Yale University Department of Music

Perceived Triad Distance: Evidence Supporting the Psychological Reality of Neo-Riemannian

Transformations
Author(s): Carol L. Krumhansl
Source: Journal of Music Theory, Vol. 42, No. 2, Neo-Riemannian Theory (Autumn, 1998), pp.
265-281
Published by: Duke University Press on behalf of the Yale University Department of Music
Stable URL: http://www.jstor.org/stable/843878 .
Accessed: 03/04/2013 14:34
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 support@jstor.org.

Duke University Press and Yale University Department of Music are collaborating with JSTOR to digitize,
preserve and extend access to Journal of Music Theory.

http://www.jstor.org

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions

PERCEIVED TRIAD DISTANCE:

EVIDENCE SUPPORTING
THE PSYCHOLOGICAL
REALITY OF NEO-RIEMANNIAN
TRANSFORMATIONS
CarolL. Krumhansl

This articleexamines two sets of empiricaldatafor the psychological

reality of neo-Riemanniantransformations.Previous research(summarized, for example, in Krumhansl1990) has establishedthe influence of
parallel, P, relative, R, and dominant,D, transformationson cognitive
representationsof musical pitch. The present article considers whether
empirical data also support the psychological reality of the Leittonweschsel, L, transformation.Lewin (1982, 1987) beganworkingwith the
D P R L family to which were added a few other diatonic operations.
Subsequently,Hyer (1989) reduced the transformationsto the family
consisting of only the D P R L transformations.Cohn (1996, 1997) has
provided an extensive theoreticalanalysis of neo-Riemanniantransformations excluding the dominant, D, transformation.Thus, one of the
issues that arises out of this literatureis whetherthe D transformationis
needed given that, as will be shown next, it is equivalentto a combination of R and L transformations.
265

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions

The parallel, P, relative, R, and Leittonwechsel, L, transformations

move betweenmajorandminortriadswith a minimalchangeof one tone.
The P transformationshiftsthe thirdof a majoror minortriadby one chromatic step holding constantthe tones relatedby a perfect fifth (e.g., C E
G to C Eb G). The R transformationshifts the fifth of a majortriad (or
root of a minor triad)by two chromaticsteps holding constantthe tones
relatedby a majorthird(e.g., C E G to C E A). The L transformationshifts
the root of the majortriad(or fifthof a minortriad)by one chromaticstep
holdingconstantthe tones relatedby a minorthird(e.g., C E G to B E G).
The dominant,D, transformation,shifts a majoror minortriadup or down
a fifth (e.g., C E G to G B D or FA C, and C EbG to G Bb D or FAb C);
in this sense it is ambiguous.The D transformationcan also be writtenas
a combinationof R and L transformations(e.g. applyingthe L transformationto C E G producesB E G, thatis, E G B; then applyingthe R transformationto E G B producesD G B, thatis, G B D). Thus, the D transformationmight be redundant.
Psychologicalconsiderations,namelynumerouspsychologicalresults
showing the effect of pitch proximity (see summariesin Bigand, Parncutt, and Lerdahl1996; Krumhansl1990), also suggest thatthe P, R, and
L transformationsmight have priorityover the D transformationbecause
the latterrequiresmore than a minimal change of one tone. Given this,
the firstthreeof these transformations,P, R, andL, might providea good
model of triaddistance as measuredin empiricalstudies. The empirical
measuresto be examinedcome fromanexperimentderivingindirectmeasures of triaddistance from probe-toneratings (Krumhansland Kessler
1982). That study also produceda geometricrepresentation,in the form
of a torus, that will be used to representvarious structuresrelatedto the
neo-Riemanniantransformations.The second set of empiricalmeasures
come from an experimentthatcollected judgmentsof chordtension in a
fixed tonal context (Bigand, Parncutt,and Lerdahl1996).
Derivation of the torus from probe-tone ratings
The toroidalconfigurationwas derivedfromprobe-toneratings(Krumhansl and Kessler 1982) throughthe following steps:1
1. Ten subjects (average years musical instruction 10.9) were presented with the following contexts in both majorand minorkeys: I
triad,IV V I cadence,VI V I cadence, andII V I cadence.Each context was followed on successive trials with each tone of the chromatic scale (in randomorder);the final tone is called a probe tone
(Krumhansland Shepard 1979). (For example, the sequence consisting of F-, G-, and C-majortriadswas followed by some order-

