Thursday, January 06, 2011

More Friends on Facebook Does NOT Equal a Larger Amygdala

Bottom image adapted from Fig. 2 of Schumann et al. (2010). Neuroanatomy of the human amygdala postmortem. Nissl-stained section of amygdala nuclei.

The amygdala is a subcortical structure located within the medial temporal lobes. It consists of a number of different nuclei, or collections of neurons delineated by commonalities in morphology and connectivity. The amygdala is best known for major roles in fear conditioning (Paré et al., 2004) and responding to emotional stimuli more generally (Phelps & LeDoux, 2005), but its functions extend beyond that.

A new study by Bickart and colleagues (2010) examined the relationship between the overall size of the amygdala in a group of 58 volunteers and the number of people in each person's social network. The authors observed a direct correlation between the two: the larger the amygdala, the larger the social network. Why did they expect such a finding? The "social brain hypothesis" (Dunbar, 1998) is cited as providing general evidence in favor of increased brain size in more social animals. However, the major references that motivated the specific hypothesis about the amygdala are book chapters, which seemed rather weak and unscholarly to me.

Predictably, a number of silly headlines appeared in the popular press...
How to Win Friends: Have a Big Amygdala?

Got a big social network? Then you probably have a large amygdala, according to a new study that found a connection between the size of this brain region and the number of social relationships a person has. The complexity of those relationships — as measured by the number of people who occupied multiple roles in a social network such as being simultaneously a friend and a co-worker — was also linked with amygdala size.
...and in blogs:
The Twitter Spot in Your Brain

Heavy Facebook users may have weighty amygdalas
But the worst headline of all (because it is patently false) is...
Study: More Friends on Facebook Equals A Bigger Amygdala In Your Brain

The number of Facebook friends you have is correlated to the size of your amygdala, the center used to process the memory of your emotional reactions in your brain, according to a new study published in Nature Neuroscience. The volume of your amydala has been connected to the size of the circle of those you come in contact with even with nonhuman primate species before, so Kevin Bickart and his coauthors took the idea and tested it out on people who interact with people on Facebook.

Does the Amygdala Have a Social Network?

First of all, the size of the amygdala has absolutely nothing to do with Facebook or any other contemporary social networking site. The scale for quantifying social network size and complexity was taken from a 1997 paper on Social Ties and Susceptibility to the Common Cold (Cohen et al., 1997), which in turn cited a book chapter from 1991. There was no such thing as Facebook or Myspace in 1997, only Geocities (1994) and (1995). As for the history of online communities, The WELL was launched in 1985 as a bulletin board system and could be considered as a proto-social networking site.

So who was included in Cohen et al.'s (1997) definition of a social network? One requirement was that the participant spoke to the individual in person or on the phone at least once every two weeks:
The Social Network Index assesses participation in 12 types of social relationships. These include relationships with a spouse, parents, parents-in-law, children, other close family members, close neighbors, friends, workmates, schoolmates, fellow volunteers (eg, charity or community work), members of groups without religious affiliations (eg, social, recreational, or professional), and members of religious groups. One point is assigned for each type of relationship (possible score of 12)1 for which respondents indicate that they speak (in person or on the phone) to someone in that relationship at least once every 2 weeks. The total number of persons with whom they speak at least once every 2 weeks (number of network members) was also assessed.
Results for the amygdalar correlations with social network size and complexity were nearly identical, so the authors focused on the former in the paper. Importantly, amygdala volume did not correlate with life satisfaction or the perceived quality of the relationships. Basically, you could have a large social network that is not of your own choosing. One participant could have a large family, many co-workers, intrusive neighbors, and few real friends, while another is gregarious and parties with different groups of friends every night of the week. This particular study does not distinguish between the two.

Were any other parts of the brain correlated with social network size? No! [which I find hard to believe]. For the entire group of 58 participants, no significant results were observed in other subcortical regions (i.e., hippocampus [which served as a control region], brainstem, nucleus accumbens, ventral diencephalon, thalamus, caudate, putamen, and globus pallidus). An exploratory whole-brain analysis of cortical thickness did not reveal any correlations at conventional levels of significance.

Are we supposed to believe that only one area of the brain is involved in maintaining social networks? I think not. Even within the article, subgroups of participants (i.e., males, all older subjects)2 showed correlations between hippocampal volume and social network characteristics. This makes intuitive sense, as a better memory might be helpful in keeping track of large numbers of people.

What about cortical regions containing spindle neurons (Nimchinsky et al., 1999), a cell type unique to humans, great apes, humpback whales, elephants and dolphins? Spindle neurons (aka von Economo neurons) are found in the anterior cingulate cortex and frontoinsular cortex. Or how about orbitofrontal cortex, with a volume that correlates with social cognitive competence (Powell et al., 2010)? None of these regions were related to network size.

What happens to people born without amygdalae due to Urbach-Wiethe disease, a rare genetic disorder? Does "the woman without fear" (and without amygdalae) have a tiny social network? How about other patients with amygdala damage? Anecdotal evidence suggests they can have close ties with their families and can even become more social after their brain injuries.3 Furthermore, Amaral et al. (2003) suggest that
...the effects of amygdala lesions in the adult and infant macaque monkey do not support a fundamental role for the amygdala in social behavior.
So small or absent amygdalae may not be a complete disaster.

