Professor Hiroshi Nomura and his team at Nagoya City University have solved one of memory's most frustrating mysteries: why the same memory can feel crystal clear one moment and maddeningly inaccessible the next, even though the information itself is safely stored in the brain.
The research, published in the journal Neuron, reveals that slow fluctuations in brain histamine neurons act like a gating mechanism, controlling whether a memory becomes accessible when needed. This finding reframes how we think about forgetting—it may not always be about losing information, but rather about the brain being in a state where stored memories are temporarily locked away.
The team focused on histamine neurons in the tuberomammillary nucleus of the hypothalamus, a region already known for regulating wakefulness. What Nomura and colleagues discovered is that these neurons project widely to memory-related brain areas including the cortex, hippocampus, and amygdala. During their experiments with awake mice, they recorded these neurons' activity and found it rose and fell slowly over tens of seconds, synchronized with changes in cortical activity, pupil size, and facial movement—signs that histamine activity reflects a broader state of the entire brain and body.
The researchers trained mice to associate a sound with a sugar-water reward, then measured how reliably they would lick in response to the sound cue. The critical finding came when they compared histamine activity before successful memory recall with activity before failed recall: histamine neuron activity was consistently higher just before trials where mice showed strong memory-guided licking responses, and lower before trials with no licking at all.
To test whether this correlation actually meant causation, the team went further with two decisive experiments. First, they used a real-time monitoring system to deliver memory cues precisely during either high- or low-histamine activity states. The difference was striking: memory-guided licking responses were about 40% higher when the sound appeared during high-histamine states compared to low-histamine states. Second, they used optogenetics to directly manipulate the neurons—suppressing them before the cue reduced memory-guided licking, while activating them increased it. Importantly, these manipulations didn't alter general licking behavior, responses to the reward itself, auditory sensitivity, or arousal, meaning the effect was specifically about memory access rather than broader changes in the animal's behavior or alertness.
The researchers traced the downstream effects to the basolateral amygdala, which plays a crucial role in learned reward associations. When mice successfully expressed their learned memory, populations of amygdala neurons more reliably reproduced the neural activity pattern associated with the cue. But when histamine neurons were suppressed beforehand, that memory-related pattern became weaker and less reliable—as if the neural signature of the memory had been dimmed.
These findings support what Nomura calls a "priming-state" model: spontaneous fluctuations in histamine activity prepare memory circuits in advance, making incoming cues either more or less likely to trigger the neural patterns needed for recall. Rather than retrieving memory being a simple process of reading out a stored trace, the brain's internal state acts as a switch that opens or closes access to information that's definitely there.
The work opens new questions. The study examined reward memories in mice, so researchers will need to determine whether this histamine gating mechanism shapes other types of memory in humans and other animals. But for anyone who has experienced that tantalizing tip-of-the-tongue moment, the message is clear: sometimes the memory isn't lost. It's just temporarily out of reach.