266

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions

ing of the tones C, CO,D,.....B.) The subjectsratedon a scale from

1 to 7 how well each probe tone "fitinto or went with"the musical
contextjust heard.The ratingswere shifted to a common tonic for
the differentmajor keys and the differentminor keys used in the
study.Because the fourcontext types producedsimilarresults,they
were averaged.This yielded a probe-toneprofilefor majorcontexts
and for minorcontexts (the numericalvalues appearin Krumhansl
1990, 37).
2. The probe-toneprofiles were then shifted and correlatedwith one
another.This yields an indirect measure of the distances between
keys, which will be takenhere to be approximatelyequal to the distances between their tonic triads.2For example, to obtain the distance from C major to F minor, the major profile was shifted to
tonic of C, the minorprofile was shifted to tonic of F, and the two
sets of twelve ratings (for the chromaticscale tones) were correlated. A correlationhas a value between -1 and 1, where 1 is the
maximumdegree of similaritybetween the probe-toneprofiles. In
the presentcontext, the correlationsconsideredare those between
C majorand all other majorand minor triads,denoted KKcor(the
numericalvalues appearin Krumhansl1990, 38).
3. The correlationsbetween probe-toneprofiles were enteredinto an
analysis programcalled multidimensionalscaling (MDS, Shepard
1962, availablein statisticalpackages such as SYSTATand SPSS).
This analysisproducesa configurationof points such thatdistances
are inversely related to the KKcor values. In other words, probetone profiles that are similar (highly correlated)are representedby
points that are close together in the space. The analysis found an
excellent fit to the KKcorvalues with a 4-dimensionalconfiguration
in which the points fall on the surface of a torus (the coordinates
appearin Krumhansl1990, 42). The interpointdistances,measured
by a Euclidean4-dimensionalmetric, will be denotedKKdist.3
4. Because a torusis the cross-productof a circle with anothercircle,
S' X S1,it can be depictedin two dimensionsas shown in Example
1. The horizontalaxis representsthe angulardistancearoundone of
the circles and the vertical axis represents the angular distance
aroundthe othercircle. In this representation,it is understoodthat
the right edge is the same as the left edge, and the top edge is the
same as the bottomedge.
The configurationwas originallyinterpretedin termsof the threemore
familiar transformationsshown in Example la: dominant, D, parallel
major/minor,P, and relativemajor/minor,R.4The questionto be considered next is whetherthe KKcorand KKdistvalues are bettermodeled if

267

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions

!D/I

P
9 All
'

DI/

D/
'.

Model 1:DPR

DI/C

/
D/

D/

(la)

bi D

D/

D/

et/id#
D/

/d#

...

#...LC# ,L.?
(lb)

L...

Model 2: PRL

E..

L..

eb

(d#

,C
e
,#
\ t

\R /

(Ic)
Model 3: DPRL

P
-

?
D,,.
f. ...
I\

-1 'd
.IJ

Oc/

L..." .

LL
L. J' /
//

.. L. .

L.../b,4
/

eb'd#

IOW
/tl

'L?R
D.C?_

/
.

Example 1
The toroidalrepresentation(Krumhansland Kessler 1982) is depicted
in two dimensions,where it is understoodthatthe left andright edges
are the same and the top and bottomedges are the same.
Example 1a shows the toroidalrepresentationof Model 1: DPR superimposed. Example lb superimposesModel 2: PRL. Example Ic superimposes Model 3: DPRL

