Scarlet & Grey
Ohio State University
School of Music


Daniel Werts

Notes by Igor Karaca


Music 829
February 14, 2002

Werts, Daniel. (1997). "Good, Fair, and Bad Chord Progressions: A Regression-Analysis of Some Psychological Chord Progression Data Obtained in an Experiment by J. Bharucha and C. Krumhansl." Music, Gestalt and Computing., Ed. Marc Leman. Berlin; Heidelberg; New York; Barcelona; Budapest; Hong Kong; London; Milan; Paris; Santa Clara; Singapore; Tokyo: Springer, 1997. 200-213.

 

Main research focus: systematic and cognitive musicology (how music generates meaning in the sensorial, perceptive, cognitive, motoric and emotional domains); multimedia technology applied to the arts.

Y(est)= 0.320 + 0.013 (X1) + 0.433 (X2) + 0.262 (X3) + 0.227 (X4) - 0.244 (X5) - 0.266 (X6)

Explanation:

-Y is Krumhansl's data.

-Y(est) is approximated value for each chord progression.

-X1 through X6 are six musical properties (factors) chosen for Wert's formula after a long period of trial and error in which many factors, all of which seemed musically relevant, were tested.

X1(Sum of Ideal Chord Weights) - most powerful factor, accounting for 54.1% of the explained variance in data. X1 = W1 + W2. W1 is the assumed psychological weight for the first chord in the progression, and W2 is the assumed psychological weight for the second chord.

X2 (Chord - Order Condition) – accounts for 13.9%. X2 can have two values: 1 (for progressions that meet the condition) and 0 (for those progression for all others)! Meeting the condition means that the pair of chords can be extracted from this model:

Am Em F Dm Bo G C

X3 (Subdominant-Dominant Condition) -- accounts for 8.4%. X3 can have two values: 1 (for progressions that meet the condition) and 0 (for those progression for all others)! Meeting the condition means that a chord pair has to contain at least one chord from S group and one from D group.

X4 (Root Movement By Upward Or Downward 5th). 7.3%. Can be 1 (for 14 chord progression) or 0 (for remaining 28)! Condition: The root of the 2nd chord has to be 5th above or below that of the 1st chord. (This has a positive effect on listeners' perception).

X5 (Problematic resoution of the leading tone). -- 7.8%. Negative factor. Can be 1 or 0. For 1 leading note cannot resolve to tonic because either 2nd chord does not contain it (Em-G) or conventionally does not (Em-F).

X6 (Parallel Minor Condition). -- 8.5%. Negative factor. 1 or 0 again. Condition: progression must be part of, or lead smoothly to, a cadence in the parralel minor key, or otherwise create an impression of that key. This threatens the stability of the original key.

In closing- Werts shows how the ideal chord weights can be derived, in the sense of being almost perfectly approximated, from set of other, seemingly unrelated, vaues.



This document is available at http://dactyl.som.ohio-state.edu/Music829E/Notes/Werts.html