Why do we feel tired after having vivid dreams?

To study how the brain distributes energy during different phases of sleep, the researchers developed a method to observe the mouse brain in its natural environment. They used a UV-curable resin to make the skull transparent, and then applied wide-field fluorescence imaging. This approach allowed them to simultaneously measure the blood volume in specific regions of the brain as a marker of fuel intake, the level of neuronal ATP as a direct source of energy for neurons, and the level of astrocyte pyruvate, a key molecule that links glucose from the blood to the brain’s energy metabolism.

During slow-wave (NREM) sleep, delta activity dominates in the brain, but theta oscillations are also noticeable. It has been found that these theta oscillations predict changes in local blood volume approximately 4 seconds later. This suggests that blood flow is flexibly adjusted to match the current neural activity, as the vessels respond quickly to metabolic demands, providing energy on demand.

The transition to REM sleep is accompanied by a significant shift in the energy regime. Approximately 50 seconds before the formal onset of REM sleep, the volume of blood in the brain begins to increase. This increase begins in the posterior cortex and then spreads forward over a period of approximately 15 seconds, indicating a pre-emptive process of metabolic preparation for the upcoming energy-intensive phase.

After the onset of REM sleep, the level of astrocyte pyruvate increases, which is consistent with increased glycolytic activity in astrocytes and increased availability of substrates for energy metabolism. At the same time, contrary to expectations, the level of neuronal ATP decreases even as blood flow increases. This creates a paradoxical situation: while energy intake increases, the amount of ATP in the neurons decreases.

Scientists have proposed several explanations for this phenomenon. It is likely that during REM sleep, neurons expend large amounts of ATP on memory-related processes, such as synaptic reorganization, communication between the hippocampus and the cortex, and extensive network switching. Another possible mechanism is a change in the way energy is transferred from astrocytes to neurons, or a shift in the way mitochondria work, which redistributes resources between different metabolic pathways. Thus, energy is not simply stored, but dynamically redistributed between vessels, astrocytes, and neurons, depending on the current tasks of the brain.

This study reveals an important principle of biological information processing. Unlike a conventional computer, the brain is forced to operate under strict metabolic constraints. Instead of evenly distributing energy, it flexibly redirects it based on its state, memory needs, and internal tasks. According to lead researcher Yusuke Takahashi, understanding how the brain balances energy inflow and outflow can explain why biological intelligence is so efficient. REM sleep serves as a natural example of how the brain reconfigures its energy economy to support complex internal processing, particularly memory consolidation.

These findings provide a new perspective on why people sometimes feel tired after a night of vivid dreams, as the intense brain activity during REM sleep requires significant energy resources. Additionally, these results contribute to our understanding of the fundamental mechanisms underlying brain function, highlighting the intricate and adaptive ways in which the nervous system manages limited energy reserves to support essential cognitive processes.