The Algorithm of The Brain

I know that the computer metaphor does not do justice to the brain, but can we conceivably come up with a universal algorithm in how the brain processes stimuli and reacts/responds to them? Further, can we then tie up those algorithmic sub-modules to actual neural subsystems/structures and neurotransmitter systems as substantiated in the physical brain?

That is what I intend to do today, but first let us list our very basic algorithm of how the brain processes stimuli and responds to it. Consider it like a flowchart with each step there being made a decision. At each step that is numbered 1, nothing further happens; at each step numbered 2, further 2 choices are available.

  1. Stimuli comes!
    1. Ignore?
    2. Attend?
      1. Unimportant?
      2. Important?
        1. Default response?
        2. Choose response?
          1. Unfeasible?
          2. Feasible?
            1. Execute response!

Now, let me unpack this a bit. The first step for the purposes of this post is an incoming stimulus. When the stimuli comes we (the brain) can be in different levels of alertness and lookout for incoming stimuli; thus the brain may miss or detect the stimuli. We may be in neuro-vegetative states like sleep and feeding and may be relaxing and miss on both threatening as well rewarding stimuli. Or we can be in a vigilant mode either on lookout for danger or say alert while ready to pounce on prey. A Vigilance system can be reliably conjectured to underlie this and indeed Locus Coerelus- Nor epinephrine (LC-NE) system may just be exactly that system that makes us alert and inhibits neuro vegetative states. Another brain structure relevant here is Amygdala which is popularly known for its role in detecting threatening stimuli, but is involved in pleasant stimuli detection too. Hypersensitiveness of this system can conceivably lead to anxiety at one end (constant lookout for trouble) and addiction (constant lookout for possible gains) at the other. One can also extend this line of reasoning and posit that differential sensitivity of this system may underlie the personality trait of Neuroticism.

Once you have noticed or attended to a stimuli what next? Not every stimuli is salient or important. The next step for the brain is to identify whether the stimuli is indeed important from a functional point of view- whether it is an indicator of, or an actual, reward or punishment. Here comes the incentive salience function of Dopamine. Dopamine neurons in say Nucleus Accumbens (NAcc) area code for whether the incoming stimuli is important or not (see work of Berridge et al) ; if its not important nothing needs to be done; however if it is important and consequential than an appropriate response needs to be executed. Activity has to ensue. Please note that though NAcc is typically thought of as part of a reward circuit, it is equally involved in determining salience of an aversive stimuli. Hypersensitiveness of this incentive salience system can conceivably lead to depression at one end (where all stimuli are important , but negatively toned or aversive) and mania at the other end (where all stimuli are important, but perceived as positively valenced). One can also extend this line of reasoning and associate differential sensitivity in this system to trait of Extraversion.

Once you have determined that the stimuli is important and needs responding, how do you determine the right response? One effortless and ‘hot’ way is to use the default response – if someone threatens you, punch them in the face! The other, more effortful, and ‘cold’ way is to choose a response from the response sets that have been activated or by overriding the default response and selecting something better. This is the selfregulation system. As a brain region, I’m sure ACC has a major role to play here- detecting conflicts between responses and also inhibiting dominant default response. In terms of neurotransmitters I see a role for Serotonin here – regulating the response, especially emotional and instinctual response. Hypersensitiveness of this system may lead to obsessions (rigid thinking) and compulsions (rigid acting) and differential sensitivity in the system may be associated with Conscientiousness.

Now, that you/ your brain has chosen the most appropriate response, one further step needs to be executed before you actually execute the action. Many readers of this blog will be familiar with the Value -Expectancy model of motivation: Value was coded by dopamine neurons using incentive salience, what about expectancy? Basically the V-E model posits that an action will be taken only if you value the outcome and are reasonably sure that you can act in such a way as to achieve the outcome. Neurons in PFC may conceivably code for outcome prediction. PFC is important to predict whether a particular course of action will lead to desired results. It is also conceivable that dopamine neurons may play an important role here. The basic idea is to predict whether you can execute the response and receive the reward or avoid the punishment and only then if the action is feasible, then execute the action. This outcome prediction module I think recruits PFC to a large extent. Hypersensitivity of this system may lead to ADHD and differential sensitivity associated with Openness to experience.

To me the above looks very neat and logical and elegant and I would love your comments regarding the same and also any contradictions you see in literature or any additional thoughts you may have.

Effecient Related Posts: