• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Schizophr Res. Author manuscript; available in PMC Feb 1, 2009.
Published in final edited form as:
PMCID: PMC2383319

Effect of Retrieval Effort and Switching Demand on fMRI Activation During Semantic Word Generation in Schizophrenia


Verbal fluency deficits in schizophrenia are difficult to interpret because the tasks are multi-factorial and groups differ in total words generated. We manipulated retrieval and switching demands by requiring alternation between over-learned sequences in which retrieval is relatively automatic (OS) and semantic categories requiring increased retrieval effort (SC). Controlled processing was also manipulated by including switching and non-switching conditions, and formal thought disorder (FTD) was assessed with the communication disorders index (CDI). The OS/SC semantic fluency paradigm was administered during fMRI to 13 patients with schizophrenia and 14 matched controls. Images were acquired on a 3 Tesla Siemens scanner using compressed image acquisition to allow for cued overt word production. Subjects alternated between OS, SC, OS-switch, SC-switch, and baseline blocks. Images were pre-processed in SPM-2, and a two-stage random effects analysis tested within and between group contrasts. There were no group performance differences. fMRI analysis did not reveal any group differences during the OS non-switching condition. Both groups produced expected activation in bilateral prefrontal and inferior parietal regions. However, during the SC condition patients had greater activation than controls in left prefrontal, right anterior cingulate, right superior temporal, bilateral thalamus, and left parietal regions. There was also evidence of patient over-activation in prefrontal, superior temporal, superior parietal, and visual association areas when a switching component was added. FTD was negatively correlated with BOLD response in the right anterior cingulate, cuneus and superior frontal gyrus during increased retrieval demand, and positively correlated with fMRI activation in the left lingual gyrus, right fusiform gyrus and left superior parietal lobule during increased switching demand. These results indicate that patients are able to successfully perform effortful semantic fluency tasks during non-speeded conditions. When retrieval is relatively automatic there does not appear to be an effect of schizophrenia on fMRI response. However, when retrieval and controlled processing demands increase, patients have greater activation than controls despite unimpaired task performance. This inefficient BOLD response may explain why patients are slower and less accurate on standard self-paced fluency tasks.

Keywords: schizophrenia, verbal fluency, semantic memory, functional imaging, brain mapping

1. Introduction

Verbal fluency tasks demand rapid generation of words belonging to a specific category (e.g., “animals”), and successful performance necessitates intact lexical and semantic knowledge, well-organized semantic networks, and efficient word access and retrieval requiring varying degrees of cognitive control (Posner & DiGirolamo, 1998; Shallice, 1988). There is a general consensus that semantic knowledge in schizophrenia is intact (Elvevåg and Storms 2003), and that semantic impairments are due to deficits in processing speed (Vinogradov et al., 2002), organization of material retrieved from the semantic store (Moelter et al., 2005), dysfunctional word access and retrieval (Himelhoch et al., 1996), or a combination of these factors (Goldberg et al., 1998; Vinogradov et al., 2002; Zakzanis et al., 2000). The current study manipulated semantic retrieval and between-category switching demands to test the hypothesis that group differences in functional magnetic resonance imaging (fMRI) blood oxygenation level dependent (BOLD) response are secondary to the high level of cognitive control required by standard speeded verbal fluency paradigms. It is predicted that patients’ fMRI response will be unimpaired during generation of words to over-learned sequence categories (e.g., days of the week) when retrieval and switching demands are low.

Previous attempts to assess word retrieval during fluency tasks dissociated clustering from switching components (Troyer, Moscovitch, Winocur, 1997; Troyer et al., 1998; Robert et al., 1998). Clustering occurs when individuals generate exemplars within a specific sub-category such as “Farm Animal” (e.g., “horse, cow, pig”), and is thought to require minimal retrieval effort due to more automatic spread-of-activation (Collins & Loftus, 1975) between strongly associated nodes in the semantic network. Switching requires increased cognitive control to generate exemplars from a new sub-category (e.g., “Jungle Animal”) and inhibit responses to the previous sub-category once it has been exhausted. Investigators have found that individuals with schizophrenia have a reduced number of switches (Robert et al, 1998) and an increased switch latency (Elvevåg, Fisher, Gurd, Goldberg, 2002), Although some investigators have found that number of clusters and cluster size (Elvevåg et al., 2002) as well as associative strength within a cluster is intact (Moelter et al., 2001), others have found that cluster size and the proportion of clustered responses can be reduced (Bozikas et al., 2005; Moelter et al., 2001), suggesting that semantic organization might be disrupted even during less effortful retrieval conditions.

Functional imaging studies using word production tasks initially supported a fronto-temporal disconnection hypothesis in schizophrenia (Friston and Frith, 1995). In healthy subjects phonemic verbal fluency increased prefrontal and decreased superior temporal lobe activity (Frith et al., 1995). With one exception (Spence et al., 2000), patients had increased superior temporal gyrus (STG) activity relative to healthy comparison subjects (Frith et al., 1995; Liddle 1997; Yurgelun-Todd et al., 1996). However, results for the prefrontal cortex (PFC) were less consistent, with evidence of PFC reductions (Curtis et al., 1998; Schaufelberger et al., 2005; Yurgelun-Todd et al., 1996), unimpaired PFC function (Frith et al., 1995; Liddle 1997; Spence et al., 1998), and reduced PFC lateralization in schizophrenia (Weiss et al., 2004). Inconsistent PFC findings have been attributed to group differences in task performance (Fletcher et al., 1998). However, investigating task performance effects has been hindered due to use of covert word production tasks because of concerns about increased head motion. A recent study that equated groups on covert word production based on number of button presses (Boksman et al., 2005) found that patients demonstrated reduced activity in the PFC and anterior cingulate gyrus (ACC), accompanied by reduced functional connectivity between the ACC and temporal cortex despite equivalent performance. Attenuated PFC and ACC activation was also seen in an overt phonemic fluency task where groups were matched on performance (Schaufelberger et al., 2005).

