Deconstructing processing speed deficits in schizophrenia: Application of a parametric digit symbol coding test
Introduction
Evidence for reduced information processing speed has been consistently observed in individuals diagnosed with schizophrenia (e.g., Chapman and Chapman, 1973, King, 1991, Mohamed et al., 1999, Nuechterlein, 1977). Recently, the evaluation of potential illness biomarkers has brought renewed attention to information processing inefficiency in schizophrenia, with particular focus on timed digit symbol coding tasks, such as the Wechsler Digit Symbol Coding Test (DSC; Wechsler, 1997). A recent meta-analysis concluded that patients with schizophrenia have the most significant impairment on the DSC relative to all common neuropsychological measures, a finding that was not sensitive to medication exposure (Dickinson et al., 2007). Furthermore, illness-related slowing predicted diagnostic status even after patients' substantial generalized cognitive deficit was taken into account (Dickinson et al., 2008). Reduced processing speed is observed in patients with schizophrenia prior to the onset of illness (Niendam et al., 2003) and is associated with clinical (Leeson et al., 2008) and functional outcomes (Brekke et al., 1997, Gold et al., 2002). Finally, processing speed deficits are present among patients' non-schizophrenic first- and second-degree relatives, suggesting that there is a vulnerability-related component to slowed processing (Glahn et al., 2007). Together, these data suggest that processing speed inefficiencies, as measured by the digit symbol coding, represent an important behavioral marker of the pathophysiology of schizophrenia.
Despite growing interest in processing speed inefficiencies (Dickinson, 2008, Morrens et al., 2007, Salthouse, 1996b), the cognitive roots of this deficit remain unclear. Two cognitive components are traditionally considered critical for determining good performance on the DSC: motor speed and relational memory (Joy et al., 2004, Salthouse, 1996b), the latter thought to be most critical when test takers do not consult the DSC code key for each item. After reviewing reports of DSC performance among healthy individuals, Joy et al. (2003) suggested that visual scanning efficiency is also critical for good performance, emphasizing its role when test takers consult the code key frequently during test administration. Additionally, they raised the possibility that among cognitively impaired individuals, these abilities may contribute differently to overall processing speed than they do among unimpaired study participants. Currently, the relative contributions of these cognitive constructs to the speed of processing deficit observed in schizophrenia are unknown. Indeed, patients with schizophrenia show impairments in motor coordination and speed (e.g., Cannon et al., 2000, Saykin et al., 1994, relational memory performance (e.g., van Erp et al., 2008, and visual scanning efficiency (e.g., Mahurin et al., 1998).
To date, published attempts to deconstruct the processing speed impairment in schizophrenia have focused almost exclusively on regression-based approaches in which performance on neuropsychological tests thought to share particular cognitive components with DSC (e.g., graphomotor speed as measured by the Symbol Copy Test; Wechsler, 1997) is used to predict digit symbol coding performance (e.g., Joy et al., 2004). Several of these quite clever studies converge to suggest that patients' motor slowing cannot account entirely for their general reduction in processing speed (Jogems-Kosterman et al., 2001, Morrens et al., 2006). Although informative with respect to links with published work, this approach has limitations, and may be somewhat misleading in regard to unique variance attributable to a given neuropsychological test (Dickinson et al., 2008).
A complementary approach involves the development of “refined behavioral tasks” (Jonides and Nee, 2005) that work experimentally to isolate, control, and/or exaggerate the role of particular information sub-processes to identify their roles in influencing the production of complex, overt behavior. Unfortunately, to our knowledge there are no published manuscripts that attempt to deconstruct the processing speed deficits in schizophrenia. To this end, we adapted a computerized digit symbol coding task (Glahn et al., 2007) in order to independently manipulate the demands placed on visual scanning efficiency and relational memory while holding decisional and motoric requirements constant. More specifically, this novel digit symbol coding task includes conditions where the set size and presentation consistency are varied over 60-s blocks of trials. Increasing the number of digit symbol pairs or set size from 3 to 6 to 9 pairs augments the demands placed upon both scanning efficiency and relational memory. In contrast, presentation consistency, designed to localize relational memory, was assessed by maintaining the digit symbol pairing throughout the block of trials (Fixed Condition) or by randomly assigning these pairing for each trial in a 60-s block (Random Condition).
In the current study, we applied this task to a large, clinically well characterized group of patients with schizophrenia and community control subjects. We predicted that despite the apparently minor influence of relational memory on cognitive processing speed when measured in a healthy sample (e.g., Joy et al., 2003), increasing subjects' capacity to boost processing efficiency by utilizing relational memory would greatly enhance the disparity between non-ill controls and a sample of schizophrenia patients.
