Perhaps the biggest puzzle researchers faced, once they had verified that the inner clock’s cycle differed from the earth’s rotational cycle, was how exposure to light and dark captured the free-running rhythm. Showing that a cycle of 12 hours of light and 12 hours of darkness entrained circadian rhythms to exactly 24 hours was not very informative.
Clever experiments presented brief periods of light, against a background of darkness, at various points in the free-running rhythm, with the stunning result that the rhythm would shift earlier, or later, or not shift at all, at specific points in the internal oscillation. For hamsters, for example, a light pulse would shift the rhythm later around the time that running-wheel activity was on the increase—early in the night for this nocturnal animal. And hours later, a light pulse would shift the rhythm earlier, as the daytime quiet period approached. But “day” and “night” did not correspond to clock time in the external world: rather “circadian time” was the reference point, defined by the animal’s pattern of free-running nocturnal activity.
Humans, of course, are awake and active mostly during the day, when it is light outdoors. Even so, when research subjects were isolated from day-night cues, a light pulse would shift their rhythm later at the start of their circadian night, and earlier near the end of their circadian night, just like night-active hamsters. The latest thinking is that light in the first half of the day speeds up the clock, while light in the second half of the day slows it down. The effect is strongest during late evening and the first part of the night, late in the night through early morning. However, the inner clock is measuring light input continuously. For a given day, the sum of speeding up and slowing down determines whether the rhythm-as-a-whole will shift later or earlier. This detailed knowledge is key to designing light therapy schedules for people whose rhythms have fallen out of sync with local time.