More dissociations and interactions within central executive functioning: A comprehensive latent-variable analysis
Introduction
A large number of theoretical and empirical studies have been conducted over the last decade to make it possible to define the functioning of the control component of working memory (Baddeley, 1986, Baddeley and Hitch, 1974), i.e. the central executive. The functional analogy suggested by Baddeley between the central executive and the supervisory attentional system in Norman and Shallice’s (1980) model of attentional control represented a first step in this direction. The attentional nature of the central executive has also been emphasized by much other subsequent research (Baddeley and Logie, 1999, Engle et al., 1999, Miyake and Shah, 1999). Baddeley (1996) then proposed several avenues of research intended to contribute to the study of the central executive, each of which potentially corresponding to a “general” function assumed by this component. He thus distinguished between the coordination of two concurrent tasks or mental processes, the alternation between response strategies (flexibility), the selective attention function, and the activation and retrieval of information in long-term memory (LTM).
This manner of addressing the study of central executive functioning raises the question of its fractionation. Several contemporary and subsequent empirical studies have indeed cast doubt on a unitary conception of the central executive. These studies used a wide variety of experimental paradigms, methodologies, and were conducted in various target populations: there are thus clinical and experimental studies in brain-damaged patients (Burgess et al., 1998, Duncan et al., 1997, Godefroy et al., 1999), correlational studies, as well as exploratory and confirmatory analyses in healthy young adults (Friedman and Miyake, 2004, Lehto, 1996, Miyake et al., 2000), healthy elderly adults (Fisk and Sharp, 2004, Hedden and Yoon, 2006, Lowe and Rabbitt, 1997, Robbins et al., 1998), demented elderly adults (Collette, Van der Linden, & Salmon, 1999), and children (Lehto et al., 2003, Levin et al., 1991, StClair-Thompson and Gathercole, 2006). All these studies suggest the existence of several control functions – or “executive functions” as they are usually called in neuropsychology – that encompass a variety of processes which interact but can nevertheless operate in a relatively independent way. While most of these studies have focused on a few specific aspects of executive functioning, some of the more recent research has attempted to provide a comprehensive account of executive functions, while also trying to gain a better understanding of their interrelationships.
Among these latter studies, we can identify two types of approach. The first is based on clinical observations and/or correlational analyses and consists in demonstrating the relationships or lack of relationships between performances on tasks which are thought to assess different executive functions. The second approach, based on hypothesis testing procedures such as confirmatory factor analysis (CFA), consists in testing theoretical models of how different control functions interact with one another.
Research conducted in the context of the first approach classically reports dissociations between performance on executive tasks and/or the presence or absence of significant correlations between such tasks. While the details of the results differ across studies (a fact which can be partially explained by the different tasks employed), researchers consistently report that the correlations between the different tasks are nonsignificant or weak, and that several independent factors are responsible for the overall performance. For example, Lehto (1996) reported that there was no relationship between the performances obtained by young adults in different tasks assessing executive and storage-and-processing functions. Similarly, Collette et al.’s (1999) study indicated that the performance of patients with Alzheimer’s disease in a series of executive tasks loaded mainly on two independent factors which were defined as representing inhibitory processes on one hand, and the coordination between storage and processing operations on the other. This kind of approach to the exploration of the central executive is, however, constrained by a number of theoretical and methodological limitations (Burgess, 1997, Rabbitt, 1997), which seriously complicate the interpretation of the results. One of these limitations concerns the assessment of the functions attributed to the central executive, since the fact that these functions are not clearly defined means that they are difficult to operationalize. Another limitation is that executive tasks are, in essence, impure measures, because they necessarily involve both executive and non-executive processes and, consequently, individual performance on these tasks may be biased by the non-executive requirements of the task. Moreover, the specific contribution of non-executive processes to performance can vary greatly from one task to another as well as from one participant to another, thus making it difficult to isolate specific executive processes.