Variability in regional activation and semantic memory performance has also been attributed to different levels of formal thought disorder (FTD) within and between patient samples. FTD refers to the disordered speech of patients with schizophrenia (Andreasen, 1982) and has long been associated with semantic memory performance (Goldberg et al., 1998) and related prefrontal cortex (PFC) and superior temporal gyrus (STG) structure (Rajarethinam et al., 2000; Shenton et al., 1997; Vita et al., 1995) and function (Fu et al., 2005; Kircher et al., 2000, 2001). These studies have generally found an inverse relation between FTD and STG volume and function (Rajarethinam et al., 2000; Shenton et al., 1997; Vita et al., 1995; Kircher et al., 2001), and a positive relationship with PFC function (Kircher et al., 2000, 2001) although no relationship was found in a study of acutely ill patients (Fu et al., 2005). Investigation of severe FTD is complicated by the fact that acutely ill patients are often unable to provide consent or comply with task demands. Rather than study acutely ill patients with prominent FTD, the current study uses the communications disorder index (CDI; Docherty et al., 1996) to detect subtle communication failures indicative of disordered thinking, which has been shown to be more strongly correlated with measures of attention, memory and executive control than traditional clinical scales (Docherty et al., 2005).

A limitation of speeded fluency tasks is that dissociation of clustering from switching deficits can be confounded by group differences in total word production (Mayr 2002). This limitation was addressed by Elvevåg and colleagues (2002) who used total word output as a covariate and modified methods for scoring clusters and switches. However, an alternative approach is to control for total word production by using a cued retrieval task (Gurd et al., 2002). In the Gurd et al. paradigm (2002) retrieval demand is manipulated by alternating between less effortful over-learned sequences (OS; e.g., “days of the week”) and more effortful semantic categories (SC; e.g., “vegetables”). Switching is manipulated by including both single category and alternating category conditions (e.g., “vegetables” versus “vegetables/fruit”). In an fMRI study of healthy volunteers Gurd et al. (2002) found that greater retrieval demand (SC minus OS contrast) increased activity in bilateral prefrontal, anterior cingulate, and cerebellar areas, whereas increased switching demand (Switch minus No Switch contrast) led to greater activity in the bilateral superior parietal cortex. The current study will utilize an overt version of this paradigm to assess the impact of retrieval and between-category switching demands and FTD on regional brain activation in healthy volunteers and individuals with schizophrenia.

2. Experimental Materials and Methods

2.1. Participants

The final sample consisted of 13 patients with schizophrenia (7 female) and 14 healthy comparison subjects (6 female) from the Schizophrenia Center at the University of Pennsylvania. This sample did not include one comparison subject who later developed a psychiatric illness and one patient with excessive motion. Groups did not differ in age (comparison subjects: mean=34.3 years, SD=9.3; patients: mean=36.2 years, SD=6.9), education (comparison subjects: mean=13.6 years, SD=1.5; patients: mean=13.5 years, SD=2.6), or parental education (comparison subjects: mean=13.0 years, SD=2.9; patients: mean=14.3 years, SD=2.9). All comparison subjects and all but two patients were right-handed.

2.2. Clinical Evaluation

Research participants underwent standardized assessment, consisting of medical, neurological, psychiatric, and laboratory tests. The psychiatric evaluation for patients included clinical assessment, structured interview (First et al., 1996), history obtained from family, care providers, and records, and scales for measuring symptoms and functioning administered by investigators trained to a criterion reliability of 0.90, intra-class correlation. All patients had a DSM-IV (APA, 1994) diagnosis of schizophrenia or schizoaffective disorder - depressed type and no history of any other disorder or event that might affect brain function.

Patients were mildly to moderately ill according to the Scale for Assessment of Negative Symptoms (SANS; Andreasen et al., 1983; mean score=36.6, SD=31.8, range=0–124), Scale for Assessment of Positive Symptoms (SAPS; Andreasen et al., 1984; mean score=19.5, SD=14.2, range=2–46), and Brief Psychiatric Rating Scale (BPRS; Overall & Gorham, 1980; mean score=33.8, SD=9.32, range=18.0–47.0). Patient speech samples were rated for FTD using the Communication Disorders Index (CDI; Docherty et al., 1996). After establishing inter-rater reliability (ICC=.85) two raters (MTB and STM) scored 10 minute speech transcripts for total communication failures, which were converted to frequency scores (per 100 words), and log transformed to normalize distributions. Patients had an average CDI score of 1.8 (SD=1.1, range 0.1–4.1), indicating mild to moderate communication failures. All patients were medicated (5 first generation, 8 second generation), with average daily dose of 415 mg/day in chlorpromazine equivalents (SD=261, range=200–1067) and 16.6 mg/day in olanzapine equivalents (SD=10.5, range=8.00–42.7). No patient was receiving anticholinergic or other psychotropic medication. Average age of onset of psychotic symptoms in the context of functional decline was 22.3±7.51 (range 13–36), with illness duration of 10.6±7.23 years.

Healthy participants underwent the same intake evaluation procedures and were administered the SCID non-patient version and the SCID-II (First et al., 1995, 1997). They had no history of illness affecting brain function or major psychiatric illness in first-degree relatives. After complete description of the study written informed consent was obtained.

