Postural Sway and Neurocognition in Individuals Meeting Criteria for a Clinical High-Risk Syndrome

Abstract

Neurocognitive deficits are implicated in individuals that meet criteria for a clinical high-risk (CHR) syndrome. Evidence in patients with schizophrenia suggest that cerebellar dysfunction may underlie neurocognitive deficits. However, little research has examined if similar associations are present in those meeting CHR criteria. This study examined associations between the MATRICs cognitive battery, postural sway (an index of cerebellar functioning), and SIPS-RC psychosis risk scores in a CHR sample (N=66). Poorer working memory and processing speed were associated with less postural control. Consistent with the cognitive dysmetria theory of schizophrenia, neurocognitive deficits are associated with cerebellar dysfunction in this critical population.

Introduction

Neurocognitive deficits are widely implicated in individuals that meet criteria for a clinical high-risk (CHR) syndrome [1]. These individuals exhibit attenuated psychotic symptoms and upwards of 35% develop formal psychosis within a two-year period [2]. Thus, those that meet criteria for a CHR syndrome are a critical population of interest for research focused on elucidating the pathophysiology of schizophrenia. Although neurocognitive deficits differentiate those that do and do not transition to psychosis in CHR samples [3,4], the causes of neurocognitive impairment are not currently well-understood, nor how impaired neurocognition is related to disease risk. Notably, the cognitive dysmetria theory of schizophrenia [5] may be a particularly promising model from which to examine these questions. For example, the cognitive dysmetria theory posits that the neurocognitive impairment observed in schizophrenia spectrum populations is a product of poor neural coordination across cerebello-thalamo-cortical circuitry [5,6]. Within this circuitry, evidence suggests the cerebellum may serve a prominent role in neurocognition [6,7]. Currently, postural sway stands as one of the most sensitive behavioral measures of cerebellar functioning available [8]. However, to date, the relationship(s) between postural sway, neurocognition, and psychosis risk have not been examined in individuals that meet criteria for a CHR syndrome.

In the present study, we examined associations between postural sway and neurocognition as assessed with the Measurement and Treatment Research to Improve Cognition in Schizophrenia (MATRICS) cognitive battery [9]. In addition, a personalized risk calculator (SIPS-RC) designed for individuals that meet CHR criteria developed by the Shanghai-At-Risk-for-Psychosis (SHARP) program [10] was used to examine associations between postural sway and psychosis risk. Consistent with work in patients implicating cerebellar dysfunction in processing speed and working memory deficits [11,12], we hypothesized that greater postural sway would be associated with poorer performance on MATRICS tests associated with these domains. Potential associations with other MATRICS test were evaluated in an exploratory fashion. Further, given recent findings demonstrating cerebello-thalamo-cortical dysconnectivity predicts time to conversion [13], we predicted greater postural sway would be associated with higher risk for psychosis.

Methods

Participants

Data for the present study were obtained from 66 individuals that met criteria for a CHR syndrome. Exclusion criteria included history of head injury, substance dependence, neurological or psychotic disorder. The Structured Clinical Interview for DSM-V Disorders (SCID; [14]) was administered to rule out a psychotic disorder and the Structured Interview for Psychosis Risk Syndromes (SIPS; [15,16]) was administered to diagnose a CHR syndrome. The SIPS defines a CHR syndrome by the recent onset of moderate to severe positive symptoms and/or a decline in functioning accompanying the presence of schizotypal personality disorder or a first-degree relative with psychosis. Informed consent was obtained in accordance with the protocol approved by the Institutional Review Board. Assent was obtained for individuals younger than 18 with informed consent obtained from their legal guardian.

Cerebellar Functioning Assessment

Postural sway was measured using an Advanced Mechanical Technology Incorporated Accusway force platform (Watertown, MA). Participants were instructed to remain as still as possible while standing in the middle of the platform. There were four conditions administered in a fixed order that were designed to increasingly challenge cerebellar function. In an effort to limit the number of statistical comparisons, we focused solely on the most difficult condition (eyes closed with feet close together, removing both sensory and trunk stability facilitators), and the one found to generate the largest sway area in individuals that meet criteria for a CHR syndrome [17]. The center of pressure (COP) area was measured using principal components analysis for use in analyses [18].

Cognitive Assessment and Psychosis Risk Calculator

To assess neurocognition, participants were administered the MATRICS cognitive battery [9], which includes 10 standardized cognitive measures. See Table 1 for a description of each cognitive test and its associated scoring. Uncorrected t-scores were used in analyses.

Table 1.