Section snippets
Participants
This study was approved by the UTHSCSA IRB, and written informed consent was obtained from all subjects prior to participation. The sample included 85 outpatients with schizophrenia and 30 healthy comparison subjects matched for age (mean age in years [s.d.]: patients, 45.83 [10.06], controls, 43.82 [10.13]), education level (mean number of years completed [s.d.]: patients, 11.83 [4.15], controls, 12.57 [2.53]), sex (number female: patients, 45/83, and controls, 16/3), and handedness. The
Results
The 2 × 2 × 3 mixed-effects ANCOVA resulted in a main effect of Diagnostic Group (F[1,112] = 58.95, p < 0.01), in which control participants displayed a greater number of correct responses than patients did, supporting Hypothesis 1 (see Fig. 2). Consistent with Hypothesis 2, subjects performed better in the fixed condition than they did in the random condition (F[1,112] = 21.21, p < 0.01). A main effect of Set Size (F[2,224] = 8.90, p < 0.01), best described by a linear contrast (F[1,112] = 11.98, p < 0.01),
Discussion
The present findings are consistent with our prediction that schizophrenia patients' failure to utilize relational memory to enhance cognitive processing efficiency may account for a substantial extent of their processing speed impairment. The group difference in processing efficiency in the fixed pair, memory-intensive condition was much greater than the difference observed in the random pair condition, when heavy demands were placed on visual scanning ability—a result that may be somewhat
Role of funding source
Funding for this study was provided by a National Alliance for Research on Schizophrenia and Depression, NARSAD, young investigator award to Dr. Glahn, and by a National Institute of Mental Health, NIMH, grant R01MH074047 to Dr. Velligan. The funding agencies had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.
Contributors
Dr. Glahn designed the experimental task. Drs. Glahn and Velligan supported subject recruitment, which was carried out by Mr. Rice and Ms. Woolsey, Chaves, and Maples, all of whom also participated in data collection and clinical rating. Dr. Reichenberg provided statistical consultation, and Drs. Glahn and Bachman conducted statistical analyses and interpreted the results. Drs. Glahn and Bachman wrote the first draft of the paper. All authors contributed to and have approved the final
Conflicts of interest
The authors declare that they have no conflicts of interest.
Acknowledgment
The authors would like to thank the research subjects for their generous participation in this study.
References (37)
- et al.
The inheritance of neuropsychological dysfunction in twins discordant for schizophrenia
Am. J. Hum. Genet.
(2000) - et al.
General and specific cognitive deficits in schizophrenia: Goliath defeats David?
Biol. Psychiatry
(2008) - et al.
Psychomotor slowing and planning deficits in schizophrenia
Schizophr. Res.
(2001) - et al.
Speed and memory in the WAIS-III Digit Symbol-Coding subtest across the adult lifespan
Arch. Clin. Neuropsychol.
(2004) - et al.
Executive-frontal lobe cognitive dysfunction in schizophrenia: a symptom subtype analysis
Psychiatry Res.
(1998) - et al.
Sensorimotor and cognitive slowing in schizophrenia as measured by the Symbol Digit Substitution Test
J. Psychiatr. Res.
(2006) - et al.
Remember and know judgments during recognition in chronic schizophrenia
Schizophr. Res.
(2008) - et al.
Brief psychiatric rating scale expanded version: how do new items affect factor structure?
Psychiatry Res.
(2005) - et al.
Neuropsychological and psychophysiological correlates of psychosocial functioning in schizophrenia
Schizophr. Bull.
(1997) - et al.(1973)
Psychophysiological dysfunctions in the developmental course of schizophrenic disorders
Schizophr. Bull.
Digit symbol coding and general cognitive ability in schizophrenia: worth another look?
Br. J. Psychiatry
Overlooking the obvious: a meta-analytic comparison of digit symbol coding tasks and other cognitive measures in schizophrenia
Arch. Gen. Psychiatry
Adjudicating neurocognitive endophenotypes for schizophrenia
Am. J. Med. Genet. B. Neuropsychiatr. Genet.
Beyond hypofrontality: a quantitative meta-analysis of functional neuroimaging studies of working memory in schizophrenia
Hum. Brain Mapp.
Cognitive correlates of job tenure among patients with severe mental illness
Am. J. Psychiatry
Cited by (49)
Neurocognitive profiles of patients with first-episode and recurrent depression: a cross-sectional comparative study from China
2021, Journal of Affective DisordersComparison of cognitive dysfunction between schizophrenia and bipolar disorder patients: A meta-analysis of comparative studies
2020, Journal of Affective DisordersResting-state hyperconnectivity within the default mode network impedes the ability to initiate cognitive performance in first-episode schizophrenia patients
2020, Progress in Neuro-Psychopharmacology and Biological PsychiatryValidation of a brain-computer interface version of the digit symbol substitution test in healthy subjects
2020, Computers in Biology and MedicineCitation Excerpt :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.