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions

the Leittonwechseltransformation,L, is addedto the model or is substitutedfor the dominanttransformation,D (becauseD can be expressedin
termsof L and R).
Modeling triad distances with D, P, R, and L transformations
All models to be considereduse the shortest-pathdistances.In other
words, the distancebetween one triadand anotherwill be takento be the
smallestnumberof transformationsneeded to move from one triadto the
other.Threemodels will be compared:
Model 1. The model with D, P, andR transformations,thatis, the original model, shown in Example 1a.
Model 2. The model with P, R, and L transformations,shown in
Example lb.
Model 3. The model with D, P, R, and L transformations,shown in
Example Ic.
Tables 1 and 2 show the numberof each type of transformationfor each
model. The values shown are the transformationsfrom C major to all
major triads (Table 1) and from C major to all minor triads (Table 2).
Under the numberof transformationsfor each triad is an example of a
shortestpath.5 Therecan be alternativeshortest-pathswith the samenumber of transformations.In most cases these have the same numberof each
kind of transformation(for example, to go from the C-majortriadto the
F-minortriad,therearetwo possibilities:RLPor PLR,bothof whichhave
one R, one L, andone P). In a few cases, two possible shortest-pathsexist
with differentnumbersof each of the transformations(for example,moving from the C-majortriadto the Db-majortriadcan be accomplishedby
PLRL, with one P, one R, and two Ls, or by RPLP,with two Ps, one R
and one L). The values in parenthesesindicate alternativeshortest-path
routes;the dataanalysis showedthese producedslightly less clearresults
and will not be consideredfurther.
All threemodels were testedusing multipleregressionwhich findsthe
best-fittinglinearfunctionpredictingthe dependentvariablefrom a number of independentvariables.For example, in testing how well KKcor
can be predicted by Model 1, KKcor is the dependentvariable that is
modeled as a weighted sum of the three variablesD, P, and R (shown in
the firstthreecolumns of Tables 1 and 2). The analysisreturnsa multiple
correlationvalue, R, indicatinghow well the dependentvariableis modeled, with 1 indicatingthat a perfect fit is obtained.R values are evaluated in terms of statistical significance, denoted p, an estimate of the
probabilityof the result for randomdata. By convention,if p < .05, the
resultis consideredstatisticallysignificant.
Table3 summarizesthe resultsof the multipleregressionanalysesfor
269

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions

Triad
Db
D
Eb
E
F
F#
G
Ab
A
Bb
B

Model 1: DPR
D
P
R
2
1
1
DDPR
2
0
0
DD
1
0
1
PR
1
1
1
DRP
1
0
0
D
0
2
2
PRPR
1
0
0
D
1
1
1
DPR
1
0
1
RP
2
0
0
DD
2
1
1

Model 2: PR L
P
R
L
1(2) 1(1) 2(1)
PLRL (or RPLP)
0(1) 2(2) 2(1)
LRLR (or RPRL)
1
1
0
PR
1
0
1
LP
0
1
1
RL
2
2
0
PRPR
1
0
1
LR
1
PL
1
1
0
RP
0(1) 2(2) 2(1)
RLRL (or LRPR)
1(2) 1(1) 2(1)

Model 3: DPRL
P
D
R
L
1
1
0
1
PLD
2
0
0
0
DD
1
0
1
0
PR
1
0
1
0
LP
1
0
0
0
D
0(2) 2(1) 2(0) 0(1)
PRPR (or DDLP)
1
0
0
0
D
1
1
0
0
PL
1
0
1
0
RP
2
0
0
0
DD
1
1
1
0

DDPR

LRLP (or PLPR)

DLP

Table 1
Triaddistancesfrom the C majortriadto all othermajortriads
measuredby shortestpathsfor Model 1: D P R, Model 2: P R L, and
Model 3: D P R L. The numbersindicatethe numberof each kind of
transformation;numbersin parenthesesindicate alternativeshortest
paths. Below these are examples of shortestpaths.
the KKcorand KKdistvalues in the first two columns. All three models
were statisticallysignificant,so the focus will be on the magnitudesof
the correlations.Model 1 produceda somewhatbetter fit to the KKcor
values than Model 2, but the two models were equivalenton the KKdist
values. Thus, substitutingthe L transformationfor the D transformation
did not substantiallychange how well the model fits the data. However,
higher correlationswere found for Model 3 than either Model 1 or 2.6
These results suggest that all four transformationsD, P, R, and L contributeto the patternin the KKcor and KKdistvalues. The advantageof
Model 3 over Model 1 shows the L transformationhas psychological
reality independentof the other transformations,D, P, and R. Similarly,
270

