Decision making can be infuriating, debilitating and often disappointing in retrospect. And yet, animals of all stripes are constantly faced with critical decisions that determine whether they survive or perish. Many of these decisions are reflexive or instinctive: for example, the decision to jump into a river to escape an approaching train, or to duck to avoid a blow to the head. However, other decisions require more deliberation. In these cases, an animal might take time to accumulate information, which it can use to evaluate prospective options and select the one that seems most likely to produce a favorable outcome. For example, when making an important decision like selecting a mate, it is important to consider all of the available evidence.
A recent study from Carlos Brody's lab, at Princeton University, USA, has combined electrophysiological and behavioral experiments to investigate how the brain integrates evidence to make informed decisions. The decision maker they studied was a humble, thirsty rat. In order to quench his thirst, the rat was asked to play a simple game. The team positioned speakers to the left and right of the rat that randomly emitted gentle clicking sounds. The rat's task was to count the number of clicks from the left and right speakers and then, after a brief pause, poke the button that corresponded to whichever side emitted more clicks. If he answered correctly, a drop of water was released. The goal of the parched rat was simply to maximize the amount of water that he received.
While the rat performed this decision-making game, Hanks and colleagues recorded from neurons in two brain areas: the posterior parietal cortex and the prefrontal cortex. Previous work in monkeys had shown that these brain areas are active during perceptual decision making. Similar to findings in the monkey, the neurons in the parietal and prefrontal cortex increased their firing as the rat gathered sensory evidence during the click task.
The team then used a clever analysis technique to determine the relationship of each neuron's firing rate to the gradual accumulation of evidence. They observed that, at any point during the decision-making process, the activity of neurons in the parietal cortex faithfully encoded the current level of accumulated evidence. However, neurons in the prefrontal cortex had a more binary response, which reflected the actual decision outcome. Thus, although neurons in both areas fired during evidence accumulation, activity in the prefrontal cortex was more predictive of the animal's behavior.
To further investigate the functional role of the prefrontal cortex, the authors used optogenetics to silence prefrontal neurons during different epochs of the click task. They discovered that decisions were affected only if the prefrontal cortex was silenced toward the end of the evidence-accumulation period. This result is consistent with the view that the frontal cortex is not directly involved in evidence accumulation, but instead contributes to decision execution. However, it remains unclear whether the parietal cortex plays a causal role in perceptual decision making. To test this hypothesis, it will be necessary to silence the parietal cortex during evidence accumulation.
For several decades, neuroscientists have recorded from brain areas that might be involved in decision making and used these data to build models of how the brain executes informed judgments. However, because most of these experiments have been performed in monkeys, it has been difficult to casually test these models through targeted manipulation of the underlying circuits. By transiently silencing specific cortical regions during an elegant decision-making task, all the evidence suggests that Hanks and colleagues have made an important step in the right direction toward understanding how the brain controls decision making.