The second kind of approach is more recent and was developed to address these limitations. The work performed by Miyake et al. (2000) represents an important starting point for such research. It uses a latent variable procedure in which the variance shared among a set of tasks selected to measure one and the same underlying construct (i.e. executive function) is statistically extracted, while the specific requirements of each task and measurement error are removed, thus minimizing the task impurity problem. This shared variance represents the latent variable of interest and it is possible to examine how different latent variables relate to one another. These authors used this technique to examine the relationships between three frequently postulated executive functions (mental set shifting, inhibition of prepotent responses, and updating the contents of working memory), and to specify how they are organized. The relationships were thus examined at the level of latent variables, i.e. what is common to multiple tasks chosen to tap the same underlying construct but not at the level of manifest variables, i.e. individual tasks. The comparison of several CFA models led the authors to conclude that the three considered functions, though moderately correlated, were clearly separable. This approach is now in widespread use to explore the relationships among executive functions themselves as well as between executive functions and other important higher-level cognitive constructs such as fluid intelligence (gf) or working memory capacity (e.g., Friedman and Miyake, 2004, Hedden and Yoon, 2006, Miyake et al., 2001, Salthouse et al., 2003).
The study presented here uses the same approach to investigate the relationships among a large set of executive functions chosen on the basis of the results obtained from a previous exploratory study (Fournier, Larigauderie, & Gaonac’h, 2004). This earlier study represented a first attempt to bring together the main functions previously subsumed under the term “central executive”, namely coordination, inhibition, retrieval from LTM, and planning (Baddeley, 1996, Collette and Van der Linden, 2002, Shallice, 1982), by exploring the factorial structure of a comprehensive sample of executive and working memory tasks that are thought to assess these functions. A principal components analysis (PCA) suggested that the performances of 65 young adults on the selected tasks loaded mainly on five factors which were defined as corresponding to (1) the ability to inhibit prepotent responses in association with strategic retrieval from LTM, (2) the ability to inhibit distracting information when selecting the relevant information, (3) the ability to coordinate the storage and processing of verbal information, (4) the ability to coordinate the storage and processing of visuospatial information, and (5) the ability to coordinate different processing operations without storage. The pattern of results demonstrated that the different facets of the central executive are based, at least in part, on distinct functions, and consequently support the executive fractionation hypothesis. However, the results also suggested that a theoretical fractionation of central executive functioning into very general functions such as coordination, inhibition, and retrieval from LTM does not correspond to any empirical reality, and that more fine-grained dissociations within these functions (e.g., verbal versus visuospatial coordination abilities) as well as specific interactions between some of these functions (e.g., inhibition and strategic retrieval abilities) should be considered. Overall, these findings were taken as the starting point for the current research and led us to focus on a set of functions relating to various domains of executive functioning. Thus, these earlier exploratory data, along with exploratory and confirmatory data from the literature, served to guide the selection, definition and assessment of the target functions in the present study, and also provided the basis for specifying an a priori model of how these functions are organized. This model was tested using CFA.
Section snippets
The central executive functions considered in the present study
Our previous study had clearly indicated that it is essential to distinguish between two coordination abilities: the ability to coordinate the storage and processing of verbal information, and the ability to coordinate the storage and processing of visuospatial information. This observation echoes a number of other studies that have reported dissociations between verbal/numerical and visuospatial storage-and-processing tasks (Friedman and Miyake, 2000, Jarvis and Gathercole, 2003, Oberauer et
Participants
The participants consisted of 180 undergraduate or graduate students at the University of Poitiers. Their age ranged from 18 to 31 years (mean age = 23.62 ± 3.24 years), and 50% were female. All the participants were native French speakers and took part voluntarily in the study.
Materials, design, and procedure
All the participants completed the 19 tasks thought to assess the six executive functions we were studying. First, we directed our choice towards tasks classically used in cognitive and neuropsychological studies and/or in
Results and discussion
The results are presented in three sections. First, the procedure for screening the data is presented. Second, descriptive statistics and Bravais–Pearson correlations are presented. Finally, the CFA analyses which test the separability of the target functions are presented.
General discussion
The goal of the present study was to examine the separability of six central executive functions: verbal storage-and-processing coordination, visuospatial storage-and-processing coordination, dual-task coordination, strategic retrieval, selective attention and shifting. The results of the CFA demonstrated that five of the six initially considered central executive functions were separable, albeit related, constructs. Because of the lack of convergent validity among dual-task measures, these
Concluding remarks
The present findings suggest that dissociations within executive functions and specific interactions between some of these functions must both be taken into account if we are to improve our knowledge and understanding of executive functioning. The results reported here, together with those from an earlier study (Fournier et al., 2004), confirm that a fractionation of central executive functioning in terms of general functions such as “coordination”, “inhibition”, and “LTM retrieval” does not
Acknowledgements
The authors are grateful to two anonymous reviewers for their helpful comments on an earlier version of this manuscript.