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions

Triad
c
c
d
d
e
f

ft
g
g#
a
bb
b

Model 1: DPR
P
R
D
1
0
0
P
2
1
1
DRPR
1
1
0
DR
1
2
00
PRP
1
1
0
DR
1
0
1
DP
2
1
0
RPR
1
1
0
DP
2
1
2
DDRPR
0
0
1
R
2
1
0
DDP
2
1
0
DDR

Model 3: DPRL
Model 2: PRL
P
R
D
R
L
P
0
0
1
0
0
1
P
P
0
1
1
1
1
1
RPL
RPL
0
1
1
0
2
1
DR
RLR
1
0
2
1
2
0
PRP
PRP
0
1
0
0
0
0
L
L
1
1
1
0
1
1
DP
LPR
2
1
2
0
0
1
RPR
RPR
1
1 1
1
0
1
DP
LRP
0(0) 1(2) 0(0)
1(2) 0(0) 2(1)
LPL (or PLP)
LPL (or PLP)
1
1
0
0
0
0
R
R
2
2
0
1(2) 2(2) 2(1)
RLRLP(or RPRPL) DDP
1
2
1
0
0
0
DL
LRL

L
0

0
0
1
0
0
0
2(1)
0
0
1

Table2
Triaddistancesfrom the C majortriadto all minortriadsmeasuredby
shortestpaths for Model 1: D P R, Model 2: P R L, and
Model 3: D P R L. The numbersindicatethe numberof each kind of
transformation;numbersin parenthesesindicate alternativeshortest
paths. Below these are examples of shortestpaths.
the advantageof Model 3 over Model 2 indicatesthat,even thoughcombinations of L and R can substitutefor D, the D transformationhas an
independentpsychological reality.
The resultsof this analysisof the two sets of values,KKcorandKKdist,
showedthatthe model incorporatingall fourtransformations,D, P,R, and
L, fit the empiricalvalues better than the models containing only three
transformations,D, P, andR, or P,R, andL. These resultshavetwo important implications.First, they establishedthe independentpsychological
statusof the L transformation.Addingthe L transformationimprovedthe
fit of the empiricalvalues over the D, P, and R model developed previ271

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions

KKcor

KKdist

BPLmus

BPLnonmus

Model 1: D P R
R = .91,p < .0001 R = .74,p = .0009
Model2: PR L
R= .91,p < .0001 R= .74, p= .0011
Model3: DPRL
R = .96, p < .0001 R = .83,p < .0001
TonalPitchSpace

R= .78,p= .0002

R =.90, p < .0001

R = .89, p < .0001

R =.82, p < .0001

R = .73, p < .0001

R =.85, p < .0001

PitchCommonality
R = .88, p < .0001 R = .81,p < .0001
HorizontalMotion

R = .83,p < .0001

R =.35, n.s.

R = .41, p < .05

R = .78, p < .0001

R = .94,p < .0001

R = .90,p < .0001
R = .97,p < .0001

R = .47, p = .02

R = .77,p = .0003

R = .87,p < .0001

Table3
Statisticalresults of the models tested.The four sets of empiricalvalues
from the studies by Krumhansland Kessler (1982) and Bigand, Parncutt, and Lerdahl(1996) were fit by Model 1: D P R, Model 2: P R L,
and Model 3: D P R L as shown in the firstthreerows of statistics.
Three additionalmodels were also tested:the tonal pitch space model
(Lerdahl1988), pitch commonality(Parncutt,1989), and the actual
pitch distancesin the chordtension study (Bigand et al. 1996).
ously (Krumhansland Kessler 1982). Psychologically, this can be explainedby the importanceof pitchproximityandthe fact thatthe L transformationinvolves shiftinga single tone by just one chromaticstep. Second, these resultsestablishedthe independentpsychological statusof the
D transformation,indicatingthat combinationsof R and L transformations cannot substitutefor the D transformation.This is consistent with
empirical results showing that the dominant relation is an important
underlying principle in psychological representationsof pitch. Before
examiningwhetherthese resultsextendto two othersets of empiricalvalues, some structureson the toroidalrepresentationwill be described.
Structures on the toroidal representation
Examples2a, b, and c show the subgroupsgeneratedby P R, P L, and
R L, respectively,on the toroidalrepresentation(Krumhansland Kessler
1982). Cohn (1997) calls these 'binary-generated'subgroups.The transformationpair P R generatesthree subgroupsof order8 that cut diagonally across the toroidal representationfrom upper left to lower right;
tonics are separatedby minor thirds.The shortest-pathdistance used in
272

