Two experiments evaluated the effect of interstimulus interval aSI) on recall of digit sequences printed on the palms. Digits were printed with ISIs of 1 or 5 sec. Half of the patterns were drawn on the left palm and half on the right. High levels of accuracy were obtained in both experiments with the 5-sec rate. The passive tactile digit span was nearly seven at the slower rate, but much less at Lsec ISIs. Heller, 1986b). The first digit of two-digit sequences was printed on the left palm and the second was printed on the right palm . Half of the numbers were written on the left palm and half on the right for even numbers , with the beginning half drawn first on the left. The extra digit was always drawn on the left palm for odd numbers . Observers were fully informed about the procedure prior to the retention task. They were told where to expect the numbers, and they were informed about orientation and sequence length. However, feedback was not given as individualsattempted recall.
Results and DiscussionTactile memory was far better when sequences were presented at a slower rate . Table I shows the proportion of sequences recalled correctly at l-sec and 5-sec ISIs.Performance was very good for long sequences at the 5-sec lSI. Data analysis showed a significant effect of rate [F(l,18) = 16.3, P < .00 1], a robust effect of sequence length [F(4,72) = 39 .9, p < .00 1], and a significant interaction [F(4,72) and that two-and three-digit sequences were equally easy to remember (p > .05) .The results of the present experiment suggest that the tactile digit span is at least six when spatial separation is employed and slow rates of presentation are used . To further quantify the normal span, an additional 10 subjects attempted serial recall of seven-digit sequences at either
MethodSubjects. There were 20 subjects in the first experiment; half were male and half were female .Stimuli and Apparatus. The stimuli of Experiment 1 were random number sequences, varying in length from 2 to 6 digits . Numbers were approximately 4 em high; they were drawn with the aid of a Lucite stylus with a . 16-cm rounded tip. An upright orientation was used, and the digits were drawn at a rapid rate.Designand Procedure. Experiment I took the form of a mixed-factor analysis of variance , with the factors being interstimulus interval (lor 5 sec) and sequence length (2-, 3-, 4-, 5-, or 6-digit sequences), with repeated measures on sequence length . Subjects were initially exposed to a practice run that consisted of the 10 digits two times each , presented in a random sequence, with feedback for an incorrect identification. The aim of this practice was to familiarize subjects with the method of writing the patterns and to ensure normal tactile perception . All subjects scored at least 80% correct (see Critchley, 1953). Subsequently. independent groups of subjects were exposed to number sequences with ISIs of either 1 or 5 sec.The subjects were first exposed to 10 two-digit sequences and then to blocks of longer and longer sequences. Spat...