- Email: [email protected]
Implicit memory functioning in schizophrenia: Explaining inconsistent findings of word stem completion tasks
Psychiatry Research 226 (2015) 347–351
Contents lists available at ScienceDirect
Psychiatry Research journal homepage: www.elsevier.com/locate/psychres
Implicit memory functioning in schizophrenia: Explaining inconsistent findings of word stem completion tasks María José Soler a, Juan Carlos Ruiz a,n, Carmen Dasí a, Inma Fuentes-Durá a,b a b
Faculty of Psychology, University of Valencia, Avenida Blasco Ibañez, 21, 46010 Valencia, Spain CIBERSAM, Spain
art ic l e i nf o
a b s t r a c t
Article history: Received 3 March 2014 Received in revised form 9 January 2015 Accepted 12 January 2015 Available online 28 January 2015
The definitive implicit memory profile of schizophrenia is yet to be clarified. Methodological differences between studies could be the reason for the inconsistent findings reported. In this study, we have examined implicit memory functioning using a word stem completion task. In addition, we have addressed methodological issues related with lexical and perceptual stimuli characteristics, and with the strategy used to calculate priming scores. Our data show similar performance values in schizophrenic patients and healthy controls. Furthermore, we have not detected significant differences in priming between the two groups, even when this parameter was calculated using three different procedures. These results are in line with those we have reported previously using the same stimuli in a word fragment completion task. Considered as a whole, our research suggests that implicit memory functioning in schizophrenia is unimpaired when assessed using word fragment or stem completion tasks. In light of this, future studies should follow standardized criteria to assess implicit memory when the sensitivity of the task employed is essential for identifying potential memory deficits in schizophrenia. & 2015 Elsevier Ireland Ltd. All rights reserved.
Keywords: Schizophrenia Implicit memory Priming Word stem completion task
1. Introduction Memory has been identified as one of the areas of impaired cognitive skills that characterize schizophrenia (Heinrichs and Zakzanis, 1998; Fiovaranty et al., 2005; Reichenberg and Harvey, 2007; Gold et al., 2009). However, memory is a complex domain, and, although areas of impairment have been identified over the last decades, the full pattern of deficits is still unclear. This uncertainty is particularly evident with respect to implicit memory. Theories of memory establish that the long-term memory system is divided into explicit and implicit memory (Graf and Schacter, 1985; Squire, 2004). Explicit memory requires an intentional retrieval of the encoded information of prior events, whereas implicit memory does not require this conscious access to encoded information. Explicit memory is usually assessed using recall and recognition tasks and is well known to be impaired in schizophrenic patients (Aleman et al., 1999; Mckenna et al., 2002; Schaefer et al., 2013). Research on procedural and implicit memory is not very extensive. Procedural memory has been explored using prototypical tests of motor skill learning, such as the pursuit rotor, and results show that it is unimpaired in schizophrenia (Clare
n
Corresponding author. Tel.: þ 34 963 864 414; fax: þ34 963 864 697. E-mail address: [email protected] (J.C. Ruiz).
http://dx.doi.org/10.1016/j.psychres.2015.01.016 0165-1781/& 2015 Elsevier Ireland Ltd. All rights reserved.
et al., 1993; Kern et al., 2010). Implicit memory has been measured through the priming effect (greater accuracy or faster performance when items have been studied previously) (Gabrieli, 1998), using tasks such as word stem completion or lexical decision (see Toth, 2000 for a complete list of implicit tests of memory), and results have been more inconsistent. The character of the priming effect has led to the classification of implicit tasks in different categories according to the nature of the target-dependent variable used in the task – accuracy or reaction time measures – and the nature of the processes involved in the tasks – conceptual or perceptual processes (see Toth, 2000 and Spataro et al., 2011 for more classification options and a detailed discussion of inconsistencies in the categorization of some tasks). Results obtained with perceptual tasks using reaction time measures (e.g. lexical decision) are mixed. The studies in question have focused on the semantic priming effect: the reduction in reaction time to the processing of a stimulus if a semantically related word is presented previous to its appearance, in comparison to when the word is not semantically related to the stimulus (Meyer and Schvaneveldt, 1971; Neely, 1991). Results have shown an absence of priming, normal priming or hyperpriming (Kiang et al., 2008; Pomarol-Clotet et al., 2008; Kreher et al., 2009; Niznikiewicz et al., 2010; Kiang et al., 2012; Pfeifel et al., 2012; Neil and Rossell, 2013). Studies using perceptual tasks with accuracy measures (word fragment completion or word stem completion) have also explored
348
M.J. Soler et al. / Psychiatry Research 226 (2015) 347–351
priming, but in these tasks priming is defined as the improvement in task performance when the subject has previously and unconsciously processed the stimuli involved in the task (in contrast to not having previously processed the stimuli). Research regarding this is also contradictory, with some authors reporting impaired implicit memory (Randolf et al., 1993; Kern et al., 2010) and others reporting preserved implicit memory in schizophrenia (Bazin and Perruchet, 1996; Perry et al., 2000). In a review by our group of studies employing perceptual tasks (Soler et al., 2011) we observed that all used a word stem completion (WSC) task and that the discrepancies in their results could have been due to differences in the methodology used, the criteria used in stimuli selection, how “priming” was defined, and participant characteristics such as IQ. Consequently, we designed a study to overcome the methodological problems of previous studies, and the results showed there were no differences between schizophrenic patients and controls in implicit memory evaluated using a WFC task. However, we used a word fragment completion (WFC) task, because, according to Bruss and Mitchell (2009), it involves only perceptual mechanisms, and so the results obtained can be considered a measure of perceptual implicit memory. The purpose of the present article is to extend the study of implicit memory functioning in schizophrenia by employing the commonly used task of WSC using standardized stimuli and defining priming according to Graf and Schacter (1985). In this way, we aim to avoid inconsistencies due to lexical and perceptual stimuli characteristics like frequency and familiarity, the number of possible solutions for a stem, or the baseline performance used to measure priming. To do this, we have: (1) designed a WSC task with stimuli selected from a normative database; (2) defined priming using an unambiguous definition; and (3) used three different criteria to assess priming. In a WSC task participants initially process a group of words with no explicit learning instructions. Next, after a short distraction task, participants have to complete a list of stems (e.g. mon _ _ -“money”). Half of the stems in the list are built from the group of previously processed words and the other half are built from new ones. All the stems include the three first letters of the word. The difference between the proportion of completed stems from processed and new words is the priming or implicit memory measure. However, variations in this standard procedure are common in the previous literature, particularly in the way performance is calculated, and, as a consequence, in the priming scores obtained. Kern et al. (2010) reported priming scores of 10.7% for patients and 18.7% for controls, which contrasted considerably from those of Randolf et al. (1993), who reported a priming of 30% for patients and 53% for controls. However, neither study indicated the number of possible solutions for the stems, which is necessary information for calculating priming. Perry et al. (2000) used stems that could be completed with at least 10 different words and found no differences between patients and controls, reporting priming values of around 50%. Indeed, most previous studies do not indicate how priming scores were calculated (e.g. Bazin and Perruchet, 1996). In summary, it should be underlined that previous studies have not specified what constituted a correct solution for the stem, which makes the comparison of their results and the measurement of priming difficult, thus hindering a definitive characterization of implicit memory in schizophrenia. When priming values are calculated the criterion used to define a hit is relevant, as the magnitude of the priming depends on it. Stems usually have more than one solution; indeed, they can be considered to be correctly completed according to any one of three criteria: 1) when it is completed with an expected word (usually that from which the stem was constructed); 2) when it is completed with any alternative that fits the stem; and 3), when an item-based baseline is
