Complex Span Versus Updating Tasks of Working Memory: The Gap Is Not That Deep

Center for Lifespan Psychology, Max Planck Institute for Human Development, Berlin, Germany.
Journal of Experimental Psychology Learning Memory and Cognition (Impact Factor: 2.86). 08/2009; 35(4):1089-96. DOI: 10.1037/a0015730
Source: PubMed


How to best measure working memory capacity is an issue of ongoing debate. Besides established complex span tasks, which combine short-term memory demands with generally unrelated secondary tasks, there exists a set of paradigms characterized by continuous and simultaneous updating of several items in working memory, such as the n-back, memory updating, or alpha span tasks. With a latent variable analysis (N = 96) based on content-heterogeneous operationalizations of both task families, the authors found a latent correlation between a complex span factor and an updating factor that was not statistically different from unity (r = .96). Moreover, both factors predicted fluid intelligence (reasoning) equally well. The authors conclude that updating tasks measure working memory equally well as complex span tasks. Processes involved in building, maintaining, and updating arbitrary bindings may constitute the common working memory ability underlying performance on reasoning, complex span, and updating tasks.

Download full-text


Available from: Oliver Wilhelm, Oct 10, 2015
1 Follower
37 Reads
  • Source
    • "At each measurement occasion different versions of the tasks were used. One of the working memory capacity tasks used in this intensive longitudinal study was a memory updating task (Oberauer et al., 2000, 2003; Schmiedek et al., 2009). "
    [Show abstract] [Hide abstract]
    ABSTRACT: In the present paper we investigate weekly fluctuations in the working memory capacity (WMC) assessed over a period of 2 years. We use dynamical system analysis, specifically a second order linear differential equation, to model weekly variability in WMC in a sample of 112 9th graders. In our longitudinal data we use a B-spline imputation method to deal with missing data. The results show a significant negative frequency parameter in the data, indicating a cyclical pattern in weekly memory updating performance across time. We use a multilevel modeling approach to capture individual differences in model parameters and find that a higher initial performance level and a slower improvement at the MU task is associated with a slower frequency of oscillation. Additionally, we conduct a simulation study examining the analysis procedure's performance using different numbers of B-spline knots and values of time delay embedding dimensions. Results show that the number of knots in the B-spline imputation influence accuracy more than the number of embedding dimensions.
    Frontiers in Psychology 07/2014; 5:687. DOI:10.3389/fpsyg.2014.00687 · 2.80 Impact Factor
  • Source
    • "Tau indexes the exponential component of the RT distribution and reflects the subset of extremely slow responses that otherwise have a strong influence on mean RT and standard deviation of RT calculation. Tau is similar conceptually to a distribution's skewness, but is considered a more reliable metric (Schmiedek et al., 2007). A latent factor was created using principal components factor analysis separately for sigma (59.87% variance accounted for; both factor loadings r ϭ .77; "
    [Show abstract] [Hide abstract]
    ABSTRACT: Objective: The current study examined competing predictions of the default mode, cognitive neuroenergetic, and functional working memory models of attention-deficit/hyperactivity disorder (ADHD) regarding the relation between neurocognitive impairments in working memory and intraindividual variability. Method: Twenty-two children with ADHD and 15 typically developing children were assessed on multiple tasks measuring intraindividual reaction time (RT) variability (ex-Gaussian: tau, sigma) and central executive (CE) working memory. Latent factor scores based on multiple, counterbalanced tasks were created for each construct of interest (CE, tau, sigma) to reflect reliable variance associated with each construct and remove task-specific, test-retest, and random error. Results: Bias-corrected, bootstrapped mediation analyses revealed that CE working memory accounted for 88% to 100% of ADHD-related RT variability across models, and between-group differences in RT variability were no longer detectable after accounting for the mediating role of CE working memory. In contrast, RT variability accounted for 10% to 29% of between-group differences in CE working memory, and large magnitude CE working memory deficits remained after accounting for this partial mediation. Statistical comparison of effect size estimates across models suggests directionality of effects, such that the mediation effects of CE working memory on RT variability were significantly greater than the mediation effects of RT variability on CE working memory. Conclusions: The current findings question the role of RT variability as a primary neurocognitive indicator in ADHD and suggest that ADHD-related RT variability may be secondary to underlying deficits in CE working memory.
    Neuropsychology 03/2014; 28(3). DOI:10.1037/neu0000050 · 3.27 Impact Factor
  • Source
    • "Although n-back (e.g., Kane et al. 2007b; Schmiedek et al. 2009) and complex span (e.g., Conway et al. 2005; Unsworth and Engle 2007a, 2007b) tasks have undergone substantial examination, testing the ways in which the parameters of these tasks influence various indices of task performance, as well as the strength of their correlations with Gf, no similar questions have been posed with regard to the measures of relational integration. "
    [Show abstract] [Hide abstract]
    ABSTRACT: This study aimed to evaluate how well fluid reasoning can be predicted by a task that involves the monitoring of patterns of stimuli. This task is believed to measure the effectiveness of relational integration-the process that binds mental representations into more complex relational structures. In Experiments 1 and 2, the task was indeed validated as a proper measure of relational integration, since participants' performance depended on the number of bindings that had to be constructed in the diverse conditions of the task, whereas neither the number of objects to be bound nor the amount of elicited interference could affect this performance. In Experiment 3, by means of structural equation modeling and variance partitioning, the relation integration task was found to be the strongest predictor of fluid reasoning, explaining variance above and beyond the amounts accounted for by four other kinds of well-established working memory tasks.
    Memory & Cognition 09/2013; 42(3). DOI:10.3758/s13421-013-0366-x · 1.92 Impact Factor
Show more