J. Bredenkamp, THE CONNECTION OF VARIOUS INVARIANCE HYPO THESES IN THE FIELD OF PSYCHOLOGY OF MEMORY, Zeitschrift fur experimentelle und angewandte Psychologie, 40(3), 1993, pp. 368-385
This article tries to connect four invariance hypotheses that, up to n
ow, have been independently investigated and hereby incorporates resul
ts of former investigations of a mental calculation expert. First, the
hypothesis of Cavanagh (1972) is described. The information-theoretic
al interpretation of this hypothesis results in the connection with th
e modified total time-hypothesis and the hypothesis of Nevelski (1970)
, which together imply Thurstone's equation for the learning curve. Th
e result of a reanalysis of some experiments published by Bredenkamp a
nd Hager (1979) was in accordance with this connection. Therefore, the
information-theoretical interpretation of Cavanagh's hypothesis was a
pplied to estimate the time per feature test - a constant in Cavanagh'
s relation - from some published experiments. The estimations were in
accordance with one another and show that each feature test lasts appr
oximately 9 ms. This value fits well with Geissler's time quantum. Rel
ating Cavanagh's hypothesis with the hypothesis of a constant duration
of memory traces (Baddeley, 1990; Schweikert & Boruff, 1986), the rei
nterpretation of the results from former experimental investigations o
f a mental calculation expert (Bredenkamp et al., 1988; Bredenkamp, 19
90) becomes possible. Finally, some suggestions are made as how the de
veloped set of hypotheses can be tested empirically.