List of MATRICS Tests with Associated Description and Scoring

Cognitive Domains and Tests	Description	Scoring
Speed of Processing
- Trail Making Test: Part A (TMT-A)	- Participants draw a line to connect a series of randomly arranged, consecutively numbered circles.	- Time needed to connect all the numbered circles.
- Brief Assessment of Cognition in Schizophrenia (BACS), Symbol-Coding	- Participants substitute one of nine symbols for its associated number.	- Number of correctly substituted symbols in 90 s.
- Category Fluency: Animal Naming	- Participants name as many animals as possible.	- Number of animals reported in 60 s.
Working Memory
- Letter-Number Span (LNS)	- Participants verbally repeat a series of letters and numbers with the letters first in alphabetical order then the numbers in ascending order.	- Total number of correct responses.
- Wechsler Memory Scale (WMS-III): Spatial Span	- Participants must replicate the pattern of taps a test administrator produces on a board of geometric shapes.	- Total number of correct replications.
Reasoning and Problem Solving
- Neuropsychological Assessment Battery (NAB): Mazes	- Participants complete a series of increasingly challenging mazes.	- Sum of scores across the mazes. Scores on each maze reflect the time required time to complete the maze.
Attention/Vigilance
- Continuous Performance Test-Identical Pairs (CPT-IP)	- Participants view a series of numbers (ranging from 2 to 4 digits) and press a response button when they observe consecutively matching numbers.	- Average of d-prime scores across three blocks.
Visual Learning
- Brief Visuospatial Memory Test-Revised (BVMT-R)	- Participants view a set of figures for 10 s and then must draw the figures accurately and in their correction location.	- Sum of scores across trials (3 trials total).
Verbal Learning
- Hopkins Verbal Learning Test-Revised (HVLT-R)	- Participants are read a list of 12 words and then verbally recall as many of the words as possible.	- Sum of scores across trials (3 trials total).
Social Cognition
- Mayer-Salovey-Caruso Emotional Intelligence Test (MSCEIT): Managing Emotions	- Participants hear a series of vignettes and identify the effectiveness of various emotion-driven responses of individuals within each story.	- Transformed sum of scores of the various responses across vignettes.

Psychosis Risk Calculator

The SIPS-RC was developed by the SHARP program [10] to provide an individualized risk score for psychosis conversion. Probability estimates for the SIPS-RC are quantified using the following SIPS structures: functional deterioration within the past year, positive and negative symptom severity, and low levels of dysphoric mood. Functional decline is measured using the Global Assessment of Functioning scale [19]. Positive and negative symptom severity are both quantified using sum scores from three domains from the positive (unusual thought content, suspiciousness, and disorganized communication) and negative (social anhedonia, expression of emotion, and ideational richness) SIPS dimensions, respectively. Dysphoric mood is measured using the dysphoric mood dimension from the general symptoms SIPS domain. Each of these four SIPS-RC components are then converted into simple binary variables (Yes/No) that can be used to determine the individualized risk estimate from the flowchart of calculated risk for CHR individuals provided in Zhang and colleagues [10].

Results

Descriptive characteristics and demographics are provided in Table 2. Pearson correlations were used to examine associations between postural sway, neurocognition, and psychosis risk. Two-tailed tests with an alpha level of .05 were used for all analyses. Greater postural sway was significantly associated with both poorer performance on LNS [r(64) = −.30, p = .01] and Category Fluency [r(64) = −.33, p = .007], and was marginally associated with poorer performance on the CPT:IP test [r(64) = −.22, p = .07]. There were no other significant associations between postural sway and neurocognition (|rs| < .13, ps > .32) (see Table 3). Increased postural sway was associated with higher SIPS-RC risk scores, but this correlation was not significant [r(64) = .18, p = .14].

Table 2.

Demographic/Clinical Characteristics and Descriptive Statistics for Postural Sway, Neurocognitive Measures, and Psychosis Risk

Age
Mean (SD)	18.61 (1.81)
Gender (%)
Male	58
Female	42
Education (yr.)
Mean (SD)	12.33 (1.79)
Parent education (yr.)
Mean (SD)	15.38 (3.08)
Alcohol Frequency
Mean (SD)	1.80 (1.60)
SCID Diagnoses (%)
Mood Disorders	30.3
Anxiety Disorders	27.3
Substance Use Disorders	17.6
Attention-Deficit/Hyperactivity Disorder	9.1
Postural Sway
Mean (SD)	103.28 (58.95)
SIPS-RC
Mean (SD)	5.21 (2.62)
TMT-A
Mean (SD)	54.00 (11.37)
BACS
Mean (SD)	54.22 (14.83)
Category Fluency
Mean (SD)	53.08 (13.46)
LNS
Mean (SD)	49.73 (8.97)
WMS-III: Spatial Span
Mean (SD)	54.54 (10.46)
NAB: Mazes
Mean (SD)	56.84 (8.60)
CPT-IP
Mean (SD)	44.94 (8.74)
BVMT-R
Mean (SD)	53.54 (9.60)
HVLT-R
Mean (SD)	51.30 (11.89)
MSCEIT
Mean (SD)	42.86 (9.92)