used (following the suggestion of Shaw (1997)). Priming for a stimulus is determined by comparing the probability of completion when it has been previously processed vs. when it has not been previously processed. This difference reflects the effect of implicit memory or the advantage that a stimulus has due to its unconscious processing. With the present study, we also set out to extend the data reported in our previous work (Soler et al., 2011), which we obtained using a word fragment completion task, by setting a word stem completion task. In this way, we have sought to improve the description of implicit memory functioning in schizophrenic patients in relation to healthy controls. 2. Method 2.1. Participants The study included 19 outpatients diagnosed with schizophrenia and 19 healthy controls, all of them native Spanish speakers. All patients were recruited from the Center for Rehabilitation and Social Integration-General Barroso in Valencia, Spain, and met the DSM-IV criteria for schizophrenia according to the Structured Clinical Interview for DSM-IV Axis I Disorder (SCID-I) (First et al., 2001), which was carried out by a trained psychiatrist. All were clinically stable, with an IQ above 85, no organic cerebral disease, no substance abuse or dependence, and no formal thought disorder. Antipsychotic medication type and dose had been stable in the previous 3 months in all patients. Subjects underwent the reduced version of the WAIS-III (Fuentes et al., 2010) to assess intellectual functioning and the Brief Psychiatric Rating Scale (BPRS) (Ventura et al., 1993) to evaluate symptomatology. An experienced, specifically trained psychologist rated the subjects' performance. Healthy control participants were recruited via advertisements in the community and were screened for exclusion criteria: history of psychotic or affective disorder; IQ below 85; substance abuse or dependence; and organic cerebral disease. Controls were matched with patients in age, gender and education. The demographic and clinical characteristics of all participants are summarized in Table 1. All participants gave their written informed consent prior to participation, after having the procedures explained to them. The study was in line with the Helsinki Declaration. 2.2. Experimental procedure The stimuli were 56 word stems selected from the Soler et al. (2009) norms, with a frequency over 0 and a familiarity range from 2.06 to 6.60 (in a 7-point scale). This database of Spanish verbal stimuli includes information relative to the target words corresponding to the stems (e.g. familiarity, frequency, number of meanings, activation and valence). This initial list was randomly divided into two lists (A and B). Both lists were statistically equivalent in frequency (list A: 11.12 (7.83); list B: 11.14 (7.02)), in familiarity (list A: 4.50 (1.16); list B: 4.90 (1.19)), in number of meanings (list A: 4.50 (1.99); list B: 3.71 (2.27)), in valence (list A: 5.383 (1.42); list B: 4.95 (1.42)), in activation (list A: 4.71 (1.08); list B: 4.87 (1.08)), and word length (list A: 6.18 (0.77); list B: 5.96 (0.88)). The stems were the three first letters of the words. The WSC task was administered individually in a quiet room. In the first phase of the task, 28 words in lowercase letters appeared one at a time in the center of a computer screen for 8 s. Participants were required to judge their knowledge on a scale of 1 (known) to 3 (unknown) using a rating sheet. Half of the participants received list A as a rated list (studied word stems) and the other half received list B. Afterwards, participants performed a 5-minute filler task in which they were asked to write the names of European countries. After the filler task, in the second phase of the task, participants had to complete 56 word stems and 2 initial fillers. The word stems were presented one at a time in the center of a computer screen for 12 s. each, in lowercase letters, with the missing letters indicated by underscores (e.g. ani _ _ _ -“animal”). Participants were instructed to write on a sheet the
Table 1 Demographic and clinical characteristics of patients and healthy controls. Schizophrenia (n¼ 19) Controls (n¼ 19) Age (years) Years of education IQ Female/Male ratio Illness onset (years) BPRS
42.26 (S.D. ¼ 6.64) 10.84 (S.D. ¼2.81) 92.26 (S.D.¼ 8.73) 7/12 17.89 (S.D. ¼ 7.51) 20.27 (S.D. ¼6.55)
t/χ2
P
44.37 (S.D. ¼ 6.93) 0.96 0.345 11.05 (S.D. ¼ 3.14) 0.22 0.830 98.58 (S.D. ¼ 11.73) 1.88 0.068 7/12
M.J. Soler et al. / Psychiatry Research 226 (2015) 347–351 first word that came to mind to successfully complete the stem. No explicit reference was made to the previously rated list. Twenty-eight of the word stems could be completed with the words from the rated list, and 28 with the remaining target words of the other list (non-studied word stems). Studied and non-studied word stems were presented randomly. The whole session lasted approximately 25 min. 2.3. Data analysis Statistical analyses were conducted using SPSS version 19.0. t tests and chisquare tests were used for comparisons of the demographic variables between groups. Analyses of variance (ANOVAs) were conducted to compare the performance of patients and controls and differences between experimental conditions (group and study condition). t tests were calculated to compare the priming scores of participants.
349
In addition, priming was calculated by subtracting the percentage of correct responses for a stem when it had previously been processed from the percentage of correct responses for the same stem when it had not previously been processed (third criterion). This difference represents the effect that the previous processing of a stem has on its subsequent completion, and it can also be considered a proxy measure of implicit memory. A t test was performed with these scores to compare priming in the two groups (patients: M ¼28.59, S.D. ¼17.74; controls: M ¼26.94, S. D.¼21.30); once again, differences did not prove to be significant, t(55) ¼ 0.52, P ¼0.602.
4. Discussion 3. Results Demographic and clinical characteristics of patients and controls are shown in Table 1. 3.1. Performance Two similar two-way analyses of variance (ANOVA) – group (patients vs. controls) and condition (studied vs. non-studied) – were performed to analyze performance. In the first, performance was defined as the percentage of correct responses, assuming that only stems completed with words from the initial list were appropriate responses (first criterion). In the second, performance was defined as the percentage of correct responses, assuming that both stems completed with words from the initial list and stems completed with words not included in the initial list were appropriate responses (second criterion). The mean percentage of correct responses using the two criteria in the two groups as a function of condition is shown in Table 2. Results using the first criterion yielded a significant effect of condition: F(1, 36) ¼84.84, P o0.0001. The percentage of completed stems was higher in the studied condition. However, neither the effect of group nor the interaction was significant. Results using the second criterion were similar. Only the effect of condition was significant, F(1, 36) ¼139.32, Po 0.0001. Once again, the studied condition showed better performance. 3.2. Priming To examine the difference between groups with respect to priming, defined as the difference between the proportion of correctly completed studied stems and the proportion of correctly completed non-studied stems, two t tests were performed using the two initial criteria to assess performance. The difference between the two groups was not significant when the first criterion – t(36) ¼0.16, P ¼0.873 – or when the second criterion – t(36) ¼ 0.09, P¼ 0.929 – was used. Table 2 shows priming scores in the two groups as a function of the criteria used to calculate them.