2.3. Test Procedures

A modified version of the OS/SC word generation paradigm (Gurd et al., 2002) was administered as a block design during fMRI. Participants were asked to generate words during 4 conditions: over-learned sequence fluency (OS) – subjects produced words from single known ordered lists (days of the week, months of the year, numbers, letters of the alphabet); semantic category fluency (SC) – subjects produced words from a single semantic category (fruits, vehicles, furniture, vegetables); OS switching (OS Switch) – subjects alternated between two known lists (e.g., “one”, “Monday”, “two”, “Tuesday”); SC switching (SC Switch) – subjects alternated between two semantic categories (e.g., “chair”, “peach”, “table”, “apple”). Each condition was repeated twice in a pseudo-random order in blocks of 7 trials each, with a 4 sec. presentation rate and 2 sec. inter-stimulus-interval (ISI). A trial began with a visual cue (e.g., “Produce Fruit”) that remained on the screen during the 4 sec. verbal response window, and was replaced with a visual cross hair during the subsequent 2 sec. ISI. Participants were instructed to respond “Skip” if unable to generate a word on a given trial. There were also 8 baseline blocks (repeat the word “Rest”) between conditions.

Unlike the original covert version of the task, a compressed image acquisition sequence (Abrahams et al., 2003) was used to provide silent periods for generation of overt responses that were recorded by the task administrator. Each run lasted 11.2 min., and was repeated once, for a total task time of 23 minutes. Task administration was triggered by the scanner and coupled to image acquisition using the PowerLaboratory® platform (Chute & Westall, 1997). All subjects correctly completed practice trials before imaging to ensure comprehension and familiarity with the response apparatus.

2.4. Image Acquisition

Data were acquired on a 3T Siemens Trio Scanner. A 5 minute magnetization-prepared, rapid acquisition gradient echo image (MPRAGE) was acquired for anatomic overlays of functional data and spatial normalization (Talairach & Tournoux, 1988). fMRI was acquired with blood oxygenation level dependent imaging (BOLD) (Bandettini et al., 1992) using a 36 slice whole-brain, single-shot gradient-echo (GE) echo-planar (EPI) sequence (TR/TE=6000/30 ms, FOV=240 mm, matrix= 64 X 64, slice thickness/gap=3/0.75 mm). A compressed pulse sequence (Abrahams et al., 2003) was used in which image acquisition occurred during the first 2 seconds of the TR, and overt word responses were generated during the subsequent 4 seconds when there was no sound from the MR gradients.

2.5. Data Analysis

Performance was indexed by measuring total number of words correctly generated (excluding “Skip” responses and non-responses). Repeated-measures multivariate analysis of variance (MANOVA) tested for effects of group (2 levels), condition (4 levels), fMRI run (2 levels), and higher order interactions. Significance threshold was set at p<.05 (two-tailed).

fMRI data were preprocessed in SPM2 (Wellcome Department of Cognitive Neurology, London, UK) using standard procedures. Images were slice-timed and motion-corrected to the median image using tri-linear interpolation. Despite overt word production, the amount of motion was small (less than 1.4 mm, and .03 degrees in all dimensions), and did not differ between groups for either translational [F(1,25) = 0.32, p = .58] or rotational parameters F(1,25) = 0.28, p = .60]. Structural images were co-registered to their median image, and spatial normalization parameters of the structural image to a standard T1 template were applied to all functional images. Resulting images were spatially smoothed (6mm FWHM, isotropic) and high-pass filtered (0.006 Hz).

Subject-level statistical analyses were performed using the general linear model in SPM2. Condition events were modeled using a modified boxcar haemodynamic response function. There were 4 event types: OS, SC, OS-switch, and SC-switch. Contrast maps were obtained through linear contrasts of event types. Random effects analyses were performed in SPM2 for within- and between-group comparisons. Within-group analyses were accomplished by entering whole brain contrasts for comparison subjects and patients separately into one-sample t-tests. Between-group analyses were accomplished by entering whole brain contrasts for each group into two-sample t-tests, with inclusive masks to restrict analysis to voxels with above-threshold activation obtained from within-group contrasts. A significance threshold of p<.05 corrected for multiple comparisons was calculated using a Monte Carlo simulation in AlphaSim (Ward, 2000) resulting in a height threshold of p<0.005, requiring a minimum of 10 voxels in a cluster. Simple regression analyses were performed in SPM2 to examine correlations between task related activation and FTD in patients. Regressions utilized inclusive masks to restrict analyses to voxels with above-threshold task activation (p<.05, uncorrected), and were then corrected for multiple comparisons as previously described.

3. Results

3.1. Behavioral Data

All participants understood instructions and completed tasks with minimal errors. As illustrated in Table 1, there were no main effects of group [F(1,25)=2.18, p=.15], fMRI run [F(1,25)=.05, p=.82], or group by fMRI run interaction [F(1,25)=.63, p=.43]. There was an effect of task condition [F(1,25)=6.29, p<.05], that did not interact with group [F(1,25)=.47, p=.50] or fMRI run [F(1,25)=1.03, p=.32]. Task effects reflected better performance during OS than SC [t(26)=−5.48, p<.001], during OS than OS-switch [t(26)=2.30, p<.05] and during OS-switch than SC-switch [t(26)= −2.59, p<.05]. The addition of a switching component did not affect SC performance [t(26)= −.67, p<.51].

Table 1
Performance During No Switching and Switching Conditions of an Overlearned Sequence (OS) versus Semantic Category (SC) Word Generation Task in Patients with Schizophrenia and Healthy Comparison Subjects.

3.2. fMRI Data

3.2.1. Effect of Retrieval Effort

Supporting our hypothesis, there were no between-group differences in fMRI activation during the OS condition (Table 2). Both groups showed bilateral activation in inferior frontal, middle frontal, and inferior parietal regions, with additional healthy control activation in the left cerebellum, and additional patient activation in the left postcentral gyrus and STG. As seen in Table 3, group differences were present for the SC minus baseline contrast. Comparison subjects had greater activation than patients in the right caudate. Patients had greater activation than comparison subjects in left prefrontal and motor cortex, right anterior cingulate gyrus, bilateral insula, right putamen, bilateral middle and STG, left inferior and superior parietal lobule, left lingual gyrus, left cuneus and precuneus, and cerebellum.

