an autodidact meets a dilettante…

‘Rise above yourself and grasp the world’ Archimedes – attribution

dyslexia is not one thing 3: problems with automaticity

leave a comment »

Q Canto: So the next hypothesised  basic source of dyslexia is ‘a failure to achieve automaticity’, that’s to say the sort of rapid, more or less unconscious processing of sounds into letters and vice versa, which probably means effective connection between brain regions or structures.

Jacinta: Perhaps because one of the structures is somehow internally dysfunctional.

Canto: wYes, and it often begins with vision. Researchers have found that many dyslexic individuals couldn’t separate two rapidly succeeding visual flickers as clearly as other individuals – an apparent processing problem. Similar research with dyslexic children found that, though they could identify stimuli initially as well as the non-dyslexic, they fell behind with added complexity and speed. This occurred more or less equally whether the stimuli were aural or visual. The connections just didn’t come ‘naturally’ to them.

Jacinta: So what about the connection between language – I mean speech, which is tens of thousands of years old – and reading and writing, a much newer development for our brains to deal with? Do dyslexic people have problems with processing good old speech? Are they slower to learn to talk?

Canto: Yes, a good question. Wolf describes research in which children with dyslexia in a number of languages, including English, ‘were less sensitive to the rhythm in natural speech, which is partly determined by how the sounds in words change through stress and ‘beat patterns’’ (Wolf, p177). Others have found breakdowns in processing in various motor tasks involving hearing and seeing. That’s to say, in the automaticity of such tasks. One psychologist who studies dyslexic children found an extensive range of problems with processing speed, especially a time gap or asynchrony between visual and auditory processing, and this observation has become commonplace.

Jacinta: But does this relate specifically to learning to speak? I’ve heard that Einstein was slow at that as a child.

Canto: Yes it’s said that he didn’t learn to speak full sentences before the age of five. But here we’re just talking about ‘naming speed’, and how it appears to use the same neurological structures as reading, as problems with one is predictive of problems with the other.

Jacinta: And the problem isn’t so much with naming per se, but the speed, the gap.

Canto: Yes, the lack of automaticity. Neurologists working in this field have developed ‘rapid automised naming’ (RAN) tasks which have become the most effective predictors of reading performance, regardless of language. Wolf herself has developed a refinement, rapid alternating stimulus (RAS), which, as the name suggests, gives more weight to attention-switching automaticity. Here’s an interesting quote from Wolf:

If you consider that the whole development of reading is directed toward the ability to decode so rapidly that the brain has time to think about incoming information, you will understand the deep significance of those naming speed findings. In many cases of dyslexia, the brain never reaches the highest stages of reading development, because it takes too long to connect the earliest parts of the process. Many children with dyslexia literally do not have time to think in the medium of print.

Jacinta: It makes me think of the unconscious, but not the Freudian one. A processing that you don’t have to think about. So that you can think about the info, not the form that encapsulates it.

Canto: Yes, and none of this explains why some have these problems with automaticity – which brings us back to neurology. Are dyslexic individuals using a different circuit from the rest of us, and does this explain their skills and abilities in other areas?  Remember the names – Einstein, da Vinci, Gaudi, Picasso… not that dyslexia guarantees genius or anything…

Jacinta: Yes, far from it, I’d say, but it’s a fascinating conundrum.

Canto: So, neurology. And this takes us to how the ‘reading brain’, a very new phenomenon, evolutionarily speaking, came into being. fMRI images appear to confirm hypotheses that the brain ‘uses older object recognition pathways in the occipital-temporal zone (area 37) to name both letters and objects’ (Wolf, p179). It’s a process described as ‘neuronal recycling’. And it takes us to brain regions associated with particular tasks. For example, the left occipital-temporal area is apparently more associated with object naming, a much older task, evolutionarily speaking, than letter naming, and one that takes up more cortical space. The more streamlined, specialised use of this region for letters, and the development of automaticity for that purpose, is a prime example of our much-vaunted neuroplasticity.

Jacinta: What they’ve called RAN is always faster for letters than objects – that’s perhaps because letters are a small, even quite tiny subset of the near-infinite set of objects.

