Evaluation of the Triple Code Model of numerical processing-Reviewing past neuroimaging and clinical findings.
The Triple Code Model needs a tune-up—math deficits often lie in the bridges between symbols, words, and quantities, not in the quantity sense itself.
01Research in Context
What this study did
Siemann et al. (2018) read every brain and behavior paper on the Triple Code Model. This model says we do math with three brain codes: seeing digits, saying number words, and picturing quantities. They pulled together past studies on kids with developmental delay and dyscalculia to see if the model still holds.
The review is narrative, not a meta-analysis, so the team simply described patterns and gaps they noticed across many years of work.
What they found
The Triple Code Model gets only mixed support. Some kids with math trouble do show weak quantity sense, but others show normal quantity sense and struggle only when symbols like 5 or + must be linked to meaning.
The authors say the model is too simple and needs new parts to explain why different kids fail at different math steps.
How this fits with other research
Schwenk et al. (2017) meta-analysis backs the review: symbolic speed, not quantity sense, best flags math difficulty. Their pooled data fit the review’s call to split symbolic from non-symbolic paths.
Peters et al. (2020) seems to clash: they found no quantity deficit in dyscalculia at all—only weak spatial skills. The difference is method. Lien used strict group matching and spatial tasks, while older papers cited by Julia mixed many test types. Once you control for spatial ability, the pure quantity effect fades, so both papers can be true.
Defever et al. (2013) and Brankaer et al. (2011) already showed kids with math or mild ID struggle most when digits must meet dots. Julia folds these early access-deficit findings into the newer story: trouble lives in the wiring between codes, not inside the quantity code itself.
Why it matters
For your next assessment, test both symbolic speed and spatial skills instead of relying on a single quantity comparison task. If a child is slow on digit-digit tasks but fine on dot arrays, target symbol-quantity links in intervention, not counting objects. The Triple Code Model is still useful as a map, but treat its paths as separate skills you can train.
Get CEUs on This Topic — Free
The ABA Clubhouse has 60+ on-demand CEUs including ethics, supervision, and clinical topics like this one. Plus a new live CEU every Wednesday.
Add a two-minute digit-digit comparison speed probe to your math assessment battery and note if the child is much slower than on dot arrays.
02At a glance
03Original abstract
UNLABELLED: This review reconciles past findings on numerical processing with key assumptions of the most predominant model of arithmetic in the literature, the Triple Code Model (TCM). This is implemented by reporting diverse findings in the literature ranging from behavioral studies on basic arithmetic operations over neuroimaging studies on numerical processing to developmental studies concerned with arithmetic acquisition, with a special focus on developmental dyscalculia (DD). We evaluate whether these studies corroborate the model and discuss possible reasons for contradictory findings. A separate section is dedicated to the transfer of TCM to arithmetic development and to alternative accounts focusing on developmental questions of numerical processing. We conclude with recommendations for future directions of arithmetic research, raising questions that require answers in models of healthy as well as abnormal mathematical development. WHAT THIS PAPER ADDS: This review assesses the leading model in the field of arithmetic processing (Triple Code Model) by presenting knowledge from interdisciplinary research. It assesses the observed contradictory findings and integrates the resulting opposing viewpoints. The focus is on the development of arithmetic expertise as well as abnormal mathematical development. The original aspect of this article is that it points to a gap in research on these topics and provides possible solutions for future models.
Research in developmental disabilities, 2018 · doi:10.1016/j.ridd.2017.11.001