Elsevier

Physiology & Behavior

Volume 92, Issues 1–2, September 2007, Pages 186-192
Physiology & Behavior

Changes in cortical activity after training of working memory — a single-subject analysis

https://doi.org/10.1016/j.physbeh.2007.05.041Get rights and content

Abstract

Working memory (WM) capacity is an important factor for a wide range of cognitive skills. This capacity has generally been assumed to be fixed. However, recent studies have suggested that WM can be improved by intensive, computerized training [Klingberg T, Fernell E, Olesen P, Johnson M, Gustafsson P, Dahlström K, et al. Computerized training of working memory in children with ADHD — a randomized, controlled trial. J Am Acad Child Adolesc Psych 2005;44:177--86]. A recent study by Olesen, Westerberg and Klingberg [Olesen P, Westerberg H, Klingberg T. Increased prefrontal and parietal brain activity after training of working memory. Nat Neurosci 2004;7:75--9] showed that group analysis of brain activity data show increases in prefrontal and parietal cortices after WM training. In the present study we performed single-subject analysis of the changes in brain activity after five weeks of training.

Three young, healthy adults participated in the study. On two separate days before practice and during one day after practice, brain activity was measured with functional magnetic resonance imaging (fMRI) during performance of a WM and a baseline task. Practice on the WM tasks gradually improved performance and this effect lasted several months. The effect of practice also generalized to improve performance on a non-trained WM task and a reasoning task. After training, WM-related brain activity was significantly increased in the middle and inferior frontal gyrus. The changes in activity were not due to activations of any additional area that was not activated before training. Instead, the changes could best be described by small increases in the extent of the area of activated cortex. The effect of training of WM is thus in several respects similar to the changes in the functional map observed in primate studies of skill learning, although the physiological effect in WM training is located in the prefrontal association cortex.

Introduction

Working memory (WM) is the ability to retain information during short periods of time. The maximum amount of information that a person can retain – the WM capacity – is an important factor for many cognitive skills, including problem solving and reasoning ability [1], [2], [3]. WM capacity increases with age during childhood [1], [4], [5], and decreases during old age [6]. The increase in capacity is thought to depend on maturation of the brain, and the decrease at older ages could be due to neuronal degeneration. These processes are paralleled by increased cortical activity in the prefrontal and parietal cortex during childhood [7] and decreased dorsolateral prefrontal activity with aging [8].

It is largely unknown to what extent practice can affect WM capacity. This question in turn depends on whether practice can induce plasticity of the neural systems underlying WM. One previous study with macaque monkeys indicates that the WM systems are plastic [9]. In that study, the animals practiced delayed-response tasks for several weeks while difficulty was gradually increased by degrading the visual stimuli. Practice was found to change the receptive characteristics of neurons in the principal sulcus in the prefrontal cortex, such that they became more resistant to the effect of stimulus degradation.

In psychological studies, there are examples of successful training of attention [10], but previous attempts to improve WM by training have only achieved moderate success.

Repeated execution of WM trials, when difficulty level is not adapted, typically leads to faster reaction times, but not an increase in WM capacity [11], [12]. Improved performance has been achieved after teaching rehearsal strategies to children with learning disabilities [13], [14], [15]. A general problem that these authors noted was the lack of generalization from trained tasks to non-trained tasks, and the lack of long-lasting effects. A case study describes a subject who could retain a large number of digits by associating to series of numbers stored in long-term memory [16]. The subject could remember more than 80 digits, but this ability did not increase his WM capacity for verbal material.

In the present study, we used a novel training paradigm in which subjects practice WM tasks for five weeks. Key features of this training include repetitive training with continuous adaptations of the difficulty level, with training lasting for weeks. We have recently been able to show that this training could improve WM in children with Attention Deficit Hyperactivity Disorder (ADHD), and that this also ameliorated the ADHD symptoms [17], [18]. In the present study we investigated the effect of this training paradigm on healthy adults, without any WM deficits. Before and after practice, the subjects performed neuropsychological tests related to control of attention, WM and reasoning which were not part of the daily training, to see whether the training effect generalized to non-trained tasks. Training-induced changes in neuronal activity were estimated by measuring brain activity with fMRI before and after training.

