Cognitive Stress and Learning Economic Order Quantity Inventory Management: An Experimental Investigation

Abstract

We use laboratory experiments to evaluate the effects of cognitive stress on inventory management decisions in a finite horizon economic order quantity (EOQ) model. We manipulate two sources of cognitive stress. First, we vary individuals’ participation in a 6-digit mixed number and letter memorization task involving their personal identification numbers. This exogenously increases cognitive load. Second, we introduce an intervention to reduce cognitive stress by only allowing participants to order when inventory is depleted. This restricts the order choice set. Increases in cognitive load negatively impact earnings with and without the intervention, with these impacts largely occurring in the first attempt of the task. With repetition, participants’ choices in all treatments trend to near-optimal policy adoption. However, only in the intervention and low cognitive load treatment do the majority of choices reach the optimal policy. We estimate the learning dynamics of order decisions using a Markov learning model. Estimates suggest increased cognitive load reduces the probability of switching to more profitable policies. Choice set complexity increases biases for smaller order size adjustments, leading to greater policy lock-in.