Fractal scaling in animal behavioral activity, where similar temporal patterns appear repeatedly over a series of magnifications among time scales, governs the complex behavior of various animal species and, in humans, can be altered by neurodegenerative diseases and aging. However, the mechanism underlying fractal scaling remains unknown. Here, we cultured C. elegans in a microfluidic device for 3 days and analyzed temporal patterns of C. elegans activity by fractal analyses. The residence-time distribution of C. elegans behaviors shared a common feature with those of human and mice. Specifically, the residence-time power-law distribution of the active state changed to an exponential-like decline at a longer time scale, whereas the inactive state followed a power-law distribution. An exponential-like decline appeared with nutrient supply in wild-type animals, whereas this decline disappeared in insulin-signaling-defective daf-2 and daf-16 mutants. The absolute value of the power-law exponent of the inactive state distribution increased with nutrient supply in wild-type animals, whereas the value decreased in daf-2 and daf-16 mutants. We conclude that insulin signaling differentially affects mechanisms that determine the residence time in active and inactive states in C. elegans behavior. In humans, diabetes mellitus, which is caused by defects in insulin signaling, is associated with mood disorders that affect daily behavioral activities. We hypothesize that comorbid behavioral defects in patients with diabetes may be attributed to altered fractal scaling of human behavior.

Fractal scaling governs the complex behavior of various animal species and, in humans, can be altered by neurodegenerative diseases and aging1. However, the mechanism underlying fractal scaling remains unknown. Here, we videorecorded C. elegans that had been cultured in a microfluidic device for 3 days and analyzed temporal patterns of C. elegans actions by fractal analyses. The residence-time distribution of C. elegans shared a common feature with those of human and mice2–4. Specifically, the residence-time power-law distribution of the active state changed to an exponential-like decline at a longer time scale, whereas this change did not occur in the inactive state. The exponential-like decline disappeared in starved C. elegans but was restored by culturing animals with glucose. The exponential-like decline similarly disappeared in insulin-signaling daf-2 and daf-16 mutants. Therefore, we conclude that insulin signaling regulates fractal scaling of C. elegans behavior. Our findings indicate that neurosensory modulation of C. elegans behavior by insulin signaling is achieved by regulation of fractal scaling. In humans, diabetes mellitus is associated with depression, bipolar disorder, and anxiety disorder5, which affect daily behavioral activities. We hypothesize that comorbid behavioral defects in patients with diabetes may be attributed to altered fractal scaling of human behavior.

Fractal scaling governs the complex behavior of various animal species and, in humans, can be altered by neurodegenerative diseases and aging1. However, the mechanism underlying fractal scaling remains unknown. Here, we videorecorded C. elegans that had been cultured in a microfluidic device for 3 days and analyzed temporal patterns of C. elegans actions by fractal analyses. The residence-time distribution of C. elegans shared a common feature with those of human and mice2–4. Specifically, the residence-time power-law distribution of the active state changed to an exponential-like decline at a longer time scale, whereas this change did not occur in the inactive state. The exponential-like decline disappeared in starved C. elegans but was restored by culturing animals with glucose. The exponential-like decline similarly disappeared in insulin-signaling daf-2 and daf-16 mutants. Therefore, we conclude that insulin signaling regulates fractal scaling of C. elegans behavior. Our findings indicate that neurosensory modulation of C. elegans behavior by insulin signaling is achieved by regulation of fractal scaling. In humans, diabetes mellitus is associated with depression, bipolar disorder, and anxiety disorder5, which affect daily behavioral activities. We hypothesize that comorbid behavioral defects in patients with diabetes may be attributed to altered fractal scaling of human behavior.