Ergodic Subspace Analysis

Published on Jan 6, 2020in Journal of Intelligence
· DOI :10.3390/JINTELLIGENCE8010003
Timo von Oertzen18
Estimated H-index: 18
(MPG: Max Planck Society),
Florian Schmiedek32
Estimated H-index: 32
(Leibniz Association)
+ 0 AuthorsManuel C. Voelkle16
Estimated H-index: 16
(Humboldt University of Berlin)
Properties of psychological variables at the mean or variance level can differ between persons and within persons across multiple time points. For example, cross-sectional findings between persons of different ages do not necessarily reflect the development of a single person over time. Recently, there has been an increased interest in the difference between covariance structures, expressed by covariance matrices, that evolve between persons and within a single person over multiple time points. If these structures are identical at the population level, the structure is called ergodic. However, recent data confirms that ergodicity is not generally given, particularly not for cognitive variables. For example, the g factor that is dominant for cognitive abilities between persons seems to explain far less variance when concentrating on a single person’s data. However, other subdimensions of cognitive abilities seem to appear both between and within persons; that is, there seems to be a lower-dimensional subspace of cognitive abilities in which cognitive abilities are in fact ergodic. In this article, we present ergodic subspace analysis (ESA), a mathematical method to identify, for a given set of variables, which subspace is most important within persons, which is most important between person, and which is ergodic. Similar to the common spatial patterns method, the ESA method first whitens a joint distribution from both the between and the within variance structure and then performs a principle component analysis (PCA) on the between distribution, which then automatically acts as an inverse PCA on the within distribution. The difference of the eigenvalues allows a separation of the rotated dimensions into the three subspaces corresponding to within, between, and ergodic substructures. We apply the method to simulated data and to data from the COGITO study to exemplify its usage.
  • References (25)
  • Citations (1)
📖 Papers frequently viewed together
2 Authors (D. Glaser, K. Friston)
8 Citations
5 Citations
78% of Scinapse members use related papers. After signing in, all features are FREE.
#1Janne Adolf (Katholieke Universiteit Leuven)H-Index: 2
#2Eiko I. Fried (LEI: Leiden University)H-Index: 26
Researchers commonly draw inferences from the group level to the individual and vice versa—that is, across levels. One of the empirical cornerstones of medicine is the clinical trial that tests the efficacy of a drug compared with placebo. If the intervention group outperforms the placebo group, the conclusion is that the drug should be prescribed for individuals with a given disorder. When are such inferences across levels defensible? In their recent paper in PNAS, Fisher et al. (1) state that ...
8 CitationsSource
#1Florian Schmiedek (MPG: Max Planck Society)H-Index: 32
#2Martin Lövdén (KI: Karolinska Institutet)H-Index: 43
Last. Ulman Lindenberger (MPG: Max Planck Society)H-Index: 86
view all 4 authors...
3 CitationsSource
#1Manuel C. Voelkle (Humboldt University of Berlin)H-Index: 16
#2Christian Gische (Humboldt University of Berlin)H-Index: 2
Last. Ulman Lindenberger (MPG: Max Planck Society)H-Index: 86
view all 4 authors...
The lead-lag structure of multivariate time-ordered observations and the possibility to disentangle between-person (BP) from within-person (WP) sources of variance are major assets of longitudinal (panel) data. Hence, psychologists are making increasing use of such data, often with the intent to delineate the dynamic properties of psychological mechanisms, understood as a sequence of causal effects that govern psychological functioning. However, even with longitudinal data, psychological mechani...
9 CitationsSource
#1Aaron J. Fisher (UC: University of California)H-Index: 18
#2John D. Medaglia (Drexel University)H-Index: 22
Last. Bertus F. Jeronimus (UG: University of Groningen)H-Index: 14
view all 3 authors...
Only for ergodic processes will inferences based on group-level data generalize to individual experience or behavior. Because human social and psychological processes typically have an individually variable and time-varying nature, they are unlikely to be ergodic. In this paper, six studies with a repeated-measure design were used for symmetric comparisons of interindividual and intraindividual variation. Our results delineate the potential scope and impact of nonergodic data in human subjects r...
147 CitationsSource
#1Michael C. Neale (VCU: Virginia Commonwealth University)H-Index: 100
#1Michael Neale (VCU: Virginia Commonwealth University)H-Index: 11
Last. Steven M. Boker (UVA: University of Virginia)H-Index: 28
view all 10 authors...