Although the present study does not compare implicit and explicit memory, the results we have obtained are in line with the hypothesis that implicit and explicit memory in schizophrenic patients have a dissociative functioning. In recent decades, research has focused on determining which memory domains are unimpaired and which are impaired (Gold et al., 2009; Kern et al., 2010) with the aim of identifying which cognitive areas need to be rehabilitated. In this framework, implicit memory does not seem to be a unique entity and can be assessed fractionally, like explicit memory (Wilson and Zangwill, 2002). This affirmation is based on dissociative results regarding the performance of patients and controls depending on the tasks used in the assessment of implicit memory, as these tasks require perceptual, conceptual or procedural stimulus processing on the part of the subject (McKenna et al., 2002). Accumulated evidence is contradictory and requires clarification. In this context, the aim of our study was to examine implicit memory functioning in schizophrenic patients while avoiding methodological factors that could lie behind the inconsistent results found in the literature. Previous studies of implicit memory using word-stem completion tasks have shown both impaired and unimpaired implicit memory functioning in schizophrenic patients in comparison with controls (p.e. Randolf et al., 1993; Bazin and Perruchet, 1996; Perry et al., 2000; Kern et al., 2010), thus obstructing the description of a definitive implicit memory profile for schizophrenia. However, our findings indicate that, when assessed using a WSC task, performance in implicit memory in schizophrenic patients is equivalent to that of healthy controls. We have performed a careful selection of the stems used in the task performed by our subjects. Contrary to most of the studies that have explored implicit memory functioning in schizophrenia, the stems used in our study were selected from a standardized database in order to control stimuli characteristics. At the same time we calculated priming by measuring performance according to the two alternatives that can be used to define a hit or a correct answer in a WSC task. Moreover, we have used a third approach to calculate priming. Our results indicate that schizophrenic patients perform similarly to controls in the WSC task independently of the criteria used
Table 2 Means and standard deviations for patients and controls on performance measures (correctly complete stem fragments percentage as a function of the criteria used to define a correct response) and priming. Correct response definition
Studied words
Non-studied words
Total
Priming
Mean
S.D.
Mean
S.D.
Mean
S.D.
Mean
S.D.
First criterion
Patients (n¼19) Controls (n¼ 19)
65.04 58.65
15.09 12.10
37.22 31.58
11.06 10.40
51.13 45.11
11.06 8.79
27.82 27.07
14.51 14.15
Second criterion
Patients (n¼19) Controls (n¼ 19)
82.33 76.50
14.51 11.80
63.35 57.14
17.33 16.79
73.97 66.82
13.06 13.21
18.98 19.36
13.57 12.04
350
M.J. Soler et al. / Psychiatry Research 226 (2015) 347–351
to define a correct response. In addition, independently of these criteria, the proportion of hits corresponding to previously processed words was significantly greater in comparison to those corresponding to previously unprocessed words. That is, there was a significant priming effect in the two groups regardless of the criteria applied, thereby suggesting that schizophrenics do not have an implicit memory deficit. However, both groups performed better in the task when the second criterion was used. In this way, priming magnitudes differed depending on the criteria used to define a hit. Priming magnitudes obtained using the first criterion (patients 27.82%; controls 27.07%) were significantly larger than those obtained using the second one (patients 18.98%; controls 19.36%). We wish to highlight that, although the criteria differed, no differences were evident between patients and controls, thus supporting the hypothesis that schizophrenic patients have unimpaired implicit memory. Moreover, when the priming for a stem was established in comparison with itself (previously processed vs. not previously processed), the results were replicated. Priming was significant in both groups (patients 28.59%; controls 26.94%), but there were no differences between them. It is noteworthy that priming magnitudes using this third criterion were similar to those obtained when a correct response was obtained using the first criterion, suggesting that the net gain due to the effect of implicit memory in a WSC task is more adequately measured when the stems have a single solution. The second objective of our study was to compare the results of WFC and WSC tasks to determine possible implicit memory differences in schizophrenia associated with the conceptual–perceptual theoretical distinction that establishes two implicit memory forms: one conceptual and the other perceptual. Although there is a debate about the “pure” nature of WFC and WSC tasks (Roediger et al., 1992; Bruss and Mitchell, 2009), WFC is usually defined as a perceptual task and WSC as a conceptual task. To allow an adequate comparison of the results of both tasks, we have used exactly the same words as in our previous study in which we employed a WFC task (Soler et al., 2011). The present results are similar to those obtained in that study, in which schizophrenic patients obtained significant priming scores, suggesting that a deficit in implicit memory is not evident when it is assessed with WFC and WSC tasks. The priming score was 31.81% in the previous study (WFC task) and 27.82% in the present study (WSC task). In both cases priming was calculated using the first criterion (the word fragments had only one solution), strengthening the idea that stems with unique solutions should be used in WSC tasks. Taken together, these results help to define implicit memory profile in schizophrenia and indicate that both conceptual and perceptual implicit memory are unimpaired in schizophrenia. It is important to note several limitations of our study. The sample size was relatively small and the participants were stable outpatients whose illness had been diagnosed years before. Future research should recruit inpatients and outpatient with a shorter duration of illness. Another limitation is that the influence of symptoms and medication on implicit memory performance in our schizophrenic patients was not considered. This is relevant, as the stage of the illness and antipsychotic medication may contribute in different ways to implicit memory functioning.
References Aleman, A., Hijman, R., de Haan, E.H.F., Kahn, R.S., 1999. Memory impairment in schizophrenia: a meta-analysis. American Journal of Psychiatry 156, 1358–1366. Bazin, N., Perruchet, P., 1996. Implicit and explicit associative memory in patients with schizophrenia. Schizophrenia Research 22, 241–248. Bruss, P.J., Mitchell, D.B., 2009. Memory systems, processes, tasks: taxonomic clarification via factor analysis. American Journal of Psychology 122, 175–189.
