The ordinary-language meaning of technical terms often wanders far from home, following paths of connotative association and denotative opportunity. We've followed the semantic travels of "passive voice" through meanings like "vague about agency", "stylistically listless", and "failure to take sides". I recently read that writers should "Use an active voice (putting things in present/future) instead of a passive voice (putting things in the past)".
The terminology of the "autism spectrum" seems to have started a similar journey through successive steps of family resemblance.
For example, Jason Calacanis ("We Live in Public (and the end of empathy)", 1/28/2009), used the term "Internet Asperger's Syndrome" to describe the reaction to a late-90s "art project" in which his friend Josh Harris "put a couple dozen cameras all over his loft and recorded the inevitable breakdown of his life with the love of his life", and set up internet chat rooms for public discussion of the results.
The commenters in the chat rooms were so "vicious", according to Calacanis, that "it took Josh five years to recover": something about the experiment "robbed the subjects — and their audience — of every last ounce of empathy". This leads Calacanis to propose what he calls "Harris' Law":
At some point, all humanity in an online community is lost, and the goal becomes to inflict as much psychological suffering as possible on another person.
And he says that he's come to "recognize a new disorder, the underlying cause of Harris' Law", Internet Asperger's Syndrome, which "affects people when their communication moves to digital", causing them to "[stop] seeing the humanity in other people", and to behave in other ways that (in his view) parallel the symptoms of Asperger's Syndrome.
The term Internet Asperger's Syndrome was recently picked up by Jonathan Kimak in a humorous piece for Cracked ("6 New Personality Disorders Caused by the Internet", 6/30/2009). Kimak writes (inaccurately) that Asperger's Syndrome is a
… rarely diagnosed but often claimed disorder is a mild form of Autism that comes with what seems to be a biological inability to show empathy for other human beings, as well as (and maybe stemming from) an inability to recognize nonverbal cues. They continually do weird, upsetting things because they don't know it's upsetting you. That part of their brain is broken.
People cringe when they hear this term because they know that a large number of the teenagers claiming Asperger's are, in fact, merely dicks.
He agrees with Calacanis's diagnosis:
Calacanis figured out that people who do all of their communicating online wind up mimicking Asperger's behaviors because they are imposing the same disadvantages on themselves. In both cases, when the ability to see nonverbal responses and facial expressions goes away, so does empathy. Soon the thing you're communicating with isn't a person, they're just a bunch of words on a screen. A bunch of words that the little bastard didn't even bother to spellcheck.
Thus Kimak ends up connecting Asperger's Syndrome with various forms of internet-mediated mob cruelty — his characteristic examples are things like "A kid commits suicide on webcam while the trolls cheer him on … Normal kids, … but get them in a chat room and suddenly it reads like the transcript to a Charles Manson parole hearing"
Ironically, it takes a certain lack of empathy to see Charles Manson's sociopathic crimes as having any similarity at all with the social awkwardness and focused, "systematizing" interests of Asperger's people. And spontaneous adolescent mob cruelty, internet-mediated or not, strikes me as having little to do with either one.
In this case, Calacanis and Kimak make the the connection between chat-room meanness and Asperger's because of the idea that the lack of non-verbal cues leads to the depersonalization of victims. But the kind of mobbing gossip that they describe — as familiar from school cafeterias as from web forums — is way outside the spectrum of Asperger's behaviors, from everything I've seen and read. And charismatic sociopaths like Manson are especially skilled in exactly the sorts of communicative manipulation that Aspies have problems with.
The current DSM-IV-TR description of Asperger's Disorder (299.80) is:
- marked impairment in the use of multiple nonverbal behaviors such as eye-to-eye gaze, facial expression, body postures, and gestures to regulate social interaction
- failure to develop peer relationships appropriate to developmental level
- a lack of spontaneous seeking to share enjoyment, interests, or achievements with other people (e.g., by a lack of showing, bringing, or pointing out objects of interest to other people)
- lack of social or emotional reciprocity
- encompassing preoccupation with one or more stereotyped and restricted patterns of interest that is abnormal either in intensity or focus
- apparently inflexible adherence to specific, nonfunctional routines or rituals
- stereotyped and repetitive motor mannerisms (e.g., hand or finger flapping or twisting, or complex whole-body movements)
- persistent preoccupation with parts of objects
The disturbance causes clinically significant impairment in social, occupational, or other important areas of functioning.
Apparently, Calacanis and Kimak have taken two points of association — the lack of non-verbal cues characteristic of text-mediated interactions, and the "lack of social or emotional reciprocity" that is a possible diagnostic indicator for Asperger's Syndrome — and made a terminological jump to a completely different sort of phenomenon, namely socially-reinforced mob cruelty in chat rooms and similar web forums. (With a nod to yet another phenomenon, namely Charles-Manson-like sociopathic charisma.) This is quite like the chain of associations at work in various folk interpretations of "passive voice".