Table 2
Local Cluster-Level Maxima of Blood-Oxygen-Level-Dependent fMRI Signal Change During Over-learned Sequence (OS) versus Baseline Word Generation in Healthy Comparison Subjects and Patients With Schizophrenia.
Table 3
Local Cluster-Level Maxima of Blood-Oxygen-Level-Dependent fMRI Signal Change During Semantic Category (SC) versus Baseline Word Generation in Healthy Comparison Subjects and Patients With Schizophrenia.

When the effect of increased retrieval demand was directly tested (SC minus OS contrast) both groups activated the left ventrolateral prefrontal cortex (VLPFC, BA 45, 47). This VLPFC activation was also seen in a previous positron emission tomography (PET) study of healthy volunteers (Gourovitch et al., 2000) that contrasted semantic fluency with a control task similar to the OS condition. Healthy subjects also activated the left anterior cingulate, left caudate and right putamen, and patients showed additional activation in bilateral middle frontal gyrus, left superior temporal gyrus, left parahippocampal gyrus and right caudate (Table 4). Between-group contrasts did not reveal any areas of greater comparison subject versus patient activation. However, the opposite contrast revealed that patients had increased activation relative to controls in the left STG, right middle temporal gyrus, right parahippocampal gyrus, and left cuneus (Figure 1).

Figure 1
Blood-Oxygen-Level-Dependent fMRI Signal Change During Semantic Category (SC) Minus Over-Learned Sequence (OS) Word Generation. Statistical parametric maps are surface-rendered on smoothed brain images to illustrate activation in patients (red color) ...
Table 4
Local Cluster-Level Maxima of Blood-Oxygen-Level-Dependent fMRI Signal Change During Semantic Category (SC) versus Over-learned Sequence (OS) Conditions in Healthy Comparison Subjects and Patients With Schizophrenia.

3.2.2. Effect of Switching Demand

Adding a switching component differentially affected fMRI activation in the two groups. When retrieval demand was low (OS Switch minus OS No Switch contrast) comparison subjects did not show any areas of increased activation. However, patients increased activity in the right putamen, right lingual gyrus and left precuneus (Table 5), and their activation was greater than comparison subjects in the right middle frontal gyrus, anterior and posterior aspects of the left caudate, right thalamus, bilateral middle temporal gyrus, left STG, and right superior parietal lobule. When the retrieval demand was high (SC Switch minus SC No Switch contrast; Table 6), addition of a switching component increased right superior parietal activation in healthy comparison subjects, consistent with previous findings (Gurd et al., 2002). Patients also activated the right superior parietal lobule, and showed additional effects in the right precuneus. Between group contrasts did not reveal any areas of greater comparison subject activation. However, patients had greater activation than comparison subjects in the left medial frontal cortex.

Table 5
Local Cluster-Level Maxima of Blood-Oxygen-Level-Dependent fMRI Signal Change During Over-Learned Sequence Switch (OS Switch) versus Over-learned Sequence No Switch (OS No Switch) Conditions in Healthy Comparison Subjects and Patients With Schizophrenia. ...
Table 6
Local Cluster-Level Maxima of Blood-Oxygen-Level-Dependent fMRI Signal Change During Semantic Category Switch (SC Switch) versus Semantic Category No Switch (SC No Switch) Conditions in Healthy Comparison Subjects and Patients With Schizophrenia.

3.2.3. Relationship with Thought Disorder

Regression analyses revealed inverse findings for FTD relationships with retrieval effort and switching demand (Table 7). There were no correlations for the OS minus baseline contrast. For the SC minus baseline contrast higher FTD was associated with lower activation in right anterior cingulate and cuneus. A negative relationship was also found between FTD and activity in the right superior frontal gyrus for the SC minus OS contrast. For the OS Switch minus No Switch contrast higher FTD was associated with greater activation in left lingual gyrus and right fusiform gyrus. There was also a positive FTD correlation for the SC Switch minus No Switch contrast in the left superior parietal lobule.

Table 7
Local Cluster-Level Maxima of Significant Correlation Between Blood-Oxygen-Level-Dependent fMRI Signal Change and Formal Thought Disorder Ratings on the Communication Disorders Index (CDI) in Patients With Schizophrenia.

4. Discussion

Under non-speeded task conditions patients with schizophrenia can successfully perform semantic word generation tasks even when word retrieval and switching demands are increased. This supports previous conclusions that reduced processing speed contributes to verbal fluency deficits in schizophrenia (Vinogradov, 2002). During fMRI patients had normal frontal and parietal lobe activation when generating words to over-learned sequences where retrieval demands were low. However, when generating words during more effortful SC conditions, patients had increased left STG activation relative to controls. This increase in patient versus control STG activation was characteristic of previous phonemic and semantic fluency studies (Frith et al., 1995; Liddle 1997; Yurgelun-Todd et al., 1996). Over-activation was also observed when a between-category switching demand was added. STG over-activation extended to PFC, middle temporal, parietal, and basal ganglia regions. Results suggest that during paced, over-learned conditions, neural mechanisms of lexical retrieval and word generation are relatively intact in schizophrenia, and that poorly modulated neural response during standard fluency tasks is secondary to well documented deficits (Braff, 1993; Liddle, 1993; Cohen & Servan-Schreiber, 1992) in cognitive control.

In the current study controlled processing demands increased at word selection and generation stages as subjects moved from OS to SC conditions. During the OS condition selection of category exemplars required minimal control because strong direct links between associated nodes in the network facilitated an automatic spread-of-activation (Collins & Loftus, 1975) between related exemplars (e.g., “Monday”, “Tuesday”). Post-retrieval monitoring demands were also low because words were generated in a fixed sequence. Previous lexical priming studies have shown that patients are relatively unimpaired in automatic aspects of lexical processing (Minzenberg, Ober & Vinogradov, 2002). In contrast, SC conditions increased selection demand as subjects had to choose semantic subcategories to guide search (e.g., household furniture), and switch to new subcategories (e.g., office furniture) once exemplars from the previous subcategory were exhausted. Increased post-retrieval monitoring was also required to avoid repetitions since exemplars were no longer generated in a fixed sequence. Addition of a between-category switching component added further post-retrieval monitoring and additional response inhibition demands.

