Specific learning disorder in mathematics: prevalence, overlap and autism

I'm not going to spend too long discussing the findings reported by Kinga Morsanyi and colleagues [1] on the topic of specific learning disorder in mathematics (SLDM) but they are potentially important. Important because, alongside some [DSM-5] diagnostic definition changes to the issue of learning disorder with maths in mind (see here), something stood out from the Morsanyi results: "About half of the children with an SLDM profile had some form of language or communication difficulty. Some of these children also had a diagnosis of autism, social, emotional, and behavioural difficulties or attention deficit and hyperactivity disorder."

Surveying some 2400 primary school children here in Blighty, researchers analysed results based on various "standardized tests of mathematics, English, and IQ." They also looked at "several demographic factors over the primary school years" and subsequently applied the DSM-5 diagnostic criteria for specific learning disorder (SLD) to their cohort.

The found that about 6% of children met the diagnostic thresholds for SLDM. Sex/gender didn't seem to make any difference to the findings, but as with many things these days (see here), SLDM was typically not a 'stand-alone' diagnosis.

The 'language or communication difficulty' bit to the Morsanyi findings is perhaps not unexpected; I'll mention that again in a bit. The 'lumping' together of various SLDs covering reading, writing and maths skills in the DSM-5 is perhaps justified to this extent. Of more interest to this blog is the idea that developmental disorders, particularly autism, are also not uncommon alongside SLDs.

Without wishing to conflate a diagnosis of SLD with anything else, I'd like to take readers back to a post I wrote some time ago on how maths ability - particularly maths problem solving ability - in the context of autism is as varied as the label itself (see here). The sweeping generalisation that many/all those on the autism spectrum somehow have some savant ability when it comes to maths is brought crashing down by cold, objective science [2]. Science that observed that nearly a quarter of those on the autism spectrum "evidenced a mathematics learning disability" compared with around 4% who "exhibited mathematical giftedness." That's quite a disparity.

What can we learn from such data? Well, preferential screening is perhaps the first lesson to be learned. Ensuring that sweeping generalisations and stereotyping about autism in particular ("what's his/her special ability?") are replaced by appropriate objective testing to determine strengths and weaknesses in areas such as maths ability. Next, a greater appreciation that SLDs can overlap with one and another [3] might also be useful. So, issues with reading and writing might also be over-represented among those with an SLD in relation to maths ability as per what the Morsanyi data implied. Finally, intervention needs to be further examined: what can we do to aid a learner with autism presenting with SLDM? I'm no expert in this area, but understand that there are groups looking at this issue (see here). Obviously such interventions need to be mindful of the diagnosis of autism and other potential comorbidities and what that could mean for learners, but I'm assuming that there are going to be individualised strategies that work as well for those on the autism spectrum as they do for those not on the autism spectrum. This includes ensuring that maths anxiety is tackled too.

Oh, and just before I go, a friendly reminder: maths is probably not just a systemising-linked ability (see here and see here).

----------

[1] Morsanyi K. et al. The prevalence of specific learning disorder in mathematics and comorbidity with other developmental disorders in primary school-age children. Br J Psychol. 2018 Jul 5.

[2] Oswald TM. et al. Clinical and Cognitive Characteristics Associated with Mathematics Problem Solving in Adolescents with Autism Spectrum Disorder. Autism Res. 2016 Apr;9(4):480-90.

[3] Moll K. et al. Specific learning disorder: prevalence and gender differences. PLoS One. 2014 Jul 29;9(7):e103537.

----------

"Systemisers are better at maths". But be careful of sweeping generalisations...

The title of this post - "systemisers are better at maths" - comes from the findings reported by Paola Bressan [1], an author not unfamiliar to this blog (see here). This time around, the author has opted for a topic a little less 'controversial' [2] than last time, in the form of an analysis of the 'empathising-systemising theory'. In particular, the notion that mathematics is typically a systemising activity (based on rules and laws) and so those who lean more towards being systemisers should be better at maths. That's the theory anyway...

