We all have memories we’d rather forget. Yet too often they bubble up into our consciousness. That gaffe at work or during an interview? A faceplant after slipping on ice on a first date? An accidental reply-all to the whole family? (Cringe).

For most, a quick jab of embarrassment, anger, or fear is all we feel and it quickly dissipates. But for people with post-traumatic disorders (PTSD) or depression, unwanted memories from their trauma can seriously derail their lives.

So how is it that these memories only sometimes invade unsuspecting minds?

A new study in the Journal of Neuroscience has some answers. By scanning the brains of 24 people actively suppressing a particular memory, the team found a neural circuit that detects, inhibits, and eventually erodes intrusive memories.

A trio of brain structures makes up this alarm system. At the heart is the dACC (for “dorsal anterior cingulate cortex”), a scarf-like structure that wraps around deeper brain regions near the forehead. It acts like an intelligence agency: it monitors neural circuits for intrusive memories, and upon discovery, alerts the “executive” region of the brain. The executive then sends out an abort signal to the brain’s memory center, the hippocampus. Like an emergency stop button, this stops the hippocampus from retrieving the memory.

The entire process happens below our consciousness, suppressing unwanted memories so that they never surface to awareness.

But what happens if memories do break into our thoughts? Here, the dACC has another task. When proactive surveillance fails, the brain region increases its alert signal to the executive—think DEFCON1—probing it to further damp down activity in the hippocampus.

“Preventing unwanted memories from coming to mind is an adaptive ability of humans,” wrote the authors, led by Dr. Michael C. Anderson at the University of Cambridge and Dr. Xu Lei at Southwest University in Chongqing, China.

Meet the Dynamic Trio

The three brain regions are familiar to memory researchers. Each functions like a government agency in a spy novel, with multiple tasks and vast intercommunications. Retrieving—or dampening down—a memory is similar to an intelligence operation.

The hippocampus is the boots-on-the-ground “operative” for fishing out a memory from neural networks. Buried deep inside the brain, the structure encodes, temporarily stores, and retrieves memories that capture the stories of our lives—the whens, wheres, and whats.

Another player is the brain’s “command center,” the prefrontal cortex (PFC). Through vast neural networks to different brain regions, including the hippocampus, this “executive” monitors the brain’s operations and is center to cognitive control. If hippocampal actions are out of whack, one part, the rDLPFC, sends an electrical “abort operations” signal and dampens hippocampal activity.

But what is providing the rDLPFC with intelligence?

Meet the enigmatic dACC, a C-shape structure that activates across multiple brain functions. Previous studies using computational modeling suggest that it carefully monitors ongoing neural processes. Like an intelligence agent scanning for signs of potential danger, it captures information “indicating a need to intensify cognitive control,” the authors explained. dACC then relays the demand to the command center, urging the executive to implement control—at least, in non-memory contexts.

The new study asked: does dACC also spy on offending memories?

Brain Scan Tag Team

How do you uncover a neural hub for controlling memory?

The trick is tracking the brain’s activity with multiple types of scans, each capturing unique aspects of brain processing. One is EEG (electroencephalogram), which uses electrodes placed on the scalp to detect brain waves—the cumulative electrical activity of neurons. Like a wide-angle surveillance camera, EEG excels at capturing electrical patterns across relatively large areas of the brain in real time, but sacrifices resolution.

fMRI is the perfect buddy-cop. Compared to EEG it’s slow to react, but offers far higher resolution. Using the two methods simultaneously provides the best of both worlds, allowing the team to peek into neural activity changes like an IMAX movie.

After obtaining the data, they can match up precise time stamps of activity changes—which they get from EEG—to their precise location on fMRI scans.

For the study, they recruited 24 volunteers, evenly split between male and female, with no history of neurological troubles or mental health issues. The volunteers then learned 68 word pairs. For example, “gate” paired with “train;” “lawn” with “beef.” One word in each pair was used as a cue; when asked, participants would try their best to remember the associated word.

Next, the volunteers went into the fMRI scanner. For some trials, after being presented with the cue—say, “gate”—they were instructed to recall the associated word, “train.” In other trials, they had to actively not think of the answer. Aptly named, the test is called Think/No-Think, or TNT, paradigm.

During the task, the team tracked and analyzed interactions among the trio of brain regions using EEG and fMRI. Finding patterns in neural network activity, they then zoned in on two specific brain wave signatures (theta power and N2 amplitude) in the dACC, which is often associated with cognitive control.

A Two-Step Dance

The dACC activity came in two bursts.

The first sparked at roughly 400 milliseconds, about the blink of an eye, and generally before a memory enters consciousness. dACC relayed information to the commander rDLPFC, which in turn ordered the hippocampus to lower its activity and stop retrieving the memory.

We can see this with decreased theta brain waves in the hippocampus, which is necessary to retrieve memories, the authors explained.

Mission complete, the entire neural circuit dampened down during the rest of the test, suggesting the neurons were happy to chill out with a job well done—no need to keep working to inhibit an already-suppressed memory.

In contrast, if the dACC signal didn’t activate in time—for example, if the person remembered the associated word even when trying not to—the region went into high alert. This “reactive alarm” skyrockets activity in the commander, rDLPFC. The region then further smothers theta waves in the hippocampus in an effort to stop intrusive thoughts. People who excelled at actively suppressing the associated word, for example, had far stronger information flow from the rDLPFC command center to the hippocampus for words they forgot, compared to those they remembered despite trying to squash the memory.

Overall, the brain has an internal two-step mechanism, proactive and reactive that helps tamper down intrusive thoughts, explained the authors. Both have dACC as their intelligence agent. When you encounter a reminder, say, the toy of a beloved pet that recently passed away, the dACC detects neural network signals generated by the cue. In two waves, it then either prevents retrieval or pushes the painful memory out of awareness.

For now, the study is limited to visual cues. Further ones will have to see if other powerful cues—like hearing the voice or smelling the perfume of a deceased loved one—also triggers dACC. But for now, we’ve found a built-in guardian angel in the brain that can “clear the mind from unwanted thoughts and hasten the demise of memories we would prefer not to have,” the authors wrote.

Image Credit: coffeeNwaffle / 7 images

Shelly Xuelai Fan is a neuroscientist-turned-science writer. She completed her PhD in neuroscience at the University of British Columbia, where she developed novel treatments for neurodegeneration. While studying biological brains, she became fascinated with AI and all things biotech. Following graduation, she moved to UCSF to study blood-based factors that rejuvenate aged brains. She is the ...

Follow Shelly: