Scandinavian Journal of Psychology, 2004, 45, 55–60 © 2004 The Scandinavian Psychological Associations. Published by Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA. ISSN 0036-5564. Blackwell Publishing Ltd Modification of performance on the Span of Apprehension Task in a group of young people with early onset psychosis TORILL UELAND 1,2 , BJØRN RISHOVD RUND 2 , NANCY ELLIOT BORG 1,† , ELISABETH NEWTON 3 , RICK PURVIS 4 and TIL WYKES 4 1 Sogn Centre for Child and Adolescent Psychiatry, Oslo, Norway 2 Institute of Psychology, University of Oslo, Norway 3 Clinical Psychology, University College London, UK 4 Department of Psychology, Institute of Psychiatry, Kings College London, UK † Deceased Ueland, T., Rund, B. R., Borg, N. E., Newton, E., Purvis, R. & Wykes, T. (2004). Modification of performance on the Span of Apprehension Task in a group of young people with early onset psychosis. Scandinavian Journal of Psychology, 45, 55–60. The aim of the study was to examine the effect of contingent monetary reinforcement and enhanced instructions on Span of Apprehension (SPAN) performance in a group of young people with early onset psychosis. Twenty-five participants (mean age 16.7) received a 3- and 12- letter version of the SPAN task six times: baseline, three x intervention, post-test, and 10-day follow-up. No significant effects of time were found in the 3-letter condition, indicating a ceiling effect for accuracy. In the 12-letter condition detection rates improved significantly reaching a maximum at the third intervention ( p < 0.001). Performance showed a temporary decline at post-test, but performance returned to the max- imum level at the 10-day follow-up. The study suggests that SPAN performance can be improved in young people with early onset psychosis using relatively simple interventions and that performance gains are relatively durable. Key words: Span of apprehension, early onset psychosis, cognitive remediation. Torill Ueland, Sogn Centre for Child and Adolescent Psychiatry, Pb.26 Vinderen, 0319 Oslo, Norway. Phone: + 47 22 02 86 62, Fax: + 47 22 02 86 41. E-mail: torill.ueland@psykiatri.uio.no It has been firmly established that cognitive deficits are a core feature of schizophrenia (Andreasen, 1999; Green, 1998; Rund & Borg, 1999). Over the past two decades it has been well documented that a majority of individuals with schizophrenia manifest deficit performance on a wide vari- ety of cognitive tasks, ranging from basic disturbances in information processing to more complex mental functions. Recently, it has been suggested that cognitive deficits are directly related to current and future social functioning (Wykes & Dunn, 1992; Green, 1996), even more so than symptoms (Wykes, 1994; Green, 1998). Consequently, a great deal of interest has been directed towards establishing whether such cognitive deficits can be modified. The initial aim of cognitive remediation studies has been to determine whether performance deficits can be modified under any conditions. Targets for intervention have been directed at various areas of cognition, where people with a diagnosis of schizophrenia tend to show impaired perform- ance, for example, verbal memory, reaction time studies, vigilance, working memory and executive functions. There is growing evidence that at least some cognitive deficits can be modified. Improvements following training have for instance been shown on the Wisconsin Card Sorting Task (WCST) (Bellack, Muesser, Morrison, Tierney & Podell, 1990; Green, Satz, Ganzell & Valav, 1992; Bellack, Weinhardt, Gold & Gearon, 2001) and on the Continuous Performance Test (CPT) (Medalia, Aluma, Tryon & Merriam, 1998). In the present study we wanted to examine an elementary measure of visual information processing that some invest- igators (Nuechterlein & Dawson, 1984; Rund & Landrø, 1990) claim may be a vulnerability indicator of schizophrenia, the Span of Apprehension Task (SPAN). The SPAN task was originally developed by Estes and Taylor (1964), to provide an index of the rate of visual information process- ing. In the SPAN task letters are flashed briefly on a screen and subjects are required to report all letters seen (full report version), or specified target letters that appear in conjunc- tion with irrelevant letters (partial report version). Employ- ment of this task has demonstrated dysfunction not only in actively psychotic patients with schizophrenia, but also in individuals vulnerable to schizophrenia, including adult patients with schizophrenia in remission and unaffected first-degree relatives of schizophrenia probands (Asarnow, Steffy, MacCrimmon & Cleghorn, 1977; Asarnow & Mac- Crimmon, 1981; Asarnow, Granholm & Sherman, 1991). Particularly the partial report version of the SPAN task has been reported to be sensitive to schizophrenia (Asarnow & Sherman, 1984). One of the major determinants of deficient performance on the partial report version of the SPAN task is proposed to be in some aspect of serial scanning processes (Asarnow et al. , 1991). Support for serial scanning being central in the SPAN task is found in studies of scanpaths that show that people in general tend to make more mistakes in the bottom