Rats and humans can optimally accumulate evidence for decision-making.

摘要: The gradual and noisy accumulation of evidence is a fundamental component decision-making, with noise playing key role as the source variability errors. However, origins this have never been determined. We developed decision-making tasks in which sensory delivered randomly timed pulses, analyzed resulting data models that use richly detailed information each trial’s pulse timing to distinguish between different mechanisms. This analysis allowed measurement magnitude accumulator’s memory, separately from associated incoming evidence. In our tasks, memory was noiseless, for both rats humans. contrast, addition new primary variability. suggest task modeling approach powerful method revealing internal properties processes.