Elsevier

Intelligence

Volume 38, Issue 6, November–December 2010, Pages 543-551
Intelligence

Intelligence, working memory, and multitasking performance

https://doi.org/10.1016/j.intell.2010.08.002Get rights and content

Abstract

Multitasking performance is relevant in everyday life and job analyses highlight the influence of multitasking over several diverse occupations. Intelligence is the best single predictor of overall job performance and it is also related to individual differences in multitasking. However, it has been shown that working memory capacity (WMC) is related to both intelligence and multitasking performance. Here we consider the simultaneous relationship among intelligence, WMC, and multitasking. The sample comprised three hundred and two applicants for air traffic control training courses. The main finding shows that intelligence and WMC are both related to multitasking, but only WMC predicts multitasking when their simultaneous relationship is considered. Furthermore, the processing and storage components of WMC predict multitasking performance. From an applied perspective it is suggested that WMC might be a relevant measure for personnel selection settings involving multitasking requirements.

Introduction

Understanding individual differences in multitasking is relevant because there is a large set of everyday life activities involving the simultaneous combination of distinguishable tasks. For better or worse we are living in a high technology society requiring multitasking skills (Hunt, 1995). Available job analyses underscore the relevance of multitasking for several occupations, like aircraft pilots, school bus drivers, or fire fighting supervisors (Peterson, Mumford, Borman, Jeanneret, & Fleishman, 1999). However, not too much is known about predictors of multitasking performance (Burgess et al., 2000, Damos, 1993, Myors et al., 1989, Rubinstein et al., 2001). Here we study concurrently two of such potential predictors, namely, intelligence and working memory capacity (WMC) considering a large and heterogeneous sample of applicants for admissions to air traffic control training courses.

A widely accepted definition of multitasking performance is the cognitive ability to perform “multiple task goals in the same time period by engaging in frequent switches between individual tasks” (Delbridge, 2000). Instead of multitasking, some researchers use the term ‘dual tasking’ or ‘task switching’ when two tasks are involved (Logan and Gordon, 2001, Monsell, 2003). In multitasking situations, different tasks are thought to interfere with one another (Monsell, 2003, Pashler, 1994). In his seminal book, Kahneman (1973) endorsed the view that available mental resources are shared by different tasks in a multitasking environment. Given that these mental resources are limited, interference among tasks is predicted. Logan (2002) claimed that this interference is produced because some mental operations cannot be divided, and, therefore, a bottleneck is generated. Nevertheless, no agreement regarding the most promising theoretical explanation is still available.

On the other hand, WMC is defined by the ability to retain information during short periods of time while performing a concurrent (and interfering) processing. WMC — i.e. the maximum amount of information that can be retained in the short term — relates to reasoning, problem solving, language comprehension, or learning, among other mental activities (Colom et al., 2005a, Colom et al., 2005b, Conway et al., 2002, Conway et al., 2005, Engle et al., 1999, Kane et al., 2004, Miyake and Shah, 1999, Miyake et al., 2001). It is a cognitive system implicating the simultaneous temporary storage and processing of any given information. Thus, for instance, the letter rotation task involves the processing requirement of verifying if a given capital letter is displayed normal or mirror-imaged, as well as the short-term storage requirement of temporarily maintaining its spatial orientation for latter recall (Shah & Miyake, 1996). The accepted definitions of WMC (Engle et al., 1999, Miyake et al., 2001) highlight (a) the dual nature of tasks measuring this construct (mental resources must be distributed between the processing and the storage components), (b) the complexity subjects must face trying to successfully combine the storage requirement and the explicit demanded concurrent processing. Therefore, coping with several tasks at once, WMC could help to successfully switch from one task to another.

Finally, intelligence is usually characterized as a very general mental ability (Gottfredson, 1997). This ability (or g) involves several general purpose mental mechanisms, like reasoning, planning, or solving problems. Intelligence can be measured by tests like the Raven Progressive Matrices Test, verbal tests that depend on figuring out the relationships between certain words when the meanings of all the words themselves are largely familiar, and so forth.

