How different brain regions contribute to memory for visual objects

Visual object memory refers to our brain’s ability to store, recognize, and recall visual information about the objects we perceive. This ability is essential for interacting with the world, influencing learning, problem solving, navigation, and social interactions. Without effective visual object memory, these activities would be nearly impossible. Therefore, many neuroscientists have been determined to uncover the mechanisms behind this essential aspect of cognition in both humans and animals.

Numerous studies of primates engaged in memory tasks have established that the anterior ventral temporal cortex (aVTC) is crucial for visual object memory. Neurons in this region can represent complex visual objects, suggesting that they may function in visual object memory even without direct visual input, relying instead on regulatory signals from higher cognitive areas. Despite this understanding, the specifics of this “top-down” regulation and the broader functional network that includes the aVTC remain unclear.

In an attempt to answer these questions, a Japanese research team conducted an in-depth investigation to shed light on this elusive problem. Led by Toshiyuki Hirabayashi, a senior research scientist at the National Institutes for Quantum Science and Technology (QST), they conducted various types of experiments on macaques performing visual memory tasks. Their latest paper was published in Nature Communications on July 10, 2024 and was co-authored by Takafumi Minamimoto of the Advanced Neuroimaging Center, QST.

The researchers first performed functional positron emission tomography scans on macaques during a visual recall task, which allowed them to pinpoint the most active brain regions by measuring minute changes in blood flow. They combined these measurements with functional magnetic resonance imaging data, previously taken on a large population of macaques, which quantified connectivity between different brain areas. They identified specific nodes within the aVTC cortex and orbitofrontal cortex (OFC) as key members of the network that governs memory for visual objects.

To support these results, they conducted chemogenetic silencing experiments. In other words, they genetically modified the macaques’ OFC using a viral vector to introduce custom-designed receptors into the neurons. These receptors prevent the neurons from firing, but only in the presence of a very specific designer drug. The team observed that the monkeys performed significantly worse on visual recall tasks when the OFC was chemically silenced, which did not impair their visual perception in any way.

However, the researchers wanted to take their analysis a step further and so explored the fine-scale details governing memory for visual objects in the aVTC and OFC.

“The identification at the macroscopic scale of brain network nodes and the subsequent understanding at the microscopic and cellular scale of the flow of causal information along the identified nodes are necessary for a complete understanding of the network mechanisms underlying object memory,” notes Hirabayashi.

To this end, they performed recordings from individual neurons in the aVTC of the same macaques used in the previous experiments, assessing memory-related activity and higher-order modulation in these neurons. They found that memory-related activity of individual aVTC neurons, but not perception-related activity, was specifically attenuated by OFC silencing. This was consistent with previously obtained behavioral results.

Moreover, similar changes in neuronal activity occurred when monkeys made a mnemonic error in the task preceding OFC silencing, suggesting the behavioral relevance of memory-related activity in individual aVTC neurons, which was supported by top-down inputs from the OFC.

These analyses allowed the team to better understand the mechanisms underlying visual short-term memory in primates. Since our brains share many functional and structural features with those of these animals, the results of this study may also help us better understand ourselves. It is worth noting that this could have important implications for medicine.

As Hirabayashi explains, “the network mechanisms discovered in non-human primates could help us understand the mechanisms of memory deficits associated with human dementia.” He adds, “Artificial neuromodulation of the network currently discovered in dementia patients could restore their visual memory functions.”

Provided by the National Institutes for Quantum Science and Technology