References (83)
The episodic buffer: A new component of working memory?
Trends in Cognitive Science
(2000)- et al.
The relationship between the ability to divide attention and standard measures of cognitive abilities
Intelligence
(2001) - et al.
Exploration of the neural substrates of executive functioning by functional neuroimaging
Neurosciences
(2006) - et al.
Brain imaging of the central executive component of working memory
Neuroscience and Biobehavioral Reviews
(2002) - et al.
The functional anatomy of inhibition processes investigated with the Hayling task
NeuroImage
(2001) - et al.
Executive dysfunction in Alzheimer’s disease
Cortex
(1999) - et al.
Memory span and general intelligence: A latent-variable approach
Intelligence
(2005) - et al.
A PET study of word finding
Neuropsychologia
(1991) - et al.
Control functions of the frontal lobes. Modularity of the central-supervisory system?
Cortex
(1999) Varieties of inhibition: Manifestations in cognition, event-related potentials and aging
Acta Psychologica
(1999)
Working memory impairments in traumatic brain injury: Evidence from a dual-task paradigm
Neuropsychologia
The unity and diversity of executive functions and their contributions to complex “frontal lobe ”tasks: A latent variable analysis
Cognitive Psychology
Task switching
Trends in Cognitive Sciences
Forest before trees: The precedence of global features in visual perception
Cognitive Psychology
Working memory capacity – facets of a cognitive ability construct
Personality and Individual Differences
The multiple faces of working memory: Storage, processing, supervision, and coordination
Intelligence
Working memory capacity and suppression
Journal of Memory and Language
Coordinating cognitive information: Task effects and individual differences in integrating information from several sources
Cognitive Psychology
Executive function and the frontal lobes: A meta-analytic review
Neuropsychological Review
Integration as a general boundary condition on retrieval-induced forgetting
Journal of Experimental Psychology: Learning, Memory, and Cognition
The capacity for generating information by randomization
Quarterly Journal of Experimental Psychology
Working memory
Exploring the central executive
Quarterly Journal of Experimental Psychology
The central executive: A concept and some misconceptions
Journal of the International Neuropsychological Society
Random generation and the executive control of working memory
The Quarterly Journal of Experimental Psychology
Testing central executive functioning with a pencil-and-paper test
Working memory
Working memory: The multiple-component model
Measuring attentional demand in continuous dual-task performance
Quarterly Journal of Experimental Psychology
A general factor involved in dual-task performance decrement
Quarterly Journal of Experimental Psychology
d2 test d’attention concentrée
Theory and methodology in executive function research
The ecological validity of tests of executive function
Journal of the International Neuropsychological Society,4
Response suppression, initiation and strategy use following frontal lobe lesions
Neuropsychologia
Structural equation modeling with LISREL, PRELIS, and SIMPLIS: Basic concepts, applications, and programming
Working memory span tasks: A methodological review and user’s guide
Psychonomic Bulletin and Review
Coefficient alpha and the internal structure of tests
Psychometrika
Age differences and divided attention: Is there a general deficit?
Experimental Aging Research
Frontal lobe deficits after head injury: Unity and diversity of function
Cognitive Neuropsychology
Individual differences in working memory capacity and what they tell us about controlled attention, general fluid intelligence, and functions of the prefrontal cortex
Cited by (86)
New Research Perspectives on the Interplay Between Genes and Environment on Executive Function Development
2023, Biological PsychiatryA one-year follow-up of the cognitive and psycho-behavioural skills in artistic gymnastics
2023, Psychology of Sport and ExerciseThe factor structure of executive function in childhood and adolescence
2022, IntelligenceCitation Excerpt :Despite this widespread support, other factor structures have been suggested. For example, Fournier-Vicente, Larigauderie, and Gaonach (2008) found more fine-grained dissociations within the central EF, with five inter-related factors (a verbal and a spatial coordination function, strategic retrieval, selective attention and shifting). And more recently, Himi, Buhner, and Hilbert (2021) showed that the three-factor model of Miyake et al. (2000) is highly inter-related with other basic cognitive skills (working memory capacity, relational integration, divided attention) and together link to a higher-order factor that would represent general cognitive ability.