sec. It might
be inferredthat the reduction of predictabilityreduced the
pitch boundary within which the auditory system could
successfully integrate the incoming information. But van
Noorden [28] found that previousknowledge of the order
of tones had no effect on the coherenceboundary.It might
be that the small frequencyjumps in Schouten's demonstration are effective in holding the streamtogether.
Heise and Miller [23] investigatedfour melodic contours: a V-shaped contour, an invertedV, and risingand falling scale patterns;the V-shaped patternswhich change direction can be thought of as being less "predictable"than
the scale patternswhich move in only one direction. Each
patternwas eleven tones long and the frequencyof the middle
tone was variable.They found that the degreeto which the
variabletone could be separatedin frequency from the rest
of the patternbefore it segregatedinto its own streamwas a
function both of the shapeand of the steepnessof the pattern.
The steepnesswas variedby keepingthe tempo constant and
varyingthe intervalbetween successivetones. As the rate of
frequency change for the entire pattern increasesover time,
so does the amount of frequency separationof the middle
tone requiredto producesegregation.Less separationof the
middle tone is required to produce segregationwith the
V-shaped patterns than with the scale patterns, possibly
indicatingthat the perceptualsystem can follow "predictable" patterns more quickly than "unpredictable"ones.
(However,certainanomaliesin their data, which will not be
discussedhere, might suggestother interpretations.)
Van Noorden found similarresultsfor patternswith as
few as three tones [28] .He investigatedthe relativetemporal
coherence of so-called linear and angularthree-tone sequences. For linear sequences, i.e., those with two tone
intervalsin the samedirection,the temporalcoherenceboundary occurs at faster tempi than were found with angularsequences in which the melodic pattern changed direction at
the middle tone. In this latter case the first and third tones
are more contiguous in frequency than are the first and
second, or second and third tones, which facilitatesa loss of
temporal coherence similarto that found by Schouten. It
should be noted that while these melodic trajectoryeffects
do seem to play a role in musicalstreamformationbeyond
the simple frequencyseparationeffect, they are certainlyconfounded by other contextual organizations.
Both of these trajectoryexamplesillustratea principle
of perceptual organizationthat uses pattern continuity,
in some form or another,as a criterionfor "source"distinction. Figure7 illustrates,however, that frequencyproximity
may sometimescompete with trajectoryorganization;Deutsch
found that simultaneousascendingand descendingscale patStephenMcAdamsand AlbertBregman:HearingMusicalStreams
terns presentedto opposite ears segregateinto upright and
inverted V-shaped melodic contours [16]. Each contour
is heardas if being presentedto one ear. The same resultwas
found by Halpern [22] for simultaneousascendingand descendingsine tone glissandi,as can be heardin TapedIllustration 8; in this case both glissandiwere presentedto both ears.
In these two examples a stream boundaryis establishedat
the pitch where the two lines cross.
a
a
High
Contour
\
I
I
1
t
--I
L
/S
Low
a
Low
.Contour
'
_
Time-+
Figure 7. This stimulus, used by Deutsch (1975), is composed of ascending and descendingscales presentedsimultaneously to opposite ears. Two V-shaped patterns (high
and low contours outlined with dashes)are perceivedrather
than the complete ascendingand descendingscale patterns.
LoudnessandContinuityEffects
Fissioncan be obtainedby alternating
sequencesof
tonessimilarin frequencyandtimbrebut differingin intensity. Figure8 illustratesthe rangeof perceptsachievedas the
amplitudelevelof tone A is variedrelativeto thatof tone B
in the alternating
sequenceABAB... . The referencelevel
fortone B usedby vanNoordenin theseexperiments[28]
was35 db SPL;the frequencyof bothtoneswas1 KHzand
each tone lasted40 msec.If the level of tone A is below
6 db SPLat 1 KHz),
the auditorythreshold(approximately
only the B streamis heardat halftempo,asmightbe expected
(seeFigure8a). Whentone A is loudenoughto be heardand
is at least 5 db belowtone B, two separatestreamsof different loudnesscan be perceived,each at half tempo, with A
being the softer stream(see Figure8b). WhenA is within 5
db of B, a "pulsing"streamis heard and neither the A nor
the B streamcan be heardindependently,i.e., temporal
coherenceis inevitablein this range(see Figure8c). As the
levelof the A toneis increasedabovethatof the B tone,three
differentperceptsmay result,dependingon the alternation
tempoof A andB. If thistempois lessthanabout13 tones/
sec., fissionis the next perceptheard.ThistimeB is the softer
stream(see Figure8d). Thusa certaindegreeof loudness
differenceallowsus to focus on eitherstreamusingonly
thisinformation.If the tempois greaterthanabout12.5-13
tones/sec.,the perceptencounteredis the "roll"effect
discoveredby vanNoorden.It soundsas if streamA, the
louderstream,werepulsingat half tempoas in the fission
percept,but the B streamsoundsasif it werepulsingat full
Page31
A Below Threshold
a -K
B
A
(
B
time
Fission
b
fA
B
A
I
B
A
B
A
B
A
B
A
B
A
B
Coherence
c
A
Fission
d
Roll
e
Continuity
tempo (see Figure8e). Thus the A streammay be heardindependently,but not the B stream. In other words,it is as if
the A tones consistedof two parts:one that combineswith
the B stream to give a full tempo roll, and another, at half
tempo, which can be perceivedseparately.At a tempo of
about 13 tones/sec. another effect emergeswhen the level
of A is about 18-30 db above that of B. This is the continuity
effect shown in Figure 8f, so named because tone B is not
heardas pulsingbut ratheras a fairly soft, continuous tone
under the louder, pulsing A stream;this is an example of a
class of effects that will be discussed below. Finally, if the
level of the A streamis incrementedstill further,this stream
completely masksthe B stream(Figure 8g). Again, this set of
loudness-based phenomena exhibits an ambiguous region
between the coherence and fission boundarieswhere one
might pay attention to either the A or B streamsindividually.
or to the AB streamas a whole. There are thus three perceptual regions for alternatingtones at the same frequency
where the tempo is above about 12.5 tone/sec.: the roll region,
the continuity region, and the temporalcoherenceregion.
Van Noorden made quantitativemeasurementsof the
fission boundary (see Figure 9). For tempi of about 2.5
to 10 tones/sec., the fission boundary is more or less horizonal, i.e., the intensity difference (AL) necessary for a
segregatedpercept does not change with tempo over this
range,but lies about 2 to 4 db on either side of the reference
tone level. For tempi less than 2.5 tones/sec., the minimum
level difference for fission increaseswith decreasingtempo
(or longer inter-tone intervalsof silence) and is symmetric
about AL = 0 (no difference in level). For tempi greater
than 10 tones/sec. the level differenceat which fission occurs
increaseswith increasingtempo but the situation is not symTempo (Tones/sec)
20
0
5
3
2
Masking
SContinuity
"
r
Masking
08.,
g
A
B,
A
Roll
Inevitable
ev le
_Coherence
I
j3
?
Page 32
--* 35 db SPL
Fission
20
Figure 8. This figure illustratesthe range of possible percepts found by van Noorden (1975) for two alternatingsine
tones differing in each case only in their intensities.Tone B
was kept at 35 db SPL throughout; both tones were 40
msec. long and had a frequency of 1 KHz. a) The level of
tone A is below the auditory threshold. b) Tone A is at
least 5 db below tone B. c) Tone A is within 5 db of tone B.
d) Tone A is louder than tone B with an alternationtempo
less than about 13 tones/sec. e) Tone A is louderthan tone
B with more than 13 tones/sec. f) Tone A is about 18-30
db louder than tone B and the tempo is still above 13 tones
/sec. g) Tone A is more than 30 db louderthan tone B. The
arrows indicate the percepts reported; a more complete
descriptionis given in the text.
LA>
LA>LB
LB
Fission
20o
LA< LB
Sub-Threshold
0 40
200
400
60o
800
Tone RepetitionTime (msec)
Figure9. The perceptsshown in Figure8 fall into different
regionsof tempo-intensity combinations.The verticaldotted
line correspondsto the durationof each tone. The level differences between tones A and B are relativeto the level of
tone B at 35 db SPL. The horizontal axis is shown as tone
repetition time on the bottom and as its reciprocal,tempo,
on the top. It should be pointed out that Figure5 and Figure
9 cannot be directly compared,becausethe alternatingtones
in Figure5 were played at different frequencies.
