Improving digit span assessment of short-term verbal memory

Area of Science:

• Neuroscience
• Cognitive Psychology
• Psychometrics

Background:

• Digit span (DS) is a standard measure of short-term verbal memory.
• Traditional DS measures have limitations in reliability and precision.
• Computerized administration and novel metrics can potentially improve DS assessment.

Purpose of the Study:

• To introduce and evaluate a new mean span (MS) metric for digit span (DS) assessment.
• To compare the psychometric properties of the MS metric with traditional DS measures.
• To explore the utility of the MS metric in understanding demographic influences on DS performance.

Main Methods:

• Two experiments were conducted using computerized auditory digit presentation.
• Performance-adapted list length adjustment was employed.
• A novel mean span (MS) metric was developed and analyzed.

Main Results:

• The MS metric demonstrated reduced variance and improved test-retest reliability compared to traditional DS.
• MS showed higher correlations with other neuropsychological tests.
• MS enhanced sensitivity in forward versus backward span comparisons and allowed for normative criteria with subdigit precision.

Conclusions:

• The new mean span (MS) metric significantly enhances the precision and reliability of digit span (DS) assessment.
• Computerized stimulus delivery and the MS metric improve the evaluation of short-term verbal memory.
• The MS metric provides a more nuanced understanding of factors influencing DS performance, such as age and education.