The word 'autism' is mentioned in the Bressan paper because of the *connection* made between systemising and autism by some big names in the autism research field. As far as I can see however, the Bressan study did not include anyone diagnosed with autism or an autism spectrum disorder (ASD) but rather relied on data from mostly university students studying "mainly psychology and engineering." Said student participants (N=201) were given a couple of measures to test maths skills (including a 'how good are you at maths?' self-report measure) alongside tests for non-mathematical intelligence and 'systemising tendencies' in the form of the Italian version of the Systemising Quotient (SQ-Short).

Results: the data for the study are freely available to anyone and everyone (see here). Bressan observed a few details. So: "Systemising was predicted by sex, with men scoring on average higher than women." Occupation, I assume meaning what participants were studying, also showed a potential relationship with systemising scores too. Then to the main event: "In a sample of students of a diverse range of disciplines, the systemising score predicted both self-assessed mathematical skills (201 participants) and mathematical intelligence (151 participants). This remained true after controlling for nonmathematical intelligence, for occupation (social vs biological/physical careers), and for sex."

These are interesting results. They offer some support for the idea that cognitive style *might* be an important factor when it comes to 'success' or not in certain types of discipline, and require further study. But...

Allowing for my 'no sweeping generalisations are required' sentiments that are typically included with my analysis of research on this blog, I'm still a little reserved on this topic. I'm reserved because on the previous occasion that I've covered maths achievement and systemising (see here), the results were a little different. The paper by Escovar and colleagues [2] for example, reported that social skills seemed to be more important to maths ability. They also observed that "mathematical achievement was not related to systemizing or the discrepancy between systemizing and empathizing" in their study. Bressan criticised this study for a potential lack of statistical power, but I'm slightly less critical of those results, particularly because Escovar et al utilised a wider range of study instruments covering a wider range of potentially important variables. I'm also inclined to mention how maths ability can sometimes be clouded by other factors. Take for example another post I wrote on maths problem solving ability and autism (see here) and how issues like attention and anxiety can play a role on performance. That and also the fact that not everyone on the autism spectrum is some sort of maths genius...

----------

[1] Bressan P. Systemisers are better at maths. Sci Rep. 2018 Aug 2;8(1):11636.

[2] Escovar E. et al. The Empathizing-Systemizing Theory, Social Abilities, and Mathematical Achievement in Children. Sci Rep. 2016 Mar 14;6:23011.

----------

Methylphenidate affecting academic performance meta-analysed

When first tweeting about the paper by Anne Fleur Kortekaas-Rijlaarsdam and colleagues [1] I kinda made a mistake. The text of the tweet read something like: "Does methylphenidate improve academic performance? A systematic review and meta-analysis... Pretty much so, but only in relatively small amounts."

Reading that tweet back to myself I realised that by saying 'but only in relatively small amounts' it's highly likely that any readers might have thought that I was talking about the dose of methylphenidate (MPH) rather than the [intended] effect on things like "math productivity (7.8% increase, p < .001); math accuracy (3.0% increase, p = .001); [and] increased reading speed (SMD .47, p < .001)." I apologise, and once again reiterate my blogging (and social media) caveat about not giving anything that looks, sounds or smells like medical or clinical advice.

I did think it worthwhile to write a short post about the Kortekaas-Rijlaarsdam findings given that (a) ADHD (attention-deficit hyperactivity disorder) is something that turns up quite a bit on this blog (see here for example), and (b) pharmacotherapy 'for ADHD' is an important intervention area which has also, unfortunately under some circumstances, been 'hijacked' for other purposes (see here). On that last point, I specifically refer to the idea that for some people, striving for a diagnosis of ADHD is merely a route to either better academic accommodations or perhaps more worryingly, to gain access to a class of medicines with some potentially significant 'cognitive-enhancing' qualities...

