In the previous post in my blog series on Pessoa's 2022 book, The Entangled Brain, we began examining subcortical regions and their potential role in the production and regulation of emotions, such as fear and anger. In this post, we round out the chapter by taking a brief look at subcortical involvement in motivation.
The origins of motivation research
What does it mean to study motivation? According to Pessoa, research on motivation involves understanding how animals go about seeking rewards. Here, the first wave of motivation research in neuroscience was sparked by Olds and Milner (1954), who placed electrodes in different brain areas of rats, before placing the rats in a box (see Fig. 1). If the rats happened to hit a lever, this would send an electrical stimulation to their brains, to which the rats would start exploring the box excitedly if they found the stimulation desirable. After pressing the lever (by accident) a couple of times, the rats would cease their search and simply start hitting the lever incessantly.
Brain regions responsible for motivation
One important detail from the above research was that, out of the 76 electrodes placed in random locations in their brains, the rats responded positively to 36 of them, whereas 11 locations elicited aversive reactions to the lever. This suggested that there might be brain regions more important for the production of motivated behaviours. The current consensus brings us to an area in the lower part of the striatum, called the nucleus accumbens (see Fig. 2). Importantly, the accumbens contains neurons that are highly sensitive to the neurotransmitter dopamine, which is widely seen as a compound important for reward-related stimuli.
Fig. 2 The nucleus accumbens, putamen, and caudate (Pessoa, 2022)
At this point, I think it's a good time to go all the way back to Chapter 3, where Pessoa dedicated a small but nontrivial section on neurotransmitters, and more specifically, dopamine. Here, he outlines that while information is transmitted electrically within neurons, this is communicated chemically between neurons via neurotransmitters. While neurotransmitters might be produced in a couple of areas, the extensive nature of brain connections allows neurotransmitters to reach all parts of the nervous system.
Take dopamine as an example. While it is produced and contained in the substantia nigra, many different brain regions have access to it. Relevant to our discussion on motivated behaviours, the network of fibers extending from the substantia nigra not only connects the dopamine-sensitive accumbens, but also reaches other parts of the midbrain (e.g., the ventral tegmental area), striatum (e.g., the caudate and putamen, see Fig. 2), and even the cortex (e.g., the parietal and temporal lobes)! Clearly, the production of motivated behaviour can't be tied to a singular brain region. Instead, involves at the very least a cluster of subcortical areas, and at the most a network of structures spanning the entire brain.
A note on dopamine
While we're on the topic of dopamine, I'd like to point out a very interesting argument from Pessoa. Like many, I've always seen dopamine as somewhat of a reward molecule. After all, self-help books and videos are often quick to point out the role of dopamine in building good habits, achieving feel-good productivity, or even in our battle against doomscrolling. But Pessoa begs to differ and states a few arguments against calling dopamine a reward molecule (these arguments are in chapter 3 of the book!).
For one, neurotransmitters like dopamine do not simply have a singular purpose -- far from only being involved in motivation and rewards, dopamine is also involved in multiple other functions! For instance, the death of dopamine-containing neurons can lead to symptoms of Parkinson's disease, while the overactivation of dopamine can cause psychotic symptoms seen in Schizophrenia. In both cases, the role of dopamine has nothing to do with rewards. And yet, why don't we refer to dopamine as a 'movement' molecule, or a 'psychosis' molecule? More broadly speaking, just like how there is likely a many-to-many function-to-structure mapping in the nervous system, it is equally as likely that a similar mapping applies to the relationship between neurotransmitters and different functions!
Reward or reward prediction error?
If you aren't convinced by the argument above, let's try one more. But first, we have to understand a concept key to the second wave of motivation research in neuroscience -- reward prediction error. Introduced by Schultz and colleagues in 1997, reward prediction error simply refers to the discrepancy between the reward that is expected and the actual reward received. Here, researchers were interested in what this discrepancy meant for learning.
Fig. 3 provides a simple illustration of a hypothetical function that drives behaviour depending on reward prediction error. When the behaviour leads to a reward that is expected, nothing new is learnt by the animal. Subsequently, there is no need to update their beliefs or change their behaviour, and they can continue doing the same actions. Meanwhile, when there is a mismatch between the actual rewards and what was predicted, there now exists a reward prediction error. If the actual rewards are superior to what was predicted, the animal is motivated to update their predictions of future rewards (specifically, expecting better rewards in the future) and to repeat more of the behaviour just executed. In other words, learning has occurred!
Fig. 3 Reward prediction error (Pessoa, 2022)
On the other hand, if the actual rewards are worse than what was expected, the animal is motivated to expect worse rewards in the future and to do less of the given action. To summarise, learning only occurs when there is a reward prediction error -- how beliefs and actions are directed depends on the direction of the error.
And here's where reward prediction error comes in as an argument against dopamine as a 'reward' molecule -- Schultz (2016) revealed that dopamineergic neuronal activity depended not on rewards per se, but on reward prediction error! In other words, these neurons were not firing whenever the animal received a reward. Rather, the intensity of firing was a function of the direction of reward prediction error (i.e., more activity when actual reward > predicted reward; less activity when actual reward < predicted reward). Here, we can see that dopamine really isn't a 'reward' molecule that is released inexplicably whenever an animal is exposed to appetitive stimuli. This has opened up many debates as to the link between dopamine, reward prediction errors, and motivation. Pessoa believes that there is still much work left to be done in this area.
Concluding remarks
Now that we've covered the subcortical regions involved in emotion and motivation, Chapter 6 will turn to cortical involvement in these functions. Stay tuned for more!
References
Olds, J., & Milner, P. (1954). Positive reinforcement produced by electrical stimulation of septal area and other regions of rat brain. Journal of Comparative and Physiological Psychology, 47(6), 419–427. https://doi.org/10.1037/h0058775
Pessoa, L. (2022). The entangled brain. Journal of Cognitive Neuroscience, 35(3), 349–360. https://doi.org/10.1162/jocn_a_01908
Schultz, W. (2016). Dopamine reward prediction error coding. Dialogues in Clinical Neuroscience, 18(1), 23–32. https://doi.org/10.31887/dcns.2016.18.1/wschultz
No comments:
Post a Comment