Sunday, August 14, 2011

On neural correlates and causation

The advent of neuroimaging techniques such as magnetic resonance imaging (MRI) has revolutionized autism research. We can now look into the brain and see the "neural correlates" of autism. But, as with any form of correlation, identifying a neural correlate doesn't necessarily mean that we have identified a neural cause.

A case in point. Earlier this week I stumbled across a press release doing the rounds of the internet, proclaiming that "Brain imaging research reveals why autistic individuals confuse pronouns". Pronouns are the words like "he", "she", "you" and "I" that can stand in for real names. Kids with autism often struggle with them (there goes another one). In particular, they'll say "you" to refer to themselves and "I" to refer to other people.

Various theories have been put forward over the years to try and explain pronoun "reversal". Leo Kanner thought it happened just because the autistic kids were echoing things other people had said. Bruno Bettelheim (he of 'refrigerator mother' fame) reckoned kids with autism didn't have a sense of self, and so "you" and "I" were indistinguishable to them. An intriguing theory, proposed more recently by Yuriko Oshima-Takane is that kids with autism don't learn how pronouns work because they don't attend to other people's conversations.

So what does brain imaging add to this debate?

The study

The study was conducted by Akiko Mizuno, a graduate student working with Marcel Just at Carnegie Mellon Uni. She tested a group of 15 high-functioning adults with autism on what is known in the trade as a first-order visual-perspective-taking task. On each trial, they saw a series of photographs in which a woman (called Sarah) first showed them a card with different pictures on each side and then asked "What can you see now?" or "What can I see now?" Participants had to press a button on the left or right to give the correct answer.

Interpreting Sarah's questions required the participants to comprehend the pronouns "you" and "I". The adults with autism were slower and less accurate at this task than non-autistic adults. They were also a little slower on control questions that didn't involve pronouns, such as "What can Sarah see?" and "Who can see the carrot?" but the group differences weren't quite as marked. This is crucial because it suggests that the adults with autism had specific problems with the pronoun condition.

These results in themselves are really interesting. They suggest that subtle difficulties with pronouns are apparent, even amongst high functioning adults with autism. It's not clear whether these individuals ever reversed pronouns themselves in their speech, and it's important to remember that the study looked at comprehension of pronouns rather than production. But it's nevertheless striking that there are group differences, even on such a simple task.

The focus, however, was on the brainy stuff.

While the participants were completing the task, their brains were being scanned using fMRI. The headline finding was that, in the autism group, there was reduced "connectivity" between two brain regions, the right anterior insula and the precuneus. Furthermore, within the autism group, there was a significant correlation between brain connectivity and reaction time. People who were slower had weaker connectivity.

Mizuno et al. imply that this is what ultimately causes pronoun reversal:
"The observed lower functional connectivity between those two neural nodes in the autism group, therefore, may result in disturbed perspective-taking processes in shifting a centre of reference between self and other."
Finer grained analyses showed that group differences in "connectivity" were observed only when Sarah asked "what can you see?" and not when she asked "what can see?". The author's explanation is as follows:
"These findings indicate that the critical disturbance... may be dysfunctional processing when recognizing the self as a referent of ‘you’, and shifting to map self onto the pronoun ‘I’."
In other words, when Sarah says "What can you see?", the participant has to translate that into "What can I see?" and this translation process is reliant on the "connectivity" between the precuneus and anterior insula [1].

Reasons to be cautious:

It's possible that Mizuno and colleagues are correct in their interpretation. In fact, I'd really like them to be right, because I've been waffling on about brain connectivity in autism for ages. But there are a number of reasons to query their conclusions.

1.  Everyone uses the term "functional connectivity" in the context of fMRI scans, but it's pretty misleading. fMRI measures brain activity indirectly via changes in blood oxygen levels. Here's an example of the time course of oxygen level changes for a control participant in Just et al.'s original fMRI "connectivity" in autism study.

The two brain regions in this figure are considered to be "functionally connected" because their activation goes up and down at roughly the same time. What isn't obvious from the figure (and is rarely acknowledged) is the fact that the changes are happening really slowly - roughly one cycle of activation and deactivation every trial.

If the two brain regions are 'talking' to each other in order to complete the pronoun task, they're doing it a much faster rate than anything fMRI can hope to measure.

2.  Since that first paper, Just and colleagues (as well as several other research groups) have published a large number of studies demonstrating changes (usually reductions) in "functional connectivity" throughout the autistic brain. Their new study adds to this impressive body of evidence. But this in turn raises a second concern.

