Computational assessment of long-term memory structures from SDA-M related to action sequences

Abstract

Assistance systems should be able to adapt to individual task-related skills and knowledge. Structural-dimensional analysis of mental representations (SDA-M) is an established method for retrieving human memory structures related to specific activities. For this purpose, SDA-M involves a semi-automatized survey of users (the “split procedure”), which yields data about users’ associations between action representations in long-term memory. Up to now this data about associations has commonly been clustered and visualized by SDA-M software in the form of dendrograms that can be used by human experts as a tool to (manually) assess users’ individual expertise and identify potential issues with respect to predefined action sequences. This article presents new algorithmic approaches for automatizing the process of assessing task-related memory structures based on SDA-M data to predict probable errors in action sequences. This automation enables direct integration into technical systems, e.g. user-adaptive assistance systems. An evaluation study has compared the automatized computational assessments to predictions made by human scholars based on visualizations of SDA-M data. The different algorithms’ outputs matched human experts’ manual assessments in 84% to 86% of the test cases.

Cognitive Interaction Technology in Sport—Improving Performance by Individualized Diagnostics and Error Prediction

“…All SDA-M procedures were executed with the QSplit SDA-M Suite v1.6 for Windows. This included the split procedure and the usual data normalization, scaling, clustering and invariance analysis steps as described by Schack ( 2012 ), as well as advanced analyses using the AMPA and CASPA algorithms (see Strenge et al, 2019 ). Generally, the available data were analyzed on two different levels:…”

“…SDA-M consists of several survey and analysis steps, which are briefly outlined in the following. The theoretical, methodical and algorithmic foundations of SDA-M have been described in detail by Schack ( 2012 ), and Strenge et al ( 2019 ) presented and exemplified recent algorithmic extensions for automatic assessment of SDA-M data concerning individual likelihoods of errors in action sequences.…”

“…Two different algorithmic approaches for predicting human error based on SDA-M data have been developed and presented by Strenge et al ( 2019 ): Analysis of Most Probable Actions (AMPA) and Correct Action Selection Probability Analysis (CASPA). These new algorithmic approaches automatized the process of assessing memory structures based on SDA-M data to predict probable errors in action sequences, which eliminated the previous need for manual assessments using dendrograms.…”

“…Strenge et al ( 2019 ) speculated that the results of AMPA or CASPA could be used by technical systems like intelligent glasses to provide anticipatory action support.…”

“…In this line of research, numerous studies found that the differing mental representation structures of experts and novices can be measured with the “ structural-dimensional analysis of mental representations” (SDA-M) method (Schack, 2012 ) and influenced by appropriate training (e.g., Heinen et al, 2002 ; Schack, 2004 ; Schack and Mechsner, 2006 ; Schack and Hackfort, 2007 ; Frank et al, 2013 , 2014 ; Schack et al, 2014b ). A methodological review and evaluation of research in expert performance in sport by Hodges et al ( 2007 , p. 164) noted that the SDA-M method “ is expected to aid in our understanding of the usually non-declarative motor representations underlying expert performance in fast, complex coordinative actions and in identifying the problems novices encounter in understanding motor problems.” Recently, Strenge et al ( 2019 ) described advanced algorithms for automatized analyses of task-related mental representation structures based on SDA-M related to action sequences. These algorithmic approaches might be useful as a component of future CIT systems, like cognitive glasses, to measure and improve human performance in sport.…”