The Neural Correlates of Abrupt Visual Learning

 Post by Megan McCullough

The takeaway

Rapid visual learning, which can be thought of as a moment of insight, is characterized by synchronized neural activity in the inferotemporal and prefrontal cortices.

What's the science?

Although most learning in adult humans requires multiple learning sessions and involves slow changes in the brain, abrupt learning refers to the circumstances in which adults learn after one or only a few exposures to a stimulus. One example of abrupt learning is recognizing a person’s face after one introduction. Previous studies have examined the role of oscillatory synchronization (brainwaves occurring at the same time) in learning over time, but its role in abrupt learning is unknown. This week in Current Biology, Csorba and colleagues aimed to study the role of synchronization of neural activity between the prefrontal (PFC) and inferotemporal (IT) regions in facilitating abrupt learning by recording neuronal activity in non-human primates.

How did they do it?

The authors recorded neuronal activity in the PFC and IT cortex of two adult rhesus macaque monkeys while they participated in an oculomotor foraging task. The task consisted of three phases: the presentation of a scene, a foraging phase, and the reward phase. First, each animal was presented with a natural image. Next, the animals were allowed to explore the image visually. Finally, the animals were rewarded when their gaze reached an unmarked reward zone and the time it took the animals to find the reward zone after being presented with the scene was recorded. This task was chosen because the learning was abrupt, performance improved significantly after only a few trials. To examine the relationship between neural activity in the regions of interest and abrupt learning, the authors measured the relationship between the local field potential (LFP) signals in each area.

What did they find?

The animals learned to recognize the images presented to them and associate them with specific reward areas, showing that this task involved abrupt visual learning. The authors found an increase in synchronization of LFPs in the PFC and IT region around the time the animals had their moment of insight. Furthermore, the synchronized activity in these two brain regions could predict the changes in performance of the monkeys. The data show that the strength of the synchronization was highest around the moment of insight but also carried into the post-learning phases of the task. This coordinated activity appears to link visual inputs with reward outcomes.

What's the impact?

This study uncovered the neural correlates of abrupt visual learning. Because the animals were allowed to freely look at natural images, the results of this study may provide a look into learning that occurs in natural settings outside of a laboratory. This research illustrates the role that coordinated activity between brain regions has in allowing quick visual learning.

Access the original scientific publication here

The Impact of Online Communication at Work on Creativit

Post by Megan McCullough

The takeaway

Videoconferencing as a means of communication inhibits the production of creative ideas. This is because the narrow visual field of those using a digital screen correlates with a narrower cognitive focus.

What's the science?

As a result of the COVID-19 pandemic, there has been a switch to full-time or hybrid remote employment. This shift is projected to outlast the pandemic, with 20% of all U.S workdays estimated to take place remotely even after the pandemic ends. Because of the impact of collaboration on workplace productivity, recent studies have examined the effect of working remotely on the quality and quantity of creative ideas. This week in Nature, Brucks and Levav examined the impact of the physical difference between remote and in-person work on generating and selecting creative ideas.

How did they do it?

The authors conducted two separate studies to examine the productivity differences between online and in-person collaboration. The first study took place in a laboratory setting. The authors randomly paired 602 college undergraduates and randomly assigned them to a virtual or an in-person condition. All pairs were tasked with generating alternative uses for a product and then selecting their most creative task. The authors recorded the number of ideas each pair generated and assigned creativity scores to each idea to measure the decision-making skills of each pair. To test the hypothesis that any differences in creativity between the two groups is due to the effect of screens on narrowing the visual scope of the user, the authors measured the ability of participants to recall props placed around the room and recorded eye gaze during the task. 

The second study involved 1,490 engineers in a realistic work setting. It was conducted to extend the findings of the laboratory study to a more realistic environment. These participants were randomly paired and assigned to one of the two groups. Then they were asked to generate product ideas and select one to submit to their company. As in the first experiment, the authors recorded the number of ideas each pair generated and assigned creativity scores to each idea.

What did they find?

