Visualizations are an important part of computational modeling, and are often the way that models are evaluated. Choosing an appropriate visual representation can surface emergent behavior of a computational model that would be difficult to otherwise infer.

Results of a model of human memory (Kittur et al., 2004) showing a reduction in decay over time (in the negative space between traces rather than the traces themselves).

 

Visualization of 10,000 runs of the memory model using randomly generated vectors; an easy way to make sense of the variance and reliability inherent in the model.