Note:Parent education is provided as a proxy for socio-economic status. SIPS-RC = Structured Interview for Psychosis Risk Syndromes- Risk Calculator; TMT-A = Trail Making Test: Part A; BACS = Brief Assessment of Cognition in Schizophrenia: Symbol-Coding; Category Fluency = Category Fluency: Animal Naming; LNS = Letter Number Sequencing; WMS-III: Spatial Span = Wechsler Memory Scale: Spatial Span; NAB: Mazes = Neuropsychological Assessment Battery: Mazes; CPT-IP = Continuous Performance Test-Identical Pairs; BVMT-R = Brief Visuospatial Memory Test-Revised; HVLT-R = Hopkins Verbal Learning Test-Revised; MSCEIT = Mayer-Salovey-Caruso Emotional Intelligence Test: Managing Emotions

Table 3.

Correlations between Postural Sway and Cognition

	TMT-A	BACS	Category Fluency	LNS	WMS-III: Spatial Span	NAB: Mazes	CPT	BVMT-R	HVLT-R	MSCEIT
Postural Sway	.05	.05	−.33 **	−.30 *	−.05	.03	−.23 †	.13	.05	.11

Note:

^*for p < .05

^**for p < .01

^†for p = .07, df = 64.

Because there is evidence that alcohol use can modulate postural control performance [20] and there are age-related changes in cognition [21], we conducted sensitivity analyses using partial correlations, controlling for alcohol frequency and age. Alcohol use was assessed using the Alcohol Use Scale [22] and ratings were numerically coded from 0 (never) to 5 (almost daily). One participant’s alcohol use was not assessed. The statistical significance of correlations did not change when controlling for alcohol use: LNS [r(62) = −.27, p = .03]; Category Fluency [r(62) = −.30, p = .02]. When controlling for age, the association between postural sway and LNS was marginally significant [r(63) = −.23, p = .06], results did not change for Category Fluency [r(63) = −.29, p = .02]. Similar findings were obtained when controlling for both alcohol use and age: LNS [r(61) = −.21, p = .095]; Category Fluency [r(61) = −.28, p = .03].

Discussion

The present findings that poorer working memory and slower processing speed were associated with impaired postural control ability suggest a link between cerebellar dysfunction and neurocognitive impairments in the pathophysiology of psychosis. These findings are consistent with several neuroimaging studies in patients with schizophrenia that demonstrate that verbal working memory and processing speed dysfunction are associated with both abnormal structure and reduced activity and connectivity within and across cerebello-thalamo-cortical regions [11,12,23–26]. Regarding the potential pathophysiological mechanisms underlying these associations, the anterior cerebellum in particular is strongly implicated in both temporal processing and postural control [27]. Consistent with a cognitive dysmetria model of psychosis [5,6], anterior cerebellum abnormalities may result in poor neural coordination across the cerebello-thalamo-cortical networks implicated in working memory and processing speed deficits in psychosis [11,12,23–26], contributing to the cognitive dysfunction observed in schizophrenia spectrum disorders. Indeed, evidence suggests that poorer temporal accuracy is associated with both abnormal anterior cerebellum resting state connectivity and symptom progression in those that meet CHR criteria [28]. Finally, although the association between greater postural sway and higher psychosis risk was not statistically significant, it is notable that the effect size was in the small to medium range [29], which points to the promise of postural sway as a potential biomarker of psychosis risk that warrants investigation in larger well-powered samples.

Although the sample size for the current study is comparable to other research in CHR populations, larger samples may reveal smaller but meaningful associations between cerebellar functioning and neurocognition. Similarly, given concomitant diagnoses are common in CHR samples and are also associated with postural control [30], larger samples would afford the ability to account for potential comorbidity effects. Furthermore, future work should focus on directly examining relationships between cerebellar network activity, neurocognition, and psychosis risk. In addition, given evidence for associations between neurological soft signs (NSS) and abnormal cerebellar-thalamic track development [31], research examining if NSS are associated with cognitive dysfunction would provide converging evidence for potential cerebellar involvement in cognitive impairment in the high-risk period.

In conclusion, consistent with the cognitive dysmetria theory of psychosis, the present findings showed links between a sensitive measure of cerebellar functioning and neurocognitive deficits in a CHR sample. Given evidence that cerebello-thalamo-cortical dysfunction predicts conversion to psychosis [13], as well as research showing that both working memory and processing speed impairment predict conversion [32], the present findings may inform the development of targeted treatments aimed at reducing neurocognitive dysfunction in high-risk populations. Indeed, there is evidence to suggest that cerebellar stimulation can ameliorate procedural learning deficits in individuals meeting CHR criteria [33], suggesting significant promise for these approaches.