A sleep-deprivation experiment found that stimulant benefits depended on the executive task being tested. A drug’s ability to sustain alertness therefore should not be treated as evidence that it restores every higher-order function.
Killgore and colleagues randomly assigned 54 healthy adults to caffeine, 600 mg; modafinil, 400 mg; dextroamphetamine, 20 mg; or placebo. Dosing occurred after 44 hours awake. Participants then completed planning, sequencing and card-sorting tasks after approximately 45–50 hours awake.
The experiment used double-blind administration in a residential sleep laboratory. Its question concerned performance during substantial sleep deprivation, not enhancement in rested people or treatment of a chronic sleep disorder.
The modafinil and dextroamphetamine groups completed the Tower of London in significantly fewer moves than placebo participants. Modafinil was also associated with greater deliberation before moves. On the Tower of Hanoi, caffeine participants used fewer moves than all three other groups, while completion speed was not significantly influenced by the stimulants.
Modafinil performed better on reported card-sorting perseveration indices. These different patterns do not support a single drug ranking applicable to every executive process.
The tasks were administered at different times during continued wakefulness. The authors explicitly cautioned against direct comparisons across tasks for that reason. Later performance could reflect both the task and the evolving state of sleep deprivation or medication effect.
Each treatment group was also modest in size, and only the chosen doses were tested. A comparison at one dose cannot establish equivalent dosing or the complete response profile of each drug.
Being more awake is not a validated substitute for planning, flexibility, response control or judgment. Nor does a laboratory advantage establish that a sleep-deprived person is safe to drive.
The useful conclusion is narrower: under this protocol, different stimulants produced different task-specific advantages. A case assessment still needs actual sleep history, exposure timing and evidence about the relevant function.
Related reading: Sleep disruption and next-day cognition in monkeys.