A model of memory impairment in schizophrenia: Cognitive and clinical factors associated with memory efficiency and memory errors Gildas Brébion a,b,c, ⁎, Rodrigo A. Bressan d , Ruth I. Ohlsen a , Anthony S. David a a Section of Cognitive Neuropsychiatry, Institute of Psychiatry, King's College London, United Kingdom b Unit of Research and Development, Parc Sanitari Sant Joan de Déu, Barcelona, Spain c CIBERSAM, Spain d Center for Neuroimaging and Cognition, University of Sao Paulo, Brazil abstract article info Article history: Received 25 January 2013 Received in revised form 5 September 2013 Accepted 14 September 2013 Available online 7 October 2013 Keywords: Memory Source monitoring Schizophrenia Processing speed Hallucinations Background: Memory impairments in patients with schizophrenia have been associated with various cognitive and clinical factors. Hallucinations have been more specifically associated with errors stemming from source monitoring failure. Methods: We conducted a broad investigation of verbal memory and visual memory as well as source memory functioning in a sample of patients with schizophrenia. Various memory measures were tallied, and we stud- ied their associations with processing speed, working memory span, and positive, negative, and depressive symptoms. Results: Superficial and deep memory processes were differentially associated with processing speed, working memory span, avolition, depression, and attention disorders. Auditory/verbal and visual hallucinations were differentially associated with specific types of source memory error. Conclusions: We integrated all the results into a revised version of a previously published model of memory functioning in schizophrenia. The model describes the factors that affect memory efficiency, as well as the cognitive underpinnings of hallucinations within the source monitoring framework. © 2013 Elsevier B.V. All rights reserved. 1. Introduction Patients with schizophrenia demonstrate cognitive dysfunction that might be as incapacitating as their positive and negative symptoms (Elvevag and Golbdberg, 2000) and that cannot be explained merely by general intellectual disability (Gray et al., 2013). A substantial mem- ory deficit is commonly observed, mostly characterised by inability to conduct effective encoding of information (Cirillo and Seidman, 2003). This deficit might be secondary to a number of factors. We previously conducted a study of various aspects of verbal memory and source memory in a schizophrenia sample (Brébion et al., 2005a). Measures of memory efficiency in terms of the amount of correct responses, as well as measures of memory errors, were tallied. Verbal memory effi- ciency measures were split into superficial encoding assessed by rote rehearsal and deep encoding assessed by semantic organisation of the target words. We integrated all our results into a model that describes the associations of various cognitive and clinical factors with specific types of memory dysfunction (see Fig. 1). According to this model, pro- cessing speed is a major cognitive underpinning of verbal memory efficiency in schizophrenia and is related to both superficial and deep encoding. Selective attention has a more restricted role, insofar as it is only related to superficial encoding. At the clinical level, the deep mem- ory processes of semantic organisation are affected by depression and by the use of medications with anticholinergic activity. Among the neg- ative symptoms, only avolition is related to verbal memory efficiency. With regard to memory errors, they can be seen as stemming from var- ious types of source memory failure. These memory errors are associat- ed with hallucinations and/or delusions. Unexpectedly, we observed in our previous study that they were also inversely associated with certain negative symptoms reflecting diminished emotion or social interaction. We conducted a further investigation into another independent sam- ple of schizophrenia patients, in an attempt to extend and refine our model. Verbal as well as visual memory was investigated. Again, we tal- lied measures of memory efficiency, split into superficial and deep encoding, and measures of memory errors. Semantically organisable lists of words were used to assess the ability to carry out deep encoding of information. To broaden the exploration of encoding depth, we also used lists of mixed high- and low-frequency words. High-frequency words can be processed relatively automatically, whereas the processing of low-frequency words is more effortful. Further, we simultaneously presented mixed black-and-white and coloured pictures in a visual rec- ognition task. Coloured pictures were assumed to draw more attention and therefore undergo deeper encoding, to the detriment of the other Schizophrenia Research 151 (2013) 70–77 ⁎ Corresponding author at: Unit of Research and Development, Parc Sanitari Sant Joan de Déu, C\Doctor Antoni Pujadas 42, 08830 Sant Boi de Llobregat, Barcelona, Spain. E-mail address: gildas.brebion@pssjd.org (G. Brébion). 0920-9964/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.schres.2013.09.009 Contents lists available at ScienceDirect Schizophrenia Research journal homepage: www.elsevier.com/locate/schres