A different chain of autism-associations emerges in Gary Stix, "The Science of Economic Bubbles and Busts", Scientific American, 6/22/2009.
But before getting to the autism part, let me take advantage of the digression-tolerant blog format to point out a lovely example of Explanatory Neurophilia from the same article:
Behavioral economists have identified a number of biases, some with direct relevance to bubble economics. . In confirmation bias, people overweight information that confirms their viewpoint. Witness the massive run-up in housing prices as people assumed that rising home prices would be a sure bet. The herding behavior that resulted caused massive numbers of people to share this belief. Availability bias, which can prompt decisions based on the most recent information, is one reason that some newspaper editors shunned using the word “crash” in the fall of 2008 in an unsuccessful attempt to avoid a flat panic. Hindsight bias, the feeling that something was known all along, can be witnessed postcrash: investors, homeowners and economists acknowledged that the signs of a bubble were obvious, despite having actively contributed to the rise in home prices.
Neuroeconomics, a close relation of behavioral economics, trains a functional magnetic resonance imaging device or another form of brain imaging on the question of whether these idiosyncratic biases are figments of an academician’s imagination or actually operate in the human mind.
Think about that for a minute. Decades of careful experimentation have demonstrated the existence of these biases, and explored their nature in elaborate quantitative detail, resulting in thousands of publications and a Nobel Prize. But all this might just be a figment of some academician's imagination — until we see the fMRI pictures to be found in Bernd Weber et al., "The medial prefrontal cortex exhibits money illusion", PNAS 106(13): 5025-5028, 2009, which show that
… areas of the ventromedial prefrontal cortex (vmPFC), which have been previously associated with the processing of anticipatory and experienced rewards, and the valuation of goods, exhibited money illusion. We also found that the amount of money illusion exhibited by the vmPFC was correlated with the amount of money illusion exhibited in the evaluation of economic transactions.
(The "money illusion" is the tendency of people to evaluate sums of money in terms of their nominal or face value instead of in terms of their purchasing power.)
I could digress further, to explain why the fMRI evidence is actually weaker, so far, than the evidence from experiments using behavioral dependent variables other than blood oxygenation level in various brain regions — and likely to remain that way for some time — but that's a topic for another post.
For now, on to what Mr. Stix has to say about the relationship of autism to financial bubbles:
One group that does not value perceived losses differently than gains are individuals with autism, a disorder characterized by problems with social interaction. When tested, autistics often demonstrate strict logic when balancing gains and losses, but this seeming rationality may itself denote abnormal behavior. “Adhering to logical, rational principles of ideal economic choice may be biologically unnatural,” says Colin F. Camerer, a professor of behavioral economics at Caltech.
The suggestion here is that the rational-choice dynamics of neo-classical economics is somehow "autistic".
This idea may be connected to the rebellion in 2000 of a group of French economics students, who wrote an open letter to their professors and administrators to complain about the domination of the neoclassical model in their curriculum, explaining that "Nous ne voulons plus faire semblant d’étudier cette science autiste qu’on essaie de nous imposer" ("We no longer want to pretend to study this autistic science that is being forced on us").
Nous avons utilisé ce terme d’ « autisme » parce qu’il nous semblait bien résumer ce que nous ressentions, notamment la fermeture totale de la discipline au monde extérieur. Or cette fermeture à l’extérieur est communément associée à l’autisme, même si, comme beaucoup d’autres personnes, nous connaissons très mal cette maladie. Selon le Petit Robert, l’autisme désigne une « attitude de détachement de la réalité extérieure accompagnée d’une vie intérieure intense ». Pour le Larousse, c’est une « perturbation affective caractérisée par un repliement du sujet sur lui-même avec perte plus ou moins importante des contacts avec le monde extérieur ». Ce seul mot parvient donc, selon nous, à caractériser le comportement du courant dominant (académiquement) en économie. Celui-ci se caractérise en effet par sa coupure au monde extérieur, d’ailleurs couplée à une « vie intérieure intense », c’est-à-dire la production à jets continus de petits modèles tous plus débiles les uns que les autres.
We have used the term "autism" because it seem to us to summarize what we were objecting to, namely the complete closure of the discipline to the outside world. Now, this closure to the outside is commonly associated with autism, although we, like many others, know very little about this illness. According to the Petit Robert dictionary, autism refers to an "attitude of detachment from exterior reality accompanied by an intense inner life". For the Larousse, it's an "affective disorder characterized by withdrawal of the subject into himself, along with a significant loss of contact with the outside world". This single word then, we feel, manages to describe the behavior of the dominant (academic) trend in economics. This is characterized in fact by its cutting of connections to the outside world, at the same time coupled with an "intense inner life", that is, the production of a continuous stream of little models, each more retarded than the last.