Canto: Yes, and here I’m going to quote a difficult passage by Wolf at  some length, and then try, with your help, to make sense of it:

…culturally invented letters elicit more activation than objects in each of the other ‘older structures’ (especially temporal-parietal language areas) used for reading in the universal reading brain. This is why measures of naming speed like RAN and RAS predict reading across all known languages. It is also why, side-by-side, the brain images of the object- and letter-naming tasks are like comparative evolutionary photos of a pre-reading and post reading brain (Wolf, p181).

Jacinta: So this is a bit confusing. Culturally invented letters are new, evolutionarily speaking. And there are older language structures used for reading. Repurposed? Added onto? A bit of renovation? And what exactly is ‘the universal reading brain’?

Canto: Good question, and a quick internet research reveals much talk of a ‘universal reading network’. Here’s a fascinating abstract from a 2020 study, some ten years after the publication of Wolf’s book. It’s entitled “A universal reading network and its modulation by writing system and reading ability in French and Chinese children”:

Are the brain mechanisms of reading acquisition similar across writing systems? And do similar brain anomalies underlie reading difficulties in alphabetic and ideographic reading systems? In a cross-cultural paradigm, we measured the fMRI responses to words, faces, and houses in 96 Chinese and French 10-year-old children, half of whom were struggling with reading. We observed a reading circuit which was strikingly similar across languages and consisting of the left fusiform gyrus, superior temporal gyrus/sulcus, precentral and middle frontal gyri. Activations in some of these areas were modulated either by language or by reading ability, but without interaction between those factors. In various regions previously associated with dyslexia, reading difficulty affected activation similarly in Chinese and French readers, including the middle frontal gyrus, a region previously described as specifically altered in Chinese. Our analyses reveal a large degree of cross-cultural invariance in the neural correlates of reading acquisition and reading impairment.

So this research, like no doubt previous research, identifies various brain regions associated with reading ability and impairment, and finds that the same automacity, or lack thereof, is associated with the same regions, such as the middle frontal gyrus, in both alphabetic and ideographic reading systems. I think this is further confirmation of the research work Wolf is citing. Of course, I don’t know much about these brain regions. A course in neurology is required.

Jacinta: But what Wolf appears to be saying in that earlier quote is that you can get brain images (via fMRI) of object naming (older brain) tasks and put them side by side with images of letter naming tasks (younger brain), and it’s like seeing the results of evolution. Sounds a bit much to me. I suppose you can see a different pattern. Isn’t fMRI based on the magnetism of iron in the blood?

Canto: Yes yes. This is complex, but of course it’s true that the neural networking required for reading and writing is much more recent than that for language – and remember that of the 7000 or so languages we know of, only about 300 have a written form, which suggests that the Aborigines, before whities arrived, and the Papua-New Guineans, who have about 700 different languages on their island, were unable to even be dyslexic, or were all dyslexic without knowing it, or giving a flying fuck about it, because they had no writing, and no wiring for reading it.

Jacinta: So it would be interesting, then, to scan the brains of those language users – and there are no humans who aren’t language users – who don’t have writing. Take for example the Australian Aborigines, who became swamped by white Christian missionaries determined to ‘civilise’ them, more or less overnight in evolutionary terms, through teaching them to read and write. And then would’ve been characterised as backward for not picking up those skills.

Canto: That’s an interesting point, but it’s the same even in ‘cradles of civilisation’ such as Britain, where the vast majority were illiterate, and encouraged to be so, 500 years ago. At that time the printing press was a new-fangled device, church services were mostly conducted in Latin, and it was convenient to keep the peasantry in ignorance and in line. And yet, when it became more convenient to have a literate population, the change appears to have been relatively seamless, dyslexia notwithstanding. So it seems that, from a neurological perspective, little change was required.

Jacinta: Yes, that’s a good point, and it points to brain plasticity. Curiouser and curiouser – so it’s not so much about evolution and genes, but relatively rapid neural developments…. to be continued…


M Wolf, Proust and the squid, 2010

Written by stewart henderson

April 21, 2023 at 5:02 pm

Leave a Reply

%d bloggers like this: