Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Jun 16;9(24):eadd4165.
doi: 10.1126/sciadv.add4165. Epub 2023 Jun 14.

Not so smart? "Smart" drugs increase the level but decrease the quality of cognitive effort

Elizabeth Bowman  1 , David Coghill  2 , Carsten Murawski  1 , Peter Bossaerts  3
Affiliations

Not so smart? "Smart" drugs increase the level but decrease the quality of cognitive effort

Elizabeth Bowman et al. Sci Adv. .

Abstract

The efficacy of pharmaceutical cognitive enhancers in everyday complex tasks remains to be established. Using the knapsack optimization problem as a stylized representation of difficulty in tasks encountered in daily life, we discover that methylphenidate, dextroamphetamine, and modafinil cause knapsack value attained in the task to diminish significantly compared to placebo, even if the chance of finding the optimal solution (~50%) is not reduced significantly. Effort (decision time and number of steps taken to find a solution) increases significantly, but productivity (quality of effort) decreases significantly. At the same time, productivity differences across participants decrease, even reverse, to the extent that above-average performers end up below average and vice versa. The latter can be attributed to increased randomness of solution strategies. Our findings suggest that "smart drugs" increase motivation, but a reduction in quality of effort, crucial to solve complex problems, annuls this effect.

PubMed Disclaimer

Figures

Fig. 1.
Fig. 1.. Task relevance, experiment design, and overall participant performance.
(A) Computationally difficult tasks are ubiquitous in everyday life. (B) Task interface with example instance (grayscale version; original in color). Items become highlighted as they are selected. (C) Timeline of experiment and Latin square randomization across four experimental sessions. (D) Proportion of correct solutions submitted, stratified by task difficulty (Sahni-k index, from low 0 to high 4); circle: estimate of proportion; bars, ±2 SE.
Fig. 2.
Fig. 2.. Performance, effort, and speed.
(A to C) Empirical cumulative distribution function under PLC (blue) and drugs (red) and pointwise 95% confidence bounds (CB; based on Greenwood’s formula). (A) Knapsack value reached as a fraction of maximal value. PLC first-order stochastically dominates drugs, implying that the chance that participants reach any value is uniformly lower under drugs than under PLC. (B) Effort is equal to the time spent until submission of solution. Drugs first-order stochastically dominates PLC, implying that the chance of spending any amount of time is uniformly higher under drugs than under PLC. (C) Effort is equal to the number of moves of items in/out of knapsack until submission of solution; drugs first-order stochastically dominates PLC, implying that the chance of executing any number of moves is uniformly higher under drugs than under PLC. (D) Probability density estimates of speed under PLC (blue) and drugs (red), where speed is equal to the number of seconds per move. Because the density under drugs is shifted to the left of that under PLC, speed tends to be higher under drugs than under PLC.
Fig. 3.
Fig. 3.. Quality of effort.
(A) Violin plots of productivity, measured as average increase in value of knapsack per item move in/out of knapsack. Stars indicate significance of differences in means based on a generalized linear model that accounts for confounding factors and participant-specific random effects for average productivity and impact of drugs (table S6); *P < 0.05 and ***P < 0.001. (B and C) Estimated participant-specific (random) deviations in productivity from mean productivity. Productivity is measured as average increase in value of knapsack per item move; random effects were estimated with a generalized linear model that accounts for confounding factors and participant-specific random effects for average productivity and impact of drugs (table S6). (B) MOD against DEX. The red line shows OLS fit, with significant positive slope (P < 0.001). (C) MPH against PLC. The red line shows OLS fit, with significant negative slope (P < 0.001). Arrows indicate range of productivity deviations under PLC (horizontal) and MPH (vertical). The range is smaller under MPH than under PLC, implying reversion to the mean. (D) Reduction in quality of first full knapsack chosen under drugs (right) relative to PLC (left). Quality is measured as overlap between number of items in chosen knapsack and optimal knapsack. Decrease in mean quality is significant at **P < 0.01, based on a generalized linear model that accounts for effect of instance difficulty and overlap with items in the Greedy solution, as well as participant-specific random effects for average quality (table S7); overlap tends to be lower under drugs than under PLC, implying lower quality of solution search.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. E. Bowman, B. Feng, C. Murawski, P. Bossaerts, Surveying the Use of Pharmaceutical Cognitive Enhancers in the Australian Financial Services Industry (SSRN Scholarly Paper 3661966, Social Science Research Network, 2020).
    1. Dietz P., Soyka M., Franke A. G., Pharmacological neuroenhancement in the field of economics—Poll results from an online survey. Front. Psychol. 7, 520 (2016). - PMC - PubMed
    1. Emanuel R. M., Frellsen S. L., Kashima K. J., Sanguino S. M., Sierles F. S., Lazarus C. J., Cognitive enhancement drug use among future physicians: Findings from a multi-institutional census of medical students. J. Gen. Intern. Med. 28, 1028–1034 (2013). - PMC - PubMed
    1. Kortekaas-Rijlaarsdam A. F., Luman M., Sonuga-Barke E., Oosterlaan J., Does methylphenidate improve academic performance? A systematic review and meta-analysis. Eur. Child Adolesc. Psychiatry 28, 155–164 (2019). - PubMed
    1. Ilieva I., Boland J., Farah M. J., Objective and subjective cognitive enhancing effects of mixed amphetamine salts in healthy people. Neuropharmacology 64, 496–505 (2013). - PubMed

MeSH terms