Computer Music Journal, Box E, Menlo Park, CA 94025
Volume 3 Number 4
metrical about AL = 0; for tempi greaterthan this, there is
the complex interactionof level differenceand tempo which
resultsin the roll, continuity, and maskingregions.
A combinationof frequency separationand level difference was used to obtain the roll and pulsationthresholds
shown for two different tempi in Figure 10. A great deal of
psychoacousticresearchhas been devoted to the boundary
between the roll and continuity regions. The "pulsation
threshold,"as this boundaryhas been termed [24, 28, 33],
requiresa greaterloudness difference at slower tempi than
does the roll threshold,as can be seen in Figure 10.
The continuity effect may operate accordingto principles very similarto those which governocclusion in vision.
The interpretationthat one object occludes anotheris given
by informationderivedfrom the intersectingedges of the objects. In Figure 11, the fragmentshave no apparentstructure
since informationregardingthe shape of the occludingstructure is not present.However,in Figure 12, with the appropriate edge information, the fragmentscan be grouped into
meaningful chunks. The same process may be postulated
to occur in audition. That is, the "edges"of a louder sound
and a contiguous softer sound might be used to induce the
perceptionthat the less prominentsound is partiallyoccluded,
or masked,by the louder. In such a case, the illusion of continuity of the softer sound "behind" the louder sound may
result.This has been found to be true for sine tones masking
sine tones, noise maskingsine tones, and noise maskingnoise
[8, 14, 24, 28. 32, 33, 34].
Tempo - 21 Tones/sec
Tempo
a
EI
,
Figure 11. These fragments have little apparent structure and
meaning since information regarding the shape of the
occluding structure is not present.
r
1
16 Tones/sec
40
il/
I
g
i
,?
IC
+-
>
30"\-0continuity
0Qo\
1
_
ro
fission
fession
\
0-400
-200
LB = 35db SPL
Figure 12. With appropriate information about the edges of
the occluding structure, the fragments of Figure 11 can be
grouped into meaningful chunks.
0
200
-400
fB = 1 kHz
-200
0
200
fB = 1 kHz
Frequency Difference (Hz.)
Figure 10. The boundaries between the fission, roll, and
continuity regions already discussed in Figures 8 and 9 are
shown here as a function of frequency and level differences.
There is an increase in slope for roll and pulsation thresholds
and an upward shift in the latter at slower tempi. The
frequency and level differences are given relative to the
frequency and level of tone B, i.e. 1000 Hz. and 35 db SPL,
respectively. 0 KHz. frequency separation corresponds to the
vertical dashed line in Figure 9. (Note that Figure 9 cannot be
mapped directly onto Figure 10; in the former illustration,
listeners were asked to try to hear fission, but in Figure 10,
the test subjects were listening for roll. The different focus on
the part of the listener results in a shift in the boundaries
shown.)
Stephen McAdams and Albert Bregman: Hearing Musical Streams
That the nature of the temporal "edge" between the
two sounds is the importantfactor was illustratedby Bregman
and Dannenbring[8]. As Figure 13 illustrates,they varied
the transitionalamplitude of a tone both leading into the
attack and coming out of the decay of a noise masker. In
reality, the tone stopped the instant the noise beganand then
began againthe instant the noise stopped. The greatestcontinuity was found when there was no changein the amplitude
of the tone. Whenthe level of the tone was either increased
or decreasedbefore the attack of the noise, the illusion of
continuity was substantially diminished. The amount of
change in judged continuity was greater for the initially
decreasingamplitude ramp than for the initially increasing
ramp.The changein the amplitudeof the tone may suggesta
possible terminationof the tone sometime duringthe noise
burst. The length of the rampwas also found to have an effect
on perceivedcontinuity. The least continuity was found when
the amplitude rampslasted 50 msec., and for longer ramps
there was less of a break in perceivedcontinuity. This may
suggestthat a very long rampwould be interpretedas no ramp
since the rate of change of intensity would approachzero.
Page 33
Timbre
db
76
91
Tone
6
db
0Noise
db-ne
76 or 66 db Tone
91 db
Noise
-
-
76db
\
6
db
Tone
/
91 db
Noise
Time -+
Figure 13. The stimuli used by Bregmanand Dannenbring
(1977) consisted of a pure tone interrupted by noise.
The level of the tone leading into and coming out of the
noise burst was varied as shown. The greatest degree of
continuity was found in the center case where there was
no change in level.
It has also been reported that alternationspeed has almost
no effect on the perceivedcontinuity [14].
An important finding in relation to this phenomenon
was reportedby Warren,et al. [34]. They noted that there
must be some degree of spectraloverlapin the two signals
to obtain the continuity effect. Withsine tones, for example,
the effect is greatestwhen the tones are identicalin frequency,
and falls off monotonically as the frequency separationon
either side of the maskerincreases.The effect was even found
to occur througha burst of noise as long as 50 seconds [34] .
Though some people think the pulsation threshold
reflects only a neuralexcitation pattern that decays gradually, with the perceptionof continuity resultingfrom overlap of excitation at smallertime intervals[24, 33] , the following demonstrationillustrates that higher-level context-dependent predictivemechanismsmay be involved.These seem
to parallel the predictive nature of the stream formation
mechanismsmentioned previously.In Taped Illustration9,
one hearsspeech that has been gated;it is difficult to understand the spoken message.Then the gaps in the speech are
filled with white noise. This effect was first verified experimentally by Millerand Lickliderin 1950 [27]. The same
sort of phenomenonwith gated music and then gated-plusnoise-filled music is presented in Taped Illustration 10.
We have chosen a familiar composition to dramatizethe
effect. It may be very difficult to recognizeit in the gated
condition, but should become easily recognizablewhen the
gaps are filled with noise. In both of these demonstrations
the amount of informationremovedand then replaced,or
inducedby the perceptualsystem, is far too largeand far too
complex to be entirely attributedto interactionbetween localized excitation patterns.
Page 34
Another musically useful dimensionthat can play a
role in streamingeffects is timbre. Timbre tends to be the
psychoacoustician'smultidimensionalwaste-basket category
for everything that cannot be labeled pitch or loudness,
includingshort-termspectralchangessuch as onset transients,
long-term spectra,those dynamicqualitieswhich a musician
would term "texture," and so on. The important thing to
notice about streamformationis that any groupof simultaneous spectralcomponentswhich move roughlyparallelto each
other in the pitch (logarithmicfrequency) domain are most
likely to resultin a fused perceptand thus to be interpreted
as emanatingfrom one source [22, 25] . So even though the
individualspectralenvelopes of two instrumentsor sources
changeover time, they each follow patternswhich are sufficiently consistent internally to be fused and yet different
enough, in most cases, to distinguishthem one from the other.
Thereare, of course, many situationsin which especiallysensitive instrumentalistscan blend their tones and become as
one voice, but these situations usually involve very slow,
smooth changeswhere onset transientscannot give the listener
the informationneededto separatethe sources.As an example,
how often is a performanceof Ravel'sBolero preciseenough
so that the instrumentsplayingthe various"harmonic"parts
are not heardindividually,but fuse into one voice? The fact
that one typicallyhears them separatelypoints to the extreme
sensitivity of the auditory system to even minute variations
in temporaland spectralelements.
Let us examine one of the many aspectsof timbrelisted
above in more detail, using a simplistic steady-state timbre
derivedfrom summedsinusoids,as was done to test the effect
of timbreon streamsegregation[25]. If the frequenciesof a
sequence of sinusoidaltones, as shown in Figure 14, are within
close proximity (such that they would form one streamif
none of them were enriched),it is possible to cause a subset of
these tones to segregateinto its own stream by inducing a
timbredifference.In this figure,the verticalline indicatesthat
timbreis createdby, or at least co-exists with, the fusion of
components into a single percept (Taped Illustration11). We
will discussthe natureof perceptualfusion later.