Intelligence and WMC are two psychological constructs extensively investigated in differential and cognitive psychology respectively (Colom et al., 2005a, Colom et al., 2005b, Colom et al., 2006b, Colom et al., 2008, Kyllonen and Christal, 1990, Unsworth and Engle, 2007). With respect to multitasking, Ben-Shakhar and Sheffer (2001) showed that high mental ability facilitates coping with these situations. König, Bühner, and Mürling (2005) have reported that intelligence and working memory predict individual differences in a multitasking situation, although working memory was a better predictor (see below). Similar results were found by Bühner, König, Pick, and Krumm (2006).

Hambrick, Oswald, Darowski, Rench, and Brou (2009) reported a study showing that WMC predicts multitasking. These researchers also measured intelligence, but, unfortunately (a) their SEM model dropped this variable from the analyses because WMC and intelligence were too highly correlated, and (b) the raw correlations among intelligence, working memory, and multitasking were not reported (their footnote 6 indicates that WMC was more highly correlated with multitasking than intelligence, suggesting that WMC is more fundamental than intelligence in accounting for individual differences in multitasking).

It is acknowledged that intelligence is the best single predictor of job related performance. Schmidt and Hunter's (2004) summary of available meta-analyses reports validity coefficients ranging from .31 to .73 (average for predicting performance on the job = .55 and average validity for predicting in job training programs = .63). These authors note that intelligence, as measured by the available standardized tests, predicts (a) the attained occupational level and (b) actual performance on the job. They report values above .50 for predicting performance in training programs, on the job, and over the achieved occupational level, noting that “relationships this large are rare in psychological research and are considered ‘large’ (Cohen, Cohen, West, & Aiken, 2003, p. 171).

The present study is based on previously reported high relationships between intelligence and working memory (Ackerman et al., 2002, Colom et al., 2004a, Colom et al., 2005a, Colom et al., 2005b, Kyllonen and Christal, 1990). Thus, for instance, Oberauer, Schulze, Wilhelm, and Süß (2005) reported a correlation of .85. Using broad and representative latent constructs (not frequently done in published studies) Colom, Abad, et al. (2005) found a correlation of .89. Given these correlation values between working memory and intelligence, it may be asked which construct is a better predictor of multitasking when both are considered concurrently.

Therefore, here we measure intelligence and WMC for predicting multitasking performance using two situations closely resembling job requirements fitting air traffic control mental or cognitive requirements. Because tasks tapping WMC and multitasking situations share coping with two or more competing concurrent cognitive requirements, we hypothesize that WMC will behave as a better predictor than intelligence. The processing and storage components of working memory are also analyzed separately, because only the latter is usually (but not always) considered as the dependent measure of interest. As concluded by Conway et al. (2005) extensive methodological review “established procedures of assigning absolute spans have various shortcomings, and so scoring procedures that exhaust the information collected with a task should be used instead” (p. 776).

Section snippets

Participants

The initial sample comprised 317 applicants for air traffic control training courses, but fifteen were excluded because of their extreme scores in some measures. Therefore, the analyzed sample included 302 participants (206 males and 96 females). Their mean age was 28.4 (SD = 6.2). All applicants were university graduates but from different educational areas (Engineering, 38.5%; Social sciences, 25.6%; Natural sciences, 13.2%; Humanities, 11%; Health 3.5%; other, 8.2%) which contributes to its

Results

Table 1 shows the descriptive statistics for the measures of interest along with the correlation matrix.

It can be seen that these measures show correlation values quite consistent with the definition of three latent factors for intelligence, working memory (WMC), and multitasking.

The analyses were conducted using AMOS 5.0 (Arbuckle, 2003) and the models were assessed by the next fit indices. The χ2/degrees of freedom ratio is first considered given that it is usually taken as a rule of thumb (

Discussion

The general finding shows that working memory capacity (WMC) and intelligence are both related to multitasking. However, SEM results did show that intelligence does not predict multitasking once its correlation with WMC is controlled for. This result is not consistent with Ben-Shakhar and Sheffer, 2001, Stankov et al., 1989, König et al., 2005, Bühner et al., 2006. These studies found that both WMC and intelligence predict multitasking, although WMC was a better predictor than intelligence.