Initial verbal fluency imaging studies (Frith et al., 1995; Liddle 1997; Yurgelun-Todd et al., 1996) found reduced PFC and increased STG activation in patients, consistent with a fronto-temporal disconnection model (Friston and Frith, 1995) that has received support from recent functional connectivity studies (Jennings et al., 1998; Meyer-Lindenberg, 2005; Wolf et al., 2007). However, we found patient increases rather than decreases in PFC activity, which does not fit the classic disconnection model. A possible explanation for the current pattern of greater patient than control increases in the PFC and other regions is the inverted-U physiological inefficiency model (Callicott et al., 2003), which postulates that patients have reduced levels of activation when poorly performing a task that exceeds their capacity and increased levels of activation when accurately performing a task that is at their capacity. The current task paradigm employed a slowly paced cued presentation initially developed for early dementia patients. It is likely that even when retrieval and switching demands were increased the task was non-demanding for healthy subjects, but challenging for patients - placing them in the over-activation portion of the inverted-U function. The absence of group performance differences in even the most demanding task conditions suggest that patients never exceeded their cognitive capacity and, therefore, never entered the downward going portion of the inverted-U function that would lead to reduced PFC response. Optimizing fMRI designs to match groups on mental effort rather than simply on level of performance is receiving increased attention (Karlsgodt et al., 2007) and is an important consideration for future studies.

Formal thought disorder has long been associated with executive dysfunction and semantic memory impairment (Kerns & Berenbaum, 2002; Spitzer, 1997), STG gray matter reductions (Rajarethinam et al., 2000), and disrupted frontal and temporal lobe function during semantic tasks (Kircher et al., 2001a,b; Shenton et al., 1997). The current study did not find any FTD relationship with STG, and found reciprocal correlations for anterior and posterior brain regions depending upon task contrast. At high levels of retrieval demand (SC minus baseline contrast) more severe FTD was associated with reduced activation in dorsal anterior cingulate gyrus (ACC) and cuneus. Previous studies have found both negative (McGuire et al., 1998) and positive (Assaf et al., 2005; Liddle et al., 1992) correlations between ACC activity and FTD. The dorsal ACC is associated with conflict detection and response monitoring (Barch et al., 2000), and it is possible that more severely thought disordered patients were less able to detect conflict resulting in attenuated ACC activity. A negative correlation with FTD was also found in the right superior frontal gyrus for the SC minus OS task comparison. This contrasts with the positive correlation found with FTD in this same region by Kircher and colleagues (2001b) who scanned subjects while they were describing Rorschach inkblots.

In contrast, positive correlations were found between posterior brain regions (lingual gyrus, fusiform gyrus and superior parietal lobule) and FTD when the effect of increased switching demand was examined (OSswitch-OS, and SCswitch-SC contrasts). The fusiform gyrus has previously shown positive associations with FTD during sentence completion (Kircher et al., 2001a) and speech production tasks (McGuire et al., 1998). This relationship was also recently observed in an fMRI semantic priming paradigm and attributed to an increased spread of activation in more thought disordered patients (Kuperberg et al., 2007). There is less of a literature on the role that the superior parietal cortex and lingual gyrus might play in FTD and semantic memory impairment. In discussing the superior parietal cortex in relation to the current task Gurd and colleagues (2002) emphasized that it plays a role in the control of attentional switching across spatial and non-spatial stimulus modes. The lingual gyrus has also been associated with top-down control of visuospatial selective attention (Hahn, Ross, and Stein, 2006). Increased activation of parietal and lingual gyrus regions in more thought disordered patients may, therefore, reflect greater need for attentional control in these subjects.

There were certain limitations that deserve mention. First, all of the patients were treated with antipsychotic medications and there is some evidence that this can increase metabolism and blood flow in the basal ganglia (Miller et al., 1998) and possibly in the prefrontal cortex (Honey et al., 1999). Medication may, therefore, have contributed to fMRI over-activation. The sample was somewhat small, which likely reduced statistical power, and included two left-handed patients, which might have influenced laterality effects. Although the slowly-paced cued word format of the current study insured that patients were able to perform the task and controlled for total word production, it did not allow us to do a more detailed analysis of task performance. An advantage of standard self-paced verbal fluency tasks is that they allow for differentiation of more automatic generation of words within a category (i.e., clustering) from the more controlled process of generating words from a new category (i.e., switching). Comparing these two processes can provide information on the organizational structure of semantic memory in healthy and clinical populations. Because clustering is a relatively automatic process, items that are clustered co-occur in much closer temporal proximity (Lockhead, Gruenewald, King, 1978) than did the responses in our current cued response format. Because of this built-in response delay, we were not comfortable interpreting the meaning of word sequences within each task and relied instead on OS and SC contrasts to make inferences about the level of automatic and controlled processing. A possible solution in future studies would be to combine the current fMRI paradigm with a comprehensive neuropsychological assessment of semantic memory.

In summary, our data suggest that it is the controlled processing demands of word production tasks that lead to group differences in brain activation between patients with schizophrenia and healthy comparison subjects even when groups are matched on behavioral performance. When patients are not under speeded conditions and can rely on overlearned retrieval processes activation in frontal and parietal regions is unimpaired. Our study also joins others (Barch et al., 1999) in demonstrating the feasibility of administering overt word production tasks during functional imaging.


We thank members of the Schizophrenia Research Center for subject accrual and members of the imaging center for assistance with fMRI data collection.