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions

Db

(2a)
PR Subgroups

Order8

Db/C#

eb

d#

u..L.../C#

(2b)
PL Subgroups
Order6

Lt..

e d#

L.

d#
d#

S..F L

Db/C# D L..
. A
A\LR
.E
L

(2c)

RL Group
Order24

R
L8
ebl:

L.
\

LF..

RR

L-

L..Ab

G
9\

\R
L .D

Dt/C#

Eb

el/d#

,R
b

Example 2
Neo-Riemannianstructuresshown on the toroidalrepresentation
(Krumhansland Kessler 1982).
Example2a shows the threePR subgroupsof order8. Example2b
shows the four PL subgroupsof order6. Example2c shows the RL
group of order24.
273

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions

f#,

Db/

C#6
A
#bAl

EDb/#

"

c+

A.Kf

T"

"Ebb

CAb

"" 0..0

"-..g#.,"-e
B
eb/d 0

bb...... .

G
b

"b
C,

IV,

Eb

"gD
\A

F*?

FLIP-LA
. -.

13b

,.X.

Example 3
The P, R, and L transformationsare superimposedon the toroidal
representation(Krumhansland Kessler 1982). The tones indicatedin
outline letters at each vertex are the tones thatexchangebetween that
vertex and neighboringvertices. Each hexagon has a common
exchangingtone, indicatedin the centerof the hexagon surrounded
by the small hexagon.
Model 1:D P R can be conceptualizedas distancealong the P R subgroup
with the D transformation(the othertransformationin the model) used to
move from the referenceC majorto the appropriateP R subgroup.The
transformationpairP L generatesfoursubgroupsof order6 thatcut across
the toroidal representationhorizontally.Tonics are separatedby major
thirds.The shortest-pathdistanceused in Model 2: P R L can be conceptualizedas distancealong the P L subgroupwith the R transformation(the
other transformationin the model) used to move from the reference C
274

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions

majorto the appropriateP L subgroup.The transformationpairR L generatesone completegroupof order24; it is the circle of fifths,with minor
keys interspersedbetween theirR and L relatedmajorkeys.
Example 3 shows the Model 2: P R L representation(see also Douthett and Steinbach 1998). The exchangingnotes are added at each vertex in outline letters.For example, the vertex C majorhas C exchanging
with B (E minor), E exchanging with Eb (C minor), and G exchanging
with A (A minor) with the L, P, R transformations,respectively.As can
be seen, the six verticesof eachhexagonhavea commonexchangingtone.
For example, the hexagon with the vertices C major,C minor,Ab major,
F minor,F major,andA minorsharethe common exchangingtone of C.
The common exchangingtone is indicatedat the centerof each hexagon
(outlined by the small hexagon). Going along the diagonals from lower
left to upperright, the circle of fifths appearsin the exchanging tones.
Going along the diagonalsfromupperleft to lowerright,threesubgroups
of order4 separatedby minorthirdsappearin the exchangingtones.Thus,
the representationderived from the exchanging tones replicates the D
transformationandthe P R subgroups(Example2a) found in the toroidal
representation.Moreover,the representationof exchanging tones suggests that in regions of the space certaintones might be psychologically
primedas potentialexchangingtones. This would account,for example,
for the readinesswith which listeners assimilate modulationsto closely
relatedkeys (Krumhansland Kessler 1982).
The discussion so far has focused on only four transformations:D, P,
R, and L. It might be of interestto considerthe geometricrepresentation
of some of the other Riemanniantransformations.Example4 shows the
six RiemannianSchritte described by Klumpenhouwer(1994). These
take a majortriadto anothermajortriad(Example4a), and a minortriad
to anotherminor triad (Example 4b). The transformationsare: Quintschritt,(Q), Gegenquintschritt(-Q), Terzschritt(T), Leittonschritt(L, to
be distinguishedfrom the use of L above and below as meaningLeittonwechsel), Ganztonschritt(G), and Kleinterzschritt(K). Comparingthe
two figuresmakesclear the complementarynatureof the transformations
for majorand minortriads.In principle,it would be possible to ascertain
using the above method whether any of these transformationshas an
independentpsychological reality from the other transformationsconsideredhere.7
Judgments of chord tension
An independentset of empiricaldataon triaddistancescomes from a
studyof musicaltension(Bigandet al. 1996).8The essentialdetailsof the
methodology are:
275

