Choosing the speed of dynamic mental simulations.

The brain continuously maintains a current representation of its immediate surroundings. Perceptual representations are often updated when the world changes, e.g., when we notice an object move. However, we can also update representations internally, without incoming signals from the senses. In other words, we can run internal simulations of dynamic events. This ability is evident during mental object rotation. These uncontroversial observations lead to an obvious question that nevertheless remains to be answered: How does the brain control the speed of dynamic mental simulations? Is there a central rate controller or pacemaker module in the brain that can be temporarily coupled to sensory maps? We can refer to this as the common rate control theory. Alternatively, the primitive intelligence within each map could tune into the speed of recent changes and use this information to keep going after stimuli disappear. We can call this the separate rate control theory. Preliminary evidence from prediction motion experiments supports common rate control, although local predictive mechanisms may cover short gaps of <200ms. The putative rate controller might turn out to be the same thing as the velocity store described in smooth pursuit eye movement models and/or the pacemaker module proposed in the cognitive timing literature. Indirect neuroimaging evidence suggests rate control is a function of the core timing system in the dorsal striatum.