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


  • Abrahams S, Goldstein LH, Simmons A, Brammer MJ, Williams SCR, Giampietro VP, Andrew CM, Leigh PN. Functional magnetic resonance imaging of verbal fluency and confrontation naming using compressed image acquisition to permit overt responses. Human Brain Mapping. 2003;20:29–40. [PubMed]
  • American Psychiatric Association. Diagnostic and statistical manual of mental disorders: DSM-IV. 4. American Psychiatric Press; Washington, DC: 1994.
  • Andreasen NC. Scale for the Assessment of Negative Symptoms (SANS) The University of Iowa; Iowa City: 1983.
  • Andreasen NC. Scale for the Assessment of Positive Symptoms (SAPS) The University of Iowa; Iowa City: 1984.
  • Andreasen NC. Should the term “thought disorder” be revised? Comprehensive Psychiatry. 1982;36:1315–1321. [PubMed]
  • Assaf M, Rivkin PR, Kuzu CH, Calhoun VD, Kraut MA, Groth KM, Yassa MA, Hart J, Pearlson GD. Abnormal object recall and anterior cingulated overactivation correlate with formal thought disorder in schizophrenia. Biological Psychiatry. 2005;59:452–459. [PubMed]
  • Bandettini PA, Wong EC, Hinks RS, Tikofsky RS, Hyde JS. Time course of EPI of human brain function during task activation. Magnetic Resonance Medicine. 1992;25:390–397. [PubMed]
  • Barch DM, Braver TS, Sabb FW, Noll DC. Anterior cingulate and the monitoring of response conflict: evidence from an fMRI study of overt verb generation. Journal of Cognitive Neuroscience. 2000;12:298–309. [PubMed]
  • Barch DM, Sabb FW, Carter CS, Braver TS, Noll DC, Cohen JD. Overt verbal responding during fMRI scanning: Empirical investigations of problems and potential solutions. NeuroImage. 1999;10:642–657. [PubMed]
  • Boksman K, Theberge J, Williamson P, Drost DJ, Malla A, Densmore M, Takhar J, Pavlosky W, Menon RS, Neufeld RWJ. A 4.0-T fMRI study of brain connectivity during word fluency in first-episode schizophrenia. Schizophrenia Research. 2005;75:247–263. [PubMed]
  • Bozikas VP, Kosmidis MH, Karavatos A. Disproportionate impairment in semantic verbal fluency in schizophrenia: differential deficit in clustering. Schizophrenia Research. 2005;74:51–59. [PubMed]
  • Braff DL. Information processing and attention dysfunctions in schizophrenia. Schizophrenia Bulletin. 1993;19:233–259. [PubMed]
  • Callicott JH, Mattay VS, Verchinski BA, Marenco S, Egan MF, Weinberger DR. Complexity of prefrontal cortical dysfunction in schizophrenia: more than up or down. American Journal of Psychiatry. 2003;160:2209–2215. [PubMed]
  • Chute DL, Westall RF. Power Laboratory. MacLaboratory, Inc; Devon, PA: 1997.
  • Cohen JD, Servan-Schreiber D. Context, cortex, and dopamine: a connectionist approach to behavior and biology in schizophrenia. Psychological Review. 1992;99:45–77. [PubMed]
  • Collins AM, Loftus EF. A spreading activation theory of semantic processing. Psychological Review. 1975;82:407–428.
  • Curtis VA, Bullmore ET, Brammer MJ, Wright I, Williams SCR, Morris RG, Sharma TS, Murray RM, McGuire PK, Wright IC, Williams SC. Attenuated frontal activation during a verbal fluency task in patients with schizophrenia. American Journal of Psychiatry. 1998;155:1056–1063. [PubMed]
  • Docherty NM. Cognitive impairments and disordered speech in schizophrenia: Thought disorder, disorganization, and communication failure perspectives. Journal of Abnormal Psychology. 2005;114:269–278. [PubMed]
  • Docherty NM, DeRosa M, Andreasen NC. Communication disturbances in schizophrenia and mania. Archives of General Psychiatry. 1996;53:358–364. [PubMed]
  • Elvevåg B, Fisher JE, Gurd JM, Goldberg TE. Semantic clustering in verbal fluency: schizophrenic patients versus control participants. Psychological Medicine. 2002;32:909–917. [PubMed]
  • Elvevåg B, Storms G. Scaling and clustering in the study of semantic disruptions in patients with schizophrenia: A re-evaluation. Schizophrenia Research. 2003;63:237–246. [PubMed]
  • First M, Gibbon M, Spitzer R, Williams JBW. Structured Clinical Interview for DSM-IV Axis II Personality Disorder. SCID II. American Psychiatric Press; Washington, DC: 1997.
  • First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical Interview for DSM-IV Axis I Disorders, Non-Patient Edition (SCID-NP) New York State Psychiatric Institute/Biometrics Research Department; New York: 1995.
  • First MB, Spitzer RL, Williams JBW, Gibbon M. Structured Clinical Interview for DSM-IV - Patient Version (SCID-P) New York State Psychiatric Institute; New York: 1996.
  • Fletcher PC, McKenna PJ, Frith CD, Grasby PM, Friston KJ, Dolan RJ. Brain activations in schizophrenia during a graded memory task studied with functional neuroimaging. Archives of General Psychiatry. 1998;55:1001–1008. [PubMed]
  • Friston KJ, Frith CD. Schizophrenia: a disconnection syndrome? Clinical Neuroscience. 1995;3:89–97. [PubMed]
  • Frith CD, Friston KJ, Herold S, Silbersweig D, Fletcher P, Cahill C, Dolan RJ, Frackowiak RS, Liddle PF. Regional brain activity in chronic schizophrenic patients during the performance of a verbal fluency task. British Journal of Psychiatry. 1995;167:343–349. [PubMed]