An American relative of this movement is called "Post-Autistic Economics", with a website and a journal called the "real world economics review (Formerly the post-autistics economics review)". (See Edward Fullbrook "The Post-Autistic Economics Movement: A Brief History".) In the French rebellion and its American echo, the idea seems to be that neoclassical economics is "autistic" because it is isolated from society and "[preoccupied] with one or more stereotyped and restricted patterns of interest that is abnormal either in intensity or focus". This is arguably insulting both to neoclassical economics and to autism, but it's plausible as a metaphorical application of the DSM-IV type of definition.
Colin Camerer's ideas about the relations between autism and rational-choice theory, in contrast, seem to include the idea that they both apply logic inappropriately in attempting to predict human behavior. (For a hilarious account of game theory mis-applied to normal human interactions — without any mention of the autism spectrum — see Dan Zettwoch's Deadlock.) This perspective is expressed at somewhat greater length in Camerer et al., "Neuroeconomics: How Neuroscience Can Inform Economics", Journal of Economic Literature XLII: 9-64, 2005. Here's how the authors discusses irrationality in the Ultimatum Game:
When players do follow the dictates of game theory, the result can be a low payoff and confusion. Consider this quote from an upset subject, an Israeli college student, whose low offer in a $10.00 ultimatum game was rejected (from Shmuel Zamir 2000):
I did not earn any money because all the other players are stupid! How can you reject a positive amount of money and prefer to get zero? They just did not understand the game! You should have stopped the experiment and explained it to them . . .
Ironically, while the subject’s reasoning matches exactly how conventional game theory approaches the game, it also sounds autistic, because this subject is surprised and perplexed by how normal people behave.
Camerer et al. discuss "Elizabeth Hill and David Sally’s (2003) extensive comparison of normal and autistic children and adults playing ultimatum games". But the Hill and Sally experiments don't support a view that the "rationality" of conventional game theory is "autistic", either in the Ultimatum Game or elsewhere.
The citation by Camerer et al. is to a 2003 UCL working paper, Elisabeth Hill and David Sally, "Dilemmas and bargains: Autism, theory-of-mind, cooperation and fairness". The research was continued, and the final published version was David Sally and Elisabeth Hill, "The development of interpersonal strategy: Autism, theory-of-mind, cooperation and fairness", Journal of Economic Psychology, 27(1) 73-97, 2006):
Mentalising is assumed to be involved in decision-making that is necessary to social interaction. We investigated the relationship between mentalising and three types of strategic games – Prisoners’ Dilemma, Dictator and Ultimatum – in children with and without autistic spectrum disorders. Overall, the results revealed less dramatic differences than expected among the normally developing age groups and the children with autism, suggesting that in these laboratory tasks, mentalising skills are not always necessary. There were, nonetheless, some important findings. Young children were more cautious about initiating cooperation than their older peers and, in bargaining situations, they were less generous in their opening unilateral grants and over-solicitous of an empowered receiver. Participants with autism did have a harder time shifting strategy between versions of the Prisoners’ Dilemma, and they were much more likely to accept low initial offers in the Ultimatum game and to refuse fair proposals. In addition, participants’ measured mentalising abilities explain intentional and strategic behaviour within the prisoners’ dilemma and the avoidance of unsuccessful ultimatum proposals.
The differences between ASD children and neurotypical controls were significant, but not explainable in terms of a difference in rationality (nor do Hill and Sally attempt any such explanation). For example, the distribution of offers in the Ultimatum Game looked like this:
The ASD kids were more likely to offer 0, 6, and 7 units than their non-ASD peers — but none of those choices are dictated by rational-choice game theory. And similarly for the acceptances:
In both cases, the explanation that Hill and Sally offers is not that the ASD children were more rational than the others, but rather than they were less likely to be able to model the mental state of others. The main evidence is the distribution of Ultimatum Game behaviors as a function not of ASD diagnosis, but of score on a test of "mentalising" ability:
In this case, the conclusions of game-theoretic "rationality" do overlap in part with the results of failure to model the likely reactions of others — but this is not because the poor mentalizers are more rational. In some respects, the poor mentalizers were even further from the "rational" standard than the good mentalizers were. I'd be willing to bet, in fact, that low scores on the "second-order false belief test" would usually correlate, especially developmentally, with low scores on test of abstract logical reasoning.
So along with the extended or figurative use of Asperger's Syndrome to mean something like "rude and cruel", we seem to be seeing an extended sense of autistic to mean "excessively rational". This is all part of the normal — and irrational — process of semantic drift. Unfortunately, like the similar drift in meaning of spastic, it's rude and cruel to the individuals in the set originally referenced.
[It's also worth noting that the word autism as a technical term has also undergone a century of semantic drift, and that the current phrase "autism spectrum" itself may refer to a collection of things that will turn out to have nothing more in common than a slightly more systematic application, by psychologists and psychiatrists, of a somewhat better-informed family-resemblance process… ]