Numerical Memory

(Editor:   Karen Brakke)


The presentation of a series of digits orally by an examiner to an examinee is a common measurement typically referred to as Digit Span in many intelligence measures. In the majority of these intelligence tests, the researcher presents the digits to the examinee in an auditory format only, at which time the examinee repeats the digits back to the examiner. The rationale for this type of task is that it provides a good measure of short-term auditory memory and attention. Because the participant must orally recall the auditory information in the proper sequence, the digit span task is often described as a sequencing task (Sattler, 1992). This Numerical Memory experiment employs a similar format to Digit Span tasks found in assessment instruments, comparing the individual's short-term memory for digits presented in an auditory versus visual format.


This study uses a within-subjects design comparing digit recall for numerical stimuli presented in an auditory and visual format. The independent variable in this study is the format in which the stimulus appears (auditory vs. visual), and the dependent variable is the length of the digit sequence that the research participant can recall.

STIMULI. The stimuli for this study consist of numerical values ranging from 0 to 9. These numerical values are presented to the research participant in one of two formats: visual or auditory. The auditory digits (AD) are voice input recorded using Microsoft Window's Sound Recorder. Each AD was edited down to approximately 1 second in duration with equivalent amplitudes. Using this procedure, 10 digits were approximately equivalent in length and intensity. The 10 visual digits (VD) were produced using an Authorware5 display icon. Each VD was produced in Times New Roman font in black text with a 175-font size. The display time for each VD was set for 1 second.

TASK. Each participant engages is two tasks. The participant is presented with either the VD or AD task, followed by the alternate task (AD or VD). The order of task presentation is randomized for each participant. Both the VD and AD tasks involve the presentation of a sequence of numbers presented in a random order. Each task begins with a sequence that is one digit in length. The length of the sequence increases by one digit after a level has been presented twice. Each sequence of digits presented to the participant is randomized and, therefore, unique. Research participants move to the next level of difficulty if they respond correctly to at least one of the two trails presented at each level. This process continues until the participants miss both trials at a given level, at which time participants will either complete the second task (AD or VD), or the experiment will be terminated.

Data Analysis

The most common analysis for this experiment is a repeated-measures test examining the effect of stimulus modality on length of digit span.

Data Format and Download

Sample data image from the Numerical Memory study
Figure 1

As illustrated in Figure 1, data are compiled in an Excel spreadsheet or in a comma delimited file with the following labels:

USERID - the value assigned to the individual participant.
CLASSID - the value affiliated with the registered course.
GENDER, AGE, DATE, are self explanatory columns.
TOTIME - elapsed number of seconds for completion of the experiment
AUD - data reflect, similar to the Digit Span WISC-III Subtest, the number of items in correct sequence (Auditory format). The trial is terminated when the participant fails two consecutive trials in any series.
VIS - data reflect, similar to the Digit Span WISC-III Subtest, the number of items in correct sequence (Visual format). The trial is terminated when the participant fails two consecutive trials in any series.


This type of task is often included in intelligence tests (i.e., WISC, WAIS) therefore it may be worthwhile to consider how the digit span memory test might be related to "intelligence". Students might also want to hypothesize how memory is influenced by external factors. If performance is related to the modality of stimulus presentation, what does that say about the reliability of digit span as an indicator of intelligence?

Equipment Requirements

Computers used in this experiment must have a sound card. Research participants must either use a earphones or a pair of speakers plugged into the "audio out" jack on the sound card.


Sattler, J.M. (1992).  Assessment of Children: Revised and Updated Third Edition.
	San Diego, CA: Jerome M. Sattler, Publisher, Inc.