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions

RiemannianSchrittsfor MajorTriads
(4a)
Db C#

c#

b #
D
)C

-Q
C

Ab
c

g#
GEb

B
eb)d#

eb d#

Bb

F#

Riemannian
Schrittsfor MinorTriads
f#
d

(4b)
Db C#

bb

c#

E
g#

e-Q
B

Eb

f#

eb d#

g
D

F#

Ab
K

eb'd#

Db C#

Bb
dbb

Example4
Shows RiemannianSchritts(Klumpenhouwer1994) superimposedon
the toroidalrepresentation(Krumhansland Kessler 1982).
Example4a shows the Schrittefor majortriads;Example4b shows the
Schrittefor minortriads.

276

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions

1. Two groups of subjects participated,fourteenmusic conservatory

studentsand fourteenmusically naive students.The data from the
two groups of studentswill be denotedBPLmusand BPLnonmus,
respectively.
2. Eachtrialbeganwith a shortsequencein C major,followed by three
chords:C major,X, C major.The X chordwas any of the twelve possible major,minor,major-minorseventh,or minor seventh chords.
Only the data for the major and minor triads will be considered
because these can be compareddirectlywith the KKcorand KKdist
values analyzedabove.
3. The subject's task was to rate the tension producedby the second
chord of each trial on a 12-point scale. The tension ratings will
be taken as an indirectmeasure of psychological distance from C
major.9
Before turningto other models consideredin Bigand et al. 1996, the
BPLmusandBPLnonmusdatawill be analyzedwith Models 1, 2, and 3.
The correlationvalues for these threemodels are shown in the two righthand columns of Table 2. As before, Models 1 and 2 achieved approximatelyequally good fits to the data,butModel 3 provideda betterfit then
either of the other models.' The advantageof Model 3 over Model 1
shows the independentpsychologicalstatusof the L transformation,while
the advantageof Model 3 over Model 2 shows the independentpsychological status of the D transformation.Comparedwith the model fits to
KKcorandKKdist,the presentcorrelationswere somewhatlower,andthe
optimalweights for the linearmodel were somewhatdifferent.However,
again, the best fittingmodel was Model 3: D P R L. In general,the musicians' data were the least well fit by any of the three models, suggesting
that furtheranalysis is needed to understandmusicians'cognitive representationof triad distances, with an eye towardunderstandingthe relationshipsbetween the differentmodels.
Other models of triad distance
The originalarticle(Bigandet al. 1996) tested a numberof othermodels. The firstmodel was Lerdahl's(1988) tonal pitch space theorywhich
consists of threecomponents.The firstcomponentis pitch-classproximity, which measuresthe numberof distinctiveelements in the basic pitch
space. The basic pitch space consists of five levels: chromatic,diatonic,
triadic,fifth, androot. In a given key region,the root is representedat five
levels, the fifth above the root at four levels, the third above the root at
three levels, other scale tones at two levels, and nonscale tones at one
level. Pitch-class proximityis the numberof distinctiveelements of the
second triadcomparedwith the first (in this case, C major).The second
277

