ARTICLE IN PRESS Action naming in anomic aphasic speakers: EVects of instrumentality and name...

Brain and Language xxx (2007) xxx–xxx

www.elsevier.com/locate/b&l

ARTICLE IN PRESS

Action naming in anomic aphasic speakers: EVects of instrumentality and name relation

Roel Jonkers ¤, Roelien Bastiaanse

Department of Linguistics, Groningen Graduate School for Behavioral and Cognitive Neurosciences, University of Groningen, The Netherlands

Accepted 9 January 2007

Abstract

Many studies reveal eVects of verb type on verb retrieval, mainly in agrammatic aphasic speakers. In the current study, two factors thatmight play a role in action naming in anomic aphasic speakers were considered: the conceptual factor instrumentality and the lexical fac-tor name relation to a noun. Instrumental verbs were shown to be better preserved than non-instrumental verbs in a group of anomicaphasic speakers but not in a group of Broca’s aphasic speakers. Name relation to a noun improved the performance of the anomic apha-sic speakers as well. Again, no eVect was found in the group of Broca’s aphasic speakers. Verbs with a name relation to a noun were betterretrieved in action naming than verbs without a name relation. These Wndings are discussed in terms of the spreading activation theory ofDell. (Dell, G. S. (1986). A spreading activation theory of retrieval in sentence production. Psychological Review 93, 283–321.)© 2007 Elsevier Inc. All rights reserved.

Keywords: Anomic aphasia; Verb Wnding problems; EVect of verb type; Instrumentality; Verb–noun name relation

1. Introduction

Recent studies on comprehension, naming, and sentenceconstruction revealed that diVerent factors inXuence verbretrieval in aphasic speakers. Meaning (Barde, Schwartz, &Boronat, 2006; Breedin & Martin, 1996; Breedin, SaVran, &Schwartz, 1998; Jones, 1984), verb morphology (Kiss,2000), the verb-argument structure (Kiss, 2000; Kim &Thompson, 2000; Thompson, Shapiro, Li, & Schendel,1994; Thompson, Lange, Schneider, & Shapiro, 1997), tran-sitivity and ergativity (Bastiaanse & van Zonneveld, 2004,2005; De Bleser & Kauschke, 2003; Jonkers, 2000; Luzzattiet al., 2001, 2002; Thompson, 2003), and thematic informa-tion (Black, Nickels, & Byng, 1991; Breedin & Martin,1996; Byng, 1988; Kiss, 2000; Luzzatti et al., 2002; Thomp-son et al., 1997) played a role in verb processing in aphasicspeakers. Most of these eVects were reported for aphasicspeakers suVering from agrammatic or Broca’s aphasia.The factors that inXuence verb retrieval relate in general to

* Corresponding author. Fax: +31 50 363 68 55.E-mail address: [email protected] (R. Jonkers).

0093-934X/$ - see front matter © 2007 Elsevier Inc. All rights reserved.doi:10.1016/j.bandl.2007.01.002

Please cite this article in press as: Jonkers, R., & Bastiaanse, R., Acand name relation, Brain and Language (2007), doi:10.1016/j.bandl.2

grammatical aspects of the verbs. The eVect of these factorsis therefore often explained by the underlying grammaticaldeWcit Broca’s aphasic speakers are assumed to have (cf. forexample Thompson, 2003).

The present study describes eVects of non-grammaticalfactors on verb retrieval, i.e., a conceptual factor (instru-mentality) and a phonological/lexical factor (name relationbetween a noun and a verb). It will be investigated whetherthese factors also inXuence verb retrieval in Broca’s aphasicsubjects, although they do not concern grammatical fac-tors. Next to the data of Broca’s aphasic subjects also thedata of anomic aphasic subjects will be presented in orderto Wnd out whether these factors play a role in verb retrievalin these subjects. The latter seems plausible, as anomicaphasic speakers are assumed to suVer from a lexicalretrieval deWcit, which is more speciWcally considered to bedue to problems in retrieving the complete phonologicalinformation of words (e.g., Ellis & Young, 1988; Kay &Ellis, 1987). SpeciWc problems in verb form retrieval havebeen described by Bastiaanse (1991), Jonkers (1998), andJonkers and Bastiaanse (1998) in anomic aphasic speakers.These authors assume that anomic speakers are able to

tion naming in anomic aphasic speakers: EVects of instrumentality007.01.002

mailto: [email protected]

mailto: [email protected]

2 R. Jonkers, R. Bastiaanse / Brain and Language xxx (2007) xxx–xxx

ARTICLE IN PRESS

activate the correct verb lemma, but that they are unable toretrieve the corresponding phonological form from thelexicon.