Clare, L., McKenna, P.J., Mortimer, A.M., Baddeley, A.D., 1993. Long-term memory in schizophrenia: what is impaired and what is preserved? Neuropsychologia 31, 1225–1241. Fiovaranty, M., Carlone, O., Vitale, B., Cinti, M.E., Clare, L., 2005. A meta-analysis of cognitive deficits in adults with a diagnosis of schizophrenia. Neuropsychology Review 17, 73–95. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B., 2001. Structured Clinical Interview for DSM-IV Axis I Disorders-Clinical Version (SCID-I). . American Psychiatric Press, Masson SA (Spanish Edition). Fuentes, I., Romero, M., Dasí, C., Ruiz, J.C., 2010. Versión abreviada del WAIS-III para su uso en la evaluación de pacientes con diagnóstico de esquizofrenia. /Short Version of the WAIS-III for use in the evaluation of patients with schizophrenia. Psicothema 22, 202–207. Gabrieli, J.D.E., 1998. Cognitive neuroscience of human memory. Annual Review of Psychology 49, 87–115. Gold, J.M., Hahn, B., Strauss, G.P., Waltz, J.A., 2009. Turning it upside down: areas of preserved cognitive function in schizophrenia. Neuropsychology Review 19, 294–311. Graf, P., Schacter, D.L., 1985. Implicit and explicit memory for new associations in normal and amnesic subjects. Journal of Experimental Psychology: Learning, Memory and Cognition 11, 501–518. Heinrichs, R.W., Zakzanis, K.K., 1998. Neurocognitive deficit in schizophrenia: a quantitative review of the evidence. Neuropsychology 12, 426–445. Kern, R.S., Hartzell, A.M., Izaguirre, B., Hamilton, A.H., 2010. Declarative and nondeclarative memory in schizophrenia: what is impaired? What is spare?. Journal of Clinical and Experimental Neuropsychology 32, 1017–1027. Kiang, M., Kutas, M., Light, G.A., Braff, D.L., 2008. An event-related brain potential study of direct and indirect semantic priming in schizophrenia. American Journal of Psychiatry 165, 74–81. Kiang, M., Christensen, B.K., Kutas, M., Zipursky, R.B., 2012. Electrophysiological evidence for primary semantic memory functional organization deficits in schizophrenia. Psychiatry Research 196, 171–180. Kreher, D.A., Goff, D., Kuperberg, G.R., 2009. Why all the confusion? Experimental task explains discrepant semantic priming effects in schizophrenia under “automatic“ conditions: evidence from event-related potentials. Schizophrenia Research 111, 174–181. Mckenna, P., Ornstein, T., Baddeley, A.D., 2002. Schizophrenia. In: Baddeley, A.D., Kopelman, M.D., Wilson, B.A. (Eds.), The Handbook of Memory Disorders. John Wiley and Sons Ltd., West Sussex, pp. 413–435. Meyer, D.E., Schvaneveldt, R.W., 1971. Facilitation in recognizing pairs of words: evidence of a dependence between retrieval operations. Journal of Experimental Psychology 90, 227–234. Neely, J.H., 1991. Semantic Priming Effects in Visual Word Recognition: A Selective Review of Current Findings and Theories. In: Besner, D., Humphreys, G.W. (Eds.), Basic Processes in Reading: Visual Word Recognition. Lawrence Erlbaum Associates, Hillsdale, NJ, pp. 264–336. Neil, E., Rossell, S., 2013. Comparing implicit and explicit semantic access of direct and indirect word pairs in schizophrenia to evaluate models of semantic memory. Psychiatry Research 205, 199–204. Niznikiewicz, M., Mittal, M.S., Nestor, P.G., McCarley, R.W., 2010. Abnormal inhibitory processes in semantic networks in schizophrenia. International Journal of Psychophysiology 75, 133–140. Perry, W., Light, G.A., Davis, H., Braff, D.L., 2000. Schizophrenia patients demonstrate a dissociation on declarative and non-declarative memory tests. Schizophrenia Research 46, 167–174. Pfeifel, S., Schiller, N.O., van Os, J., Riedel, W.J., Vlamings, P., Simons, C., Krapbendam, L., 2012. Electrophysiological correlates of automatic spreading of activation in patients with psychotic disorder and first-degree relatives. International Journal of Psychophysiology 84, 102–112. Pomarol-Clotet, E., Oh, T.M.S.S., Laws, K.R., McKenna, P.J., 2008. Semantic priming in schizophrenia: systematic review and meta-analysis. The British Journal of Psychiatry 192, 92–97. Randolf, C., Gold, J.M., Carpenter, C.J., Goldberg, T.E., Weinberger, D.R., 1993. Implicit memory in patients with schizophrenia and normal controls: effects of task demands and susceptibility to priming. Journal of Clinical and Experimental Neuropsychology 15, 853–866. Reichenberg, A., Harvey, P.D., 2007. Neuropsychological impairments in schizophrenia: integration of performance-based and brain imaging findings. Psychological Bulletin 133, 833–858. Roediger III, H.L., Weldon, M.S., Stadler, M.L., Riegler, G.L., 1992. Direct comparison of two implicit memory tests: word fragment and word stem completion. Journal of Experimental Psychology: Learning, Memory, Cognition 18, 1251–1269. Schaefer, J., Giangrande, E., Weinberger, D.R., Dickinson, D., 2013. The global cognitive impairment in schizophrenia: consistent over decades and around the world. Schizophrenia Research 150, 42–50. Shaw, R.J., 1997. Unprimed stem completion is only moderately predicted by word frequency and length. Behavior Research Methods, Instruments, & Computers 29, 401–424. Soler, M.J., Dasí, C., Ruiz, J.C., 2009. Datos normativos de 269 fragmentos de palabras españolas a partir de la base de Dasí, Soler y Ruiz (2004)/Normative data of 269 fragments of Spanish words from the Dasi, Soler and Ruiz' database (2004). Psicológica 30, 91–117. Soler, M.J., Ruiz, J.C., Vargas, M., Dasí, C., Fuentes, I., 2011. Perceptual priming in schizophrenia evaluated by word fragments and word stem completion. Psychiatry Research 190, 167–171.
M.J. Soler et al. / Psychiatry Research 226 (2015) 347–351
Spataro, P., Cestari, V., Rossi-Arnaud, C., 2011. The relationship between divided attention and implicit memory: a meta-analysis. Acta Psychologica 136, 329–339. Squire, L.R., 2004. Memory systems of the brain: a brief history and current perspective. Neurobiology of Learning and Memory 82, 171–177. Toth, J.P., 2000. Nonconscious Forms of Human Memory. In: Tulving, E., Craik, F.I.M. (Eds.), The Oxford Handbook of Memory. Oxford Univ. Press, New York, pp. 245–261.
351
Ventura, J., Green, M.F., Shaner, A., Liberman, R.P., 1993. Training and quality assurance with the Brief Psychiatric Rating Scale: “The drift busters”. International Journal of Methods in Psychiatric Research 3, 221–244. Wilson, B.A., Zangwill, O., 2002. Assessment of Memory Disorders. In: Baddeley, A.D., Kopelman, M.D., Wilson, B.A. (Eds.), The Handbook of Memory Disorders. John Wiley and Sons Ltd., West Sussex, pp. 617–636.
