More words, less words: Verbal fluency as a function of ‘positive’ and ‘negative’ schizotypy

www.elsevier.com/locate/paid

Personality and Individual Differences 39 (2005) 705–713

More words, less words: Verbal fluency as a functionof �positive� and �negative� schizotypy

Elias Tsakanikos a,*, Gordon Claridge b

a Institute of Psychiatry, King�s College London, Estia Centre, 66 Snowfields, London SE1 3SS, UKb Department of Experimental Psychology, University of Oxford, South Park Road, Oxford OX1 3UD, UK

Received 10 March 2004; received in revised form 9 February 2005; accepted 22 February 2005

Available online 8 April 2005

Abstract

Positive and negative symptoms of schizophrenia have been differentially associated with irregularities in

verbal production, suggesting the involvement of different underlying mechanisms in psychotic symptom-

atology. In view of that, the present investigation examined whether the amount of verbal production

would be also differentially associated with negative and positive symptoms of psychometric schizotypy

in a sample of college students (N = 190). The participants were tested on a typical verbal fluency testand completed the O-LIFE schizotypy scales. The analyses revealed that decreased verbal fluency was asso-

ciated with increased levels of negative schizotypy in participants who scored one standard deviation above

the mean on Introvertive Anhedonia. In contrast, increased verbal fluency was associated with increased

levels of positive schizotypy in participants who scored one standard deviation above the mean on Unusual

Experiences. The obtained results are discussed in terms of the proposal that psychotic-like unusual expe-

riences, like hallucinations, may be the product of a higher automatic spreading activation among stored

lexical units, a mechanism which seems to account for the previously reported link between positive schizo-

typy and creativity.� 2005 Elsevier Ltd. All rights reserved.

Keywords: Automatic spreading activation; Creativity; Executive functioning; Hallucinations; Phonological (letter)

fluency; Positive/negative schizotypy; Individual differences in speech production

0191-8869/$ - see front matter � 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.paid.2005.02.019

* Corresponding author. Tel.: +44 20 7378 3221; fax: +44 20 7378 3222.

E-mail address: [email protected] (E. Tsakanikos).

mailto:[email protected]

706 E. Tsakanikos, G. Claridge / Personality and Individual Differences 39 (2005) 705–713

1. Introduction

Abnormalities in verbal communication, such as irregularities in speech production, are fre-quent in people with schizophrenia. Dramatic fluctuations between increase (pressure of speech)and decrease (poverty of speech) in the amount of verbal production have been related to linguis-tic impairments in schizophrenia, and have been consistently associated with different aspects ofthe psychotic symptomatology (Harvey, 2000). Despite the presence of a wide range of commu-nication abnormalities in schizophrenia, factor analytic studies have identified verbal productivityas a main dimension of communication disorder (Harvey et al., 1992).

Poor performance on verbal fluency, as assessed by the ability to generate word lists under cer-tain stimulus constraints, has been associated with negative symptoms of schizophrenia, such asflat affect, anhedonia, and impoverished speech (Allen, Liddle, & Frith, 1993; Howanitz, Cicalese,& Harvey, 2000). Verbal fluency is thought to reflect executive functioning, such as cognitiveswitching, rule monitoring, and inhibition of inappropriate responses (Lezak, 1995). Impairedexecutive functioning has not only been associated with deficient frontal lobe functioning inschizophrenia (Crow, 1992), but also with frontal lobe dysfunction in other patient groups(Beatty, Goodkin, & Beatty, 1989; Joyce & Robbins, 1991). Tests of verbal fluency are generallyconsidered to be sensitive to frontal lobe dysfunction (Lezak, 1995).

Evidence suggests that verbal fluency is differentially affected in psychosis. Although poor per-formance has been associated with negative symptomatology (Allen et al., 1993; Howanitz et al.,2000), increased verbal fluency has been associated with positive psychotic symptoms, such as hal-lucinations (Kerns, Berenbaum, Barch, Banich, & Stolar, 1999; Lindamer & Whitman, 1997).This latter finding is consistent with theoretical proposals that hallucinations are produced as aresult of an increased automatic spreading activation between stored lexical items (David,1994; Hoffman & McGlashan, 1997). People who experience hallucinations may generate an in-creased amount of words during verbal fluency tests due to such an enhanced lexical activation.