In some studies, it has been shown that anomic orXuent aphasic speakers showed a better performance onnoun retrieval than on verb retrieval (Berndt, Haendiges,Burton, & Mitchum, 2002; Jonkers, 1998; Kambanaros &van Steenbrugge, 2006; Kohn, Lorch, & Pearson, 1989;Shapiro & Caramazza, 2003; Williams & Canter, 1987).However, other studies revealed the opposite pattern(Miceli, Silveri, Nocenti, & Caramazza, 1988; Miceli, Sil-veri, Villa, & Caramazza, 1984; Zingeser & Berndt, 1990).Therefore, it is unclear whether a relation between a verband a noun inXuences verb retrieval positively or nega-tively. Consequently, in the present study it is investigatedwhether a conceptual or phonological relation in namebetween a noun and a verb facilitates or inhibits verbretrieval in the anomic aphasic speakers and for compari-son in Broca’s aphasic speakers as well. Results on actionnaming will be compared with object naming scores inboth aphasic groups.

The eVect of a conceptual relation between verbs andnouns will be studied by focusing on so-called instrumentalverbs. Instrumental verbs are verbs referring to actions forwhich an instrument (not being a body part) is required inorder to perform the action. It is assumed that this instru-ment is part of the conceptual representation of the verb, asis illustrated by the following examples [example 1 is basedon the representation of the verb to clean as given by Jack-endoV (1990)]:1

(1) to clean:[CAUSE([ ]i,[INCH[NOT BE([SPOTS], [ONd[ ]j])]])]

(2) to polish:[[CAUSE([ ]i,[INCH[NOT BE([SPOTS], [ONd[] j])]])([BY[instrumentRAG])]

(3) to mop:[CAUSE([ ]i,[INCH[NOT BE([SPOTS],[ONd[ ]j])]])([BY[instrumentMOP])]

The diVerence between the more generic verb to cleanand the more speciWc verbs to polish and to mop is the pres-ence of an instrument in the conceptual representation.This part of the conceptual representation is necessary inorder to activate the intended verb.

The verbs to polish and to mop diVer with respect to thepresence of the instrument in the phonological form of theverb. The verb to mop incorporates the instrument phono-logically, whereas the verb to polish does not. Expressingthe instrument in a prepositional phrase is therefore stillpossible when to polish is used as in (4), while it is odd withthe verb to mop as in (5):

1 In these examples i is the actor, j is the patient or theme; d denotes thata determiner is necessary.


(4) The man polishes the mirror with a rag.(5) The man mops the Xoor with a mop.

Bastiaanse (1991) studied the eVect of instrumentality onverb retrieval in an experimental study in which twoBroca’s aphasic speakers and two subjects with anomicaphasia participated. She found no eVect of instrumentalityas such, but reported an eVect of name relation to a nounwithin the group of instrumental verbs. Clear patterns pertype could, however, not be found as the number of sub-jects was too small.

Kemmerer and Tranel (2000) reported a positive eVectof name relation (homophony) between verbs and nouns onaction naming for English in a large group of aphasicspeakers. However, they performed no analysis for theeVect of name relation in diVerent types of aphasic subjects.

Recently, Kambanaros and van Steenbrugge (2006)reported a study to the eVect of instrumentality and namerelation between a verb and a noun in bilingual Greek-English anomic aphasic speakers. They found a positiveeVect of instrumentality on action naming in both lan-guages. No eVect of name relation was found on the Greekversion of the action naming test, but name relation playeda negative role on the English version. No eVects werefound for both factors on a verb comprehension task.