As mentioned before, Mizuno et al. looked at connectivity between two brain regions - the right anterior insula and the precuneus. Importantly, this was the only pair of regions they considered looked at [2]. Based on their previous findings, there's a fair chance that they could have chosen any number of brain regions and would have found "underconnectivity" between them too. They may be right and this is the only connection that relates to pronoun comprehension difficulties. But we don't know this.

3.  The claim is that differences in "connectivity" are responsible for difficulties in comprehending pronouns. But it could just as easily be the other way around. People with autism struggle to comprehend pronouns, so they have to work harder or (as the reaction time data suggests) for longer, so it's no surprise that their brain activity while they're doing the task is different.

Brains vs Minds

Neuroimaging studies have provided many important insights into the workings of autistic brains. But sometimes, it's easy to be seduced by the fancy gadgets, the pretty pictures, and the funny words and think that brain imaging is somehow more scientific than good old-fashioned cognitive psychology (as exemplified by Mizuno et al.'s reaction time data), or that it offers privileged insights into the autistic mind. 

Neural correlates are just that - correlations. All the usual caveats apply.


[1].  Unfortunately, Mizuno et al. don't report whether the same effect is apparent in the reaction time data. If the people with autism had specific difficulty comprehending the word "you", they should be slower on this condition than control participants.

[2].  As far as I can tell, there is no direct evidence from previous studies that the precuneus and right anterior insula are involved in pronoun comprehension, so effectively Mizuno et al. are relying on a hunch. And in their own analyses, they show that, while the right anterior insula is one of 7 brain regions activated by the task, the precuneus isn't.


Mizuno A, Liu Y, Williams DL, Keller TA, Minshew NJ, & Just MA (2011). The neural basis of deictic shifting in linguistic perspective-taking in high-functioning autism. Brain : a journal of neurology PMID: 21733887

Further reading:


  1. Thank you for providing a superb analysis of this paper. When I read the paper I was both intrigued by the the tasks Mizuno et al used, and disappointed at the somewhat predictable conclusions. I am so glad you are adding appropriate words of caution. It can't be said often enough that the imaging tools are still at a very crude level, and strong causal inferences are not appropriate. All they do is provide a hypothesis, hopefully an interesting hypothesis, which should lead to interesting predictions for new experiments.

  2. FWIW I have to make a small, perceptible conscious effort to get pronouns right because to me all stuff is the same, there is no 'me' and 'other'.

    I'm not sure your chicken/egg query 3 has useful meaning, unless you can resolve epiphenomenalism.

  3. Glad to see once more that correlation is not causation.

    The next task would be to get rid of the ghost in the machine, as Gilbert Ryle did...but perhaps I am jumping too much into conclusions. ;)

  4. Nice critique. It seems that today functional connectivity is as sexy as simple fMRI activation was 10 years ago or so. You usually can't publish a simple activation study in a high-end journal anymore, but it's still not clear that functional connectivity is always a step up.

  5. Thanks for this, I did see the article doing the rounds.

    Watching my own child struggle with pro nouns, and more often recently and spontaneously applying them appropriately, I find this intriguing. My son is almost 3 so over time we shall see if his apparent realisation of a 'you' and a 'me, sticks. It's looking promising. And as I said he picked this up with no assistance, only observation of language, and correction in the course of normal life events, meaning it is nothing we have addressed in therapy. MY sense has always been a bit like Kanner's in that he is echoing words. Perhaps this is simplistic, but it does fit with observations of my own child's speech development.

    I wonder, when they were undertaking the study using high functioning adults if they took the time to ask these individuals why they thought pro noun use was a challenge for them?

  6. Wow, that's really interesting --- both that pronoun confusion persists into adulthood (I had assumed pronoun reversal was just a phase we went through as children; trying to think if I have to expend any extra energy figuring out pronouns like nerkul above and I don't think I do ...) and that the autistic test-takers' difficulty answering the questions was *specific* to questions phrased using pronouns!

    (I found that especially interesting because this test covers some of the same ground as tests that have previously been taken to indicate a characteristically autistic failure to mentalize ... and I could see people interpreting these results this way, like, "oh, the autistic people couldn't adopt 'Sarah's' point of view, that's the problem!"

    But now I wonder, how much of the previously-documented difficulty mentalizing might really be semantic confusion?)