Are irregularities in verbal fluency primary or secondary features of psychosis? Cognitive irreg-ularities in schizophrenia can be either viewed as cognitive endophenotypes (primary features), i.e.cognitive markers intermediating between a putative genetic disposition and clinical symptoms, oras epiphenomena of the psychotic symptoms, side effects due to medication, and/or effects of hos-pitalization (secondary features). The presence of verbal fluency abnormalities in healthy partic-ipants who score highly on psychometric measures of schizotypy would support the possibilitythat these irregularities are primary features. On the contrary, a negative result would suggest thatthese irregularities are secondary features. Unfortunately, because of conflicting evidence in therecent literature, it is unclear whether or not verbal fluency irregularities, akin to those associatedwith negative and positive symptoms of schizophrenia, can be observed in positive and negativeschizotypy as well.

In some recent studies, positive schizotypy has been associated with increased verbal fluency(O�Reilly, Dunbar, & Bentall, 2001; Weinstein & Graves, 2002). Considering that fluency is typ-ically regarded as a main facet of creativity, the authors of these studies have interpreted this asso-ciation as evidence of a biological link between creativity and �positive� psychotic-like features (seealso Miller & Chapman, 1983; Poreh, Whitman, & Ross, 1993; Schuldberg, 1990). The reportedassociation between positive schizotypy and increased verbal fluency seems to accord with thehypothesis that hallucinations are the product of an increased automatic spreading activation

E. Tsakanikos, G. Claridge / Personality and Individual Differences 39 (2005) 705–713 707

between related lexical units (David, 1994; Hoffman & McGlashan, 1997). However, other studieshave failed to demonstrate a similar association between positive schizotypy and verbal fluency(Barrantes-Vidal et al., 2002; Dinn, Harris, Aycicegi, Greene, & Andover, 2002). In addition,in contrast with the schizophrenia literature (Allen et al., 1993; Howanitz et al., 2000), none ofthe above schizotypy studies (Barrantes-Vidal et al., 2002; Dinn et al., 2002; O�Reilly et al.,2001; Weinstein & Graves, 2002) revealed any specific association between poor performanceon verbal fluency and negative schizotypy.

One possible explanation for the above inconsistent results might be related to the sample size.With the exception of the Barrantes-Vidal et al. study, which employed a sample of adolescentswith mean age = 13.4 years (a methodological pitfall, considering that the schizotypy scalesemployed were designed for adults), in studies with adults (Dinn et al., 2002; O�Reilly et al.,2001; Weinstein & Graves, 2002) the largest sample size was 103 participants. Therefore, theaim of the investigation was to re-examine the relationship between verbal fluency and positive/negative schizotypy in a larger adult sample. Moreover, any systematic relationship between ver-bal fluency and schizotypy might not be necessarily linear. For instance, a differential associationbetween positive/negative schizotypy and verbal fluency may be detectable only for the extremescorers. Consequently, the second aim was to look for evidence on possible non-linear markers.To evaluate the above possibilities, 190 participants were tested on a typical verbal fluency test(Lezak, 1995) and completed a multi-dimensional schizotypy inventory (O-LIFE; Mason,Claridge, & Jackson, 1995).

2. Method

2.1. Participants

One hundred and ninety students, recruited from the University College London (UCL) cam-pus, took part in this study. Their mean age was 23.5 years, 99 were female, and they were unpaidvolunteers. English as a second language was an exclusion criterion. Four participants weredropped from the analysis due to a large number of unanswered questions in the O-LIFE.