The new software package OpenMx 2.0 for structural equation and other statistical modeling is introduced and its features are described. OpenMx is evolving in a modular direction and now allows a mix-and-match computational approach that separates model expectations from fit functions and optimizers. Major backend architectural improvements include a move to swappable open-source optimizers such as the newly written CSOLNP. Entire new methodologies such as item factor analysis and state space mo...
312 CitationsSource
#1Timo von Oertzen (MPG: Max Planck Society)H-Index: 18
#2Andreas M. Brandmaier (MPG: Max Planck Society)H-Index: 15
Last. Siny Tsang (UVA: University of Virginia)H-Index: 4
view all 3 authors...
Ωnyx is a free software environment for creating and estimating structural equation models (SEM). It provides a graphical user interface that facilitates an intuitive creation of models, and a powerful back end for performing maximum likelihood estimation of parameters. Path diagrams in Ωnyx can be exported to OpenMx, lavaan, and Mplus to allow an easy transition to script-based SEM software. Also, OpenMx scripts can be transformed into a path diagrams in Ωnyx. In addition to the graphical user ...
59 CitationsSource
#1Annette Brose (MPG: Max Planck Society)H-Index: 17
#2Manuel C. Voelkle (MPG: Max Planck Society)H-Index: 16
Last. Florian Schmiedek (MPG: Max Planck Society)H-Index: 32
view all 5 authors...
This study tested whether the structure of affect observed on the basis of between-person (BP) differences is equivalent to the affect structures that organize the variability of affective states within persons (WP) over time. Further aims were to identify individual differences in the degree of divergence between the WP and BP structure and examine its association to dispositional and contextual variables (neuroticism, extraversion, well-being and stress). In 100 daily sessions, 101 younger adu...
37 CitationsSource
#1Manuel C. Voelkle (MPG: Max Planck Society)H-Index: 16
#2Annette Brose (MPG: Max Planck Society)H-Index: 17
Last. Ulman Lindenberger (MPG: Max Planck Society)H-Index: 86
view all 4 authors...
The vast majority of empirical research in the behavioral sciences is based on the analysis of between-person variation. In contrast, much of applied psychology is concerned with the analysis of variation within individuals. Furthermore, the mechanisms specified by psychological theories generally operate within, rather than across, individuals. This disconnect between research practice, applied demands, and psychological theories constitutes a major threat to the conceptual integrity of the fie...
72 CitationsSource
#1Florian Schmiedek (MPG: Max Planck Society)H-Index: 32
#2Martin Lövdén (MPG: Max Planck Society)H-Index: 43
Last. Ulman Lindenberger (MPG: Max Planck Society)H-Index: 86
view all 3 authors...
People often attribute poor performance to having bad days. Given that cognitive aging leads to lower average levels of performance and more moment-to-moment variability, one might expect that older adults should show greater day-to-day variability and be more likely to experience bad days than younger adults. However, both researchers and ordinary people typically sample only one performance per day for a given activity. Hence, the empirical basis for concluding that cognitive performance does ...
25 CitationsSource
#1Manuel C. Voelkle (MPG: Max Planck Society)H-Index: 16
#2Johan H. L. Oud (Radboud University Nijmegen)H-Index: 21
Last. Ulman Lindenberger (MPG: Max Planck Society)H-Index: 86
view all 4 authors...
This article has 3 objectives that build on each other. First, we demonstrate how to obtain maximum likelihood estimates for dynamic factor models (the direct autoregressive factor score model) with arbitrary T and N by means of structural equation modeling (SEM) and compare the approach to existing methods. Second, we go beyond standard time series analysis (T large and N D 1) and conventional SEM (N large and T D 1 or small) by integrating both approaches. The resulting combined model offers a...
16 CitationsSource
Cited By1
#1Florian Schmiedek (MPG: Max Planck Society)H-Index: 32
#2Martin Lövdén (KI: Karolinska Institutet)H-Index: 43
Last. Ulman Lindenberger (MPG: Max Planck Society)H-Index: 86
view all 4 authors...
Over a century of research on between-person differences has resulted in the consensus that human cognitive abilities are hierarchically organized, with a general factor, termed general intelligence or "g," uppermost. Surprisingly, it is unknown whether this body of evidence is informative about how cognition is structured within individuals. Using data from 101 young adults performing nine cognitive tasks on 100 occasions distributed over six months, we find that the structures of individuals' ...