Dissecting the contribution of recent reward versus recent performance history on cognitive effort allocation

Decision-making not only involves choosing what action to take but also the meta-decision of how much of our limited resources, including cognitive effort, to invest. Therefore, we must weigh the costs of cognitive effort against potential rewards. This cost-benefit trade-off can be operationalised by the opportunity costs of the environment, approximated by its reward rate. Here, we replicate the finding that people exert more cognitive effort when reward rate is low in a classic cognitive control (Simon) task. However, we provide an important nuance to this result. Cognitive effort allocation was better explained by participants’ recent performance history rather than reward history. Furthermore, we found that participants were not sensitive to the reward at stake in this task design, which may explain the finding that long-range reward magnitude explained behaviour less well. These results emphasise that, when investigating the role of environmental reward rate in cognitive effort allocation, it is crucial to ensure participants are sensitive to the reward at stake. This data collection consists of the task code, raw and unprocessed data, and analysis scripts used for this paper.