  • Fu CHY, Suckling J, Williams SCR, Andrew CM, Vythelingum GN, McGuire PK. Effects of psychotic state and task demand on prefrontal function in schizophrenia: An fMRI study of overt verbal fluency. American Journal of Psychiatry. 2005;162:485–494. [PubMed]
  • Goldberg TE, Aloia MS, Gourovitch ML, Missar D, Pickar D, Weinberger DR. Cognitive substrates of thought disorder, I: the semantic system. American Journal of Psychiatry. 1998;155:1671–1676. [PubMed]
  • Gourovitch ML, Kirkby BS, Goldberg TE, Weinberger DR, Gold JM, Esposito G, Van Horn JD, Berman KF. A comparison of rCBF patterns during letter and semantic fluency. Neuropsychology. 2000;14:353–360. [PubMed]
  • Gurd JM, Amunts K, Weiss PH, Zafiris O, Zilles K, Marshall JC, Fink GR. Posterior parietal cortex is implicated in continuous switching between verbal fluency tasks: an fMRI study with clinical implications. Brain. 2002;125:1024–1038. [PubMed]
  • Hahn B, Ross TJ, Stein EA. Neuroanatomical dissociation between bottom-up and top-down processes of visuospatial selective attention. NeuroImage. 2006;32:842–853. [PMC free article] [PubMed]
  • Himelhoch S, Taylor SF, Goldman RS, Tandon R. Frontal lobe tasks, antipsychotic medication, and schizophenia syndromes. Biological Psychiatry. 1996;39:227–229. [PubMed]
  • Honey GD, Bullmore ET, Soni W, Varatheesan M, Williams SCR, Sharma T. Differences in frontal cortical activation by a working memory task after substitution of riseridone for typical antipsychotic drugs in patients with schizophrenia. Proceedings of the National Academy of Sciences, USA. 1999;96:13432–13437. [PMC free article] [PubMed]
  • Jennings JM, McIntosh AR, Kapuv S, Zipursky RB, Houle S. Functional network difference in schizophrenia: A rCBF study of semantic processing. Neuroreport. 1998;9:1697–1700. [PubMed]
  • Karlsgodt KH, Glahn DC, van Erp TG, Therman S, Huttunen M, Manninen M, Kaprio J, Cohen MS, Lonnqvist J, Cannon TD. The relationship between performance and fMRI signal during working memory in patients with schizophrenia, unaffected co-twins, and control subjects. Schizophrenia Research. 2007;89:191–197. [PubMed]
  • Kerns JG, Berenbaum H. Cognitive impairments associated with formal thought disorder in people with schizophrenia. Journal of Abnormal Psychology. 2002;111:211–224. [PubMed]
  • Kircher TTJ, Bulimore ET, Brammer MJ, Williams SCR, Broome MR, Murray RM, McGuire PK. Differential activation of temporal cortex during sentence completion in schizophrenia patients with and without formal thought disorder. Schizophrenia Research. 2001a;50:27–40. [PubMed]
  • Kircher TTJ, Liddle PF, Brammer MJ, Williams SCR, Murray RM, McGuire PK. Neural correlates of formal thought disorder in schizophrenia. Archives of General Psychiatry. 2001b;58:769–774. [PubMed]
  • Kuperberg GR, Deckersbach T, Holt DJ, Goff D, West WC. Increased temporal and prefrontal activity in response to semantic associations in schizophrenia. Archives of General Psychiatry. 2007;64:138–151. [PubMed]
  • Liddle PF. Dynamic neuroimaging with PET, SPET, or fMRI. International Review of Psychiatry. 1997;9:331–337.
  • Liddle PF. The psychomotor disorder: Disorders of the supervisory mental processes. Behavioral Neurology. 1993;6:5–14. [PubMed]
  • Liddle PF, Friston KJ, Frith CD, Hirsch SR, Jones T, Frackowiak RS. Patterns of cerebral blood flow in schizophrenia. British Journal of Psychiatry. 1992;160:179–186. [PubMed]
  • Lockhead GR, Gruenewald P, King M. Holistic vs. attribute repetition effects in classifying stimuli. Memory and Cognition. 1978;6:438–445. [PubMed]
  • Mayr U. On the dissociation between clustering and switching in verbal fluency: comment on Troyer, Moscovitch, Winocur, Alexander and Stuss. Neuropsychologia. 2002;40:562–566. [PubMed]
  • McGuire PK, Quested DJ, Spence SA, Murray RM, Frith CD, Liddle PF. Pathophysiology of “positive” thought disorder in schizophrenia. British Journal of Psychiatry. 1998;173:231–235. [PubMed]
  • Meyer-Lindenberg A, Poline JB, Kohn PD, Holt JL, Egan MF, Weinberger DR, Berman KF. Evidence for abnormal cortical functional connectivity during working memory in schizophrenia. American Journal of Psychiatry. 2001;158:1809–1817. [PubMed]
  • Miller DD, Andreasen NC, O’Leary DS, Rezai K, Watkins GL, Boles Ponto LL, Hichwa RD. Effect of antipsychotics on regional cerebral blood flow measured with positron emission tomography. Neuropsychopharmacology. 1998;18:323–324. [PubMed]
  • Minzenberg MJ, Ober BA, Vinogradov S. Semantic priming in schizophrenia: a review and synthesis. Journal of the International Neuropsychological Society. 2002;8:699–720. [PubMed]
  • Moelter ST, Hill SK, Hughett P, Gur RC, Gur RE, Ragland JD. Organization of Semantic Category Exemplars in Schizophrenia. Schizophrenia Research. 2005;78:209–217. [PubMed]
  • Moelter ST, Hill SK, Ragland JD, Lunardelli A, Gur RC, Gur RE, Moberg PJ. Controlled and automatic processing during animal word list generation in schizophrenia. Neuropsychology. 2001;15:502–509. [PubMed]