In both the laboratory and field experiments, the authors found that pairs in the virtual groups generated fewer total ideas and fewer creative ideas. There was no statistically significant impact of condition on the ability to select an idea. One hypothesis as to why the difference occurs is that narrowing one’s visual focus to a screen also narrows cognitive focus. The data supported this hypothesis: virtual pairs spent more time looking at their partners and less time looking around the room. The ability to recall the props placed around the room and increased gaze around the room were correlated with an increase in creative ideas.

What's the impact?

This study found that there are differences in the generation of creative ideas between colleagues who collaborate in person compared to those who collaborate through videoconferencing. Videoconferencing groups were less effective at generating creative ideas than their in-person counterparts. This research suggests that there is an advantage to in-person work when it comes to creativity and idea generation. As companies move forward in developing remote work policies after the pandemic, this area of research will become important in the formation of those policies. 

Mental Disengagement From Navigation Degrades Spatial Codes for the Environment

Post by Lani Cupo

The takeaway

When animals navigate an environment, the hippocampus forms a spatial code based on sensory cues and motion. Mentally disengaging from navigation leads to the degradation of place codes, even when physical motion through the environment is still occurring, suggesting that internal state, not just external information, is critical to forming a spatial code.

What's the science?

Previous research in rodent navigation has established the existence of “place cells” — neurons in the hippocampus that respond to specific positions or directions in the external environment that allow animals to navigate. Specific patterns of activity among these neurons can be viewed as a spatial code, however, it is yet unknown how mental engagement impacts the activation of these spatial codes. This week in Nature Neuroscience, Pettit, Yuan, and colleagues investigated the role of mental engagement in activating spatial codes in mice by examining their behavior and neuronal activity during reward-based navigation tasks in a virtual environment. 

How did they do it?

The authors constructed a virtual environment for male adult mice, with each mouse placed on top of a spherical treadmill. The heads of the mice were restrained to allow for concurrent cellular imaging, however the mice could freely rotate the treadmill. Motion was captured with optical sensors and the information was relayed to a projection on a screen in front of the mouse’s head displaying an environment with visual cues that the mice were trained to recognize. Mice could achieve water rewards in certain “reward zones” of the virtual environment by licking a spout in front of them. Cell imaging was achieved with a method known as two-photon microscopy: lasers are shined on cells, and, because the mouse strains express fluorescent calcium indicators in neurons, light is emitted back upon neuronal activation and can be recorded during waking behavior. This allows the researchers to examine when neurons fire, linking neuronal activation with mouse behavior. In this study, the authors quantified the degree to which mice were engaged or disengaged via lick-based metrics, based on how spatially selective and abundant mouse licks were.

What did they find?

While some sessions included almost only engaged trials, other sessions included larger proportions of disengaged trials, and these usually occurred together at the end of sessions, indicating that mice switched from engaged to disengaged behavior. This could indicate satiety as they received about 1 mL of water. During engaged sessions, neural activity formed specific sequences or place codes. However, the activity of the population of cells differed during disengaged trials. Regardless of whether the trial was engaged or disengaged, the mouse was moving through the environment, indicating that the change in neural activity was not due to placement in the environment, but was instead associated with the mouse’s altered behavior reflecting mental disengagement, even when matching trials on variables such as running speed. Furthermore, when considering the activity of all neurons at a population level, there was no difference between engaged and disengaged trials, suggesting a degradation of the spatial code, not just a general alteration in neural activity. Finally, by examining neuronal activity in streaks of engaged and disengaged trials, the authors found that the shift in activity happened in less than a minute.

What's the impact?

This study found that the hippocampal place codes that are associated with rodent navigation of the environment degrade when mice mentally disengage from a goal-directed task. The authors’ findings suggest that beyond sensory cues and motion information, mental engagement is required to establish hippocampal spatial maps of the surrounding environment. These findings challenge the established idea that spatial maps form automatically in rodent hippocampi and demonstrate that internal state impacts neural encoding of the external environment.