Effects of psychotropic drugs on digit substitution: comparison of the computerized symbol-digit substitution and traditional digit-symbol substitution tests

Abstract

The digit-symbol substitution test (DSST), performed with paper and pencil or computerized, is widely used to reveal decrements in human attention and cognition. We programmed sets of adjustable tasks (digit-symbol, digit-digit, symbol-digit, symbol-symbol) into a microcomputer and compared the symbol-digit substitution (SDST) and the digit copying test (DDCT) with the traditional DSST in two placebo-controlled double-blind studies of psychotropic drugs with pre-trained young healthy subjects. Performances were measured before drug intake and several times after it; matched, different codes were used at consecutive tests. DSST and SDST substitutions remained at the baseline level after placebo, while the simple DDCT performance improved during the placebo session. The prolonged (3 min) test was not exhaustive because interim counts at 90 s predicted the final performance well. In Trial I, 15 mg diazepam orally reduced DSST and SDST functions in a similar way, but it also impaired simple copying in the DDCT, though to a lesser extent. Ebastine, an H(1)-antihistamine, proved inert alone and failed to increase the effects of diazepam on these variables. In Trial II, 7.5 mg zopiclone, 0.4 mg suriclone and 50 mg chlorpromazine, alone and in combinations, impaired the DSST performance in the manner expected. The drug effects were similar in the SDST, and somewhat less in the DDCT, while the substitution errors were subject related and not altered significantly by any treatment. The simple correlation matrices (Pearson, Spearman), confirmed by analysis of covariance, showed that the results of DSST correlated fairly well with those of SDST after zopiclone, chlorpromazine and their combination, but not after suriclone or its combination with chlorpromazine. The DDCT results correlated with those of the substitution tests when analysing pooled baseline values, but not when analysing the performances after drug intake. Subjective visual analogue variables correlated poorly or not at all with objective performances. Our results suggest that manual dexterity in these computerized tests might contribute significantly to the total impairment of performance in response to different drugs. The DSST and SDST matched each other fairly well in their sensitivity to drug effects, yet this similarity may depend on the drug used.