All mouse experiments were approved by the Institutional Animal Care and Use Committee at UT Southwestern Medical Center (Protocol #2013-0035 and #2016-10376-G).

Bmal1 KO mice are commonly used for studying the functional significance of the circadian rhythm because this single gene knockout can disrupt molecular circadian oscillations entirely. However, Bmal1 KO mice have a significantly reduced lifespan and experience premature aging. Therefore, they are inappropriate for studies that require long-term behavioral recordings. Although Per1/2/3 KO mice have disrupted molecular circadian rhythms, no notable health issues have been reported, hence their use in this study.

In the initial investigation, 8 Per1/2/3 KO mice (4 males, 4 females; 5-8 months old; C57BL/6J or C57BL/6J and C57BL/6N mixed backgrounds) were measured in a variety of conditions. Mice were first exposed to 12-hour light, 12-hour dark cycles for 7 days (LD, light intensity ~450 lux at cage level) and then kept in constant darkness for 11 days (DD). Subsequently, methamphetamine was administered through drinking water for 26 days while they remained in constant darkness (MDD). The mice were then exposed to constant light for 22 days while methamphetamine administration continued (MLL, light intensity ~170 lux at cage level). Subsequently, methamphetamine was removed while still in constant light for 54 days (LL, light intensity ~170 lux at cage level). The final 24 days of the experiment were conducted in constant darkness for 24 days (FDD). This group is referred to as "Multiple conditions" in the dataset

To better understand the multiscale behavioral changes following exposure to methamphetamine across elongated measurement intervals, five additional cohorts of mice were investigated. A cohort of heterozygous PER2:luciferase knockin mice (7 mice) and one wild-type littermate (1 mouse) were used as a comparison point (5 males, 3 females; 1.3-9.5 months old; C57BL/6J background). These mice were kept in constant darkness for 95 days without methamphetamine. General cage activity was recorded with an infrared motion detector in the cage without a running wheel for the first 10 days, then in a cage containing a locked running wheel for 11 days. After that, the running wheel was unlocked and both general activity and wheel running activity were recorded for 20 days. The wheel was then locked once again for a further 21 days before being unlocked for the final 30 days. Here, the last 30 days of running wheel activity were analysed. This group is referred to as "Control" in the dataset.

Heterozygous PER2:luciferase knockin mice were used as the control group due to their elongated recordings in constant darkness, which enabled coupling analysis. To confirm that the power results observed in these knockin mice were consistent with those in pure wild-type cohorts, two groups of wild-type mice recorded over shorter intervals were analysed.

In one of these experiments, 5 wild-type male mice (5 weeks old at the beginning of the experiment; strain C57BL/6N) were included. Mice were first exposed to 12-hour light, 12-hour dark cycles for 7 days. Subsequently, they were kept in constant darkness for the duration of the recordings. On the twenty-second day, the mice were moved into cages containing a locked wheel while remaining under DD. On day 33, the running wheels were unlocked, and the running wheel activity was measured for the subsequent 20 days until they were once again locked. This procedure resulted in a 20-day recording of running wheel activity in wild-type mice in DD. The time-localised powers of these mice are summarised in the Supporting Information. This group is referred to as "wild-type 20 days."

Additionally, 6 male mice were analysed; all were between 7 and 8 weeks old at the beginning of the experiment (strain: C57BL/6J). The mice were exposed to 12-hour light/dark cycles for seven days. A constant darkness protocol was then initiated. After 21 days, the mice were switched to a cage with a wheel to measure movement. They remained in this cage for another 21 days in constant darkness, and the wheel allowed their movement to be tracked during this interval. The time-localised power for each of the six mice measured under these conditions is shown in the Supporting Information. This group is referred to as "wild-type 21 days."

A cohort of Per1/2/3 KO mice (3 males, 4 females; 3.5-8.5 months old; C57BL/6J background with cfos-shGFP transgene) were initially exposed to constant light (~220 lux at cage level) for 26 days, then kept in constant darkness for 65 days without methamphetamine exposure. A male mouse was excluded as the entire recording length was not completed. The 65 days of running wheel activity in constant darkness were analysed in this manuscript. This group is referred to as "Per1/2/3 KO DD."

Another cohort of Per1/2/3 KO mice (1 male, 4 females; 4.5-5.5 months old; C57BL/6J or C57BL/6J and C57BL/6N mixed background) were kept in DD for 27 days without methamphetamine before being exposed to methamphetamine for the subsequent 101 days. The first 65 days of activity during methamphetamine administration were analysed and are referred to as "Per1/2/3 KO DD MA."

Each mouse was housed individually in a plastic cage (length × width × height: 29.5 × 11.5 × 12.0 cm) containing running wheels with a diameter of 11 cm. Wheel revolutions were continuously recorded every minute by the ClockLab system (Actimetrics, Wilmette, IL, USA). As described above, general cage activity was monitored with a passive infrared sensor placed above the cages without a running wheel or with locked running wheels; however, those data were not analysed in the current study. The cages were placed in light-tight ventilated cabinets, and the temperature, humidity, and light intensity inside the cabinet were recorded every 5 minutes by Chamber Controller software (Actimetrics, Wilmette, IL, USA). The white LEDs inside the cabinet were controlled by the Chamber Controller software. Cages and water bottles were changed once every three weeks. An infrared viewer was used to perform maintenance in the dark without exposing mice to visible light. For methamphetamine administration, water bottles were replaced with drinking (tap) water containing 0.005% methamphetamine. Water bottles containing methamphetamine were changed once every three weeks. During the experiment, mice had ad libitum access to food and regular or methamphetamine water.