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions

componentis triadproximitywithin a key, measuredas distancearound

the diatonic circle of fifths (C-G-D-A-E-B-F-C).The third space is the
distances between keys or regions, measuredon the chromaticcircle of
fifths (C-G-D-A-E-B-F#-Db-Ab-Eb-Bb-F-C).
The distances measuredin
all threespaces are summedto give the predicteddistancesfrom C major
to each triad.Any triadcan, of course, be representedin a numberof different key regions, so Lerdahluses a shortest-pathcriterion.For example, the shortestpathfrom C majorto A majoris V/d.
Table 3 shows the correlationsbetween the tonal pitch space predictions and the four data sets considered here. For all except BPLmus,
Model 3: D P R L gives a betteraccountof the data;the two models are
essentially equivalent for the BPLmus data. This result suggests that,
comparedto nonmusicians,musiciansmore stronglyweigh a hierarchical model of embeddedtone, triad,and key (region)distancesin judging
chordtension.In summary,of the variousmusic-theoreticmodels of triad
distance, the strongestempiricalsupportis for Model 3: D P R L, with
the exceptionjust noted of thejudgmentsof chordtension madeby musicians.
A psychoacoustically-basedmodel of chord tension was also tested
(Bigandet al. 1996). It is based on knownpropertiesof sensory processing, independentof cognitive representationsthat are acquiredthrough
learning.The model, called pitch commonality,was developedfrom previous work on virtualpitch (Terhardt1974; Parncutt1989). Accordingto
this approach,the perceivedpitchesof a triadarenot limitedto the notated
pitches and their harmonics.Virtual pitches are also perceived to the
extent that the spectral (physically present) pitches approximatea harmonic series above the virtualpitch. So, for example, an Eb-majortriad
weakly implies the pitch C. The pitch commonalitymodel computesthe
predictedpitch salience of each X chordandcomparesthese (by correlation) with the predictedpitch salience of the referenceC-majortriad.The
resultingpitchcommonalitypredictionswere correlatedwiththe four sets
of empiricalvalues, as shown in the next line of Table3. The resultswere
very similarto those for the tonalpitch space model (Lerdahl1988). That
is, for all except BPLmus,Model 3: D P R L gave a betteraccountof the
data;the two models were essentially equivalentfor the BPLmus data.
Overall,Model 3: D P R L providedthe most accuratepredictionsfor the
data.
The final analysis consideredthe voicing of the chords in the Bigand
et al. (1996) experiment.The independentvariableused in this model,
horizontalmotion,is the sum of the sizes of the intervalscoveredby each
voice when passing from the C-majorchordto the X chord.This is computed from actual pitches, not abstractpitch classes. The horizontal
motionvalues,when correlatedwith the fourempiricalsets of values,produced the resultsshown in the last row of Table3. As would be expected,
278

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions

given that these values are based on the particularsof the Bigand et al.
(1996) experiment,the correlationswere low for the KKcorand KKdist
values. The correlationbetween horizontalmotion and the BPLnonmus
data was relativelyhigh, althoughlower than Model 3: D P R L. This is
interestingin the presentcontext because it shows a directeffect of tone
distance on the chord tension judgments. Horizontal motion added to
Model 3: D P R L resultedin a very good fit of the BPLnonmusdata,11
suggestingthattone distancemeasuredbothin termsof actualpitchesand
abstractpitch classes (as embodied in Model 3: D P R L ) are psychologically highly salient for nonmusicians.
The analysisof the datafrom the Bigandet al. (1996) studyfound different results for musicians and the nonmusicians.For the musicians,
threemodels performedapproximatelyequally:the D, P,R, andL model,
the tonal pitch space model, and the pitch commonalitymodel. That the
tonal pitch space model performedrelativelywell for these datasuggests
thatmusiciansemploy a hierarchicalmodel of embeddedtone, triad,and
key (region) distancesin judging chordtension. Thatthe pitch commonality model also performedwell reinforcesParncutt's(1989) claim that
music-theoreticconstructssuch as those tested in the D, P, R, and L and
tonal pitch space models have a psychoacousticbasis.12 For the nonmusicians,the model using the fourtransformations,D, P,R, andL, provided
the best fit of the empiricalvalues, as it had for the Krumhansland Kessler (1982) values. In addition,the nonmusicians'data correlatedquite
stronglywith the Bigandet al. (1996) measureof horizontalmotion.This
suggests thatnonmusiciansareespecially sensitiveto pitchproximity,the
principle underlyingthe neo-Riemanniantransformations,measuredin
both actualand abstractpitch class distances.13

