sensesworth

Sensesworth is a conceptual metric used in cognitive science and artificial intelligence to quantify the value of a sensory input in driving perception, inference, or action. It represents the expected contribution of a given sensory channel to task performance under specific conditions, often expressed as a unitless weight or score.

Origin and etymology: the term blends senses and worth and has appeared in theoretical discussions and some modeling papers since the 2010s. It is not an officially standardized unit, but rather a flexible concept used to compare the relative importance of different sensory signals within a model or experiment.

Applications: in multimodal perception and sensor fusion, sensesworth guides the weighting of inputs from vision, audition, touch, and other channels, especially under uncertainty. In robotics and AI, it informs active sensing strategies and resource allocation by prioritizing channels with higher sensesworth. In psychology, it serves as a descriptive proxy for cue weighting in perceptual decision-making experiments.

Limitations and criticism: the measure is context-dependent and difficult to compare across tasks or individuals. Because it is typically derived from a model, there is a risk of circular reasoning if the same data are used to estimate and validate the metric. There is no universal definition or units, which can hinder cross-study comparability.

See also: cue weighting, sensor fusion, salience, information value, decision theory.