Psychophysiology (2002) 39, 38-48
Stefan Koelsch1, Erich Schröger2, & Thomas C. Gunter1
1 Max Planck Institute of Cognitive Neuroscience, Leipzig 2 Institute of General Psychology, Leipzig
ko_sc_gu02.pdf (223 KB)
If you have not installed the Adobe Acrobat Reader to read this pdf-file click on the button below:
During listening to a musical piece, unexpected harmonies may evoke brain responses which are reflected electrically as an early right anterior negativity (ERAN) and a late frontal negativity (N5). In the present study we demonstrate that these components of the event-related potential can be evoked pre-attentively, i.e. even when a musical stimulus is ignored. Both ERAN and N5 differed in amplitude as a function of music-theoretical principles. Besides, ERPs indicated a pre-attentive build-up of musical context. Participants had no special musical expertise, results thus provide evidence for an automatic processing of musical information in "non-musicians". ERPs are strongly reminiscent to ERPs elicited during the processing of language.
Fig. 2: Ignore condition, in-key chords. Grand-average potential map of the position-dependent amplitude effect which is suggested to reflect to buildup of musical context (difference-ERPs: chords from the fifth position subtracted from first position, view from top, interpolated over the time-window from 540-600 ms after stimulus onset; blue areas indicate negative potentials; all maps are presented with nose-reference).
|
ERAN (150-210 ms) |
N5 (540-600 ms) |
|
|
|
|
|
|
Fig. 3: Ignore condition, ERPs to chords at the fifth position. Scalp-distribution of effects elicited by Neapolitan chords at the fifth position (potential maps of difference-ERPs: in-key chords subtracted from Neapolitan chords) for an early (150-210 ms, left) and a late (540-600 ms, right) time interval; blue areas indicate negative potentials. When nose-reference is used, both ERAN and N5 invert polarity at centro-temporal electrodes.
|
ERAN (150-210 ms) |
N5 (540-600 ms) |
|
|
|
|
|
|
Fig. 4: Ignore condition, ERPs to chords at the third position. Scalp-distribution of effects elicited by Neapolitan chords at the third position (potential maps of difference-ERPs: in-key chords subtracted from Neapolitan chords) for an early (150-210 ms, left) and a late (540-600 ms, right) time interval. The effects elicited by Neapolitan chords at the third position are distinctly larger compared to the fifth position. Because Neapolitan chords are a subdominant variation, Neapolitan chords are harmonically inappropriate when presented at the fifth position of the sequence (where only a tonic chord is appropriate), but fairly suitable at the third position (where a subdominant is expected). The amplitude-modulation of effects thus reflects the degree of "music-syntactic" incongruity of Neapolitan chords. Notably, the degree of "music-syntactic" incongruity refers to a complex rule-based system (described by the theory of harmony of the major-minor tonal system); moreover, the degree of incongruity was processed under a condition where participants ignored the musical stimuli. With this respect, results indicate a pre-attentive musicality of the human brain.
|
N5 (540-600 ms)
|
|
|
First half |
Second half |
|
|
|
|
|
|
Fig. 6: Ignore condition, split half. During the first half of the block, the N5 elicited at the fifth position had a bilateral scalp-distribution (left), whereas the N5 was lateralized to the right during the second half (right). Potential-maps were calculated by subtracting ERPs of in-key from Neapolitan chords, ERPs were interpolated over a time-window from 540-600 ms.
|
ERAN (180-250 ms)
|
|
|
3rd position |
5th position |
|
|
|
|
|
|
Fig. 9: Attend condition (detect harmonically inappropriate chords), potential-maps of the ERAN elicited by Neapolitan chords at the third (left) and fifth (right) position (difference-ERPs: in-key chords subtracted from Neapolitan chords, interpolated over the time-interval from 150-210 ms). As in the first block, the ERAN is distinctly larger when elicited at the fifth position. The amplitude of the ERAN does virtually not differ between both the ignore and the attend condition