2.2. Schizotypy measures

The Oxford–Liverpool Inventory for Feelings and Experiences (O-LIFE) consists of 105 itemsselected on the basis of factor-analytic studies of older schizotypy scales (Bentall, Claridge, &Slade, 1989; Mason et al., 1995). The inventory includes four schizotypy scales: Unusual Experi-ences (positive schizotypy) consists of items assessing magical thinking, unusual perceptual aber-rations, and hallucinatory experiences (alpha coefficient = 0.89); Cognitive Disorganization(disorganized schizotypy) scale reflects the disorganized aspect of psychosis, and consists of itemsassessing difficulties with concentration, decision-making and social anxiety (alpha coeffi-cient = 0.87); Introvertive Anhedonia (negative schizotypy) reflects the negative symptoms of psy-chosis, and consists of items assessing the lack of enjoyment from social contact, physicalactivities, coupled with aversion to emotional and physical intimacy (alpha coefficient = 0.82);the last scale, Impulsive Non-conformity, consists of items assessing aggressive, anti-social and


impulsive behaviour (alpha coefficient = 0.77). Impulsive Non-conformity is largely based on thePsychoticism scale, Introvertive Anhedonia has high loadings on Extraversion, and CognitiveDisorganization has high loadings on Neuroticism of the Eysenck Personality Questionnaire(EPQ: Eysenck & Eysenck, 1975), although Unusual Experiences does not have high loadingon any of the EPQ scales.

Attesting to the experimental validity of these scales, high-schizotypy scorers, as identified bythe O-LIFE sub-scales, demonstrate neuro-cognitive irregularities akin to those seen in schizo-phrenic patients (Rawlings & Goldberg, 2001; Steel, Hemsley, & Jones, 1996; Tsakanikos & Reed,2003).

2.3. Verbal fluency

A phonological (letter) fluency task, frequently used in neuropsychological research (Lezak,1995), was employed to assess verbal fluency. The task consisted of three 1 min trials. In each trial,the participants were asked to generate as many words as possible that started with a certain letter(C, F, and L), avoiding names, non-English words, and derivatives of the same word. The totalnumber of correct responses generated was the dependent measure.

3. Results

Data from Table 1 presents the means, the standard deviations, and the inter-correlationsbetween the O-LIFE scales, and verbal fluency. In terms of the O-LIFE scales, the overall pattern(inter-correlation and descriptive statistics) appeared similar to that reported in the original study(Mason et al., 1995), as well as to that reported in later studies (e.g., Rawlings & Goldberg, 2001;Tsakanikos & Reed, 2003). The only exception was the absence of a significant correlationbetween Impulsive Non-conformity and Introvertive Anhedonia (in the original study,r = �0.10, p < 0.05). However, comparison of the results between the two studies using the Fisherz method (Rosenthal & Rosnow, 1984) showed that the two correlation coefficients were notsignificantly different from each other (p > 0.10). Contrary to what was expected, there was no sig-nificant correlation between schizotypy scales and verbal fluency.

Despite the absence of a statistically significant correlation between schizotypy scales and ver-bal fluency, it could be still argued that non-linear relationships between these measures may exist.

Table 1

Inter-correlations between schizotypy scales and performance on verbal fluency

M SD 1 2 3 4 5

Unusual Experiences 9.72 5.92 –

Cognitive Disorganization 11.71 5.31 0.49** –

Introvertive Anhedonia 4.83 3.69 0.15* 0.09 –

Impulsive Non-conformity 10.17 4.07 0.39** 0.27** �0.04 –

Verbal Fluency 41.64 9.57 0.02 �0.12 �0.05 �0.09 –

* p < 0.05 (two-tailed).** p < 0.01 (two-tailed).


For example, particular trends between verbal fluency and different schizotypy measures may existonly for participants with very high schizotypy scores. To investigate the above possibility, verbalfluency performance was plotted as a function of positive and negative schizotypy (Fig. 1).

Inspection of the upper graph (Fig. 1, panel A) suggests that, for participants who scored onestandard deviation above the mean on positive schizotypy (Unusual Experiences > 15), verbal flu-ency appeared to increase as a function of their schizotypy scores. For these extreme high scorers(N = 33), there was a statistically significant positive correlation between Unusual Experiences

2524232220191817161514131211109876543210

verb

aly

fluen

cy

60

50

40

30

panel A

20171514131211109876543210

verb

al fl

uenc

y

60

50

40

30

panel B

Fig. 1. (A) Positive schizotypy (�Unusual Experiences�) and (B) Negative schizotypy (�Introvertive Anhedonia�).


and verbal fluency (r = 0.35, p < 0.05). However, there was no significant correlation betweenUnusual Experiences and verbal fluency (r = 0.03, p > 0.80) for participants scoring one standarddeviation below the mean (Unusual Experiences < 5).