What sets the Kortekaas-Rijlaarsdam findings apart from other reviews of the potential nootropic (cognitive enhancing) abilities of something like MPH is their focus on resolving the issue of "whether there are improvements in core academic skills or just improvements in academic productivity" in the context of "the mediating or moderating effects of symptom improvements, demographic-, design- and disorder-related variables."

From a starting number of 148 full-text research articles, the collected texts were screened for eligibility and some 34 were eventually included for meta-analysis ("quantitative synthesis"). These articles were included because they "provided information about either accuracy or productivity scores for math, reading or spelling, or a combination of these" or could at least be calculated on the basis of their included data. Various mediating and moderating variables were also thrown into the statistical mix - "age, gender, percent diagnosed with ADHD-inattentive subtype, and study characteristics: release system, trial duration, and titration method" - and results reported.

As per previous sentences, various aspects of maths performance showed an improvement that correlated with MPH use, and reading speed but not accuracy also came out as potentially showing a relationship with medicine usage. Important too was the information that: "None of our mediators or moderators influenced MPH effects on math and reading accuracy or productivity."

So, following meta-analytic scrutiny of some 1700 children, it looks like there is a small but potentially relevant effect from MPH use of some academic abilities, at least in the short-term ("between 1 and 7 days"). The authors do well not to stray too much into speculating mode when it comes to the cause of any nootropic effect but instead to include the need for more research "to isolate groups of patients who may benefit more or less from MPH and to reveal its mechanism of action." Can't argue with that.

And going back to the topic of cognitive enahncers, an interesting article from a few years back on what they may or may not be doing to elements of our University student body...

Music to close... Sia and Chandelier. Seemingly always playing outside my dojo....

----------

[1] Kortekaas-Rijlaarsdam AF. et al. Does methylphenidate improve academic performance? A systematic review and meta-analysis. Eur Child Adolesc Psychiatry. 2018 Jan 20.

----------

Language and motor skills: preschool predictors of academic achievement in autism

A fairly quick post for your reading delight today as I bring the paper by Miller and colleagues [1] to your attention suggesting that: "Early intervention targeting language and motor skills may improve later achievement in this population."

'This population' referred to a small cohort (N=26) of children diagnosed with an autism spectrum disorder (ASD) who were examined "at the approximate ages of two, four, and ten" years with regards to their academic achievement and the variables that might be important to 'successful' achievements.

Including some familiar names when it comes to the concept of 'outcome' in relation to autism ('optimal outcome' that is), researchers determined a few potentially important relationships from their collected data: "Preschool verbal abilities significantly predicted school-age reading comprehension" and "early motor functioning predicted later math skills."

I'm not entirely surprised that infancy verbal (talking) abilities might play a role in later reading comprehension but I was rather more intrigued by the observation potentially linking motor skills to later maths abilities. Yes, I get that children learn to count on their fingers (and toes) and no doubt this and other scenarios might influence the connection between the two, but it strikes me that this connection requires quite a bit more study [2]. Indeed, welcoming the idea that motor skills are an important issue with regards to autism (see here) and that maths ability is 'as varied as the label of autism is itself' (see here) the idea that the archetypal all-rounder that is the occupational therapist (OT) might have a key role here is rather interesting (see here).

I'm also minded to suggest that a certain sport/discipline that I'm particularly fond of on this blog (the martial arts) might have some rather far-reaching 'mathematical' effects if one considers it's application to autism and motor functioning...

To close, Marvel are going full-strength with their next Wolverine film instalment titled 'Logan'...

----------

[1] Miller LE. et al. Preschool predictors of school-age academic achievement in autism spectrum disorder. Clin Neuropsychol. 2016 Oct 5:1-22.

[2] Pitchford NJ. et al. Fine Motor Skills Predict Maths Ability Better than They Predict Reading Ability in the Early Primary School Years. Front Psychol. 2016 May 30;7:783.

