Valentin Bragin, Marina Chemodanova, Aysik Vaysman, Ilya Bragin, Eugene Grinayt, Marina Ruditser. Stress Relief and Memory Training Center, Brooklyn, NY, USA. Poster presented at the IPA 2008 European Meeting in Dublin, Ireland. 2008 April 8-11
Background: It is well known that depression is accompanied by cognitive deficits, including working memory. The N-back task has been widely accepted as a tool to study working memory. In this task, subjects are instructed to monitor a series of stimuli and to identify if the current stimulus is identical to the previous stimulus.
Objective: To assess working memory in people with depression by using a computerized N-back task with multiple stimuli (words, numbers, shapes, pictures and textures).
Methods: In this study, 32 medically ill patients with depression (education – 13.94 ± 2.79 years) were evaluated by MMSE, MADRAS and a modified computerized N-back task. MMSE scores ranged from 23 to 30. We used a 2-back task for assessment of different stimuli (words, numbers, shapes, pictures, and textures) which were presented in 5 blocks, each block consisting of only 1 type of stimulus. The same sequence of blocks was presented for each individual. The number of stimuli per block was 40. Targets (30% of total signals) were presented for a second time for recognition. Reaction time (RT), performance (recognized targetsx100/ total target) and errors ((missed targets + false positive targets) x100/total stimuli) were obtained from the 2-back test. Analysis included descriptive statistics, T-tests, and Spearman correlations.
Results: MADRS score (18.72 ± 9.10) was consistent with mild to moderate depression. Subjects demonstrated their own individual performance profile for words, numbers, shapes, pictures, and textures. Performance for words, numbers, shapes, pictures and textures was 89 ± 12, 82 ± 12, 76 ± 20, 81 ± 20 and, 62 ± 24, respectively. Mean reaction times ranged from 1039 ms to 1503 ms. Statistically significant differences (p<0.05) were found between all performances except people and numbers, shapes and numbers and shapes and people. For performances, significant correlations were only found between words and pictures (r=.343, p=0.029) and shapes and textures (r=0.450, p=0.01). For errors, significant correlations were only found between words and pictures (r=0.464, p=0.007) and between shapes and pictures (r=0.447, p=0.01).
Conclusions: This study reveals that N-back task may be used as a practical tool for initial screening of working memory profile in patients with depression. Based on our data, the best performance was for the words and numbers.