McGill brain study finds fear blurs the line between danger and safety in mice

The work, led by researchers at McGill's Douglas Research Centre and the University of Toronto-Scarborough, used augmented reality with real-time brain imaging to watch how mice stored memories of threatening and harmless environments. ScienceDaily reported the findings on 3 November.

Trained to link one setting with danger and another with safety, the animals did not build increasingly distinct neural patterns for the two. Their hippocampal activity began to overlap instead. The researchers argue that response may be useful: retrieving threat information faster when the surroundings are ambiguous.

"Our findings suggest fear may blur the boundary between danger and safety rather than sharpen it," said senior author Mark Brandon, an associate professor in McGill's Department of Psychiatry and a researcher at the Douglas, according to ScienceDaily. "This could reshape how scientists think about anxiety and PTSD."

Teleporting mice between rooms

The setup is the part that stands out. The team built an arena in which sights and sounds could shift the animals between virtual environments without moving them physically, while miniature microscopes tracked hundreds of neurons in the hippocampus, the region tied to learning and memory. Robert Rozeske, assistant professor in the University of Toronto's Department of Psychology and lead author, said the method offered "a moment-by-moment view of the brain switching between competing maps of dangerous and safe memories," something he said conventional fear conditioning experiments cannot easily provide.

The clinical claim is the one to treat carefully. The study was in mice, and the link to anxiety and PTSD in people is a hypothesis the researchers raise, not a result they demonstrate. ScienceDaily's account does not report human data or a trial.

Other brain work this week

The fear study is one of several neuroscience results in the past week. On 29 October, ScienceDaily reported a McGill-led study in Science Advances that measured beta-band brain activity in 25 people with schizophrenia and 25 healthy controls, finding weaker and slower signals in the patient group. Those with the largest abnormalities also had more trouble judging the timing of sights and sounds and more disorganized speech.

Also on 29 October, ScienceDaily covered a Neurology study of 12,772 adults in Brazil followed for about eight years, which found the heaviest consumers of low- and no-calorie sweeteners had a 62% faster decline in memory and thinking, roughly 1.6 extra years of brain aging. The authors, including Claudia Kimie Suemoto of the University of São Paulo, stressed the result is an association, not proof of cause. On 27 October, Yale's James McPartland and colleagues published work in Nature Mental Health on EEG responses to faces in about 400 autistic children, using all 128 electrodes rather than a handful of sites. Separately, UVA's Roger Abounader described a preclinical glioblastoma approach combining focused ultrasound, nanoparticles and microRNAs, in a ScienceDaily report on 1 November.

Outside the lab, Aspen Neuroscience and Bambusa both raised money in moves Endpoints News reported on 30 October. Google and the Stanford Accelerator for Learning launched "Polar Peaks," an AI literacy quest for students aged 11 to 14 built around connectomics, on 26 October.

Live Science reported on 31 October that a month of twice-weekly VR training helped students at the University of Trento lucid dream, in a study published 28 September in Frontiers in Psychology.

The McGill result does not yet change treatment. It changes what researchers look for in the hippocampus.