Inspection of the lower graph (Fig. 1, panel B), suggested that verbal fluency appeared todecrease as a function of negative schizotypy for participants who scored one standard deviationabove the mean (Introvertive Anhedonia > 8). For those extreme scorers (N = 28), there was astatistically significant negative correlation between Introvertive Anhedonia and verbal fluency(r = �0.46, p < 0.05). For those participants who scored on negative schizotypy one standarddeviation bellow the mean (Introvertive Anhedonia < 3), there was no significant correlation be-tween Introvertive Anhedonia and verbal fluency (r = 0.01, p > 0.90). The obtained results suggestthat verbal fluency, at least for those who scored very high on schizotypy, was positively associ-ated with positive schizotypy, and negatively associated with negative schizotypy.1

4. Discussion

In line with the schizophrenia literature (Allen et al., 1993; Howanitz et al., 2000; Kerns et al.,1999; Lindamer & Whitman, 1997), the present data suggested that different schizotypy dimen-sions were differentially associated with verbal productivity, as assessed by a typical letter fluencytask. However, the relationship between verbal fluency and schizotypy was not linear; these asso-ciations were present only for those participants who scored one standard deviation above themean. This finding seems to support quasi-dimensional, but not fully dimensional views of schizo-typy (see Claridge, 1997, for a discussion). In addition, the fact that differentiated verbal fluencyirregularities were observed in a student sample as a function of psychometric schizotypy suggeststhat such irregularities are cognitive markers (primary features) rather than epiphenomena (sec-ondary features) of psychosis.

During the phonological (letter) task, the participants were asked to generate as many words aspossible by following certain rules, i.e. the words should start with a certain letter, and derivativesof the same words, names, and non-English words should be avoided. Consequently, adequateword production in this task relied largely on executive functions, such as maintaining �on-line�these rules and monitoring performance accordingly, while inhibiting inappropriate responses(Lezak, 1995). In the present study, negative schizotypy was found to be associated with decreasedverbal fluency, following the pattern of the negative symptoms of schizophrenia (Allen et al.,1993; Howanitz et al., 2000). Given that negative symptoms of schizophrenia have been associatedwith impaired executive functions (Crow, 1992), it is possible that similar subtle impairments hada detrimental effect on verbal influence in participants scoring highly on negative schizotypy. Fur-thermore, considering that Introvertive Anhedonia has high loadings on Extraversion, the presentfindings appear consistent with some earlier studies on the relationship between Extraversion andverbal fluency (Eysenck, 1974a, 1974b).

However, apart from executive functions, another cognitive process that is highly relevant toperformance on a verbal fluency task is the speed of lexical retrieval. A particularly increased acti-

1 The same analyses run for extreme scorers on Cognitive Disorganization and Impulsive Non-conformity, but no

evidence for similar non-linear markers were found.


vation of stored lexical units would, therefore, be expected to enhance performance on a letter flu-ency task. Certain positive symptoms of schizophrenia, such as hallucinations, have been attrib-uted to such an increased automatic spreading activation of related lexical items2 (David, 1994;Hoffman & McGlashan, 1997). Given that positive symptoms of schizophrenia have been associ-ated with increased verbal fluency (Kerns et al., 1999; Lindamer & Whitman, 1997), it is possiblethat both positive symptoms and increased verbal fluency may be products of an enhanced auto-matic spreading activation. Such an account could also accommodate the finding that high levelsof positive schizotypy are associated with increased verbal fluency, as has been shown in previousstudies (O�Reilly et al., 2001; Weinstein & Graves, 2002) and replicated in the presentinvestigation.