Contents lists available at ScienceDirect
Psychiatry Research journal homepage: www.elsevier.com/locate/psychres
Implicit memory functioning in schizophrenia: Explaining inconsistent findings of word stem completion tasks María José Soler a, Juan Carlos Ruiz a,n, Carmen Dasí a, Inma Fuentes-Durá a,b a b
Faculty of Psychology, University of Valencia, Avenida Blasco Ibañez, 21, 46010 Valencia, Spain CIBERSAM, Spain
art ic l e i nf o
a b s t r a c t
Article history: Received 3 March 2014 Received in revised form 9 January 2015 Accepted 12 January 2015 Available online 28 January 2015
The definitive implicit memory profile of schizophrenia is yet to be clarified. Methodological differences between studies could be the reason for the inconsistent findings reported. In this study, we have examined implicit memory functioning using a word stem completion task. In addition, we have addressed methodological issues related with lexical and perceptual stimuli characteristics, and with the strategy used to calculate priming scores. Our data show similar performance values in schizophrenic patients and healthy controls. Furthermore, we have not detected significant differences in priming between the two groups, even when this parameter was calculated using three different procedures. These results are in line with those we have reported previously using the same stimuli in a word fragment completion task. Considered as a whole, our research suggests that implicit memory functioning in schizophrenia is unimpaired when assessed using word fragment or stem completion tasks. In light of this, future studies should follow standardized criteria to assess implicit memory when the sensitivity of the task employed is essential for identifying potential memory deficits in schizophrenia. & 2015 Elsevier Ireland Ltd. All rights reserved.
Keywords: Schizophrenia Implicit memory Priming Word stem completion task
1. Introduction Memory has been identified as one of the areas of impaired cognitive skills that characterize schizophrenia (Heinrichs and Zakzanis, 1998; Fiovaranty et al., 2005; Reichenberg and Harvey, 2007; Gold et al., 2009). However, memory is a complex domain, and, although areas of impairment have been identified over the last decades, the full pattern of deficits is still unclear. This uncertainty is particularly evident with respect to implicit memory. Theories of memory establish that the long-term memory system is divided into explicit and implicit memory (Graf and Schacter, 1985; Squire, 2004). Explicit memory requires an intentional retrieval of the encoded information of prior events, whereas implicit memory does not require this conscious access to encoded information. Explicit memory is usually assessed using recall and recognition tasks and is well known to be impaired in schizophrenic patients (Aleman et al., 1999; Mckenna et al., 2002; Schaefer et al., 2013). Research on procedural and implicit memory is not very extensive. Procedural memory has been explored using prototypical tests of motor skill learning, such as the pursuit rotor, and results show that it is unimpaired in schizophrenia (Clare
n
Corresponding author. Tel.: þ 34 963 864 414; fax: þ34 963 864 697. E-mail address: [email protected] (J.C. Ruiz).
http://dx.doi.org/10.1016/j.psychres.2015.01.016 0165-1781/& 2015 Elsevier Ireland Ltd. All rights reserved.
et al., 1993; Kern et al., 2010). Implicit memory has been measured through the priming effect (greater accuracy or faster performance when items have been studied previously) (Gabrieli, 1998), using tasks such as word stem completion or lexical decision (see Toth, 2000 for a complete list of implicit tests of memory), and results have been more inconsistent. The character of the priming effect has led to the classification of implicit tasks in different categories according to the nature of the target-dependent variable used in the task – accuracy or reaction time measures – and the nature of the processes involved in the tasks – conceptual or perceptual processes (see Toth, 2000 and Spataro et al., 2011 for more classification options and a detailed discussion of inconsistencies in the categorization of some tasks). Results obtained with perceptual tasks using reaction time measures (e.g. lexical decision) are mixed. The studies in question have focused on the semantic priming effect: the reduction in reaction time to the processing of a stimulus if a semantically related word is presented previous to its appearance, in comparison to when the word is not semantically related to the stimulus (Meyer and Schvaneveldt, 1971; Neely, 1991). Results have shown an absence of priming, normal priming or hyperpriming (Kiang et al., 2008; Pomarol-Clotet et al., 2008; Kreher et al., 2009; Niznikiewicz et al., 2010; Kiang et al., 2012; Pfeifel et al., 2012; Neil and Rossell, 2013). Studies using perceptual tasks with accuracy measures (word fragment completion or word stem completion) have also explored
348
M.J. Soler et al. / Psychiatry Research 226 (2015) 347–351
priming, but in these tasks priming is defined as the improvement in task performance when the subject has previously and unconsciously processed the stimuli involved in the task (in contrast to not having previously processed the stimuli). Research regarding this is also contradictory, with some authors reporting impaired implicit memory (Randolf et al., 1993; Kern et al., 2010) and others reporting preserved implicit memory in schizophrenia (Bazin and Perruchet, 1996; Perry et al., 2000). In a review by our group of studies employing perceptual tasks (Soler et al., 2011) we observed that all used a word stem completion (WSC) task and that the discrepancies in their results could have been due to differences in the methodology used, the criteria used in stimuli selection, how “priming” was defined, and participant characteristics such as IQ. Consequently, we designed a study to overcome the methodological problems of previous studies, and the results showed there were no differences between schizophrenic patients and controls in implicit memory evaluated using a WFC task. However, we used a word fragment completion (WFC) task, because, according to Bruss and Mitchell (2009), it involves only perceptual mechanisms, and so the results obtained can be considered a measure of perceptual implicit memory. The purpose of the present article is to extend the study of implicit memory functioning in schizophrenia by employing the commonly used task of WSC using standardized stimuli and defining priming according to Graf and Schacter (1985). In this way, we aim to avoid inconsistencies due to lexical and perceptual stimuli characteristics like frequency and familiarity, the number of possible solutions for a stem, or the baseline performance used to measure priming. To do this, we have: (1) designed a WSC task with stimuli selected from a normative database; (2) defined priming using an unambiguous definition; and (3) used three different criteria to assess priming. In a WSC task participants initially process a group of words with no explicit learning instructions. Next, after a short distraction task, participants have to complete a list of stems (e.g. mon _ _ -“money”). Half of the stems in the list are built from the group of previously processed words and the other half are built from new ones. All the stems include the three first letters of the word. The difference between the proportion of completed stems from processed and new words is the priming or implicit memory measure. However, variations in this standard procedure are common in the previous literature, particularly in the way performance is calculated, and, as a consequence, in the priming scores obtained. Kern et al. (2010) reported priming scores of 10.7% for patients and 18.7% for controls, which contrasted considerably from those of Randolf et al. (1993), who reported a priming of 30% for patients and 53% for controls. However, neither study indicated the number of possible solutions for the stems, which is necessary information for calculating priming. Perry et al. (2000) used stems that could be completed with at least 10 different words and found no differences between patients and controls, reporting priming values of around 50%. Indeed, most previous studies do not indicate how priming scores were calculated (e.g. Bazin and Perruchet, 1996). In summary, it should be underlined that previous studies have not specified what constituted a correct solution for the stem, which makes the comparison of their results and the measurement of priming difficult, thus hindering a definitive characterization of implicit memory in schizophrenia. When priming values are calculated the criterion used to define a hit is relevant, as the magnitude of the priming depends on it. Stems usually have more than one solution; indeed, they can be considered to be correctly completed according to any one of three criteria: 1) when it is completed with an expected word (usually that from which the stem was constructed); 2) when it is completed with any alternative that fits the stem; and 3), when an item-based baseline is
