Schizophrenia Research 156 (2014) 133–134

Contents lists available at ScienceDirect

Schizophrenia Research

j ourna l homepage: www.e lsev ie r .com/ locate /schres

Letter to the Editor

Mathematics achievement scores and earlypsychosis in school-aged children

Dear Editors,

Although evidence suggests that the typical age of onset of psy-chosis is in late adolescence into the mid-twenties (Kessler et al.,2007), those with psychosis vulnerability can exhibit earlier symp-toms at a sub-diagnostic level as well as premorbid characteristicsincluding academic difficulties. Within the United States, standard-izedmeasures of math achievement are routinely given as part of an-nual academic assessments and can therefore serve as one potentialtool for examining cognitive vulnerability in the premorbid period.Although studies have examined the relation between early symp-toms and general academic achievement (Ang and Tan, 2004;Dickson et al., 2012), to date, no study has specifically examinedthe relationship between these symptoms and standardized mathachievement. Given that working memory is integral for mathemat-ics computations (Meyer et al., 2010) and that impairments in thisdomain have characterized psychosis risk and schizophrenia pa-tients (Fusar-Poli et al., 2012; Lee and Park, 2005), this is a potential-ly important area of inquiry.

In the present investigation, a large sample of 486 school-agedchildren (M age = 10.99, SD = 2.49, 50% male) from the ColoradoLearning Disabilities Research Center was assessed to determine ifmath disorder was associated with psychosis items endorsed by par-ents and/or teachers (see Table 1). Recruiting and testing proceduresfor this study are described in detail in previous papers (see Willcuttet al., 2005). All participants had complete full-scale IQ (FSIQ), read-ing and math achievement, and behavioral data (Child BehavioralChecklist, see below).

Math achievement was assessed by the math subtest of the WideRange Achievement Test, Revised (WRAT-R; Jastak and Wilkinson,1984; M = 97.20, SD = 17.66). Per current research-area standards(see Mazzocco, 2008), children with math scores below the 11th per-centile were considered to be Math Learning Disabled (MLD). Basedon these cutoffs, 88 of the children were classified as MLD, and 398were classified as non-MLD. Psychosis symptoms were assessed using4 items taken from the Child Behavioral Checklist (CBCL; Achenbach,1991; Achenbach et al., 2003). The 4 items, chosen on the basis ofhigh face validity, included: 1) hears sounds or voices that aren'tthere (M = 0.01, SD = 0.09), 2) sees things that aren't there (M =0.01, SD = 0.10), 3) strange behaviors (M = 0.09, SD = 0.37), and4) strange ideas (M = 0.05, SD = 0.22). With respect to categoricalratings, children were classified as “psychosis-positive” if any of the 4aforementioned items were rated as “very true or often true” by bothparent and teacher. The current sample yielded 15 psychosis-positivechildren (3.09%) out of the 486 total participants who had CBCLratings. A continuous measure of overall symptom frequency wasalso calculated by summing the total scores of the 4 items (M = 0.16,

http://dx.doi.org/10.1016/j.schres.2014.03.0270920-9964/© 2014 Elsevier B.V. All rights reserved.

SD = 0.52). As the items showed weak internal consistency(Cronbach's α= .30), items were also evaluated independently in con-tinuous analyses.

The demographic and control/target variable differences betweenMLD/non-MLD as well as psychosis-positive/non-positive groups arepresented in Table 1. The odds ratio of being in the psychosis-positivegroup if one met criteria for MLD relative to one who was non-MLDwas 3.16 (95% CI = 1.10–9.13, z = 2.13, p = .03). In other words, achildwithMLDwas 3.16 timesmore likely to have exhibited a psychosissymptom relative to a child who was non-MLD. From a continuousperspective, Spearman correlations indicated that low math achieve-ment was linked with elevated symptoms for the entire sample (r =− .13, p ≤ .01) and the magnitude of the effect increased in thepsychosis-positive group alone (r = − .22, p = .22; Note: the lowpower for the supplementary analysis may have contributed to thisnull result). The strange behaviors (r = − .13, p ≤ .01) and strangeideas (r = − .10, p ≤ .05) items were most strongly correlated withmath achievement whereas the other two items were not significant.