It could be argued that the positive association between verbal fluency and positive schizotypyis also predicted by the �sheer inhibition hypothesis� (Spitzer, 1993). According to this hypothesis,activation tends to spread more diffusely and more rapidly due to inadequate mechanisms thatcontrol spreading activation and re-activation in the semantic networks. Unlike the automaticspreading activation account, which seems limited to semantically and/or phonetically related lex-ical units (David, 1994; Hoffman & McGlashan, 1997), the sheer inhibition hypothesis is not lim-ited to certain types of units (Spitzer, 1993). Nevertheless, the two accounts are not necessarilyincompatible; for example, an over-activation of phonetically related lexical units, which has beenproposed to be responsible for the production of acoustic hallucinations (David, 1994; Hoffman &McGlashan, 1997), could be due to inefficient control mechanisms of spreading activation and re-activation in the semantic networks (Spitzer, 1993).

In summary, the present investigation showed that personality traits that phenomenonologi-cally resemble the negative and positive symptoms of schizophrenia were differentially relatedto verbal fluency, suggesting different underlying mechanisms. Enhanced performance on verbalfluency was associated with high levels of positive schizotypy, i.e. increased reports of hallucina-tion-like experiences, delusional ideation, and perceptual aberrations, a pattern of results pre-dicted by theories of increased automatic spreading activation in the semantic networks(David, 1994; Hoffman & McGlashan, 1997; Spitzer, 1993). Given that verbal fluency is a mainfacet of creativity, the obtained results reinforce the notion that the link between �pathological�personality traits and creativity is likely due to common cognitive features (Eysenck, 1993). Thisnotion is consistent with the possibility of a �healthy schizotypy� (McCreery & Claridge, 2002),which may account for the persistence of schizophrenia-related genes in the population despitetheir many dysfunctional aspects.

Acknowledgments

Thanks are due to Carmen Chan, Emma Coakley, Helen Fisher, Suzanne Gage, Jacqlin Iac-covou, Samaira Khan, James MacDonald, Hannah Pratt, Alexandra Suzuki, Jessica Webb,and Grace Wong for their help with the data collection.

2 For instance, a neutral sound activates automatically a similar sounding word, which in turn activates instantly and

simultaneously other semantically or/and phonetically related words etc.


References

Allen, H. A., Liddle, P. F., & Frith, C. D. (1993). Negative features, retrieval processes and verbal fluency in

schizophrenia. British Journal of Psychiatry, 163, 769–775.

Barrantes-Vidal, N., Fananas, L., Rosa, A., Caparros, B., Riba, M. D., & Obiols, J. E. (2002). Neurocognitive,

behavioural and neurodevelopmental correlates of schizotypy clusters in adolescents from the general population.

Schizophrenia Research, 61, 293–302.

Beatty, W. W., Goodkin, D. E., & Beatty, P. A. (1989). Frontal lobe dysfunction and memory impairment in patients

with chronic progressive multiple sclerosis. Brain and Cognition, 11, 73–86.

Bentall, R. P., Claridge, G., & Slade, P. D. (1989). The multidimensional nature of schizotypal traits: a factor analytic

study with normal subjects. British Journal of Clinical Psychology, 28, 363–375.

Claridge, G. (1997). Schizotypy: theoretical background and issues. In G. Claridge (Ed.), Schizotypy: implications for

illness and health (pp. 3–18). Oxford: Oxford University Press.

Crow, S. F. (1992). Dissociation of two frontal lobe syndromes by a test for verbal fluency. Journal of Clinical and

Experimental Neuropsychology, 14, 327–339.

David, A. S. (1994). The neuropsychological origins of auditory hallucinations. In A. S. David & J. Cutting (Eds.),

Neuropsychology of schizophrenia (pp. 269–313). Hillsdale, NJ: Lawrence Erlbaum Associates.

Dinn, W. M., Harris, C. L., Aycicegi, A., Greene, P., & Andover, M. S. (2002). Positive and negative schizotypy in a

student sample: neuro-cognitive and clinical correlates. Schizophrenia Research, 56, 171–185.

Eysenck, H. J. (1993). Creativity and personality: suggestions for a theory. Psychological Inquiry, 4, 147–178.

Eysenck, H. J., & Eysenck, S. B. G. (1975). Manual of the Eysenck Personality Questionnaire. London: Hodder and

Stoughton.

Eysenck, M. W. (1974a). Extraversion, arousal, and retrieval from semantic memory. Journal of Personality, 42,

319–331.