used (following the suggestion of Shaw (1997)). Priming for a stimulus is determined by comparing the probability of completion when it has been previously processed vs. when it has not been previously processed. This difference reflects the effect of implicit memory or the advantage that a stimulus has due to its unconscious processing. With the present study, we also set out to extend the data reported in our previous work (Soler et al., 2011), which we obtained using a word fragment completion task, by setting a word stem completion task. In this way, we have sought to improve the description of implicit memory functioning in schizophrenic patients in relation to healthy controls. 2. Method 2.1. Participants The study included 19 outpatients diagnosed with schizophrenia and 19 healthy controls, all of them native Spanish speakers. All patients were recruited from the Center for Rehabilitation and Social Integration-General Barroso in Valencia, Spain, and met the DSM-IV criteria for schizophrenia according to the Structured Clinical Interview for DSM-IV Axis I Disorder (SCID-I) (First et al., 2001), which was carried out by a trained psychiatrist. All were clinically stable, with an IQ above 85, no organic cerebral disease, no substance abuse or dependence, and no formal thought disorder. Antipsychotic medication type and dose had been stable in the previous 3 months in all patients. Subjects underwent the reduced version of the WAIS-III (Fuentes et al., 2010) to assess intellectual functioning and the Brief Psychiatric Rating Scale (BPRS) (Ventura et al., 1993) to evaluate symptomatology. An experienced, specifically trained psychologist rated the subjects' performance. Healthy control participants were recruited via advertisements in the community and were screened for exclusion criteria: history of psychotic or affective disorder; IQ below 85; substance abuse or dependence; and organic cerebral disease. Controls were matched with patients in age, gender and education. The demographic and clinical characteristics of all participants are summarized in Table 1. All participants gave their written informed consent prior to participation, after having the procedures explained to them. The study was in line with the Helsinki Declaration. 2.2. Experimental procedure The stimuli were 56 word stems selected from the Soler et al. (2009) norms, with a frequency over 0 and a familiarity range from 2.06 to 6.60 (in a 7-point scale). This database of Spanish verbal stimuli includes information relative to the target words corresponding to the stems (e.g. familiarity, frequency, number of meanings, activation and valence). This initial list was randomly divided into two lists (A and B). Both lists were statistically equivalent in frequency (list A: 11.12 (7.83); list B: 11.14 (7.02)), in familiarity (list A: 4.50 (1.16); list B: 4.90 (1.19)), in number of meanings (list A: 4.50 (1.99); list B: 3.71 (2.27)), in valence (list A: 5.383 (1.42); list B: 4.95 (1.42)), in activation (list A: 4.71 (1.08); list B: 4.87 (1.08)), and word length (list A: 6.18 (0.77); list B: 5.96 (0.88)). The stems were the three first letters of the words. The WSC task was administered individually in a quiet room. In the first phase of the task, 28 words in lowercase letters appeared one at a time in the center of a computer screen for 8 s. Participants were required to judge their knowledge on a scale of 1 (known) to 3 (unknown) using a rating sheet. Half of the participants received list A as a rated list (studied word stems) and the other half received list B. Afterwards, participants performed a 5-minute filler task in which they were asked to write the names of European countries. After the filler task, in the second phase of the task, participants had to complete 56 word stems and 2 initial fillers. The word stems were presented one at a time in the center of a computer screen for 12 s. each, in lowercase letters, with the missing letters indicated by underscores (e.g. ani _ _ _ -“animal”). Participants were instructed to write on a sheet the
Table 1 Demographic and clinical characteristics of patients and healthy controls. Schizophrenia (n¼ 19) Controls (n¼ 19) Age (years) Years of education IQ Female/Male ratio Illness onset (years) BPRS
42.26 (S.D. ¼ 6.64) 10.84 (S.D. ¼2.81) 92.26 (S.D.¼ 8.73) 7/12 17.89 (S.D. ¼ 7.51) 20.27 (S.D. ¼6.55)
t/χ2
P
44.37 (S.D. ¼ 6.93) 0.96 0.345 11.05 (S.D. ¼ 3.14) 0.22 0.830 98.58 (S.D. ¼ 11.73) 1.88 0.068 7/12
M.J. Soler et al. / Psychiatry Research 226 (2015) 347–351 first word that came to mind to successfully complete the stem. No explicit reference was made to the previously rated list. Twenty-eight of the word stems could be completed with the words from the rated list, and 28 with the remaining target words of the other list (non-studied word stems). Studied and non-studied word stems were presented randomly. The whole session lasted approximately 25 min. 2.3. Data analysis Statistical analyses were conducted using SPSS version 19.0. t tests and chisquare tests were used for comparisons of the demographic variables between groups. Analyses of variance (ANOVAs) were conducted to compare the performance of patients and controls and differences between experimental conditions (group and study condition). t tests were calculated to compare the priming scores of participants.
349
In addition, priming was calculated by subtracting the percentage of correct responses for a stem when it had previously been processed from the percentage of correct responses for the same stem when it had not previously been processed (third criterion). This difference represents the effect that the previous processing of a stem has on its subsequent completion, and it can also be considered a proxy measure of implicit memory. A t test was performed with these scores to compare priming in the two groups (patients: M ¼28.59, S.D. ¼17.74; controls: M ¼26.94, S. D.¼21.30); once again, differences did not prove to be significant, t(55) ¼ 0.52, P ¼0.602.
4. Discussion 3. Results Demographic and clinical characteristics of patients and controls are shown in Table 1. 3.1. Performance Two similar two-way analyses of variance (ANOVA) – group (patients vs. controls) and condition (studied vs. non-studied) – were performed to analyze performance. In the first, performance was defined as the percentage of correct responses, assuming that only stems completed with words from the initial list were appropriate responses (first criterion). In the second, performance was defined as the percentage of correct responses, assuming that both stems completed with words from the initial list and stems completed with words not included in the initial list were appropriate responses (second criterion). The mean percentage of correct responses using the two criteria in the two groups as a function of condition is shown in Table 2. Results using the first criterion yielded a significant effect of condition: F(1, 36) ¼84.84, P o0.0001. The percentage of completed stems was higher in the studied condition. However, neither the effect of group nor the interaction was significant. Results using the second criterion were similar. Only the effect of condition was significant, F(1, 36) ¼139.32, Po 0.0001. Once again, the studied condition showed better performance. 3.2. Priming To examine the difference between groups with respect to priming, defined as the difference between the proportion of correctly completed studied stems and the proportion of correctly completed non-studied stems, two t tests were performed using the two initial criteria to assess performance. The difference between the two groups was not significant when the first criterion – t(36) ¼0.16, P ¼0.873 – or when the second criterion – t(36) ¼ 0.09, P¼ 0.929 – was used. Table 2 shows priming scores in the two groups as a function of the criteria used to calculate them.
