A45 Sensory overload

Topic

An insect's nervous system has a limited capacity to process the environmental stimuli reaching it at any given moment; sensory overload occurs precisely when the combined intensity of these environmental stimuli exceeds the nervous system's processing capacity.

Each of the insect's sensory channels—sight, hearing, smell, magnetoreception—has a maximum limit on the amount of information it can process per unit of time, as established by the Shannon-Hartley theorem regarding communication channels. When the volume of information arriving via a given channel exceeds this maximum limit, the sensory channel collapses, rendering the insect unable to distinguish the signals that truly matter from the surrounding background noise.

This collapse also has a direct neuronal cause: the insect's sensory neurons respond to stimulus intensity according to a power law known as Stevens's power law—a relationship that saturates in the face of very intense light or sound stimuli. Once the neuronal response reaches this saturation point, the insect can no longer discriminate between two equally intense stimuli, even if they are actually quite distinct from one another.

The insect's ability to discriminate a real signal from background noise can be formalized using signal detection theory, which depends on two simultaneous factors: the variability of environmental noise from moment to moment, and the average separation between the typical intensity of the real signal and that of the noise. When environmental noise becomes more variable, or when the separation between signal and noise narrows, the insect's discrimination ability declines, and the error rate—confusing signal with noise—rises increasingly sharply.

In a highly disturbed environment, it is not merely a single sensory channel—a single modality such as sight or hearing—that becomes saturated in this way; rather, multiple sensory modalities of the insect can become saturated simultaneously. Under normal conditions, the insect's nervous system optimally combines information from its various sensory modalities, and this combination reduces uncertainty compared to what any single sense would provide on its own. However, when multiple modalities become saturated simultaneously, the errors associated with each cease to be independent and become correlated; this correlation results in a combined estimate that is ultimately more uncertain than one derived from a single sense alone.

ID:2562