Scientists at Rockefeller University report on studies through which they were able to visualize and track the activity of individual neurons in the entire brains of zebrafish larvae as the animals repeatedly made “turn left or turn right” choices in a behavioral experiment. The resulting frame-by-frame view of each decision in the making was detailed to the point that, even 10 seconds before the fish made their “choice,” the researchers could predict what the animal’s next move would be, and when they would execute it.
The findings surprisingly highlighted a key role for the cerebellum—a region recognized for its role in coordination, balance, and fine-tuning of movements—in shaping up the animals’ decisions. “This was surprising,” said research lead Alipasha Vaziri, PhD, head of the Laboratory of Neurotechnology and Biophysics, adding that a few studies in recent years have pointed in the same direction. “I think we might find more generally that the cerebellum is involved in more cognitive brain functions than what we have traditionally thought.”
Vaziri and colleagues report their findings in Cell, in a paper titled, “Cerebellar Neurodynamics Predict Decision Timing and Outcome on a Single-Trial Level.”
Animals interact with their environment in a goal-directed manner, and through complex behaviors that are refined over time, the authors explained. Carrying out such behaviors requires motor planning that comes before the actual activity. So while many goal-directed decisions that we make, whether they be to turn a light on or off as we enter and leave a room, or open a kitchen cupboard to take a cup or plate, may appear to be “automatic” and require little thought, these seemingly insignificant actions involve the activity of millions of neurons and complex interactions between different brain regions. It’s a dynamic process that is too complex to observe fully in real-time even in simple organisms, and “how these regions and their interactions with brain-wide activity drive action selection is less understood,” the investigators noted.
Scientists have traditionally had two options for study. Either closely observe how subsets of neurons fire, which limits and narrows the view of the whole picture, or look at the whole brain activity while averaging the data over multiple trials, to reduce noise. Averaging, however, leads to loss of some of the details. As the team stated, “ … obtaining high spatiotemporal resolution access to whole-brain neuroactivity is currently not possible in mammalian brains, while studying complex goal-directed behaviors, motor planning, and decision making in non-vertebrates is challenging, and results may not be directly transferrable to vertebrates.”
“We wanted to understand how decisions unfold on a trial-by-trial basis,” said Vaziri. Zebrafish represent a useful model for investigating brain function. The organisms’ brains have a high level of physiological and genetic homology to the human brain, and conform to basic vertebrate brain organization. Zebrafish larvae are also transparent, which makes it possible to harness sophisticated microscopy technology for carrying out whole-brain in vivo optical recording of neuroactivity at high speed and at cellular resolution. For their newly reported study, the team paired advanced statistical methods with their recently developed imaging technique, light field microscopy (LFM) to simultaneously track the activity of every neuron in the brains of zebrafish larvae as they made decisions. But first, they had to teach the organisms a new behavior, and one that was goal-oriented, not simply a reflexive reaction.
They explained, “In the present study, we have combined high-speed volumetric calcium imaging based on light-field microscopy (LFM)—a technique that uniquely enables simultaneous capturing of neuronal network activity on the scale of the whole-brain at single-cell resolution—with a new operant conditioning paradigm in larval zebrafish, relief of aversive stimulus by turn (ROAST), to investigate the neuronal basis of goal-directed action selection.”
The goal of this behavioral trial, from the fish perspective, was to get relief from heat. The researchers slightly warmed the water around the fish using a laser, and switched off the laser only when the fish made a tail movement to the right. After about 15 repeats, the fish had mastered the trick. They responded to their warming surroundings about 20 seconds after the laser came on. About 80% of the time, the fish remembered to flip their tail in the correct direction. To avoid any direction bias, the whole experiment was also repeated by teaching the fish to turn their tails to the left.
During the period between switching on the laser and the zebrafish larva making a responsive movement, the researchers tracked the activity state of about 5,000 of the most active neurons across the animal’s entire brain. They then identified which activity patterns reflected the brain sensing the heat or moving the tail, and which appeared decision related. Noticeably, they observed that about 10 seconds before a fish made a movement, its neuronal activity pattern differed markedly dependent on whether it was going to make a correct or an incorrect turn. With this data in hand, the researchers could then look at the brain state of any larva, and 80% of the time surmise correctly what it was about to do. The team was even able to predict the specific time at which the animals would initiate the turn, as well as its direction, in each trial.
Having identified which clusters of neurons corresponded to different aspects of the task, the researchers then mapped the neurons to their anatomical regions. “This allowed us to see what brain regions were involved in what aspects of the task as the decision unfolded in each trial,” Vaziri said. While several brain regions participated in transforming the sensory information into decision and action, one region, the cerebellum, stood out. Further analyses showed that rate of activity of neurons in the cerebellum determined the exact timing of the tail movement.
The results showed that there was a lopsided activity in the two hemispheres of the cerebellum, starting from heat onset and gradually increasing until the fish moved its tail, and this predicted the direction in which the fish was about to make a turn. This discovery pointed to a key role for the cerebellum in decision making. “The cerebellum also exhibited a surprisingly strong level of preparatory neuroactivity, which we found to mainly originate from the granule cell population, and which allowed us to predict both turn direction and the time point of turn initiation ~10 seconds before movement for individual trials,” the investigators stated.
The team suggests that their work extends the traditional view of the cerebellum in motor control, and provides support for it playing a role in more cognitive functions, “while expanding the existing frameworks of ‘ramping-to-threshold’ for decision making to the level of individual trials that has not been possible due to technical limitations.”