Although the present study does not compare implicit and explicit memory, the results we have obtained are in line with the hypothesis that implicit and explicit memory in schizophrenic patients have a dissociative functioning. In recent decades, research has focused on determining which memory domains are unimpaired and which are impaired (Gold et al., 2009; Kern et al., 2010) with the aim of identifying which cognitive areas need to be rehabilitated. In this framework, implicit memory does not seem to be a unique entity and can be assessed fractionally, like explicit memory (Wilson and Zangwill, 2002). This affirmation is based on dissociative results regarding the performance of patients and controls depending on the tasks used in the assessment of implicit memory, as these tasks require perceptual, conceptual or procedural stimulus processing on the part of the subject (McKenna et al., 2002). Accumulated evidence is contradictory and requires clarification. In this context, the aim of our study was to examine implicit memory functioning in schizophrenic patients while avoiding methodological factors that could lie behind the inconsistent results found in the literature. Previous studies of implicit memory using word-stem completion tasks have shown both impaired and unimpaired implicit memory functioning in schizophrenic patients in comparison with controls (p.e. Randolf et al., 1993; Bazin and Perruchet, 1996; Perry et al., 2000; Kern et al., 2010), thus obstructing the description of a definitive implicit memory profile for schizophrenia. However, our findings indicate that, when assessed using a WSC task, performance in implicit memory in schizophrenic patients is equivalent to that of healthy controls. We have performed a careful selection of the stems used in the task performed by our subjects. Contrary to most of the studies that have explored implicit memory functioning in schizophrenia, the stems used in our study were selected from a standardized database in order to control stimuli characteristics. At the same time we calculated priming by measuring performance according to the two alternatives that can be used to define a hit or a correct answer in a WSC task. Moreover, we have used a third approach to calculate priming. Our results indicate that schizophrenic patients perform similarly to controls in the WSC task independently of the criteria used
Table 2 Means and standard deviations for patients and controls on performance measures (correctly complete stem fragments percentage as a function of the criteria used to define a correct response) and priming. Correct response definition
Studied words
Non-studied words
Total
Priming
Mean
S.D.
Mean
S.D.
Mean
S.D.
Mean
S.D.
First criterion
Patients (n¼19) Controls (n¼ 19)
65.04 58.65
15.09 12.10
37.22 31.58
11.06 10.40
51.13 45.11
11.06 8.79
27.82 27.07
14.51 14.15
Second criterion
Patients (n¼19) Controls (n¼ 19)
82.33 76.50
14.51 11.80
63.35 57.14
17.33 16.79
73.97 66.82
13.06 13.21
18.98 19.36
13.57 12.04
350
M.J. Soler et al. / Psychiatry Research 226 (2015) 347–351
to define a correct response. In addition, independently of these criteria, the proportion of hits corresponding to previously processed words was significantly greater in comparison to those corresponding to previously unprocessed words. That is, there was a significant priming effect in the two groups regardless of the criteria applied, thereby suggesting that schizophrenics do not have an implicit memory deficit. However, both groups performed better in the task when the second criterion was used. In this way, priming magnitudes differed depending on the criteria used to define a hit. Priming magnitudes obtained using the first criterion (patients 27.82%; controls 27.07%) were significantly larger than those obtained using the second one (patients 18.98%; controls 19.36%). We wish to highlight that, although the criteria differed, no differences were evident between patients and controls, thus supporting the hypothesis that schizophrenic patients have unimpaired implicit memory. Moreover, when the priming for a stem was established in comparison with itself (previously processed vs. not previously processed), the results were replicated. Priming was significant in both groups (patients 28.59%; controls 26.94%), but there were no differences between them. It is noteworthy that priming magnitudes using this third criterion were similar to those obtained when a correct response was obtained using the first criterion, suggesting that the net gain due to the effect of implicit memory in a WSC task is more adequately measured when the stems have a single solution. The second objective of our study was to compare the results of WFC and WSC tasks to determine possible implicit memory differences in schizophrenia associated with the conceptual–perceptual theoretical distinction that establishes two implicit memory forms: one conceptual and the other perceptual. Although there is a debate about the “pure” nature of WFC and WSC tasks (Roediger et al., 1992; Bruss and Mitchell, 2009), WFC is usually defined as a perceptual task and WSC as a conceptual task. To allow an adequate comparison of the results of both tasks, we have used exactly the same words as in our previous study in which we employed a WFC task (Soler et al., 2011). The present results are similar to those obtained in that study, in which schizophrenic patients obtained significant priming scores, suggesting that a deficit in implicit memory is not evident when it is assessed with WFC and WSC tasks. The priming score was 31.81% in the previous study (WFC task) and 27.82% in the present study (WSC task). In both cases priming was calculated using the first criterion (the word fragments had only one solution), strengthening the idea that stems with unique solutions should be used in WSC tasks. Taken together, these results help to define implicit memory profile in schizophrenia and indicate that both conceptual and perceptual implicit memory are unimpaired in schizophrenia. It is important to note several limitations of our study. The sample size was relatively small and the participants were stable outpatients whose illness had been diagnosed years before. Future research should recruit inpatients and outpatient with a shorter duration of illness. Another limitation is that the influence of symptoms and medication on implicit memory performance in our schizophrenic patients was not considered. This is relevant, as the stage of the illness and antipsychotic medication may contribute in different ways to implicit memory functioning.
References Aleman, A., Hijman, R., de Haan, E.H.F., Kahn, R.S., 1999. Memory impairment in schizophrenia: a meta-analysis. American Journal of Psychiatry 156, 1358–1366. Bazin, N., Perruchet, P., 1996. Implicit and explicit associative memory in patients with schizophrenia. Schizophrenia Research 22, 241–248. Bruss, P.J., Mitchell, D.B., 2009. Memory systems, processes, tasks: taxonomic clarification via factor analysis. American Journal of Psychology 122, 175–189.
