Validation of a brain-computer interface version of the digit symbol substitution test in healthy subjects

Highlights

• A novel SSVEP BCI-based DSST was implemented and was test on 12 healthy people.

• Performance of SSVEP BCI-based DSST was comparable with that of computerized DSST.

• SSVEP BCI-based DSST was reliable for repeatedly evaluating cognitive function.

• SSVEP BCI-based DSST can potentially be used for severely motor-impaired individuals.

Abstract

Digit symbol substitution test (DSST), which is a valid and sensitive tool to assess human cognitive dysfunction, has been widely used in clinical neuropsychology. Although several versions of DSST are currently available, most of the existing DSST versions rely on examinees' intact motor function. This limits their utility in severely motor-impaired individuals. A brain-computer interface (BCI) version of DSST was implemented in this study. Steady-state visual evoked potential (SSVEP) was adopted to build the BCI. Nine symbols in the proposed SSVEP BCI-based DSST were designed with clearly different shapes for decreasing measurement errors due to misidentified symbols. To reduce practice effect, furthermore, the digit-symbol pairs of each trial were different. A two-target SSVEP BCI was designed to judge whether the digit-symbol probe in the center of the user interface matched one of the nine digit-symbol pairs above the user interface. All 12 examinees were able to perform the tasks using the proposed SSVEP BCI-based DSST with 96.17 ± 4.18% averaged accuracy, which was comparable with that of computerized DSST. Furthermore, for examinees participating in both offline and online experiment, the accuracies of the online and offline experiments were comparable, supporting that the proposed BCI-DSST was reliable for repeatedly evaluating examinees’ cognitive function over time. These results verified that the proposed SSVEP BCI-based DSST was feasible and effective for healthy subjects.

Introduction

Cognitive impairment is common in various brain diseases, such as schizophrenia, cerebrovascular disease, and major depressive disorder (MDD). These deficits usually decrease quality of daily life and increase additional burden to both patients and society. Therefore, assessing patients’ cognitive function is very important. Currently, information processing speed is one of the most frequently used tests for cognitive impairment assessment and is also a core aspect for neuropsychological testing.

Digit Symbol Substitution Test (DSST) and Symbol Digit Modalities Test (SDMT) have been heavily utilized in measuring switching attention and information processing speed. In the DSST, numbers are presented and examinees are required to fill in symbols that match the numbers according to the 9 digit-symbol pairing information within a limited amount of time. SDMT is essentially the same as DSST, except in reverse; examinees need to match the corresponding number for each symbol when a series of symbols are presented. Although DSST and SDMT involve multiple cognitive processes, containing attention, perceptual/processing speed, visual scanning, and memory, they may be better seen as measuring more general cognition [1]. DSST and SDMT have been shown to be sensitive to cognitive deficit in various neuropsychiatric populations, including schizophrenia [[1], [2], [3]], MDD [4], and Alzheimer's disease (AD) [5]. For example, DSST is a key element for evaluating the cognitive deficit in schizophrenia when compared to the more widely used neuropsychological measures such as word list learning, span tasks, and card sorting [6]. In addition, DSST seems to have better diagnostic value of AD patients from healthy controls than Digit Span and Trail Making task [5].

Currently, multiple versions of DSST are available, such as written, oral, and computerized versions [1,2,7,8]. These versions of DSST have their own advantages. The written DSST using paper and pencil has culture-free design and low literacy requirement. The oral DSST can reduce the dependence on upper limbs, making it possible to measure patients with upper extremity mobility impairments. Recent advances in computer technology make it possible for developing computerized cognitive tests. In computerized cognitive tests, examinees usually need to provide button-press responses and computers can automatically record the behavioral results. Compared with both written and oral versions, the computerized DSST requires less administrative work of examiners and then reduces random measurement error caused by the examiners to some extent. Both written and computerized versions require examinees with intact upper limb function since the two versions need examinees to use the upper limb writing or pressing buttons to finish these tasks. In the oral version the examinee is asked to verbally indicate the answer. Thus, the existing versions of DSST are not suitable for individuals with upper limb dysfunction or non-verbal function. Attention is critical cognitive function in motor skills learning [9]. Moreover, previous study has shown that early deficit of cognitive function can interfere with the intervention effect of motor-impaired patients [10]. Therefore, evaluating cognitive function has a pivotal role in the rehabilitation of motor-impaired patients [11]. To evaluate the cognitive function of severely motor-impaired patients, the present study attempted to develop a novel brain-computer interface (BCI)-based DSST (BCI-DSST). BCI technology can directly connect the human brain with external devices through brain activity. Even though patients with motor impairment cannot send motor commands, they are still capable of sending intent commands by BCIs. Therefore, a BCI-based DSST is a promising method for measuring cognitive function of severely motor-impaired patients.

In this study, a steady-state visual evoked potential (SSVEP) BCI-based DSST was developed. First, nine symbols in the proposed SSVEP BCI-based DSST were designed with clearly different shapes for decreasing measurement errors due to misidentified symbols. Second, the digit-symbol pairs of each trial were different for reducing the practice effect. Third, a two-target SSVEP-based BCI was designed to judge whether the digit-symbol probe in the center of the user interface matched one of the nine digit-symbol pairs above the user interface. Thus, the proposed SSVEP BCI-based DSST can be used for individuals without intact upper limb function. The experiment results of this study verified that the proposed SSVEP BCI-based DSST was feasible and effective for healthy subjects.