But Is Bigger Better?
Findings from our study provoke the question, “Is a bigger amygdala better?” To answer this question, we must consider what “bigger” means and what a bigger amygdala might be “better” for. Striedter suggests that bigger means better connected, so that a brain region with more volume (in cross-species or within species comparisons) has an enhanced ability to modulate processing in its target regions. From cross-species comparisons in nonhuman primates, researchers propose that a bigger amygdala might provide processing advantages for visual signals from conspecifics...

...Yet it is far from clear that the cross-species comparisons can be generalized to infer that a bigger amygdala is better in humans, particularly when it comes to social functioning.

When Bigger Is NOT Better

One prominent example is the finding of larger amygdalae in children (and adults) with autism (Howard et al., 2000; Mosconi et al., 2009). However, the literature on this issue is variable (and voluminous), so I won't discuss it further here. More consistent are observations of increased amygdala volumes in generalized anxiety disorder (Etkin et al., 2009; Schienle et al., 2010). In rats, chronic stress causes hypertrophy (enhanced dendritic arborization) of pyramidal and stellate neurons in the basolateral nucleus of the amygdala (Vyas et al., 2002).

A further interpretive conundrum is presented by the variety of conditions that are associated with increased amygdala volume: first-episode patients with nonschizophrenic psychoses, women high in harm avoidance, learning disabled adolescents at high risk of schizophrenia, adopted Romanian adolescents who experienced severe early institutional deprivation, and political conservatism.4 As explained by Tebartz van Elst et al. (2007):
Like often in psychiatric research, these findings seem to be contradictory and thus frustrating at first glance. However, if a dimensional rather than a categorical approach is chosen, all these findings might be integrated based on the assumption that the amygdala volume status might reflect the dominant mode of emotional information processing. Psychopathological features such as depressed mood, anhedonia, lack of drive, phobic anxiety, and rumination might characterize a mode of emotional information processing that is associated with enlarged amygdala volumes, whereas symptoms like emotional instability, dysphoria, irritability, aggression, and psychotic anxiety might go along with reduced amygdala volumes.
Correlation does not equal causation. If the findings are replicable, we don't know if having a large social network increases the size of the amygdala [perhaps due to the stress and anxiety such social pressures entail?], or if having a hefty amygdala causes one to form a large social network.

Broken Social Scene - Fire Eye'd Boy (WATCH on YouTube)


1 The notion that any one person could fulfill all 12 social roles is absurd. How can anyone simultaneously be a spouse, parent, parent-in-law, child, other close family member, close neighbor, friend, workmate, schoolmate, fellow volunteer, member of a group without religious affiliations, AND member of a religious group?? Obviously, some of these are mutually exclusive.

2 The 58 participants were a diverse group ranging in age from 19-83 yrs (mean = 52.6), which I found strange for such an exploratory study. The demographics were quite unbalanced as well: males (n=36) vs. females (n=22); young (n=19) vs. old (n=35) [which does not add up to 58].

3 Anecdotal evidence from Sophie Scott:
@sophiescott: it's odd, isn't it. I worked with DR, who had bilateral amygdala damage. Hard to assess her network but she was very close...

@sophiescott: ...with her extended family. Another amygdala patient, SE, had become very social since his head injury.

@sophiescott: there were clear effects of their brain damage on their actual social interactions, but they still had them.
4 This study was published in a newspaper, not in a peer reviewed journal (see Left Wing vs. Right Wing Brains).


Amaral DG, Bauman MD, Schumann CM. (2003). The amygdala and autism: implications from non-human primate studies. Genes Brain Behav. 2:295-302.

Bickart, K., Wright, C., Dautoff, R., Dickerson, B., & Barrett, L. (2010). Amygdala volume and social network size in humans. Nature Neuroscience DOI: 10.1038/nn.2724

Cohen S, Doyle WJ, Skoner DP, Rabin BS, Gwaltney JM Jr. (1997). Social ties and susceptibility to the common cold. JAMA 277:1940-4.

Dunbar RIM. (1998). The social brain hypothesis. Evol. Anthropol. 6:178–190.

Etkin A, Prater KE, Schatzberg AF, Menon V, Greicius MD. (2009). Disrupted amygdalar subregion functional connectivity and evidence of a compensatory network in generalized anxiety disorder. Arch Gen Psychiatry 66:1361-72.

Howard MA, Cowell PE, Boucher J, Broks P, Mayes A, Farrant A, Roberts N. (2000). Convergent neuroanatomical and behavioural evidence of an amygdala hypothesis of autism. Neuroreport 11:2931-5.

Mosconi MW, Cody-Hazlett H, Poe MD, Gerig G, Gimpel-Smith R, Piven J. (2009). Longitudinal study of amygdala volume and joint attention in 2- to 4-year-old children with autism. Arch Gen Psychiatry 66:509-16.