A musicaluse of timbre-basedstreamingmight work as
follows: Figure 15 shows an isotimbralmelodic phrase containingthree sub-melodiesthat are slowly to emergeas contrapuntal melodies on their own. Each sub-melody is marked
with a differentsymbol under the notes that are to belong to
it. Now we might cause the "X" melody to slowly changein
timbreto a point whereits timbrediffers sufficiently from the
rest of the streamto induce separation.At this point there
would be two contrapuntallines distinguishedby their respective timbresas discussedby Wessel [35]. Alternatively,several
simultaneous timbral modifications might cause an equal
number of streams to emerge as separate contrapuntal
melodies, as is illustratedin the second part of Figure 15 by
the "X" and "O" streams.Note that as these two imbedded
melodies emerge,the originalmelody is changedsince several
of its elements have left and joined other groups. (It is
importantto note that certaintypes of inharmonicenrichment
of a tone might changethe pitch as well as the timbre,which
would be anothercase altogether).An understandingof stream
formation based on timbre modification is importantif one
intends to use multitimbralmelodies compositionally, as
Computer Music Journal, Box E, Menlo Park, CA 94025
Volume 3 Number 4
3f4
3f2
f4 "
" .f4"
Two Timbres
Streams
One
Stream
OneTimbre
Time-
Time-
Figure14. Timbredifferencesbetween groupsof tones have been shown by McAdams(1977) to influencesegregationof those
groups. A repeating4-tone sequence can be decomposed into two streamswhen two of the tones are enriched with their
respectivethird harmonics.The dotted lines indicatethe streamperceptsand the verticalsolid lines representthe fusion and
timbre of the complex tone (cf. Taped Illustration11).
Varise and Schoenberg,amongothers, have done. A problem
potentially ariseswhen largechangesin both timbreand pitch
occur simultaneously.It becomes very difficult for the listener
to follow the "melody" line if the tempo is above the listener's
fission boundaryfor that set of events. One's attention may
tend to drift from one timbre set to anotherand relegatethe
rest to unattendedstreamsratherthan following the sequence
the composerintended.
o
xx
-
I •
S
I/li
w'_
?vi
X
XA
x
I
-h
•
I/I
.IF
#t,•
#
l
Figure 15. This figure illustratesthe possibleeffect of changing
the timbre of different sub-groupsof a melody line. The notes
markedwith X's and O's in the originalstreamare shown on
separatestaves below to indicatethe rhythmictransformation
that takes place after segregationoccurs. (Taken from the
A MajorPrelude,Well-TemperedClavier,Book I, by J.S.
Bach).
Stephen McAdams and Albert Bregman: Hearing Musical Streams
TimbralTrajectories
This leads into a discussion of the effects of timbral
trajectorieson streamsegregation.As was mentioned before,
if two glissandiof identical spectralstructurecross in frequency, it is reportedlyeasierfor listenersto hearhigh and low Vshaped contours than to hear the ascendingand descending
paths actually taken by each glissando. However, if the
glissandido not cross, or if they differ in spectralstructure,
it becomes possible to segregatethem and hear the risingand
fallinglines separately.In Taped Illustration12 and in Figure
16, harmonics3, 4, and 5 of two different "missing"fundamentals are modulatedexponentiallyin opposite directions.It
is easy to hear them separatelysince neither the spectral
components nor the "missing"fundamentalscross. In the next
example (see Figure 17) two different spectralstructuresare
used in the same frequencyrange.The descendingglissandois
composed of harmonics 10, 11, and 12 and the ascending
glissando is composed of harmonics 3, 4, and 5 (Taped
Illustration 13). One might argue that it is not a timbre
differencethat allows one to distinguishthese patternsbut the
fact that the pitches do not cross, as is illustratedin Figure 17
by the dotted lines which show the "missing"fundamentals.
Halpern [22] tested for this by using the same harmonic
composition (see Figure 18) as in the last example and crossing
the pitches of the glissandi,which would correspondto the
"missing"fundamentals,rather than crossing the frequency
components. The risingand falling glissandiwere still perceived, which suggeststhat the segregatingeffect was based on
spectralcontent ratherthan on pitch (TapedIllustration14).
For comparison,listen again to the pure tone crossingin
Taped Illustration8. Some pre-tests done for both Halpern's
Page 35
4f,
3f,
2f
1
5f2
LL
0
4f
LL
3f,
0
Time
Time
Time
Time
-"
[22] and McAdams'[25] studies bore out the intuition that
it is necessary to use constant ratio relationshipsamong
components to attain fusion and constant timbre. Further
researchon the fusion of harmonicsfor the buildingof timbre
is currently in progressin Bregman'slaboratoryat McGill
University.
TimbreMovementVersusPitch Movement
One pitch phenomenonhas a scientifichistory which is
almost coextensivewith that of pitch researchin general.This
phenomenonhas been termedthe "missing"fundamentalfor
reasonsthat will become obvious. It is possible to construct
a tone from severalsinusoidalcomponentswhose frequencies
are integer multiples of some fundamentalfrequency which
may or may not be actually present in the final signal's
spectrum.For example, frequenciesof 1000, 1250, 1500, and
1750 Hz would representthe 4th, 5th, 6th and 7th harmonics,
respectively,of a fundamentalat 250 Hz. Whetheror not a
spectralcomponent is presentat the fundamentalfrequency,
the perceived pitch almost always corresponds to the
fundamental.Of interest in our discussionof timbre and
streamingis the fact that tonal events with variousspectral
structurescan elicit the same "missing"fundamentalas long as
the componentsare integermultiplesof that fundamentaland
are within a certain so-called "existence region" [29] of
harmonicsthat can actually contribute to the pitch. Thus
different tonal events can elicit the same pitch but with
different timbres.
Some of the first author'spersonalexperiencein working with the pitch of complex tones has led him to the conclusion that it is possible for changesin pitch and timbre to contribute competinginformationto the overallperceivedmovement in tone sequences.Figure 19 illustratesthat is is possible
for timbreto get "sharper"or "brighter"[see for example 1,
18, 20], i.e. the spectralcomponents move to a higher frequency range,while the pitch moves in the opposite direction.
In Figure 19, while both complexes will elicit pitches correspondingto the "missing"fundamental,i.e. Fl and F2, the
Page 36
f2
Figure 16 (above, left). Harmonics3, 4, and 5 of two different
"missing"fundamentalsare modulated exponentially in frequency in opposite directions. Neitherthe frequenciesnor the
fundamentalsof the two glissandi cross each other. Two
separate glissandi are heard (cf. Taped Illustration 12).
Figure 17 (above, right).An ascendingglissandocomposed of
harmonics3, 4, and 5 of one "missing"fundamentaland a
descendingglissandocomposed of harmonics10, 11, and 12 of
another "missing"fundamentalare crossed in frequency, but
the pitches of the "missing"fundamentalsdo not cross. Two
different glissandiare heard (cf. Taped Illustration 13).
Figure 18 (below). Here the same harmonicstructureas in
Figure 17 is used but now the pitches of the "missing"fundamentalscross each other and the partialsdo not. Two separate
streamsare still heard(cf. Taped Illustration14).
lof2
5f,
4 fi
C
3fl
0,
uL
0)
0
Ts
m
f
Tm-
Computer Music Journal, Box E, Menlo Park, CA 94025
Volume 3 Number 4
~T28f2
7f2
5f,
t TI4f,
3f,
U-
f0I
T-f2
Time
Figure19. It is possibleto construct two tones whose respective spectral "centersof gravity" (see text) and "missing"
fundamentalpitches move in opposite directionswhen these
tones are presentedin succession.
-
lOf9
"centerof gravity"of the spectralenvelope,TI, of the first
toneis lowerthanthatof the secondtone,T2. Spectral"center of gravity"maybe roughlydefinedas the weightedmean
over
of the spectralcomponentsin anacousticeventaveraged
the durationof the event,theirweightsbeingderivedfrom
theirrelativeamplitudes.
