The problem of stability has long been a limiting factor in developing neural networks that can grow in size and complexity. Outside of particular, narrow parameter ranges, changes in activity can easily result in total loss of control. Human cognition must have reliable means of acting to stay within the stable ranges of sensitivity and activation. Learning is one such mechanism, and population dynamics are another. Here, we focus on another, often overlooked stability mechanism: cellular homeostasis through metabolism dynamics. We ran a visual change detection experiment designed to strain network stability while minimizing any learnable patterns. We fit the data using models with and without cellular energy levels as a factor, finding that the model influenced by its past history of energy use was a closer fit to the human data.