279

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions

NOTES
1. Onlythe directlyrelevantaspectsof the methodswill be described.
2. This is justifiedgiven the similarratingsfor the tonic triadsand the cadences,
althoughit mightbe objectedthatthisconfusestonictriadswiththekeys schematicallyrepresented
by thecadences.Thesimilaritybetweentheprobe-toneprofiles
for triadsandcadencescan be understood,however,becausethe latterall ended
on the tonictriad.
3. The differencesbetweenKKcorandKKdist,althoughminor,may be important
thatabstractsthepatbecausethelatteraredistancesin a geometricrepresentation
ternsunderlyingthe correlations.
4. Thedominanttransformation,
D, is shownonly formajortriads;it appliesequally
to minortriads.
is the firsttransfor5. The conventionwill be usedthatthe left-mosttransformation
mationapplied.
thattheimprovement
of Model3
6. Thefollowinganalysiswas doneto demonstrate
was not simplya matterof addingone moreindependentvariable(andthusreducing the degreesof freedomby one). ForModel 1, 2, and3, theunweightedsumof
the independentvariableswas correlatedwith the KKcorandKKdistvalues.All
thesecorrelationshavethe samenumberof degreesof freedom.Thisproducedthe
followingresults:Model 1, correlationof SUM(D,P,R)= .91 and .88 for KKcor
andKKdist,respectively.Model 2, correlationof SUM(P,R,L)= .73 and .78 for
KKcorandKKdist,respectively.Model3, correlationof SUM(D,P,R,L)= .95 and
.95 forKKcorandKKdist,respectively.Again,by thismeasureModel3 fitthevalues consistentlybetterthanthe othertwo models.
and
7. Kopp(1995) criticizedLewinfor privilegingthe P, R, andL transformations,
triadsof the samemode(shownin Example
emphasizedinsteadthe third-related
4).
8. I am gratefulto EmmanuelBigandfor providingthe datafor the musiciansand
nonmusiciansin the study(Bigandet al. 1996).
9. This is supportedby the significantcorrelationsbetweenBPLmusandBPLnonmus data,on the one hand,andthe KKcorandKKdistvalues,on the other.The
correlationsfor the BPLmuswere r = .81 and .83 for KKcorandKKdistvalues,
correlationsfor the BPLnonmuswerer = .73 and
respectively;the corresponding
.77. Thesesignificantcorrelationsshowthatthesedifferentempiricalmeasurestap
of triaddistance.
intothe sameor similarcognitiverepresentations
10. To demonstratethat the improvementof Model 3 was not simply a matterof
addingone moreindependentvariable(andthusreducingthe degreesof freedom
by one), the unweightedsumof theindependentvariableswas correlatedwiththe
BPLmusandBPLnonmusvaluesforModel 1, 2, and3. Thisproducedthe following results:Model 1, correlationof SUM(D,P,R)= .74 and .75 for BPLmusand
BPLnonmus,respectively.Model2, correlationof SUM(P,R,L)= .68 and .78 for
BPLmusandBPLnonmus,respectively.Model3, correlationof SUM(D,P,R,L)=
.82 and .78 for BPLmusand BPLnonmus,respectively.Again, by this measure
Model3 fit the valuesbetterthanthe othertwo modelsexceptfor the equalfits of
Model2 andModel3 for theBPLnonmusdata.
11. The multiplecorrelationpredictingBPLnonmususingModel3 D P R L andhorizontal motion was R = .95, p < .0001

280

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions

12. Thatthesethreemodelshavestrongunderlyingcommonalitiesis supportedby the

factthatthetonalpitchspaceandthe pitchcommonalitymodelscanbe quitewell
accountedfor by Model3 D P R L, withmultiplecorrelationsof R = .94 and.91,
respectively.
13. I am gratefulto RichardCohn,FredLerdahl,andJohnCloughfor commentson
an earlierversionof this manuscript.

281

This content downloaded from 128.84.127.82 on Wed, 3 Apr 2013 14:34:27 PM

All use subject to JSTOR Terms and Conditions