  • Myer-Lindberg AS, Olsen RK, Kohn PD, Brown T, Egan MF, Weinberger DR, Berman KF. Regionally specific disturbance of dorsolateral prefrontal-hippocampal functional connectivity in schizophrenia. Archives of General Psychiatry. 2005;62:379–386. [PubMed]
  • Overall JR, Gorham DR. The brief psychiatric rating scale. Journal of Operational Psychiatry. 1980;11:48–64.
  • Paulsen JS, Romero R, Chan A, Davis AV, Heaton RK, Jeste DV. Impairment of the semantic network in schizophrenia. Psychiatry Research. 1996;63:109–121. [PubMed]
  • Posner MI, DiGirolamo GJ. In: Conflict, target detection, and cognitive control. The Attentive Brain. Parasuraman R, editor. MIT Press; Cambridge: 1998.
  • Rajarethinam RP, DeQuardo JR, Nalepa R, Tandon R. Superior temporal gyrus in schizophrenia: a volumetric magnetic resonance imaging study. Schizophrenia Research. 2000;41:303–312. [PubMed]
  • Robert PH, Lafont V, Medecin I, Berthet L, Thauby S, Baudu C, Darcourt G. Clustering and switching strategies in verbal fluency tasks: comparison between schizophrenics and healthy adults. Journal of the International Neuropsychological Society. 1998;4:539–546. [PubMed]
  • Schaufelberger M, Senhorini MC, Barreiros MA, Amaro E, Jr, Menezes PR, Scazufca M, Castro CC, Ayres AM, Murray RM, McGuire PK, Busatto GF. Frontal and anterior cingulate activation during overt verbal fluency in patients with first episode psychosis. Rev Bras Psiquiatr. 2005;27:228–232. [PubMed]
  • Shallice T. From neuropsychology to mental structure. Cambridge University Press; Cambridge: 1988.
  • Shenton ME, Kikinis R, Jolesz FA, Pollak SD, LeMay M, Wible CG, Hokama H, Martin J, Metcalf D, Coleman M. Abnormalities of the left temporal lobe and thought disorder in schizophrenia: a functional magnetic resonance imaging study. American Journal of Psychiatry. 1997;154:1676–1682. [PubMed]
  • Spence SA, Hirsch SR, Brooks DJ, Grasby PM. Prefrontal cortex activity in people with schizophrenia and control subjects: Evidence from positron emission tomography for remission of ‘hypofrontality’ with recovery from acute schizophrenia. British Journal of Psychiatry. 1998;172:316–323. [PubMed]
  • Spence SA, Liddle PF, Steefan MD, Hellewell JSE, Sharma T, Friston KJ, Hirsch SR, Frith CD, Murray RM, Deakin JFW, Grasby PM. Functional anatomy of verbal fluency in people with schizophrenia and those at genetic risk: Focal dysfunction and distributed disconnectivity reappraised. British Journal of Psychiatry. 2000;176:52–60. [PubMed]
  • Spitzer MA. Cognitive neuroscience view of schizophrenic thought disorder. Schizophrenia Bulletin. 1997;23:29–50. [PubMed]
  • Talairach J, Tournoux P. Co-planar Stereotaxic Atlas of the Human Brain: 3 Dimensional Proportional System: An Approach to Cerebral Imaging. Thieme Medical Publishers Inc; New York: 1988.
  • Tallent KA, Weinberger DR, Goldberg TE. Associating semantic space abnormalities with formal thought disorder in schizophrenia: Use of triadic comparisons. Journal of Clinical and Experimental Neuropsychology. 2001;23:285–296. [PubMed]
  • Troyer AK, Moscovitch M, Winocur G. Clustering and switching as two components of verbal fluency: evidence from younger and older healthy adults. Neuropsychology. 1997;11:138–146. [PubMed]
  • Troyer AK, Moscovitch M, Winocur G, Alexander MP, Stuss D. Clustering and switching on verbal fluency: the effects of focal frontal- and temporal-lobe lesions. Neuropsychologia. 1998;36:499–504. [PubMed]
  • Vinogradov S, Kirkland J, PooleJ H, Drexler M, Ober BA, Shenault GK. Both processing speed and semantic memory organization predict verbal fluency in schizophrenia. Schizophrenia Research. 2002;59:269–275. [PubMed]
  • Vita A, Dieci M, Giobbio GM, Caputo A, Ghiringhelli L, Comazzi M, Garbarini M, Mendini AP, Morganti C, Tenconi F, Cesana B, Invernizzi G. Language and thought disorder in schizophrenia: brain morphological correlates. Schizophrenia Research. 1995;15:243–251. [PubMed]
  • Ward BD. AFNI AlphaSim Documentation. Medical College of Wisconsin; 2000. Simultaneous inference for fMRI data.
  • Weiss EM, Hofer A, Golaszewski S, Siedentopf C, Brinkhoff C, Kremser C, Felber S, Fleischhacker WW. Brain activation patterns during a verbal fluency test – a functional MRI study in healthy volunteers and patients with schizophrenia. Schizophrenia Research. 2004;70:287–291. [PubMed]
  • Wolf DH, Gur RC, Valdez JN, Loughead J, Elliott MA, Gur RE, Ragland JD. Alterations of fronto-temporal connectivity during word encoding in schizophrenia. Psychiatry Research: Neuroimaging. 2007;154:221–232. [PMC free article] [PubMed]
  • Yurgelun-Todd DA, Waternaux CM, Cohen BM, Gruber SA, Camper DBA, Renshaw P. Functional magnetic resonance imaging of schizophrenia patients and comparison subjects during word production. American Journal of Psychiatry. 1996;153:200–205. [PubMed]
  • Zakzanis KK, Troyer AK, Rich JB, Heinrichs W. Component analysis of verbal fluency in patients with schizophrenia. Neuropsychiatry Neuropsychology and Behavioral Neurology. 2000;13:239–245. [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...