Hierarchical structure of the cognitive processes in schizophrenia: the fundamental role of processing speed N. Ojeda a, b, ⁎, J. Peña a , D.J. Schretlen c, d , P. Sánchez e, f , E. Aretouli a , E. Elizagárate b, e, f , J. Ezcurra e , M. Gutiérrez b, f, g a Department of Psychology, University of Deusto, Bilbao, Spain b CIBERSAM, Centro de Salud Biomédica en Red de Salud Mental, Spain c Department of Psychiatry and Behavioral Sciences, The Johns Hopkins University School of Medicine, Baltimore, MD, USA d Russell H. Morgan Department of Radiology and Radiological Science, The Johns Hopkins University School of Medicine, Baltimore, MD, USA e Treatment Resistant Psychosis Unit, Hospital Psiquiátrico de Alava, Vitoria, Spain f Department of Neuroscience, Psychiatry Section, School of Medicine and Odontology, University of the Basque Country, Vizcaya, Spain g Hospital Santiago, Vitoria, Spain abstract article info Article history: Received 23 June 2011 Received in revised form 9 December 2011 Accepted 12 December 2011 Available online 5 January 2012 Keywords: Chronic schizophrenia Processing speed Neurocognition Objective: Decreased processing speed (PS) is a key feature of schizophrenia with respect to cognition, func- tional outcome and clinical symptoms. Our objective was to test whether PS slowing mediates other neuro- psychological deficits among patients with chronic schizophrenia. Method: One hundred patients with schizophrenia and 53 healthy adults completed a series of neuropsycho- logical measures that assess six cognitive domains. In addition to PS these included attention, verbal memory, visual memory, working memory, and executive functioning. Confirmatory factor analysis (CFA) was used to evaluate the fit of the 6-factor model. The cognitive performances of both groups were compared before and after controlling for the effect of PS, but also after controlling for the effect of each cognitive factor at a time. Finally, the PS-related variance was removed and the effect of the other cognitive factors was tested again. Results: CFA supported the hypothesized 6-factor cognitive structure. As expected, the patients and controls differed on all cognitive measures. However, after controlling for the effects of PS, group differences on the other five cognitive factors decreased substantially. Controlling for other factors produced smaller attenua- tion of group differences, and these effects were also partially accounted for by decreased PS. Conclusions: PS deficits account for most of the differences in cognition between patients with schizophrenia and healthy controls. PS slowing appears to be a core feature of schizophrenia, one that underlies impair- ments of working memory, executive functioning, and other abilities. © 2011 Elsevier B.V. All rights reserved. 1. Introduction Nearly two decades ago, Salthouse (1993) suggested that proces- sing speed (PS) was a core cognitive process that underlies cognitive decline in normal aging. According to this hypothesis, normal age- related decrements in mental PS mediate the declines observed in many other cognitive domains, such as working memory (WM) and episodic memory. The hypothesis that some basic cognitive processes (such as PS or attention) underlie higher cognitive processes (i.e., memory or executive functioning) was also later examined in schizo- phrenia (Hartman et al., 2003; Holthausen et al., 2003; Brebion et al., 2006; Rodriguez-Sanchez et al., 2007). Although several of these studies have emphasized the role of PS, this hypothesis has not been extensively investigated in large heterogeneous samples of patients with chronic schizophrenia. Furthermore, the differential influence of PS on cognitive performance has not been adequately demonstrated. There are several reasons to believe that PS deficits may be a core feature in schizophrenia. First, a recent meta-analysis has shown that patients with schizophrenia show larger deficits in PS than in other neurocognitive domains (Dickinson et al., 2007). Second, PS is also impaired in people at high risk of schizophrenia (Niendam et al., 2007) and psychosis (Jahshan et al., 2010). Third, slowed PS has also been found among unaffected relatives of patients with schizo- phrenia (Appels et al., 2003; Wang et al., 2007), further suggesting that PS deficits may be a possible endophenotype. In another line of research, PS deficits have been associated with several clinical and functional outcome measures in schizophrenia (McClure et al., 2007; Ojeda et al., 2008; Sanchez et al., 2009; Ojeda et al., 2010). For example, PS deficits have been related to global psy- chosocial functioning (Milev et al., 2005; Ojeda et al., 2008), social Schizophrenia Research 135 (2012) 72–78 ⁎ Corresponding author at: Department of Psychology, University of Deusto, Avda. Universidades, 24, 48007 Bilbao, Spain. Tel.: +34 94 413 90 00x2702; fax: +34 94 413 90 89. E-mail address: nojeda@deusto.es (N. Ojeda). 0920-9964/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2011.12.004 Contents lists available at SciVerse ScienceDirect Schizophrenia Research journal homepage: www.elsevier.com/locate/schres