Van Noordeninvestigatedthe possibilityof streaming
basedon
basedon pitchcontiguity,as opposedto streaming
the
continued
due
to
i.e.
frequencycontiguity, continuity
presenceof spectralcomponents.Figure20 showsthe stimuli
he used.Thefirstconsistedof two differentthree-component
complextoneswithharmonics3, 4, and5 for the firsttone
andharmonics8, 9, and10 for the secondtone. Bothtones
havethe samefundamental
frequency.Herean attemptwas
to streamtogether.In the
fundamentals
madeto get "missing"
secondstimulusa puretone of frequencyf alternatedwitha
complexcontainingharmonics3 through10 of the same
fundamental
frequency;thiswasan attemptto streama pure
of the samefrequency.He
fundamental
tonewitha "missing"
foundthatit wasnot possiblewitheitherstimulusto obtain
a coherentstreampercept.However,it was necessarythat
if one
thesestimulihavevastlydifferentspectralstructures:
were to partiallyoverlapthe spectralstructuresof the
respectivetones, it would be impossibleto separatethe
basedon frequencycontiguityfromthat
perceptof streaming
basedon pitch contiguity.For example,if the firsttone
consistedof harmonics3f, 4f, and5f, andthe secondtone
consistedof harmonicsSf, 6f, and7f, howis one to knowif
lOf
-
f-••-
8ftC
c5fTu4f
-
5f
-
-
3fi-
Time -+
Time-
Figure20. Van Noorden (1975) tried alternatingtwo tone complexes of differingharmonicstructurebut identical"missing"
fundamentaland also tried alternatinga sine tone with an 8-tone complex whose "missing"fundamentalwas at the same pitch
as the puretone. He was not able to obtain streamingof the fundamentalsin the first example or of the sine tone and the
"missing"fundamentalin the second example.
andAlbertBregman:
Musical
Streams
StephenMcAdams
Hearing
Page37
the streamingeffects reportedare due to a repetition of 5f or
to the perceptionof a repeating"missing"fundamental,f. A
primaryresult of this arrangementof harmonicswas that
van Noorden used substantiallydifferent timbresin the two
tones, and this condition may have contributedsubstantially
to the preventionof streamformation.Equallylikely is the
possibilitythat streamformationmechanismsdo not use pitch
information but use frequency information. The following
findingsmay lend supportto the formernotion.
Our work on timbre effects in stream formation [25]
found an anomalousresult that may have been due to pitch
effects interactingwith timbreeffects. In Figure21, two of
the test cases used in that study are shown. In the first, the
higher tone pair of the repeating four-tone sequence is
enrichedwith the thirdharmonicand in the second case the
lower pairis similarlyenriched.The first case was found to
segregatesignificantlymore than the second. It is McAdams'
contention that this is the result of an interaction between
perceived pitch "height" and timbral "brightness."This
"brightness"seems to be relatedprimarilyto spectral"center
of gravity."In our examplewith the enrichedhigh pair, the
pitch "height"is alreadygreaterthan that of the low pairand
the addition of a highercomponent increasesthe "brightness"
of its timbre.The perceptualintegrationof these two factors
may facilitate the segregationof the tone pairsby resultingin
an overallincreasein perceived"height"of the tones in the
high pair. However,in the second condition where the low pair
was enriched, the increasein "brightness,"interactingwith
perceivedpitch "height," apparentlyserved to increasethe
overall perceived "height" of the lower pair. This would
bring it into closer perceptualproximity to the higher pair
and thus attenuate the effect of the difference in timbre
between the two pairs.But this condition still evoked more
segregationthan two isotimbralcontrol cases, where all of the
tones were composed of either one or two components. Of
interest here, again, is the fact that pitch and timbre movement may work againsteach other.
While these timbre changes have proven to be a
consistent irritant to the experimenterpursuingthe elusive
"missing"fundamental,the implied interrelationshipsbetween
timbral"brightness"and pitch "height"may proveinteresting
to investigatein terms of streamformationand to apply (both
experimentally and compositionally) to melodic sequence
construction. Grey [18] has delineated a three-dimensional
timbre space roughly defined by 1) spectral distribution,
2) spectralfluctuation and synchronicityof higherharmonic
transients, and 3) low-amplitude, high-frequency attack
transients.How does the "distance"between events in this
timbre space affect the temporal coherence and fission
boundaries?Do the effects obtained vary differently along
these different dimensionsassociated with timbre? Research
by Wessel [35] and Grey and Gordon [19, 20] has provided
informationwhich would allow the composer to deal with
timbre as a compositional parameter.This informationalso
allows a certaindegreeof predictionof the perceptualrelationships between sounds resultingfrom such timbralvariations.
Fusion,Timbre,andFrequencyStreaming
is an
Theideaof competingperceptualorganizations
and
excitingone formusiccompositionandtheory.Bregman
Pinker[10] investigated
the competitionbetweensequential
Low PairEnriched
WeakerSegregation
HighPairEnriched
StrongSegregation
C
0
u,
L
u.
t
Cr
ca
E
a)
LL
a)
L_
TmtTe
..-00:>
_____-
r
%No
00
0.0.
?
Time •
Time
-
Figure21. One resultof McAdams'study (1977) suggestedthat there is an interactionbetween pitch "height"and timbral"sharpness" (see text). In a repeating4-tone sequence,one of the pairsof tones was selectively enrichedby addingthe third harmonic.
A greaterdegreeof segregationof the high and low streamswas found for the formercase. The dashed lines indicatethe twostreamperceptsand the dotted lines indicatea potential one-streampercept.The verticalsolid lines representthe fusion and
timbreof the 2-tone complex.
Page 38
Computer Music Journal, Box E, Menlo Park, CA 94025
Volume 3 Number 4
(or frequency) organizationsand simultaneous(or timbral)
organizationsin the formation of auditory streams. The
stimulusused by Bregmanand Pinkerwas a sine tone alternatingwith a two-tone complex. In Figure22, tones A and
B would representthe sequentialorganizationand tones B and
C would represent the simultaneous organization. The
harmonicity and synchronicity of tones B and C in the
complex were variedas was the frequencyseparationbetween
the sine tone A and the uppercomponent B of the complex.
The rationalefor varyingthese two parameterswas as follows.
Tones with frequency relationshipsderived from simple
ratios, i.e. those that exhibit "consonance,"should tend to
fuse more readily than combinationsconsideredto be dissonant. While the evidence for this was very weak in the
Bregmanand Pinker study, work currently in progressin
Bregman'slaboratorystrongly suggeststhat this is indeed
the case. The new evidence further suggeststhat the effect
of harmonicityitself is relativelyweak and may be overridden by strongerfactors such as frequency contiguity and
synchronicity of attack and decay. Tones with synchronous
and identically shaped attack and decay ramps are more
likely to fuse than those with asynchronousor dissimilar
attacks and decays [12, 15]. This may be a major cue in
being able to parse out the different instrumentsplaying
together in an orchestra since they all have substantially
different attack characteristics.In addition, there is a very
low probabilityof severalpeople preciselysynchronizingtheir
attacks.
In light of this work, one might make the following
predictionsfor the perceptionof the stimuli used by Bregman
and Pinker: 1) Sequential streamingis favored by the
frequencyproximity of tones A and B (as we have illustrated
in the earlierexamples). 2) The simultaneous(or timbral)
fusion of tones B and C is favoredby the synchronyof their
attacks. 3) These two effects "compete"for tone B's membership in their respectiveperceptualorganizations.4) Finally,
when tone B is "captured"by tone A, it is removedfrom the
timbral structureand tone C sounds less rich. Thus, it is
reasonedthat if the two simultaneoussine tones B and C are
perceivedas belonging to separatestreams,they should be
heardas sine tones. But if they are heardas one stream,they
should sound like one rich tone.
It would be appropriateto introduce the notion of
"belongingness"at this point, since we talk of tones belonging
to streams,and of frequency components and the timbre
resultingfrom their interactionbelongingto a perceivedtonal
event. "Belongingness"(a term used in the perceptualliterature of Gestalt psychology) may be consideredas a principle
of sensory organizationwhich serves to reconstructphysical
"units"into perceptualevents by groupingsensoryattributes
Stimulus
C
B
LL
t
4-
Time-
Percepts
:3
A
A
Cr
B
C
\
B
,
C
A & B Stream
C Pure
B
C
B
C
A & B Segregate
C Rich
Time-
Time -
Figure22. The competition between sequentialand simultaneousorganizationsin the formationof auditorystreamsis shown
here. Tone B can belong either to the sequential organizationwith tone A or to the simultaneousorganizationwith tone C
but not to both at the same time. Bregmanand Pinker (1978) variedthe frequency separationsbetween tones A and B and
between tones B and C and also variedthe relativesynchronyof onset of tones B and C. The dotted lines in the figure indicate
the streamperceptsand the verticalsolid lines representthe fusion of tones B and C (cf. Taped Illustration15).
StephenMcAdamsand AlbertBregman:HearingMusicalStreams
Page39
of thoseeventsinto unifiedpercepts.As Bregman[2] points
is a necessaryoutcomeof any process
out, "belongingness
whichdecomposes
mixtures,sinceanysensoryeffectmustbe
source."In this case,whenthe
assignedto someparticular
tones B and C are segregated,the timbre
simultaneous
resultingfromtheirinteractionstillexistsandcanbe heardif
onelistensforit, but it is not perceptually
assignedto either
of the tonesB or C, andthusdoesnot affectthe perception
of them.Thenatureof a streamis suchthatits qualitiesare
dueto the perceptual
featuresassignedto it.