Acknowledgment

The research referred to in this article was partially supported by the project AENA-UAM/785001.

References (65)

  • P.C. Kyllonen et al.

    Reasoning ability is (little more than) working-memory capacity

    Intelligence

    (1990)
  • S. Monsell

    Task switching

    Trends in Cognitive Sciences (Regular Ed.)

    (2003)
  • D. Peña et al.

    Solution strategies as possible explanations of individual and sex differences in a dynamic spatial task

    Acta Psychologica

    (2008)
  • C.L. Reeve et al.

    Survey of opinions on the primacy of g and social consequences of ability testing: A comparison of expert and non-expert views

    Intelligence

    (2008)
  • L. Stankov et al.

    Competing tasks: Predictors of managerial potential

    Personality and Individual Differences

    (1989)
  • N.A. Taatgen

    A model of individual differences in skill acquisition in the Kanfer–Ackerman air traffic control task

    Cognitive Systems Research

    (2002)
  • D. Aaronson et al.

    Extensions of Grier's computational formulas for for A′ and B″ to below-chance performance

    Psychological Bulletin

    (1987)
  • P.L. Ackerman et al.

    Individual differences in working memory within a nomological network of cognitive and perceptual speed abilities

    Journal of Experimental Psychology: General

    (2002)
  • J.L. Arbuckle

    AMOS 5.0

    (2003)
  • I.I. Bejar

    From adaptive testing to automated scoring of architectural simulations

  • J. Botella et al.

    El rendimiento en situación de doble tarea como medida de la capacidad para la tarea primaria [Performance in double task settings as a measure of capacity in the primary task]

    Estudios de Psicología

    (2000)
  • M. Bühner et al.

    Working memory dimensions as differential predictors of the speed and error aspect of multitasking performance

    Human Performance

    (2006)
  • B.M. Byrne

    Structural equation modelling with LISREL, PRELIS, and SIMPLIS: Basic concepts, applications, and programming

    (1998)
  • D.L. Coffman et al.

    Using parcels to convert path analysis models into latent variable models

    Multivariate Behavioral Research

    (2005)
  • J. Cohen et al.

    Applied multiple regression/correlation analysis for the behavioural sciences

    (2003)
  • R. Colom et al.

    Sex differences in verbal reasoning are mediated by sex differences in spatial ability

    Psychological Record

    (2004)
  • R. Colom et al.

    The assessment of spatial ability with a single computerized test

    European Journal of Psychological Assessment

    (2003)
  • R. Colom et al.

    Complex span tasks, simple span tasks, and cognitive abilities: A reanalysis of key studies

    Memory & Cognition

    (2006)
  • M.J. Contreras et al.

    Is static spatial performance distinguishable from dynamic spatial performance? A latent-variable analysis

    The Journal of General Psychology

    (2003)
  • M.J. Contreras et al.

    Sex differences in dynamic spatial ability: The unsolved question of performance factors

    Memory & Cognition

    (2007)
  • A.R.A. Conway et al.

    A latent variable analysis of working memory capacity, short-term memory capacity, processing speed, and general fluid intelligence

    Intelligence

    (2002)
  • A.R.A. Conway et al.

    Working memory span tasks: A methodological review and user's guide

    Psychonomic Bulletin & Review

    (2005)
  • Cited by (77)

    • Polychronicity at work: Work engagement as a mediator of the relationships between job outcomes

      2020, Journal of Hospitality and Tourism Management
      Citation Excerpt :

      Work engagement has an impact on the motivations of employees and is positively related to their job performances/satisfaction in the hospitality context (Xu et al., 2018). Such employees perform very well at their job, and work engagement plays an intervening role between their future work and task performance (Colom et al., 2010; Mittal & Bienstock, 2019). Work engagement has been explored by various researchers to cater employees' performance in the restaurant industry (SHARMA & Gursoy, 2018), whether the proactively changed individual's work environmental affects their own and colleagues' work engagement.

    View all citing articles on Scopus
    View full text