Computerized analysis of error patterns in digit span recall

Abstract

We analyzed error patterns during digit span (DS) testing in four experiments. In Experiment 1, error patterns analyzed from a community sample of 427 subjects revealed strong primacy and recency effects. Subjects with shorter DSs showed an increased incidence of transposition errors in comparison with other error types and a greater incidence of multiple errors on incorrect trials. Experiment 2 investigated 46 young subjects in three test sessions. The results replicated those of Experiment 1 and demonstrated that error patterns of individual subjects were consistent across repeated test administrations. Experiment 3 investigated 40 subjects from Experiment 2 who feigned symptoms of traumatic brain injury (TBI) with 80% of malingering subjects producing digit spans in the abnormal range. A digit span malingering index (DSMI) was developed to detect atypical error patterns in malingering subjects. Overall, 59% of malingering subjects with abnormal digit spans showed DSMIs in the abnormal range and DSMI values correlated significantly with the magnitude of malingering. Experiment 4 compared 29 patients with TBI with a new group of 38 control subjects. The TBI group showed significant reductions in digit span. Overall, 32% of the TBI patients showed DS abnormalities and 11% showed abnormal DSMIs. Computerized error-pattern analysis improves the sensitivity of DS assessment and can assist in the detection of malingering.

Keywords:

• Verbal
• Memory
• Short-term
• Working
• Malingering
• Wechsler Adult Intelligence Scale
• Adaptive
• Digit span
• Traumatic brain injury
• Concussion

Acknowledgments

This research was supported by the VA Research Service and Award R01ES014038 from the National Institute of Environmental Health Sciences. The content is solely the responsibility of the authors and does not necessarily represent the official views of the funding agencies. We would also like to acknowledge the assistance of the neuropsychology data collection staff: Nerene Lynskey, Kataraina George, and John McKeogh in New Zealand and Ben Edwards, Zoe Doss, David Doss, and Oren Poliva in Martinez.

Notes

¹The CCAB includes the following computerized tests and questionnaires: finger tapping, simple reaction time, symbol–digit, Stroop, digit span forward and backward, phonemic and semantic verbal fluency, card sorting, verbal list learning, spatial span, trail making, symmetry detection, design fluency, the Wechsler Test of Adult Reading (WTAR), visual feature conjunction, the Paced Auditory Serial Addition Task (PASAT), the Cognitive Failures Questionnaire (CFQ), the posttraumatic stress disorder (PTSD) symptoms checklist, and a traumatic brain injury (TBI) questionnaire.

²The difference of 0.8 digits in the criterion ranges of Figure 6 and Figure 7 reflects learning effects due to repeated digit span testing in Experiment 3 as well as the use of a slightly older and less well educated control population in Experiment 4. See Woods et al. (2010) for further details.

Figure 7. Comparison of combined mean span (MS; forward span, FS + backward span, BS) and DSMI (digit span malingering index) scores for 29 TBI (traumatic brain injury) patients (ο) and 39 matched control subjects (×) in Experiment 4. Dashed vertical line: abnormality criteria (p = .05 of controls) for total digit span. Dashed horizontal line: abnormality criteria (7%) for the DSMI.