In TapedIllustration15 one canheara casein whichA
is closeto B in frequencyandC is asynchronous
with B. Then
a caseis heardwhereA is furtherawayfromB in frequency
andC is synchronous
with B. TonesB andC havethe same
in bothcases.Listenforboththe A-B streamand
frequencies
the richnessof toneC. Thelistenersin Bregman
andPinker's
C asbeingricherwhenB andCwere
studyreportedperceiving
andthisjudgedrichnessdroppedoff with an
synchronous,
increasein asynchrony,
i.e. as C eitherprecededor followed
B by 29 or 58 msec.As the frequencyseparationbetween
A andB wasincreased,Cwasreportedlyperceivedas being
rich.
increasingly
The Role of Context in DeterminingTimbre
Thesefindingsindicatethat the perceivedcomplexity
of a momentof soundis context-dependent
(see [19] as
anotherexampleof the trendtowardviewingtimbreas
dependingon context).Contextmaybe suppliedby a number
of alternative
thatcompeteformembership
of
organizations
elementsnot yet assigned.Timbreis a perceivedpropertyof a
streamorganization
ratherthanthe directresultof a particular
In otherwords,
waveform,and is thuscontext-dependent.
two frequencycomponents
whosesynchronous
andharmonic
wouldcausethemto fuseunderisolatedcondirelationships
tions may be perceivedas separatesine tones if another
organization
presentsstrongerevidencethat they belongto
separatesequentialstreams.
A verycompellingdemonstration
of the decomposition
of a timbreorganization
by alternatefrequencystreaming
is illustratedin Figure23. Whentone A is
organizations
presentedby itselfit elicitsa timbre,denotedTA.If thistone
is precededby tone B elicitingtimbreTB,andsucceededby
tone C elicitingtimbreTC,one notices that timbreTA
andis replacedby timbresTBandTC.
completelydisappears
Here the highestand lowest componentsof tone A are
streamedwith thoseof tone B andsubsequentlyassumea
timbreidenticalto thatof tone B. Also,the innercomponents
of tone A streamwith thoseof tone C and assumea like
timbre(TapedIllustration16).
An importantquestionconcerningthe assignment
of
timbreand pitchto a tonaleventarises.Bothtimbreand
pitchhavebeen foundto be context-dependent[10, 21,
respectively].But each may be determinedby different
as such they may be
ongoingcontextual-organizations;
consideredto be associatedperceptualdimensions
of a sound
butmaynot be inextricably
boundto one another.Howrelevantto musictheory,then,arestudiesthatdealwiththe perceivedpitchandtimbreof tonesin isolation?Thisquestionis
not meantto insinuatethatsensoryandpsychophysical
experimentationareuseless.Farfromit. If we thinkin termsof
the experienceof musicat differentlevelsof
investigating
processing(Figure24), we see how importantall of these
areasof researcharein buildingthe wholepicture.Theraw
physicalinputis modifiedby the limitsof the senseorgan
whoseoutputis stillfurthermodifiedby cognitiveprocesses.
Butby studyingsteady-state,or at leastrelativelysimple
signals,we can find the limitsand interactionsof the sense
organsandperipheral
processes,suchas temporalandspectral
resolution,lateralinhibition,andmasking,whichlimitsaffect
the final percept.Beyondthese, the centralperceptual
processessuch as pitch extraction,timbrebuildup,and
coherenceand fissionmodify the initialneuralresultof
stimulationof the sensorysystem.Furtherinteractionsin
higherbrainprocessessuchas attentionalprocesses,memory
andcomparison
of pitch,timbreandloudness,contextextrac-
One Timbre
:TA
Two Timbres
:1
A
versus I:
i.
B
11
12.
A
C
Figure23. The first partof this figureshows a repeatingtone (A) consistingof 4 harmonics.This tone would elicit a certain
timbre percept,TA. In the second part this tone is precededby tone B, consistingof the top and bottom harmonics,and is succeeded by tone C consistingof the two innerharmonics.Tone B elicits timbreTB and tone C elicits timbre
TC. However,due to
the streamingof tone A's components with those of tones B and C, TA totally disappearsand is replacedby
TB and TC (cf.
Taped Illustration16).
Page40
ComputerMusicJournal,Box E, MenloPark,CA 94025 Volume3 Number4
PhysicalEnvironment
Levelsof
Processing
Temporaland Spectral Resolution
LateralInhibition
Masking
"Sensory"
etc.
Pitch Extraction
Timbre Buildup
Coherenceand Fission Limits
"Perceptual"
etc.
sequenceof tonesmay segregateinto two or moreseparate
coherent.In thiscase,we would
streamswhichareindividually
perceiveseveralsimultaneousmelodiesratherthan one.
definitionof melodyis tenable,
If suchan operational
we mustthenquestionhowit is thatsomeof the extensionsof
elementsotherthanpitch for "melodic"materialarevalid
For example,whena composerusestimbreas
perceptually.
thematicmaterial,do we stillperceivethe sequenceasmaintaininga temporalcontinuity?Sometimescoherenceis mainandsometimesit canbe very
tainedby sensitiveperformers
difficultto perceive.Of course,we havenot includedother
elementswhichaffectthe perceptionof melody,suchas
butwe areonly attemptingto
harmonicstructure,
underlying
convey the notion that temporalcoherenceshouldbe
consideredessentialto melodyformation.Conversely,one
mayuse the principlesof fissionto developrulesfor creating
in sequencesof acousticevents.
polyphonyandcounterpoint
AttentionandMusicalStructure
Attention
Memory
Context Extraction
Form and Texture Integration
etc.
*
Figure24. This block diagramsuggestsa possiblearrangement
of the processingof acoustic informationat different interconnected levels of the auditorysystem.
tion, andformandtextureintegration,
sculptthe transduced
into meaningful
information
percepts.It is felt thatall of these
levelsof processingfeed into eachotherin a sort of heterarchical(asopposedto a hierarchical)
system.Thepointbeing
madeis that in the frameworkof musicwhereall of these
complexinteractionsareof greatimportance,the context
thatis createdmaybe the essentialdeterminant
of themusical
resultof a givensound.Onesoundis potentiallyperceivable
in
a greatnumberof ways,dependingon its context.
Melody
Thisleadsus to believethatthe fundamental
perceptual
elementin musicmay be the "melody"ratherthan the
isolatedtone.Orin the terminologyof auditoryperception,
the fundamental
structureis the auditorystream.Thisis not,
of course,a newnotion;but anempiricalapproachmayallow
us to clarifyand delimitthe conceptto the extentthat we
maypredictthe perceptualresults.That,in the mindof the
firstauthor,is the primaryconcernof the composer.Letus,
then,examinemelodyandits relationto attention.
For our purposes,we can think of melodyas a
connectedandorderedsuccessionof tones [28]. It follows
thenthattemporalcoherenceis necessaryfor a sequenceof
tonesto be perceivedas a whole.On the otherhand,a
Stephen McAdams
nindAlh"rt Rrtnmn-
oRa,;
.
...A--
It hasbecomeapparentto the firstauthorthatmusical
bound
structure,as it is perceivedin real-time,is inextricably
willrevealthat
to attentionalprocesses.A bit of introspection
thereareat leasttwo kindsof attentionalprocesses,which
we mightcallactiveor willfulattention,andpassiveor automaticattention.Onemightwillfullydirectone'sattentionto
some objector sequenceof events,such as listeningto
eventswithina pieceof music.Or, someunusual
particular
suchas the
eventmightattractone'sattentionunexpectedly,
honkinghornof an oncomingcaryou hadnot noticedasyou
steppedinto the streetabsorbeddeepin thought.Thathorn
demandsyour attentionandin all probabilitygetsit in a
bothkindsof attentionmayparticipate
hurry.In particular,
in the processof listeningto auditorystreams.Forinstance,
in Figure5 whena sequenceof tonesliesabovethe temporal
coherenceboundary,no amountof activeattentioncan
extractthe perceptof one coherentstream.Hereperception
is limitedby passiveattentionalprocesses[28]. Thishas
forcomposerswhointendto usefast
importantconsequences
melodicsequences,sinceit suggeststhat therearetempiat
whichthe listenermaynot be ableto followas a melodythe
of the attentional
regardless
sequenceyou haveconstructed,
willpowerinvoked.An exampleof theseeffectsmaybe found
in the sequencespresentedat differentratesin CharlesDodge's
Earth's MagneticFields.Withoutpretendingto know the
composer'sintent,it canbe amusingto listento the same
sequencedecomposeand re-integrateitself duringvarious
sequenceone can relax
tempochanges.In a multi-streamed
attentional effort with the result that attention might
randomlyalternateamongthe availablestreams.Or one might
selectively focus attention on any one of them individually
and even play them againstone another.