  7. Semantic confusion: good point!

    I listened to an informative your-tube segment where a person gave words to that possibility in a way I'd not heard expressed before.

    He spoke of practicality in teaching atypical learners. He said he, and others like him "fundamentally have difficulty in drawing inference from representational models". Picture "you". Can you? The meaning must be inferred. It is not a concrete must be reinterpreted each use.

  8. >>>"The psychology of self — the thought of one's own identity — is especially important in social interaction, a facet of behavior that is usually disrupted in autism," said Just, a leading cognitive neuroscientist and the D.O. Hebb Professor of Psychology at CMU who directs the CCBI. "Most children don't need to receive any instruction in which pronoun to use. It just comes naturally, unless a child has autism."<<<

    Sounds kinda voodoo-ish. Did I say that outloud? Sorry...

  9. It's important to remember that any internal representation of the universe is only a very poor model, nonautistic and autistic equally, and "semantic confusion" is a less useful way of thinking about it than trying to imagine an internal representation where the semantics are consistent. In fact that goes for imagining why anyone does anything. Always assume internal consistency. And don't worry about cause and effect - those are philosophical issues.

    For me the really interesting product of autism research is it helps define an angle between two human viewpoints from which we can construct a theodolite to measure aspects of nature that no one of us can measure directly. That's incredibly cool. Know the difference between autistic and nonautistic and we'll be able to see with far greater precision all human cognition, and monkeys and beavers and ants and crystaline space aliens (in that order of increased precision, I guess). We're talking giant leaps in epistemics.

    I'd appreciate it if more autism researchers were aware of their big picture reasons for doing what they do, beyond that someone gave them a grant.

  10. Pronoun reversal is common in typically developing children even in multi lingual settings.

    Even in autism pronoun reversal usually resolves over time conisistent with tranitory pronon reversal in many typicaly developing children.

  11. Thanks to all for the comments.

    RAJ: More on pronouns coming very shortly...

  12. It occurs to me a similar confusion is endemic to autism. Initially, using the Denver (?) screening, my son was sent on for testing because of his lack of understanding of prepositions. He could not put the bead "in" the bottle, nor the paper "on" the desk.

    In high school, he continued with an inability to recognize prepositions in a sentence until a visual "box" scenario was set up by a special needs teacher. He was to visualize a rabbit and a box. If he used the word and it made sense, ie, "Put the rabbit___ the box", it was a preposition.

    Thinking in pictures is just one way of understanding language. Some think in words, some in action, by doing or physically manipulating symbols, lending concrete understanding rather than fleeting representations. How this happens amazes me. How can a kinesthetic learner be articulate? But they are.

  13. Forgive me. It just occured to me how specific language is to functioning that is perceived as "impaired".

    Just as "I"-"you" confusion is psychologically determined to indicate a lack of sense of self, prepositional confusion exemplifies the idea that autistic children are "lost in space", ie, use of weighted vests and blankets to ground them, give vestibular input.

    My son also lacked (lacks?) a sense of time. I wonder if this is indicated by the language he uses. Does the language understanding our children lack give us clues as to their sensory awareness? Are they lacking input to the senses and indicating it through language?

    Time...if you are not grounded in time, constant change must make you feel dizzy, in a lack of ability to predict anything. That's why sameness would be a comfort.

    What does the lack of theory of mind tell us? What does the lack of empathy tell us? (In the supposition that both theories have merit.)

    As far as the lack of connectedness that you espouse belief in...I tend to agree. I "saw" it once in Ben.Thoughts began and then did not carry through. I figured this must happen to him hundreds of times a day. He must stop and start over continually. But yet, somehow, he is encoding and keeping enough information to have a very high IQ.

    Sorry... diarrhea of the thoughts.

  14. In am a general ed teacher and a lay person in this area. For Review and comment. Thanks, Peter

    As I understand it, a pronoun, like nouns refer to person, place, thing, idea/concept. Thus, this would be information of the mind.

    Connectivity refers to the conduits, the neural networks which pass the information of the mind within and between regions.

    Question. Do sensory event experiences cause both network development and information of the mind libraries?

    For example, I teach Lego Robotics within a social team context. Conversation between members to complete the build, program, and testing is ongoing in the I/you relationship.

    My hypothesis is that the multiple event experiences of team interaction can contribute to both network connectivity speed and library building speed re the distinguishing processing speed between concept application.