Our results indicate that children who are low achieving in mathe-matics (i.e., MLD) exhibit a significantly greater likelihood of beingrated positively for psychosis behaviors relative to childrenwhoare typ-ically achieving in math, perhaps due to the working memory deficitsthat underlie both disorders. Given that children are routinely beingscreened for math achievement, this may be one means of examiningpremorbid functioning in a broader population. In a related point, con-tinued development in this area may allow for teachers and parents tomore easily identify childrenwhomay be good candidates for addition-al in-depth screening (e.g., screening for a prodromal syndrome). Futurestudies would benefit from the use of empirically validated psychosisscreeners (e.g., Prodromal-Questionnaire, Loewy et al., 2012; Communi-ty Assessment for Psychic Experiences, Stefanis et al., 2002) and largersample sizes. In addition, it would also be beneficial to determinewhether other common developmental disabilities that are frequentlyassessed for in school settings may also be tied to elevated risk forpsychosis symptomatology.

Role of fundingThis work was supported by National Institute of Child Health and Human Develop-

ment Grant # P50 HD 27802. Dr. Richard K. Olson is the Director of the Center and Dr.Erik Willcutt is the Primary Investigator.

ContributorsMs.Wu and Dr. Mittal prepared and analyzed the data, conceptualized the study, and

drafted the manuscript. Dr. Willcutt attained funding and assisted with the manuscript.Dr. Pennington also assisted with the manuscript.

Conflict of interestThere are no conflicts of interest to report.

AcknowledgmentsNone.

Table 1Demographics of the MLD versus non-MLD and psychosis-positive versus non-positive participants.

Math learning disabled (MLD) Psychosis

MLD(n = 88)M (SD)

Non-MLD(n = 398)M (SD)

ta df Positive(n = 15)M (SD)

Non-positive(n = 471)M (SD)

ta df

Gender (% male) 42% 50% 1.94 1 48% 73% 3.94 1Age 11.27 (2.78) 10.93 (2.42) −1.17 484 10.89 (1.95) 10.99 (2.51) 0.15 484FSIQ 95.90 (12.99) 110.44 (12.16) 10.03⁎⁎⁎ 484 105.93 (13.11) 107.87 (13.54) 0.55 484WRAT math achievement 72.63 (8.82) 102.65 (14.15) 19.09⁎⁎⁎ 484 91.80 (17.20) 97.38 (17.66) 1.21 484PIAT-R reading achievement 90.13 (12.10) 106.51 (11.11) 12.32⁎⁎⁎ 484 98.80 (15.24) 103.69 (12.84) 1.44 484CBCL DSM-oriented ADHD subscale 5.63 (3.89) 3.12 (3.28) −6.26⁎⁎⁎ 484 5.67 (3.87) 3.51 (3.50) −2.34⁎ 484CBCL hears sounds 0.02 (0.15) 0.01 (0.07) −1.67 484 .000 (.000) 0.009 (0.09) 0.36 484CBCL sees things 0.01 (0.11) 0.01 (0.10) −0.11 484 .000 (.000) 0.01 (0.10) 0.40 484CBCL strange behavior 0.20 (0.55) 0.07 (0.32) −3.19⁎⁎ 484 1.89 (0.52) 0.03 (0.18) −35.09⁎⁎⁎ 484CBCL strange ideas 0.11 (0.35) 0.03 (0.18) −3.98⁎⁎⁎ 484 0.27 (0.59) 0.04 (0.20) −3.94⁎⁎⁎ 484CBCL 4 psychosis item total 0.35 (0.80) 0.11 (0.42) −3.12⁎⁎ 484 2.13 (0.83) 0.09 (0.36) −20.51⁎⁎⁎ 484

Note: Analyses of covariance (ANCOVA) were used to control for FSIQ, reading achievement, and ADHD symptom severity when examiningMLD and non-MLD group differences in psy-chosis symptom severity. Analyses indicated theMLD and non-MLD group differences were still significant after controlling for FSIQ, reading achievement, and ADHD symptom severity.

a For gender, statistic is a χ2 instead of t-test.⁎ p b .05.⁎⁎ p b .01.⁎⁎⁎ p b .001.