Van Noorden reported that the temporal coherence
boundary (the boundarybelow which all tones may belong
to one stream)is not affected by previousknowledge of the
sequence, and consideredit to be a function of a passive
attentionalmechanism,given that the listener "wantsto hear
coherence."However,what happensbetween the boundaries
of temporalcoherenceand fission dependsupon a numberof
factors, such as context, and seems to be underthe influence
of attention. Dowling [17], for example, reported that if
listeners knew beforehandwhi:h melodies were being interleaved, they could, with a bit of practice,extract the approPage 41
priate melody even at very small separationsof the ranges
traversedby each melody. It is currentlyassumedthat this
ability would degenerate at faster tempi. In addition,
van Noorden found an effect at very fast tempi where it was
virtuallyimpossibleto hear sequenceswithin a rangeof two
or three semitonesas other than temporallycoherent. At very
fast tempi (about 12.5 tones/ sec.) the tones of such narrow
patternsare not heardas separatemembersof a sequence,but
actually mergeinto a continuouslyripplingtexture, as one can
hear in Taped Illustration 17. There are thus attentional
limits in the ability of the auditorysystem to tracka sequence
of events. Whenevents occur too quickly in succession, the
system uses the variousorganizationalrulesdiscussedin this
article to reorganizethe events into smallergroups. It may
then track events within a particulargroup if the listener is
payingattention to it, but this narrowingof focus necessarily
causesa loss of information.One resultis the inability to make
fine temporalorderjudgmentsbetween streams.These organizationalmechanismsreflect the tendency of the auditorysystem to simplify things in the face of excessivecomplexity. In
the example where the fast sequenceof tones mergesinto a
continuous "ripple,"the auditorysystem is unable to successfully integrateall of the incominginformationinto a temporal
structureand simplifies the situation by interpretingit as
texture (see also [31] ). Thus the auditory system, beyond
certaintempi, may interpretthe sequenceas a single event and
assignto it the texture or timbre createdby its spectraland
temporalcharacteristics.
An understandingof (or intuition about) these organizational processescan lead to new dimensionsof control over
musicalstructure.For example, one might construct contrapuntal sequences that play across variousstream boundaries
and through different borderlineregions between temporal
coherenceand fission. (It may be that composerssuch as Bach
were alreadyusingperceptualambiguityconsciously in their
work.) Any or all of the relevantmusicalparametersmight be
used to accomplishthis. Then, an appropriateuse of events
that vie for or demandthe listener'sattention can be used by
the composer to "sculpt" the attentional processes of the
listener. Since some events seem more strikingto some persons
than to others, this attentionalsculpturein time would lead
different listeners through different paths of auditory
experience. Further,perceptionis bound to vary from time to
time within a single person, so the experience would be
different with each listening. A composition of sufficient,
controlled complexity might thus be perceptuallyinfinite for
a givenlistener.
Conclusion
An attempt has been made here to point out that
composersand music theorists should thoroughlyexamine the
relationshipbetween the "musical"principlesthey use and
espouse, and the principles of sensory, perceptual, and
cognitive organizationthat operate in the human auditory
system. Manyof the principlesdiscussedin this articleextend
to higher-levelperceptualanalysis of musical context and
structureand may well representa scientific counterpartto
some extant music theoretical principles. In other cases,
though, this groupof phenomenasuggestsperceptualorganizations which have little relation to methods currently used
to construct or analyze musical structure. To ignore the
evidence from the real life system in developinga theory of
Page42
musicor a musicalcompositionis to take the chanceof
one'sworkto the realmof whatmightbe termed
relegating
"papermusic."
Acknowledgments
This paperis basedon a workshopentitled"The
PerceptualFactoringof AcousticSequencesinto Musical
Streams"deliveredat the 1978 International
Computer
MusicConference,Evanston,Illinois,andpublishedin the
of the Conference.Theauthorswouldlike to
Proceedings
thankDr.LeonvanNoordenforhismanyhelpfulsuggestions
andvaluablecriticismsof the manuscript;
Figures5, 8, 9,
10 and 20 were redrawn,with permission,from
Dr. vanNoorden'sthesis.Thepresentversionof this article
was preparedwhileMr.McAdamswasa GraduateFellow
of the NationalScienceFoundation.Dr. Bregman's
research
has been supportedby grantsfromthe NationalResearch
Councilof Canada,the QuebecMinistryof Education,and
the McGillUniversityFacultyof GraduateStudiesand
Research.Muchof the researchusedthe facilitiesof the
Laboratoryof the McGillUniversity
Computer-Based
of
Music
Department Psychology.The editorsof Computer
Journalwouldliketo thankMr.McAdams
for preparing
the
in thisarticle.
illustrations
Appendix1
of TapedIllustrations
Description
1. A repeatingsequenceof threehightones(1600, 2000,
with threelow tones(350, 430, 550
2500 Hz.)is interspersed
andCampbell(1971). a) At a tempoof
Hz.)usedby Bregman
five tones/sec.the sequenceis perceivedas one streamof
alternating
highandlow tones(cf. Figurela). b) At a tempo
of ten tones/sec.the sequencesegregates
into one
perceptually
streamof high tones and one streamof low tones (cf.
Figurelb).
2. Anotherrepeatingsix-tone sequenceis playedat a
tempoof ten tones/sec.,but the highertripletis closerin
frequencyto the lower.ToneF maybe perceivedasbelonging
to eitherthe highstreamor the low streamdependingon the
listener'sfocus.Note that tone F cannotbelongto both
streamsat once(cf. Figure2).
3. A repeatingsix-tonesequencestartsat a slow tempo.
As the tempois gradually
increased,the sequenceis progresinto
smaller
sivelydecomposed
perceptualstreamsuntilit is
no longerpossibleto followthe tonaleventswhichmergeinto
the perceptof timbreor texture(cf Figure3).
4. Usingthe sameinitialsequenceasin TapedIllustration
betweentemporallyadjacenttones
3, the frequencyseparation
is graduallyincreased.The samesort of decomposition
into
smaller streams occurs. The limits in this example are
determinednot by temporalresolutionbut by the audible
frequencyrange(cf Figure3).
5.
The tones of two familiarmelodies are interleavedin the
same frequencyrange.As the frequencyrangeof one melody
is shifted away from that of the other, the melodies become
recognizable.
6.
a) In the first part of this example (see Figure4), tones
in one frequencyrange,forma separate
A andB, alternating
in anotherfrequency
streamfromtonesX andY, alternating
Y aremovedinto the
tones
X
the
and
In
second
part,
b)
range,
proximityof the frequenciesof tones A andB with a sub-
ComputerMusicJournal,Box E, MenloPark,CA 94025 Volume3 Number4
sequent perceptualchange.Now tones A and X stream,as do
tones B and Y. Note also that the rhythmsof the respective
streamschangebetween the two examples.
7.
This example demonstratesthe streamingpotential of
a sine wave used for frequency modulation of a sine tone.
The progressionheardis from a very slow, trackablemodulation, through a point where the high and low peaks of the
modulated sound segregatedand then to the point where a
timbre is perceived. The example then returnsto the initial
modulation frequency.
8.
Two sine waves are modulated exponentially in
frequency. One ascends in frequency, the other descends in
frequencyand they cross at one point. Most listenersreport
hearinghigh and low V-shaped contours as opposed to hearing
the full glissandi.
9.
First, gated speech is heard. Note the difficulty in
discerningthe message. However,when the gaps are filled
with white noise, the spoken message sounds continuous
and a pulsingnoise is heardin the background.
10. The same demonstrationas in Taped Illustration9 is
used on musicalmaterial.It is difficult to induce the information lost in the gaps until those gaps are filled with white
noise.
11. The first part of this example is a repeatingfour-tone
pattern,all of the tones of which have the same timbre. This
representsa pattern usually perceivedas being temporally
coherent. However,when the thirdharmonicis addedto the
second and fourth tones of the pattern, thus changingtheir
timbre, two streamsof differing timbre may be heard. The
frequencies of the fundamentals are not changed (cf.
Figure 14).
