instrument, or physics skill). We want to be able to tell teachers that there
are principled things that the ycan do to increase the frequenc yof those 0.01
percent increments in learning.
Music is no exception to this, and music teachers are continuall yinquiring of
psychologists how psychological insights can inform their work. It is their per-
ception that musical expertise is taught and acquired with great difficulty. They
speak of ‘‘tone-deaf’’ children (usuall ychildren unable to sing in tune); the y
speak of the difficult yof teaching sight reading, of teaching rh ythm, of teaching
good intonation on a string instrument, and so on.
M yearl yresearch on the skill of sight reading has been summarized else-
where (Sloboda, 1984). That research was carried out under the influence of the
previousl ypublished work of Chase and Simon (1973) on chess perception.
Their research showed that, like playing chess, reading of music depended on
an abilit yto pick up various sorts of patterns in the stimulus. For instance,
good sight readers were found to be much more prone than poor sight readers
to a sort of ‘‘proofreader’s error’’ (Sloboda, 1976a) whereb ynotational mistakes
out of character with the genre were automaticall ycorrected back to what the
genre would have predicted. Their abilit yto use music structure to ‘‘chunk’’
notes could account for their superior short-term memor yfor notation (Halpern
& Bower, 1982; Sloboda, 1976b).
Encouraging as it was to find results for music that so clearl yparalleled
Chase’s findings, I became progressivel ymore disheartened as I talked about
those results to groups of teachers. The question the yall asked was of what
prescriptions I would draw from m yresults for the teaching of sight reading,
and after some hand waving I reall yhad to admit that there were no pre-
scriptions that I could draw at that time. I did not know how one could teach
children to ‘‘see’’ structures.
Since then I have come to realize that in order to ‘‘see’’ musicall ysignificant
structures, one first must be able to ‘‘hear’’ those structures, and I have learned
from reading some excellent recent research that the process of coming to
‘‘hear’’ musical structure is a process that occurs quite naturall yfor the majorit y
of children as a function of normal enculturation. For instance, Zenatti (1969)
showed that children at age 7 show a distinct memor yadvantage for sequences
conforming to rules of normal tonal progression, as compared with atonal
sequences. This advantage is not shared b ychildren of age 5. Similar results
were obtained from studying children’s songs (Dowling, 1982, 1988; Gardner,
Davidson, & McKernon, 1981). There is a definite age progression from tonal
inconsistenc yand instabilit ytoward conformit yto the norms of the tonal
culture.
An experiment I conducted earlier (Sloboda, 1985a) showed that the pro-
gressing of the abilit yto discriminate between ‘‘legal’’ and ‘‘illegal’’ sequences
did not seem to depend on children’s receiving an ysort of formal music
instruction. Almost no children at age 5 made meaningful discriminations,
whereas almost all 11-year-olds made discriminations in accordance with those
of adults (and music harmon ytextbooks). The children who were receiving
formal music lessons did not fare better than other children.
Although man yexperiments with adults have shown cognitive differences
between musicians and nonmusicians, some studies have shown little differ-
568 John A. Sloboda