134 Letter to the Editor

References

Achenbach, T.M., 1991. Integrative Guide for the 1991 CBCL/4-18, YSR, and TRF profiles.Department of Psychiatry, University of Vermont Burlington.

Achenbach, T.M., Rescorla, L.A., McConaughey, S., Pecora, P., Wetherbee, K., Ruffle, T.,2003. The Achenbach system of empirically based assessment. Handbook of psycho-logical and educational assessment of children: Personality, behavior, and context. , 2,pp. 406–432.

Ang, Y.G., Tan, H.Y., 2004. Academic deterioration prior to first episode schizophrenia inyoung Singaporean males. Psychiatry Res. 121 (3), 303–307.

Dickson, H., Laurens, K.R., Cullen, A.E., Hodgins, S., 2012. Meta-analyses of cognitive andmotor function in youth aged 16 years and younger who subsequently developschizophrenia. Psychol. Med. 42, 743–1498.

Fusar-Poli, P., Deste, G., Smieskova, R., Barlati, S., Yung, A.R., Howes, O., & Borgwardt, S.,2012. Cognitive functioning in prodromal psychosis: a meta-analysis. Arch. Gen. Psy-chiatry 69 (6), 562–571.

Jastak, S., Wilkinson, G.S., 1984. Wide Range Achievement Test. Revised: AdministrationManual, Wilmington, DE.

Kessler, R.C., Amminger, G.P., Aguilar‐Gaxiola, S., Alonso, J., Lee, S., Ustun, T.B., 2007. Age ofonset ofmental disorders: a review of recent literature. Curr. Opin. Psychiatry 20 (4), 359.

Lee, J., Park, S., 2005. Working memory impairments in schizophrenia: a meta-analysis. J.Abnorm. Psychol. 114 (4), 599.

Loewy, R.L., Therman, S., Manninen, M., Huttunen, M.O., Cannon, T.D., 2012. Prodromal psy-chosis screening in adolescent psychiatry clinics. Early Interv. Psychiatry 6 (1), 69–75.

Mazzocco, M., 2008. Introduction: Mathematics ability, performance, and achievement.Dev. Neuropsychol. 33 (3), 197–204.

Meyer, M.L., Salimpoor, V.N., Wu, S.S., Geary, D.C., Menon, V., 2010. Differential contribu-tion of specific working memory components to mathematics achievement in 2ndand 3rd graders. Learn. Individ. Differ. 20 (2), 101–109.

Stefanis, N., Hanssen, M., Smirnis, N., Avramopoulos, D., Evdokimidis, I., Stefanis, C.,Verdoux, H., Van Os, J., 2002. Evidence that three dimensions of psychosis have a dis-tribution in the general population. Psychol. Med. 32 (2), 347–358.

Willcutt, E.G., Pennington, B.F., Olson, R.K., Chhabildas, N., Hulslander, J., 2005. Neuropsy-chological analyses of comorbidity between reading disability and attention deficit hy-peractivity disorder: in search of the commondeficit. Dev.Neuropsychol. 27 (1), 35–78.

Sarah S. WuDepartment of Psychology and Neuroscience,

University of Colorado at Boulder, 345 UCB, Boulder,CO 80309-0345, United States

Corresponding author.E-mail address: sarah.wu@colorado.edu.

Vijay MittalDepartment of Psychology and Neuroscience,

University of Colorado at Boulder, 345 UCB, Boulder,CO 80309-0345, United States

Center for Neuroscience, University of Colorado at Boulder, 345 UCB,Boulder, CO 80309-0345, United States

Bruce PenningtonDepartment of Psychology, University of Denver, 2155 S. Race St., Denver,

CO 80210-4638, United States

Erik G. WillcuttDepartment of Psychology and Neuroscience,

University of Colorado at Boulder, 345 UCB, Boulder,CO 80309-0345, United States

7 November 2013