Eysenck, M. W. (1974b). Individual differences in speed of retrieval from semantic memory. Journal of Research in

Personality, 8, 307–323.

Harvey, P. D. (2000). Formal thought disorder in schizophrenia: characteristics and cognitive underpinnings. In S.

Tommoy & P. D. Harvey (Eds.), Cognition in schizophrenia: impairments, importance and treatment strategies

(pp. 107–125). New York: Oxford University Press.

Harvey, P. D., Lenzenweger, M. F., Keefe, R. S. E., Pogge, D. L., Serper, M. R., & Mohs, R. C. (1992).

Empirical evaluation of the factorial structure of clinical symptoms in schizophrenia. Psychiatry Research, 44,

141–151.

Hoffman, R. E., & McGlashan, T. H. (1997). Synaptic elimination, neurodevelopment, and the mechanism of

hallucinated �voices� in schizophrenia. American Journal of Psychiatry, 154, 1683–1689.

Howanitz, E., Cicalese, C., & Harvey, P. D. (2000). Verbal fluency and psychiatric symptoms in geriatric schizophrenia.

Schizophrenia Research, 42, 167–169.

Joyce, E. M., & Robbins, T. W. (1991). Frontal lobe functioning in Korsakoff and non-Korsakoff alcoholics: planning

and spatial working memory. Neuropsychologia, 29, 709–723.

Kerns, J. G., Berenbaum, H., Barch, D. M., Banich, M. T., & Stolar, N. (1999). Word production in schizophrenia and

its relationship to positive symptoms. Psychiatry Research, 87, 29–37.

Lezak, M. (1995). Neuropsychological assessment. Oxford: Oxford University Press.

Lindamer, L. A., & Whitman, R. D. (1997). Frontal lobe functioning and communication deficits as a function of

positive and negative symptoms in schizophrenia. Archives of Clinical Neuropsychology, 12, 358–359.

Mason, O., Claridge, G., & Jackson, M. (1995). New scales for the assessment of schizotypy. Personality and Individual

Differences, 18, 7–13.

McCreery, C., & Claridge, G. (2002). Healthy schizotypy: the case of out-of-the-body experiences. Personality and

Individual Differences, 32, 141–154.

Miller, E. N., & Chapman, L. J. (1983). Continued word association in hypothetically psychosis prone college students.

Journal of Abnormal Psychology, 92, 468–478.

O�Reilly, T., Dunbar, R., & Bentall, R. (2001). Schizotypy and creativity: an evolutionary connection? Personality and

Individual Differences, 31, 1067–1068.


Poreh, A. M., Whitman, D. R., & Ross, T. P. (1993). Creative thinking abilities and hemispheric asymmetry in

schizotypal college students. Current Psychology, 12, 344–352.

Rawlings, D., & Goldberg, M. (2001). Correlating a measure of sustained attention with a multidimensional measure of

schizotypal traits. Personality and Individual-Differences, 31, 421–431.

Rosenthal, R., & Rosnow, R. (1984). Essentials of behavioral research. New York: McGraw Hill.

Schuldberg, D. (1990). Schizotypal and hypomanic traits, creativity, and psychological health. Creativity Research

Journal, 3, 218–230.

Spitzer, M. (1993). The psychopathology, neuropsychology, and neurobiology of associative and working memory in

schizophrenia. European Archives of Psychiatry and Clinical Neuroscience, 243, 57–70.

Steel, C., Hemsley, D. R., & Jones, S. (1996). Cognitive inhibition and schizotypy as measured by the Oxford–Liverpool

Inventory of Feelings and Experiences. Personality and Individual Differences, 20, 769–773.

Tsakanikos, E., & Reed, P. (2003). Visuo-spatial perceptual processing and dimensions of schizotypy: figure-ground

segregation as a function of psychotic-like features. Personality and Individual Differences, 35, 703–712.

Weinstein, S., & Graves, R. E. (2002). Are creativity and schizotypy products of a right hemisphere bias? Brain and

Cognition, 49, 138–151.

Documents

More words, less words: Verbal fluency as a function of ‘positive’ and ‘negative’ schizotypy