“…In fact, many of the investigations included in the metaanalysis of number of A trials employed similar training trials to the A side (Landers, 1971;Harris, 1973;Butterworth, 1975Butterworth, , 1976Butterworth, , 1977Bremner & Bryant, 1977;Bremner, 1978b;Frye, 1980;Butterworth, Jarrett & Hicks, 1982;Appel & Gratch, 1984;Bjork & Cummings, 1984). Furthermore, other studies in the meta-analysis employed training trials either to a center location (Gratch et al, 1974;Schuberth et al, 1978;Horobin & Acredolo, 1986) or to an unspecified location (Evans & Gratch, 1972;Cummings & Bjork, 1983;Sophian & Wellman, 1983;Sophian, 1985;Sophian & Sage, 1985;Sophian & Yengo, 1985). All of this suggests that Wellman et al's previous conclusion regarding the lack of influence of the number of A trials may be incorrect.…”