----------

Miller LE, Burke JD, Troyb E, Knoch K, Herlihy LE, & Fein DA (2016). Preschool predictors of school-age academic achievement in autism spectrum disorder. The Clinical neuropsychologist, 1-22 PMID: 27705180

Is empathising rather than systemising linked to maths achievement?

"Contrary to our hypothesis, we found no relationship between systemizing and math achievement after controlling for domain general abilities and no relationship between the systemizing brain type (greater discrepancy between systemizing and empathizing) and math achievement."

That quote taken from the study published by Emily Escovar and colleagues [1] (open-access available here) provides some blogging fodder today. Based on some ideas proposed in autism research circles that "mathematics is purported to be an example of an ability requiring systemizing" and some rather sweeping claims about facets of autism being explained "as a high tendency to systemize" researchers set about testing the idea that being a good systemiser might make for good maths ability. I might add that previous research has demonstrated that a diagnosis of autism (and comorbidities) does not necessarily equal 'maths genius' (see here).

Drawing on data from a cohort of 112 "typically developing children" (authors words not mine) aged between 7 and 12 years old, researchers set about assessing various functions. "Intelligence was assessed using the full-scale intelligence quotient (FSIQ) of the Wechsler Abbreviated Scale of Intelligence (WASI)." Various facets of maths ability were also included for study as a function of the use of "the Woodcock Johnson III, Form A" and specifically "the Calculation, Math Fluency, and Applied Problems subtests as measures of math achievement." Researchers also sensibly included a measure of maths anxiety - the Scale for Early Mathematics Anxiety - as part of their protocol. Alongside the use of the Social Responsiveness Scale (SRS) to provide some background on presented autistic-like traits, the main event was the that: "The primary guardian of each child completed the Combined Empathy Quotient-Child (EQ-C) and Systemizing Quotient-Child (SQ-C)."

Results: from the point of view of proponents of the whole systemising-empathising thing with gender in mind, there was some good news: "There was a marginally significant gender difference on SQ-C, with boys scoring higher than girls. In contrast, boys and girls differed significantly on EQ-C with girls scoring higher than boys." This follows other work in this area (see here). But it was not all one-way traffic as per the opening statement to this post and the conclusion that "SQ-C is not an independent predictor of math achievement in children."

One particularly interesting detail did emerge from the study findings: "results demonstrate that empathizing is a previously unknown predictor of math skills in TD [typically developing] children." This was specifically based on the results of the Calculation skills subtest but was independent of gender (as an effect) and neither was explained by maths anxiety scores. Some further analysis taking into account SRS scores was undertaken and the idea that "children with higher social abilities tended to have lower math skills" emerged. Finally, just in case you might think that 'autistic behaviours' might be the driving force between empathising and maths ability, the authors have something to say about this too: "the link between empathizing and math achievement may be related to social awareness and cognition rather than autistic behaviors."

I am cautious that the Escovar findings do require replication and further investigation before any grand claims are made about autism, empathising/systemising and maths ability; not least because of the cohort used and, although a decent sized number, the requirement for even greater participant numbers. That other categories of behaviour can seemingly affect maths performance should also not be forgotten [2]. That being said, I'd like to think that there could be some really important issues to come from such work if the relationships proposed holds out. Not least is the idea that reinforcing empathising skills *might* have a knock-on effect for something like maths ability and perhaps equally important, the possibility of a downside to being a classroom 'social butterfly' from a maths perspective...

----------

[1] Escovar E. et al. The Empathizing-Systemizing Theory, Social Abilities, and Mathematical Achievement in Children. Sci Rep. 2016 Mar 14;6:23011.

[2] May T. et al. The role of attention in the academic attainment of children with autism spectrum disorder. J Autism Dev Disord. 2013 Sep;43(9):2147-58.

