actions occur automatically through the spreading activation of neural signals from sensors and affect systems to motors.
actions are events, or changes over time within an environment, and are therefore detectable by the senses via signals produced when contrasts are measured over space and time by the sensory system.
affective responses to either internal or external events yield signals that spread to neurons. these signals prompt and/or inhibit the activation of neurons specifically related to motor apparatuses and therefore act as a motivational force behind all goal orientation.
additionally, certain affective signals result in the strengthening of synaptic connections between neurons based on recent firing patterns, enabling a learning process that is reactive to events experienced in the affective domain.
the automatic association of actions and the changes they bring about in affective and sensory domains result in a switch from undirected action to inference-based directed responses to experienced events, as predictions with respect to actions and the effect they are likely to have on the current environment lead to simulations that are internally experienced and bring about an awareness of affordances, or opportunities to perform actions that trigger events which change the current state of the environment to some alternative, provide viable paths to potentially more favorable states which when converted to goals are used to guide actions toward targets.
the interactions between sensory, affective, and motor signals create a feedback loop resulting in an adaptive and self-regulating neural system. the system is embodied and requires an external environment in which it can sense and act in order to function, as well as a set of if-then-like rules for yielding affect signals in response to certain events. these rules produce the neuromodulatory functions that guide attention and learning within the system.
Building Intelligence 