Nimchinsky EA, Gilissen E, Allman JM, Perl DP, Erwin JM, Hof PR. (1999). A neuronal morphologic type unique to humans and great apes. Proc Natl Acad Sci 96:5268-73.

Paré D, Quirk GJ, Ledoux JE. (2004). New vistas on amygdala networks in conditioned fear. J Neurophysiol. 92:1-9.

Phelps EA, LeDoux JE. (2005). Contributions of the amygdala to emotion processing: from animal models to human behavior. Neuron 48:175-87.

Powell JL, Lewis PA, Dunbar RI, García-Fiñana M, Roberts N. (2010). Orbital prefrontal cortex volume correlates with social cognitive competence. Neuropsychologia 48:3554-62.

Schienle A, Ebner F, Schäfer A. (2010). Localized gray matter volume abnormalities in generalized anxiety disorder. Eur Arch Psychiatry Clin Neurosci. Sep 5. [Epub ahead of print].

Schumann CM, Bauman MD, Amaral DG. (2010). Abnormal structure or function of the amygdala is a common component of neurodevelopmental disorders. Neuropsychologia Oct 13. [Epub ahead of print].

Tebartz van Elst L, Ebert D, Hesslinger B. (2007). Amygdala volume status might reflect dominant mode of emotional information processing. Arch Gen Psychiatry 64:251-2

Vyas A, Mitra R, Shankaranarayana Rao BS, Chattarji S. (2002). Chronic stress induces contrasting patterns of dendritic remodeling in hippocampal and amygdaloid neurons. J Neurosci. 22:6810-8.

Subscribe to Post Comments [Atom]


At January 06, 2011 8:06 AM, Anonymous Anonymous said...

Judging by the age range of the sample, and being familiar with this group's past work (Alzheimer's), I'm guessing these subjects were taken from the control group of an aging study. Makes you wonder how hypothesis-driven this study really was.

At January 06, 2011 10:14 AM, Anonymous Ciarán said...

This is an very well-written deconstruction, the like of which I seldom read on blogs, although I think you are being a bit hard on the authors.

This paragraph is particularly harsh, for a number of reasons:

First of all, the size of the amygdala has absolutely nothing to do with Facebook or any other contemporary social networking site. The scale for quantifying social network size and complexity was taken from a 1997 paper on Social Ties and Susceptibility to the Common Cold (Cohen et al., 1997), which in turn cited a book chapter from 1991. There was no such thing as Facebook or Myspace in 1997, only Geocities (1994) and (1995). As for the history of online communities, The WELL was launched in 1985 as a bulletin board system and could be considered as a proto-social networking site.

- the authors didn't mention Facebook in the article at all. You can hardly blame other journalists or bloggers for putting two and two together, but you shouldn't repeat this error.

- if you think that Cohen et al.'s (1997) measure is inappropriate, you should suggest a better alternative. I'm no expert but it's been cited a couple of hundred times, and recently.

- But the first sentence is most telling: "the size of the amygdala has absolutely nothing to do with Facebook or any other contemporary social networking site". Besides the fact that the authors never claim such, it's pretty obvious from this sentence that you have prejudged this research from the beginning. Unless you've actually carried out a study, or can point to another study, which looked at Facebook and the amygdala, you cannot say this with any certainty at all.

This is a pity, because it seems to have distracted you from other problems with the paper, such as the size of the sample with respect to the ages of the participants (would have been better to simply stick with one age group). I am also unclear as to why so much material was left in the 'Supplementary' section, which is quite useful, especially with regard to the hippocampus.

The suggestion that a person with a larger or more complex social network might have a different brain structure to someone with a smaller or simpler one is not really that controversial. However, while this paper has not, to my satisfaction clearly demonstrated that, neither has your post rebutted it.

At January 06, 2011 10:59 AM, Blogger said...

Excellent post. Very well said and well researched. Thank you for this. If nothing else, now I have a good place to point people to when I'm asked about this paper.

At January 06, 2011 11:57 AM, Anonymous R. said...

The notion that any one person could fulfill all 12 social roles is absurd....Obviously, some of these are mutually exclusive.

Is there an unspoken assumption that "child" must mean a person who is a legal minor rather than simply a person in relationship to at least one of their parents? If not, then none of the roles are mutually exclusive.

At January 08, 2011 9:24 AM, Blogger jonathan said...

As a person with autism, I have experienced first hand what having social deficits is like. Amygdala size may have nothing to do with social deficits, but there is apparently evidence of defects in the amygdala (and other areas of the limbic system) in persons with autism, so there might be some sort of casual relationship.

Have there been similar studies done on correlations between social networks and the pars opercularis and other mirror neuron areas and other areas of the frontal lobes? I wonder about that.

At January 10, 2011 10:19 AM, Blogger Neuroskeptic said...

Good point about the amygdala enlargement in autism issue - not many people know that, or at least, not many people cite it...

At February 10, 2011 6:21 PM, Anonymous Anonymous said...

great article. 10/10


Post a Comment

<< Home

eXTReMe Tracker