Clare, L., McKenna, P.J., Mortimer, A.M., Baddeley, A.D., 1993. Long-term memory in schizophrenia: what is impaired and what is preserved? Neuropsychologia 31, 1225–1241. Fiovaranty, M., Carlone, O., Vitale, B., Cinti, M.E., Clare, L., 2005. A meta-analysis of cognitive deficits in adults with a diagnosis of schizophrenia. Neuropsychology Review 17, 73–95. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B., 2001. Structured Clinical Interview for DSM-IV Axis I Disorders-Clinical Version (SCID-I). . American Psychiatric Press, Masson SA (Spanish Edition). Fuentes, I., Romero, M., Dasí, C., Ruiz, J.C., 2010. Versión abreviada del WAIS-III para su uso en la evaluación de pacientes con diagnóstico de esquizofrenia. /Short Version of the WAIS-III for use in the evaluation of patients with schizophrenia. Psicothema 22, 202–207. Gabrieli, J.D.E., 1998. Cognitive neuroscience of human memory. Annual Review of Psychology 49, 87–115. Gold, J.M., Hahn, B., Strauss, G.P., Waltz, J.A., 2009. Turning it upside down: areas of preserved cognitive function in schizophrenia. Neuropsychology Review 19, 294–311. Graf, P., Schacter, D.L., 1985. Implicit and explicit memory for new associations in normal and amnesic subjects. Journal of Experimental Psychology: Learning, Memory and Cognition 11, 501–518. Heinrichs, R.W., Zakzanis, K.K., 1998. Neurocognitive deficit in schizophrenia: a quantitative review of the evidence. Neuropsychology 12, 426–445. Kern, R.S., Hartzell, A.M., Izaguirre, B., Hamilton, A.H., 2010. Declarative and nondeclarative memory in schizophrenia: what is impaired? What is spare?. Journal of Clinical and Experimental Neuropsychology 32, 1017–1027. Kiang, M., Kutas, M., Light, G.A., Braff, D.L., 2008. An event-related brain potential study of direct and indirect semantic priming in schizophrenia. American Journal of Psychiatry 165, 74–81. Kiang, M., Christensen, B.K., Kutas, M., Zipursky, R.B., 2012. Electrophysiological evidence for primary semantic memory functional organization deficits in schizophrenia. Psychiatry Research 196, 171–180. Kreher, D.A., Goff, D., Kuperberg, G.R., 2009. Why all the confusion? Experimental task explains discrepant semantic priming effects in schizophrenia under “automatic“ conditions: evidence from event-related potentials. Schizophrenia Research 111, 174–181. Mckenna, P., Ornstein, T., Baddeley, A.D., 2002. Schizophrenia. In: Baddeley, A.D., Kopelman, M.D., Wilson, B.A. (Eds.), The Handbook of Memory Disorders. John Wiley and Sons Ltd., West Sussex, pp. 413–435. Meyer, D.E., Schvaneveldt, R.W., 1971. Facilitation in recognizing pairs of words: evidence of a dependence between retrieval operations. Journal of Experimental Psychology 90, 227–234. Neely, J.H., 1991. Semantic Priming Effects in Visual Word Recognition: A Selective Review of Current Findings and Theories. In: Besner, D., Humphreys, G.W. (Eds.), Basic Processes in Reading: Visual Word Recognition. Lawrence Erlbaum Associates, Hillsdale, NJ, pp. 264–336. Neil, E., Rossell, S., 2013. Comparing implicit and explicit semantic access of direct and indirect word pairs in schizophrenia to evaluate models of semantic memory. Psychiatry Research 205, 199–204. Niznikiewicz, M., Mittal, M.S., Nestor, P.G., McCarley, R.W., 2010. Abnormal inhibitory processes in semantic networks in schizophrenia. International Journal of Psychophysiology 75, 133–140. Perry, W., Light, G.A., Davis, H., Braff, D.L., 2000. Schizophrenia patients demonstrate a dissociation on declarative and non-declarative memory tests. Schizophrenia Research 46, 167–174. Pfeifel, S., Schiller, N.O., van Os, J., Riedel, W.J., Vlamings, P., Simons, C., Krapbendam, L., 2012. Electrophysiological correlates of automatic spreading of activation in patients with psychotic disorder and first-degree relatives. International Journal of Psychophysiology 84, 102–112. Pomarol-Clotet, E., Oh, T.M.S.S., Laws, K.R., McKenna, P.J., 2008. Semantic priming in schizophrenia: systematic review and meta-analysis. The British Journal of Psychiatry 192, 92–97. Randolf, C., Gold, J.M., Carpenter, C.J., Goldberg, T.E., Weinberger, D.R., 1993. Implicit memory in patients with schizophrenia and normal controls: effects of task demands and susceptibility to priming. Journal of Clinical and Experimental Neuropsychology 15, 853–866. Reichenberg, A., Harvey, P.D., 2007. Neuropsychological impairments in schizophrenia: integration of performance-based and brain imaging findings. Psychological Bulletin 133, 833–858. Roediger III, H.L., Weldon, M.S., Stadler, M.L., Riegler, G.L., 1992. Direct comparison of two implicit memory tests: word fragment and word stem completion. Journal of Experimental Psychology: Learning, Memory, Cognition 18, 1251–1269. Schaefer, J., Giangrande, E., Weinberger, D.R., Dickinson, D., 2013. The global cognitive impairment in schizophrenia: consistent over decades and around the world. Schizophrenia Research 150, 42–50. Shaw, R.J., 1997. Unprimed stem completion is only moderately predicted by word frequency and length. Behavior Research Methods, Instruments, & Computers 29, 401–424. Soler, M.J., Dasí, C., Ruiz, J.C., 2009. Datos normativos de 269 fragmentos de palabras españolas a partir de la base de Dasí, Soler y Ruiz (2004)/Normative data of 269 fragments of Spanish words from the Dasi, Soler and Ruiz' database (2004). Psicológica 30, 91–117. Soler, M.J., Ruiz, J.C., Vargas, M., Dasí, C., Fuentes, I., 2011. Perceptual priming in schizophrenia evaluated by word fragments and word stem completion. Psychiatry Research 190, 167–171.
M.J. Soler et al. / Psychiatry Research 226 (2015) 347–351
Spataro, P., Cestari, V., Rossi-Arnaud, C., 2011. The relationship between divided attention and implicit memory: a meta-analysis. Acta Psychologica 136, 329–339. Squire, L.R., 2004. Memory systems of the brain: a brief history and current perspective. Neurobiology of Learning and Memory 82, 171–177. Toth, J.P., 2000. Nonconscious Forms of Human Memory. In: Tulving, E., Craik, F.I.M. (Eds.), The Oxford Handbook of Memory. Oxford Univ. Press, New York, pp. 245–261.
351
Ventura, J., Green, M.F., Shaner, A., Liberman, R.P., 1993. Training and quality assurance with the Brief Psychiatric Rating Scale: “The drift busters”. International Journal of Methods in Psychiatric Research 3, 221–244. Wilson, B.A., Zangwill, O., 2002. Assessment of Memory Disorders. In: Baddeley, A.D., Kopelman, M.D., Wilson, B.A. (Eds.), The Handbook of Memory Disorders. John Wiley and Sons Ltd., West Sussex, pp. 617–636.