12. Two glissandiwith identical spectralstructuresmoving
in opposite directionsareheard separatelysince none of their
spectralcomponents cross in frequency(cf. Figure 16).
13. Two glissandiwith differingspectralstructuresmoving
in opposite directions are made to cross each other in
frequency. Theirdifferingtimbresstill allow the listener to
segregatethem perceptually(cf Figure 17).
14. The same timbres used in Taped Illustration 13 are
arrangedsuch that their pitches cross but none of the
frequencycomponents cross. They are still heardas separate
glissandi(cf. Figure 18).
15. In the first part of this example (see Figure22) tones A
and B are close to each other in frequencywhile tones B and
C are asynchronous(tone C leads tone B by 29 msec.). Tones
A and B form a sequentialstreamand tone C is perceivedas
being fairly pure. In the second part, tone A's frequency is
separatedfrom that of tone B and tones B and C are synchronousin attack and decay. Tone A forms a streamby itself
and tones B and C fuse into a single,richerpercept.
16. First, a repeatingfour-component tone is heard. Note
the timbreof this tone (see Figure23). Then this tone A (at
the same repetition rate) is precededby a tone B consisting
of the top and bottom components of tone A and is followed
by a tone C consistingof the two inner componentsof tone A.
Two streamsare formed and the components of A are split
into separatestreamswith tones B and C. Accompanyingthis
streamorganizationis a decompositionof tone A's timbreinto
the timbresof tones B and C.
17. In this example a sequenceof tones very close to each
other in frequency is slowly acceleratedin tempo until the
percept of one continuous, ripplingtone is achieved. Then
the tempo is decelerateduntil the individualtones can be
heard once again.
StephenMcAdamsand AlbertBregman:HearingMusicalStreams
References
1. Bismarck,G. von. (1974). "Sharpnessas an attribute
of the timbre of steady sounds."Acustica, Vol. 30,
pp. 159-172.
2. Bregman,A. S. (1978a). "Askingthe 'what for' question
in auditory perception," in Kubovy, M. and J.
Pomerantz, eds., Perceptual Organization. Hillsdale,
N. J.: LawrenceEarlbaumAssociates.
3. Bregman,A. S. (1978b). "Auditorystreaming:Competition among alternativeorganizations."Perc. and Psychophys., Vol. 23, pp. 391 -398.
4. Bregman,A. S. (1978c). "Auditorystreamingis cumulative." J. Exp. Psych./HumanPerc. Perf., Vol. 4, pp.
380-387.
5. Bregman,A. S. (1978d). "The formation of auditory
streams."in J. Requin,ed., Attention and Performance
VII. Hillsdale,N. J.: LawrenceEarlbaumAssociates.
6. Bregman,A. S. and J. Campbell(1971). "Primaryauditory streamsegregationand the perceptionof orderin
rapid sequences of tones." J. Exp. Psych, Vol. 89,
pp. 244-249.
7. Bregman,A. S. and G. L. Dannenbring(1973). "Theeffect
of continuity on auditory streamsegregation."Perc.
andPsychophys., Vol. 13, pp. 308-312.
8. Bregman,A. S. and G. L. Dannenbring(1977). "Auditory
continuity and amplitudeedges." Can.J. Psych./Rev.
Can.Psych.,Vol. 31, pp. 151-159.
9. Bregman,A. S. and G. L. Dannenbring(1978). "Auditory
streaming."Sci. Am., in press.
10. Bregman,A. S. and S. Pinker(1978). "Auditorystreaming
and the building of timbre." Can.J. Psych., Vol. 32,
pp. 19-31.
11. Bregman,A. S. and S. I. Rudnicky (1975). "Auditory
segregation:streamor streams?"J. Exp. Psych./Human
Perc. Perf., Vol. 1, pp. 263-267.
12. Bregman,A. S. and H. Steiger(1978). "Auditorystreaming and verticallocalization:interdependenceof 'what'
and 'where'decisionsin audition." Unpub. manuscript,
McGillUniv. Dept. of Psych., Montreal,Quebec,Canada.
13. Dannenbring,G. L. and A. S. Bregman(1976). "Effect
of silence between tones on auditory stream segregation." J. Acous. Soc. Am., Vol. 59, pp. 987-989.
14. Dannenbring,G.L. and A.S. Bregman(1976)). "Stream
segregationand the illusion of overlap."J. Exp. Psych./
HumanPerc.Perf., Vol. 2, pp. 544-555.
15. Dannenbring,G. L. and A. S. Bregman(1978). "Streaming vs. fusion of sinusoidalcomponents of complex
tones." Perc. and Psychophys.,Vol. 24, pp. 369-376.
16. Deutsch, D. (1975) "Two-channellistening to musical
scales."J. Acous. Soc. Arrt, Vol. 57, pp. 1156-1160.
17. Dowling, W. J. (1973). "The perception of interleaved
melodies." Cog. Psych., Vol. 5, pp. 322-337.
18. Grey, J. M. (1977) "Multidimensionalperceptualscaling
of musicaltimbres."J. Acous. Soc. Am., Vol. 61, pp.
1270-1277.
19. Grey, J. M. (1978), "Timbrediscriminationin musical
pattern."J. Acous. Soc. Am, Vol. 64, pp. 467-472.
20. Grey, J. M. and J. W.Gordon(1978). "Perceptualeffects
of spectralmodificationson musicaltimbres."J. Acous.
Soc. Am., Vol. 63, pp. 1493-1500.
21. Guilford,J. M. and H. M. Nelson (1936). "Changesin the
pitch of tones when melodies are repeated."J. Exp.
Psych., Vol. 19, pp. 193-202.
(continued on p. 60)
Page43
(Buxton article continued from p. 25)
Mathews, M. (1969) The Technology of Computer Music,
Cambridge:MITPress.
Murray,D., Beauchamp,J., and Loitz, G. (1978) "Usingthe
PLATO/TI980A Music Synthesis System: The
PLACOMP Language," Proceedings of the 1978 International Computer Music Conference, Roads (Comp.),
Evanston: Northwestern University Press, Vol. 1:
151-166.
New EnglandDigitalCorporation(1978). SynclavierInstruction Manual,Norwich, Vermont: New EnglandDigital
Corp.,PO Box 305.
Newman, W. and Sproull, R. (1979) Principles of Interactive
Computer Graphics (Second Edition), New York:
McGraw-Hill.
Patel, S. (1979) "Score Editor Design: The Foundations of
Scriva," Toronto: unpublished manuscript, SSSP/
CSRG,Universityof Toronto.
Pulfer, J. (1972) MUSICOMP Manual, Ottawa: N.R.C. Radio
and ElectricalEngineeringDivision.
Reeves, W., Buxton, W., Pike, R., and Baecker,R. (1978)
"Ludwig:an Exampleof InteractiveComputerGraphics
(McAdamsand Bregmanarticle continuedfrom p. 43)
22. Halpern,L. D. (1977). "The effect of harmonic ratio
relationshipson auditory stream segregation."Unpub.
undergrad.researchreport, McGill University Dept. of
Psych., Montreal,Quebec, Canada.
23. Heise, G. A. and G. A. Miller(1951). "An experimental
study of auditory patterns."Am. J. Psych., Vol. 64,
pp. 68-77.
24. Houtgast,T. (1972). "Psychophysicalevidence for lateral
inhibition in hearing."J. Acous. Soc. Am., Vol. 51,
pp. 1885-1894.
25. McAdamsS. E. (1977). "The effect of tone quality on
auditory stream segregation."Unpub. honours thesis,
McGillUniversityDept. of Psych., Montreal,Quebec,
Canada.
26. Miller,G.A. and G.A. Heise (1950). "The thrill threshold." J. Acous. Soc. Am., Vol. 22, pp. 637-638.
27. Miller,G. A. and J. C. R. Licklider(1950). "Intelligibility
of interruptedspeech." J. Acous. Soc. Am., Vol. 22,
pp. 167-173.
28. van Noorden, L. P. A. S. (1975). Temporalcoherencein
the perception of tone sequences. Ph.D. Diss., Tech.
Hogeschool,Eindhoven,The Netherlands,Pub. by The
Institute of Perception Research, Eindhoven, The
Netherlands.
29. Ritsma, R. J. (1962). "Existence region of the tonal
residue.PartI."J. Acous. Soc. Am., Vol. 34, pp. 12241229.
30. Schouten, J. F. (1962). "On the perceptionof sound and
speech: Subjectivetime analysis." Fourth Int'l. Cong.