----------

Escovar E, Rosenberg-Lee M, Uddin LQ, & Menon V (2016). The Empathizing-Systemizing Theory, Social Abilities, and Mathematical Achievement in Children. Scientific reports, 6 PMID: 26972835

Maths problem solving ability and autism: as varied as the label itself

I'm bringing the paper by Tasha Oswald and colleagues [1] to the blogging table today and a reiteration of the important idea that within the vast heterogeneity that surrounds the presentation of autism (or even the autisms) so too there is quite a lot of variability in the presentation of strengths and weaknesses; indeed, not so dissimilar to life in general.

Mathematics problem-solving was the specific area that Oswald et al have zoomed in on, with the aim of assessing: "the relative proportions of youth with ASD who demonstrate giftedness versus disability on applied math problems."

Researchers reported that of the 27 "high-functioning adolescents with ASD [autism spectrum disorder]" (their words not mine), just over a fifth of participants (22%) "evidenced a mathematics learning disability" whilst only 4% presenting with "mathematical giftedness." These results, by the way, were based on "standardized measures of math problem solving, perceptual reasoning, verbal ability, and test anxiety." When it came to predicting which factors might be relevant to maths problem solving ability, the authors reported a sliding scale: "perceptual reasoning, followed by verbal ability and test anxiety, then diagnosis of ASD" taking into account results from age and cognitively matched controls (n=27) also.

Outside of the idea that not everyone diagnosed with an ASD is Rainman (indeed that whilst autism is over-represented in savant skills, this does not mean that autism = savant [2]), there are some interesting lessons to be potentially learned from the quite preliminary Oswald findings. Not least is the suggestion that results "highlight possible targets of intervention for students with ASD struggling with mathematics" based on those proposed predictors of maths ability and where further educational focus might be directed. Indeed, the specific idea that 'test anxiety' might for example be a factor, ties in well with other peer-reviewed evidence hinting at how the effects of different types of anxiety may be far-reaching when it comes to autism (see here).

I might also forward the idea that whilst autism was the focus on the Oswald data, one should perhaps be mindful of how other comorbidity can similarly affect academic attainment. May and colleagues [3] talked about how aspects of attention could very well affect skills such as mathematics performance with autism in mind and specifically how: "attentional switching difficulties are linked with poorer mathematics outcomes in ASD." Without hopefully confusing attentional switching with something like a diagnosis of attention-deficit hyperactivity disorder (ADHD) (although there is overlap), I will bring in the important data hinting that ADHD difficulties may indeed be over-represented when it comes to autism (see here) and what impact this might have on maths abilities in at least some cases of autism. Further, the question of whether 'tackling' said ADHD might have some interesting effects on areas like maths ability with autism in mind, bearing in mind pharmacotherapy might not be the only suggested avenue?

I'm gonna close with reference to the paper by De Marco and colleagues [4] detailing just how impressive the human mind can be...

----------

[1] Oswald TM. et al. Clinical and Cognitive Characteristics Associated with Mathematics Problem Solving in Adolescents with Autism Spectrum Disorder. Autism Res. 2015 Sep 29.

[2] Treffert DA. & Rebedew DL. The Savant Syndrome Registry: A Preliminary Report. WMJ. 2015 Aug;114(4):158-62.

[3] May T. et al. The role of attention in the academic attainment of children with autism spectrum disorder. J Autism Dev Disord. 2013 Sep;43(9):2147-58.

[4] De Marco M. et al. Brief Report: Two Day-Date Processing Methods in an Autistic Savant Calendar Calculator.  J Autism Dev Disord. 2015. Oct 17.

----------

Oswald TM, Beck JS, Iosif AM, McCauley JB, Gilhooly LJ, Matter JC, & Solomon M (2015). Clinical and Cognitive Characteristics Associated with Mathematics Problem Solving in Adolescents with Autism Spectrum Disorder. Autism research : official journal of the International Society for Autism Research PMID: 26418313