Developing a Model of the Insular Cortex and Emotional Regulation: Part 1

Background to Creating A Blog Model of the Insular Cortex

I’ve been reviewing a lot of information on this blog which can be used in lots of ways. What i’ve wanted to do is to try and put some of this information together to come up with something new and potentially useful. I’ve chosen to develop a model of the Insular Cortex which is rather simple and speculative, based as it is on 3 research findings. As pointed out in an earlier article, my hope is to use this blog to create an interaction with the readers to drive the model forwards. If I have succeeded in communicating the model effectively then it means that people with little or no background in science would be able to contribute meaningfully. For this reason, the wording in posts on the Insular Cortex model will be different from others on this blog in that they will be aimed at a general audience.

The Insular Cortex is potentially quite an important part of the brain as it receives lots of connections from other parts of the brain. There is some evidence to show that it might play a role in emotions and if this is the case then it could play an important role in many common psychiatric conditions. However we can’t jump to any conclusions. Indeed the idea behind building this model is to use the data to guide the theory of the model and then to test it conceptually with data from other studies. If the model is successful then it could be used to make predictions which can be tested. My hope is that while the model will initially be simple and possibly not realistic or useful, that by incorporating findings from different studies it can be refined. These refinements should help it’s ability to make accurate predictions. That’s the idea – whether it bears up in practice is another matter.

The ‘Open Model’

The model is what I would refer to as an ‘open model’ in that it can be contributed to and developed by the community. How this is achieved technically will most likely become clearer with time. As a simple example, if people want to leave comments on the model, their names, if they want can be included as a list of contributors to the model. Problems that are identified can be used to modify the model, for example by changing relationships within the model.

Brief Overview of the Insular Cortex

So i’ll begin with a brief overview of the insular cortex – not too much at this stage as we’ll be filling the model out over time.

Insular Cortex Schematic
Insular Cortex Schematic

Here is a schematic coronal section of the brain which is meant to convey some of the relationships of the Insular Cortex. In coronal sections, the brain is sliced from front to back (or back to front).  In the schematic diagram above, the Insular Cortex is seen on the right hand side of the diagram. If you were to look at a brain, you would see ridges (gyri) and grooves (sulci). At the very front of the brain is the Frontal Cortex and at the very back is the Occipital Cortex. You wouldn’t be able to see the Insular Cortex though because it’s tucked away underneath. If you look again at the diagram, you can trace the contour of the cortex moving from the top of the diagram to half-way down when it dips in. This is where the Insular Cortex is located. So if you were seeing a whole brain, you would need to move back part of the overlying cortex to be able to see it.

Right next to the Insular Cortex as we move inwards (medially) there is the Claustrum. There is a paper by Francis Crick in 2005, where he writes that this area might be strongly related to consciousness as it receives information from many areas of the brain and also sends information to many areas! For the moment though, the insular appears to have a very important neighbour! There is also the Lentiform Nucleus a bit further in. This contains two areas – the Putamen and the Globus Pallidus both of which form part of the Basal Ganglia which is involved in lots of different functions including thinking, emotions and movements. If we move even further inwards we come across the thalamus which acts as a relay for information passing between the cerebral cortex and the areas below the cortex. Also in the diagram we can see the red nuclei which are involved in the regulation of arms movements during walking. There are other relationships of the insular cortex and it’s also useful to note that even if some brain areas are close to each other, they don’t necessarily communicate with each other!

A Study on Anxiety and the Insular Cortex

There are lots of studies looking at the function of the Insular Cortex. Here though, I will focus on a few studies which form the basis for the model.

The first was covered on the blog earlier and looked at GABA receptors in panic disorder. GABA is a neurotransmitter in the nervous system – it’s a chemical that is involved in the transfer of information between cells in the brain (neurons). Indeed it can acts as an inhibitor – roughly speaking it would dampen down the firing of neurons. However GABA also acts to increase the firing of neurons. Whether it inhibits or excites neurons depends on the receptors on the neuron. So the finding in this study was that in people who get panic attacks, there are less GABA receptors in the Insular Cortex.

Now the first thing we need to do when building a model is to create a set of assumptions. These assumptions will be based on research findings. From the study above it’s necessary to ask – what conclusions can we reasonably draw from this? Well, there were people with panic compared to people without. There were only a small number of people in the study – 11 people with anxiety and 21 controls – so the results might be different if it was repeated with more people. I’m going to give an E number to the assumptions. Let me explain.

The E Number

Much like the STT (steps to treatment) score used in this blog, the E number will be a rough measure – indeed a judgement call. The E Number stands for the quality of the evidence and runs from 0 to 10. 10 would be very strong evidence and 0 the opposite. This is very similar to ideas that have been around in evidence-based medicine for some time where a study of a certain type might be considered more seriously than another type of study.

Deriving an Assumption about Anxiety, GABA and the Insular Cortex

Bearing in mind that we have the E number, it’s now possible to write a statement as follows

1. ‘People with Panic Disorder have less GABA receptors in the Insular Cortex than people without Panic Disorder’

Since this was a small study, I judge the evidence to be more towards the lower end of the scale – i’ll arbitrarily choose a score of 2. If the model isn’t realistic then this might be one area to focus on. Now the next step is to go one step further and say

2. ‘Panic Disorder results from/in less GABA receptors in the Insular Cortex’.

Now we’re on really shaky grounds here because the study wasn’t designed to look at cause and effect so we’re going beyond the data. So i’m going to give this a score E = 0 which means if the model is shaky this definitely needs a revisit.

Borderline Personality Disorder and the Insular Cortex

I won’t get embroiled in the whole issue of personality disorders here as it’s a very complex area with many aspects to the diagnosis. The idea behind the concept of a personality disorder is that people can develop personalities which cause distress to themselves or others and that it’s possible to identify such personality types using a set of criteria. In the Borderline Personality type, a person broadly speaking would have difficulties with relationships. Classically a person with borderline personality traits is thought of as viewing people as either all good or all bad and can also change their views of the same person. There are a number of other features but the area of relationships is the one that is relevant in the study that is considered here. This is on the basis of a review of a Science Podcast where a study was discussed in which people with Borderline Personality Disorder were playing games with other people in which each partner had to invest in the other. I haven’t read the original study and so for each E number i’m going to assign it a value of X, which basically means unknown. As soon as i’ve read the original paper i’ll be able to fill in the values but i’m including this study because along with the other studies it gave me the idea for this model (and it shows the possible benefits of listening to podcasts!). Anyway there is a review of the study to be found here. The gist of the study was that if someone had borderline personality disorder they were less likely to increase activity in the anterior insular cortex on the basis of how much they received from other people. This was interpreted as meaning that the people with borderline personality disorder were less trusting in others and therefore unperturbed by any changes in the amount they received from others. So what assumptions can we draw here

3. ‘The Anterior Insula Cortex is activated when people are receiving gift amounts from other people’ E = x

There is a difficulty with this assumption in that the imaging findings are equated with increased activity whereas there are some tenuous arguments which suggest that an area of the brain can function equally well without needing to register increased activation (as information is encoded in patterns of firing rather than the overall intensity of firing patterns).

In keeping with the spirit of the study, we can make a further assumption

4. ‘The Anterior Insula Cortex is activated when other people are relating to the individual’

This involves reducing the gifts from other people to a general commentary on relationships. Again E = x but here we are going beyond the data again as this is a conceptual leap from the specifics of the study paradigm.

There are two last assumptions

5. ‘People with Borderline Personality Disorder have difficulties with emotional regulation and have less activation in the Anterior Insula when receiving gifts from other people’ E = x

6. ‘Decreased activation in the Anterior Insula leads to emotional difficulties in relationships in people with Borderline Personality Disorder’ E = 0 (Going beyond the data but again in the spirit of the study).

The Final Piece of the Jigsaw – Temperature, The Insula Cortex and Gifts

In another podcast review (again I will return to look at the original study at a later date), Lawrence Williams discussed his research on physical temperature and perceptions of relationships. He found in his study that if people felt hot temperatures rather than cold (using a temperature pad) they were more likely to choose for others rather than themselves. Williams argument is that the Insular Cortex could be integrating emotional information and sensory information although it is speculation since this study did not involve imaging of the Insular Cortex and so E=0.

7. ‘The Insular Cortex integrates Emotional and Sensory Information’ E = 0

Summarising the Assumptions

So just to summarise we now have the following assumptions

1. ‘People with Panic Disorder have less GABA receptors in the Insular Cortex than people without Panic Disorder’

2. ‘Panic Disorder results from/in less GABA receptors in the Insular Cortex’.

3. ‘The Anterior Insula Cortex is activated when people are receiving gift amounts from other people’ E = x

4. ‘The Anterior Insula Cortex is activated when other people are relating to the individual’

5. ‘People with Borderline Personality Disorder have difficulties with emotional regulation and have less activation in the Anterior Insula’

6. ‘Decreased activation in the Anterior Insula leads to emotional difficulties in relationships in people with Borderline Personality Disorder’

7. ‘The Insular Cortex integrates Emotional and Sensory Information’ E = 0

Furthermore I’ll choose the assumptions that are the most interesting (although some are not backed up by  evidence)

2. ‘Panic Disorder results from/in less GABA receptors in the Insular Cortex’.