Acous., Copenhagen.Reports II, pp. 201-203.
31. Steck, L. and P. Machotka (1975). "Preferencefor
musicalcomplexity: Effects of context." J. Exp. Psych./
HumanPerc.Perf., Vol. 104, pp. 170-174.
32. Thurlow, W. R. (1957). "An auditory figure-ground
effect." Am. J. Psych., Vol. 70, pp. 653-654.
33. Verschuure, J. (1978). Auditory excitation patterns:
(continued on p. 63)
Page 60
in a Score Editor,"Proceedingsof the 1978 International ComputerMusic Conference, Roads (Comp.),
Evanston: Northwestern University Press, Vol. 2:
392-409.
Ritchie, D. M. and ThompsonK. (1974) "The UNIX TimeSharingSystem," Communicationsof the ACM, 17.7:
365-375.
Smith, L. (1978) MUSIO UserManual, Stanford: work in
progress,CCRMA,Dept. of Music,StanfordUniversity.
Smith, L. (1972) "SCORE-A Musician's Approach to
ComputerMusic,"Journal of the Audio Engineering
Society, 20.1: 7-14.
An ExperimentalAid for the
Tanner,P. (1972) "MUSICOMP,
Composition and Production of Music," ERB-869,
Ottawa: N.R.C. Radio and Electrical Engineering
Division.
Tucker, W. H., Bates, R. H. T., Frykberg,S. D., Howrath,
R. J., Kennedy,W. K., Lamb,M. R. and Vaughan,R. G.
(1977) "An InteractiveAid for Musicians,"International
Journalof Man-MachineStudies, 9: 635-651.
Vercoe, B. (1975) "Man-ComputerInteractionin Creative
Applications," Cambridge:unpublished manuscript,
Studio for ExperimentalMusic,M.I.T.
U.S. POSTALSERVICE
STATEMENT OF OWNERSHIP, MANAGEMENT AND CIRCULATION
(Required by 39 U.S.C 3685)
1.TITLE OF PUBLICATION
COMPUTER MUSIC JOURNAL
i
NO
A. PUBLICATION
9
51 6 .
3
2. DATE OF
0
28 Sep 19779
.
F.LI.
four
quarterly
_
4.LOCATION
OFKNOWN
OFFICe
OFPUBLICATION
State
andZIPCode)
(Notprinters)
(Street.
City,
County,
1263 El Camino Real,
Menlo Park,
CA 94025
.
.20.00
$.
(San Mateo County)
ORGeNeRAL
5.LOCATION
OFTHE
BUSINESS
OFTHE
HeAD0UARTeRS
PUBLISHERS
(.%of
OFFICES
printer.;
same
6.
NJAMESAND COMPLETEADDRESSES OF PUBLISHER,EDITOR AND
MAOAGING EDITOR
PUBLISHER (Name and Address)
Willard J. Holden, 1263 El Camino Real,
(N'ame
andAddress,
EDITOR
Curtis
Roads
CA 94025
Menlo Park,
(same)
andAddress)
EDITOR
MANAGING
(NOme
Curtis
Roads (same)
7. OWNER (if owned hy a corporation its name and address must be stated and also immediatel? thereunder the namesand addreses of tock
holders owning or holding i percent or more of total amount of stokh If not owned bN a corporation, the names and addressesof the individual
owners must be given If owned by a partnership or other unincorporated firm. its name and address, as well as that of each individual must
be given.)
NAME
s
People
.
ADDRESS
o.
Lomputer
-I263
Lamino
Ei
Reai,
Tenino
yark,
LAS94U025
KNOWNBONDHOLDERS,MORTGAGEES AND OTHERSECURITY HOLDERSOWNINGOR HOLDING PERCENTOR MOREOF
TOTAL AMOUNTOF BONDS, MORTGAGESOR OTHER SECURITIES(If there are none. so state)
NAME
S
ADDRESS
none
9. FOR COMPLETIONBY NONPROFITORGANIZATIONSAUTHORIZEDTO MAIL AT SPECIAL RATES (Section 132.122, PSM)
The purpos, function, and nonprofit natus of this organization and the exempt status for Federal income tax purpoeis (Check one)
must
of
EDDURING
(Ithanged,
AVE
0TCHAG2
[]A
CHANGEDhDURINt
submit
change
explanation
PRECEDING
MONTHS
12MONTHSwith
PRECEDING
thispublisher
NO.
COPIESOF SINGLEI
tAVERAGE
statement.)
EXTENT AND NATURE OF CIRCULATION
ISSUE DURINGCOPIES
PRECEDING
NEAREST
TO
ISSUE PUBLISHED
EACH ACTUAL
NO.
12 MONTHS
FILING DATE
,2
___
______________
_
10.
4,0'
840,
3,300
B. PAID CIRCULATION
NDOR CANDCOIERUNT
2.
853
625
1,984
1,800
2,837
2,425
RE
MAILSUBSCRIPTOONS
. TOTAL
PADCIRCULATION
(SuM O10oand102)
BYMAIL.
CARRIER
OROTHER
O.FREE
DISTRIBUTION
MEANS
SA
E-. COMPLIMEN._TARY,*
ANDOT4EFREE
neo
.n
n.TOTAL
Da
OI (SumtoC aond
1.
OFCEt
2. CRETI
N
SPOILED
LEFTOVER.
UNACCOUNTED,
USE.
M EI AGENTS
CFRO
3,634
i
2,525
1,051
775
0
155
G. TOTAL
(Sum
n
ofE. FI and2--oouldequalnetpressrunhow n,300
in A)ue
II.I certify that the statementsmadeby met
above arecorrectand complete.
RT
100
797
COPlE
,840
84o
3,300
N
P
12, FOR COMPLETIONBY PUBLISHERSMAILINGAT THE REGULAR RATES (Section l
;.Postal Service Manual)
39 U. S C. 3626 provides in pertinent part: "No person who wouldhive been enttledto mail matter under formersitlon 4359 of this title
%hall
such
matter
under
mall
atthe
rates
this
unless
hefies
provided
with
the
Service
awritten
for
subsectlon
annually
Postal
permlsion
requesR
to mail matter at such rates,"
In accordance with the provisions of this statute I hereby requst prm ilsion to mall the publlcatlon named in Item 1 at the phased posts"
rates prently authorlzed by 39 U. S. C. 3626.
/
M-94?yl V
1sY =Ir
Computer Music Journal, Box E, Menlo Park, CA 94025
Volume 3 Number 4
(McAdamsand Bregman continuedfrom p. 60)
The significanceof the pulsation thresholdmethod for
the measurementof auditorynonlinearity.Ph.D. Diss.,
Erasmus University, Rotterdam, The Netherlands.
34. Warren,R. M., C. J. Obusek,and J. M. Ackroff (1972).
"Auditory induction: Perceptual synthesis of absent
sounds."Science, Vol. 176, pp. 1149-1151.
35. Wessel,David L. (1979). "TimbreSpace as a Musical
Control Structure."ComputerMusic Journal, Vol. 3,
No. 2, pp. 45-52.
MUSIC-11
Sound Synthesis System
Runs Under Unix, RT-11 Or RSX-11M
On Any PDP-11 Or LSI-11 With EIS
We are licensed to distribute
this software system developedby the
M.I.T. ExperimentalMusic Studio
ComjmirMusic
Features:
is
available
in
microform
Please send me additional information.
University Microfilms International
300 North Zeeb Road
Dept. P.R.
Ann Arbor,MI48106
U.S.A.
18 Bedford Row
Dept. P.R.
London, WC1R4EJ
England
Name
Institution
Street
City
State
-Very High ExecutionSpeed
-Powerful And Intuitive OrchestraLanguage
-DynamicMemoryAllocation
-VarietyOf Synthesis Methods,IncludingLinear
Predictionand FastAdditiveSynthesis
For Detailed Information And Rental Rates
Call Or Write
Digital Music
Systems, Inc.
306 Court Street
Brooklyn, New York 11231
(212) 643-1077
Zip
DMX-1000
SIGNAL PROCESSING COMPUTER
For Real-Time All-Digital Sound Synthesis
16-bit minicomputer specially
designed for audio signal
processing
Produces audio output with a
self-contained D/A converter
be controlled by any
.................May
general-purpose computer
Completely programmable
i0,l:::iii
Very
fast
Interfaces available for many
popular computers
For More Information
Call or Write
Inc.
Music
Digital
Systems,
306 Court Street, Brooklyn, New York 11231
(212) 643-1077