Previous neuroimaging studies have observed practice-related decreases in activity in the prefrontal and cingulate cortex. Raichle et al. [19] showed that when subjects were asked to generate a verb from a noun, repeated presentation of the same noun resulted in less prefrontal and cingulate activity than initially observed. Reduced prefrontal activity has also been observed in several tasks involving declarative [20] and implicit encoding into long-term memory [21], as well as during repeated presentation of WM trials [22]. However, these were all studies of within-session effects of repeated task performance, but did not reflect the acquisition of skill. In studies of acquisition of motor and perceptual skills, practice has been shown to increase task-related activity [23], [24] which is consistent with the literature from skill acquisition in primates [25], [26]. Another study of skill acquisition found that several days of practice on reading mirror-reversed text significantly improved performance and also increased task-related brain activity in the inferior temporal cortex, cerebellum, striatum and left premotor/inferior prefrontal region [27]. Our hypothesis was that improvement after practice on WM tasks for several weeks would be akin to skill acquisition, and thus induce increased task-related activity in the prefrontal cortex.

Neuroimaging data can either be analysed by combining the results from several subjects, a so called group analysis, or by looking at activations in single subject. The former provide results that can be generalized to a larger population. The single-subject analysis, however, can sometimes give greater anatomical detail about the changes in brain activity.

We have previously reported the results from group analyses in experiments evaluating the effect of training of WM [28]. In the present study we will report the single-subject analysis from a subgroup of the subjects included in the study by Olesen et al.

Section snippets

Subjects

Three healthy, male, right-handed volunteers (AP, DH, IK), aged 23, 20 and 22 years, participated in the training. A control group consisting of eleven healthy adult subjects, five men and six women, mean age 25.8 SEM 1.5 years, undertook repeated testing of the neuopsychological test battery with a five week test–retest period to provide a baseline for comparing the test–retest improvements from the other subjects. The study was approved by the local ethics committee at the Karolinska Hospital.

Behavioral data from pre-and post-training tests

Subjects AP, DH and IK practiced 20, 24 and 30 days. Performance during training was continuously recorded, and showed a gradual and statistically significant improvement on the trained WM tasks (Fig. 2A). Test–retest improvement of the subjects on the neuropsychological tasks was compared to test–retest improvement in the control group. Training significantly improved performance on a computerized visuo-spatial WM task (each subject P < 0.001, all three were at ceiling), Span-board (Fig. 2B,

Discussion

The present results show that practice of WM tasks over several weeks induces a gradual improvement in performance (Fig. 2A). This improvement also generalized to a non-practiced visuo-spatial WM task and a non-practiced reasoning task (Fig 2B and C); effects which persisted for months. In agreement with the hypothesis, the training-induced significant increases in WM-related activity in the prefrontal cortex (Table 1, Table 2; Fig. 3) in all three subjects. This finding is also in agreement

Concluding remarks

In summary, the slow training-induced changes in performance were associated with increased prefrontal activity in all three subjects. This change appeared to comprise a change in the area of activated cortex, similar to the changes in functional maps previously observed after training of motor and sensory tasks. The effect of training of WM is thus in several respects similar to the plastic effect of skill learning, although the physiological effect of WM training is located in the prefrontal

References (40)

  • M. Van der Linden et al.

    Age-related differences in updating working memory

    Br J Psychol

    (1994)
  • T. Klingberg et al.

    Increased brain activity in frontal and parietal cortex underlies the development of visuo-spatial working memory capacity during childhood

    J Cogn Neurosci

    (2002)
  • B. Rypma et al.

    Isolating the neural mechanisms of age-related changes in human working memory

    Nat Neurosci

    (2000)
  • M.M. Sohlberg et al.

    Evaluation of attention process training and brain injury education in persons with acquired brain injury

    J Clin Exp Neuropsychol

    (2000)
  • M.W. Kristofferson

    Effects of practice on character-classification performance

    Can J Psychol

    (1972)
  • C.J. Phillips et al.

    Effects of practice on recognition memory of mildly mentally retarded adults

    Am J Ment Defic

    (1984)
  • E.C. Butterfield et al.

    On the theory and practice of improving short-term memory

    Am J Ment Defic

    (1973)
  • A.L. Brown et al.

    Keeping track of changing variables: effects of rehearsal training and rehearsal prevention in normal and retarded adolescents

    J Exp Psychol

    (1973)
  • C. Hulme

    Working memory and severe learning difficulties

    (1992)
  • K.A. Ericsson et al.

    Acquisition of a memory skill

    Science

    (1980)
  • Cited by (0)

    View full text