STM-Suite, an Online Platform for the Assessment of Memory Functions Discriminates among Subgroups of Children with Different Types of Specific Learning Disorders

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In summary, the manuscript describes a study investigating differences in short-term memory (STM) between groups of typically developing children (TD) and children with specific learning disorders (SLD). In general, the goal of providing an assessment tool with balancing task characteristics to measure STM could improve research on STM. However, the complex scoring procedure and ceiling as well as floor effects in tasks have not been thoroughly discussed. I have a couple questions, comments and suggestions that I hope to be helpful for improving the manuscript.

·       The main aim is described to be the development of a new test battery, but the results mostly investigate questions about differences in STM in children with learning disorders. It would be helpful for the reader to communicate better what the main goal is and how the analysis is informing this aim.

·       The description of existing batteries to measure STM is an important basis for the current study. It could make this part even stronger if the authors would add a table with an overview including strengths and weaknesses of the different tasks and their newly developed battery. Furthermore, the item selection seems to be done very thoroughly.

·       Unfortunately, I do not understand the description about the lists on page 8. Is the 3-item list without AS followed by another one without AS or one with AS?

·       The scoring used is rather complex. Did the authors compare different scoring procedures?

·       Imputation for oral-output tasks would not be necessary. Why did the authors decide to impute less reliable data instead of using observed scores? This procedure seems to introduce more unreliability without improving data quality in any other way.

·       The normalizing transformation: Ceiling and floor effects are known to be highly problematic (attenuated reliability and validity). These constrained test scores question the usefulness of the test. The distribution of the manifest scores for the different tasks are not reported making it unknown to the reader how many scores are affected. Furthermore, the usage of a normalizing transformation is only mentioned in the Appendix. It should be made clear that all scores used were not raw scores. Additionally, it is still unclear what scores are presented in the figures showing “Mean STM score” for the clusters. The ceiling and floor effects are not mentioned as a limitation of the study and the usefulness of the task battery. No reliability of scores and tasks have been reported.

·       The cluster analysis is a data driven approach based on the conducted tasks. However, children in the SLD group also have an official diagnosis. It would strengthen the results if the cluster analysis would be compared to diagnosis.

·       Why did so many children from the Typically Developing group have to be excluded due to missing data? Why was the data missing?

·       Please indicate whether the samples of the previous and current manuscript overlap.

·       Discussion: The results of the group comparison have not been related to existing literature. It is completely unclear if these are brand new insights or if other studies have found similar differences or if the current results can be related to some known differences in TD SLD groups.

·       Previous findings suggest that isolated training, e.g. of memory, is less effective than promoting academic skills (Grigorenko et al., 2020). To what extent are the results of the current study still relevant for interventions? The conclusion mostly refers to the tasks as exercise material. Based on the current study it is completely unknown whether these tasks would be useful training material. This could be discussed as future directions, but is not a conclusion based on the presented study.

·       I would highly encourage the authors to make their research more transparent, e.g. by making used material open access, providing the syntax used for estimation, and making their data accessible for other researchers.

Minor:

·       STM abbreviation in abstract used before introducing it.

·       The abstract mentions a five-cluster solution while the main text describes a seven or after exclusion four-cluster solution.

·       AS abbreviation is used in task description without introduction.

Comments on the Quality of English Language

Sometimes the used words seem to be not quite fitting. For example calling the new task battery a "device" seems weird. Additionally, sometimes the grammar is not quite correct making it harder to understand the overall very interesting paper.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

Dear Editor

Overall this manuscript describes an appropriate study design and outcome to summarise the online platform made available for assessing memory functions to possibly discriminate among subgroups of children with learning disabilities

The cohort is of a reasonable size and the outcome provides some differentiation in understanding the learning preferences and difficulties among subgroup of individuals. 

There is potential usefulness in deploying this study to characterise the LD children by analysing the short term memory bases tests. 

1 minor query

can we explain why the mathematical disorder tends to be not too dissimilar from TD group in STM performances? Line 663

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I appreciate the authors’ effort to respond to all my questions and comment. Their implementation of substatial revisions (especially adding more details to understand the methods) strengthened the manuscript substantially. The authors project to provide a platform to assess STM (and maybe train it in the future) is very admirable. I would like to suggest that the authors consider publishing their comparison of different scoring procedures because this is an issue that is relevant to the entire field of memory research and could certainly be interesting to many other scientists.