6. ‘Decreased activation in the Anterior Insula leads to emotional difficulties in relationships in people with Borderline Personality Disorder’

7. ‘The Insular Cortex integrates Emotional and Sensory Information’

The Model

On the basis of the three assumptions outlined in the previous section, I propose the following model:

The Insular Cortex processes emotions particularly anxiety. This occurs in the Anterior Insular Cortex. The number of GABA receptors in the Anterior Insular Cortex determines the intensity of the emotions experienced. The less GABA receptors there are, the more intensely will emotions be experienced and conversely if the number of receptors is increased. Sensory information is also processed in the Anterior Insular Cortex and influences the intensity of the emotions experienced independently of the GABA receptors. In this manner, the Insular Cortex would act as the interface between emotions and bodily sensations discussed by William James in the nineteenth century. The sensory information would have a complex relationship with emotions while at the level of GABA receptors the relationship would be more straightforward (a few extra assumptions here but these will soon be reworked).

Model of Anterior Insular Cortex Function
Model of Anterior Insular Cortex Function


The model above is a very early attempt to produce something which is testable and which integrates data from different studies. There are a number of criticisms here already

1. A number of the assumptions have little evidence at this stage to support them

2. The Limbic Cortex is the region usually associated with emotional processing and has a large evidence base to support it in this role.

The interested reader is encouraged to contribute to the model whether this be through criticisms, additional studies of interest and new ideas about how the model should be modified (supported by analysis). Contributors will be included in a list of people who have contributed to the model unless they specifically request otherwise. Disagreements can be resolved by branching of models.


The comments made here represent the opinions of the author and do not represent the profession or any body/organisation. The comments made here are not meant as a source of medical advice and those seeking medical advice are advised to consult with their own doctor. The author is not responsible for the contents of any external sites that are linked to in this blog.


  1. The insula, eh? That’s a mighty ambitious model, given that it’s activated in nearly every neuroimaging study under the sun. I actually wanted to tackle the insula myself in a theoretically oriented blog post, but I found the task too daunting. But you have to start somewhere. As a more manageable project, back in August I took the major activation focus in R insula from that BPD trust study and plugged the coordinates into the BrainMap database. Here’s the result: . At the time, I also wanted to read two review articles by A.D. (Bud) Craig on interoception – [but never got around to it.] There was a panel on the anterior insula at the Cognitive Neuroscience Society meeting this year, and Craig gave quite the talk.


  2. Hi Neurocritic,

    Thanks for your contributions – can I include you as a list of contributors? I’ve really enjoyed reading your blog. The model here will be owned by the community – I thought this would be an interesting online experiment. As it is, it’s very crude but hopefully it should improve with time. I’ll get both those papers by the way and review them on the site and then put some suggestions as to how they might be incorporated

    Kind regards



  3. Hi Justin,

    Uh, sure, you can list me as a contributor. I’m not sure how much constructive input I’ll have, though. As a general function, it seems that the insula might be sensitive to task difficulty across multiple domains, in which case it could be a direct reflection of autonomic responsiveness. Anxiety and panic, of course, would be extreme examples of this.

    Thanks for reading my blog.

    The Neurocritic


  4. Hi, I started looking up things about the anterior insula after reading an article in Scientific American Mind. Thanks for your work in summing up most of what I’ve been finding on the ‘net. There was also something about its role in perception of touch, and something about people whose anterior insulas were damaged by strokes no longer wished to smoke. Oh, and about meditation increasing the size of the anterior insula. Which is really puzzling to me– most sources seem to see anterior insula activity as a positive thing, but when it’s deactivated then there’s less addiction! It’s quite a confusing part of the brain!


  5. Hi Anonymous!
    Thanks for your comments and for your references to other relevant studies. I’ve seen some of these other studies and I suspect that there will be many more to follow as this part of the brain becomes a new focus for research. Probably some of the difficulties of interpreting imaging data arise from what I think is an absence of simple relationship between visible activity and function of one part of the brain. There is a chance that at the macroscopic level we might not be seeing the ‘real information’ where single unit recording (e.g. intraoperatively during surgical procedures) might be more useful. If we do assume however that the macroscopic observations are valid there is still the issue of what this information is coding for and that probably should be informed by neuroanatomical and neurophysiological considerations. Thus this direct visual information needs to be abstracted and interpreted in terms of connections and factors influencing image data. There are also ‘networks’ involved in different experiences and some behaviours/experiences used in experimental paradigms are usually a complex assortment of multiple phenomenon.


  6. Normally I don’t read post on blogs, however I wish to say that this write-up very forced me to try and do so!
    Your writing taste has been surprised me. Thanks, very nice article.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s