Jonkers and Bastiaanse (2006), however, showed thatinstrumentality and name relation between a verb and a nounmay have an eVect on verb comprehension scores as well, butthey took the data of non-Xuent and Xuent aphasic speakersinto account. Instrumentality did have an eVect on the com-prehension scores of the non-Xuent aphasic speakers: instru-mental verbs were signiWcantly better understood than non-instrumental verbs. This was, however, only found whenthere was no name relation between the verb and the instru-ment. No inXuence of instrumentality was found for theXuent aphasic speakers. The latter group did show a signiW-cantly worse performance on instrumental verbs with a namerelation between the instrument and the verb as compared toinstrumental verbs without such a name relation.

In the Wrst paragraph of this section it was mentionedthat, in general, eVects of verb type on verb retrieval relate togrammatical factors. One clear exception to this pattern isthe study from Breedin et al. (1998; see also Barde et al.,2006). These authors presented data from eight aphasic sub-jects who showed a selective impairment to verb retrieval,but six of these patients were better in retrieving semanti-cally complex verbs than semantically simpler verbs. Seman-tically complex verbs are verbs that have a higher number ofsemantic features than semantically simple verbs, e.g., to runhas more semantic features than the semantically simplerverb to go. The authors point out that this shows thatsemantic features can facilitate verb retrieval, a Wnding thatis not in line with the view that the dissociations that were foundin aphasic subjects between verb and noun retrieval could beexplained by assuming a lexical organization according togrammatical category, as for example has been suggested byCaramazza and Hillis (1991). Breedin et al. (1998) therefore


R. Jonkers, R. Bastiaanse / Brain and Language xxx (2007) xxx–xxx 3

ARTICLE IN PRESS

opt for an explanation that relates the verb retrieval prob-lems to problems in accessing verb lemmas.

Kambanaros and van Steenbrugge (2006) describe twomodels to discuss possible eVects of name relation andinstrumentality on verb retrieval. These eVects could at Wrstbe explained by using the framework of the spreading acti-vation model (see e.g., Dell, 1986). This model has beenadapted to explain the problems of aphasic speakers inword retrieval (Dell, Schwarz, Martin, SaVran, & Gagnon,1997; Foygel & Dell, 2000) or to describe the relationbetween input and output phonological processing (Martin& SaVran, 2002).

Within the spreading activation theory an eVect ofinstrumentality could be explained by assuming co-activa-tion of the instrument when activating the verb, leading to alower activation threshold for the verb. A positive eVect ofname relation is expected due to upward activation at thephonological form level facilitating the activation of thephonological form of the target verb. Kambanaros and vanSteenbrugge (2006) found, however, for English, a negativeeVect on verb retrieval with instrumental verbs that had aname relation to a noun. The spreading activation theorycannot account for a negative eVect of name relation or theabsence of such an eVect as was found for the Greek ver-sion of the action naming test. Therefore, Kambanaros andVan Steenbrugge used a model that does not assumeupward activation to discuss their results, namely Levelt’stwo-stage model (Levelt, 1989; Levelt, Roelofs, & Meyer,1999). Dependent on the results, in the current study thedata will be explained in terms of either Levelt’s two-stagemodel or Dell’s (1986) spreading activation theory.

2. Methods study 1

2.1. Participants

This study included 19 anomic aphasic speakers and 15Broca’s aphasic speakers. The participants were selectedfrom a larger group of 66 aphasic subjects that took part ina study to verb and noun retrieval at word and sentencelevel (Jonkers, 1998). Participants who did not suVer fromthe two mentioned aphasia types were excluded from thestudy.

All participants were native speakers of Dutch. Thediagnosis of anomic or Broca’s aphasia was based on theclassiWcation of the Dutch version of the Aachen Apha-sia Test (Graetz, De Bleser, & Willmes, 1992) and wasconWrmed by clinical judgment of the spontaneousspeech of the aphasic subjects by someone who did notknow them.

Subject data are presented in Appendix A. Fifteen ano-mic aphasic speakers and all Broca’s aphasic speakerssuVered from a single stroke. In two of the anomic aphasicspeakers, the aphasia was caused by closed-head injury andin two cases by an infectious disease. Mean age of the ano-mic aphasic speakers was 57 (range 24–79). For the Broca’saphasic speakers the mean age was 58 (range 38–78). The


aphasic participants did not have major auditory or visualproblems.

The data of a group of 15 non-brain-damaged speakers,selected to match both aphasic groups in age, gender, andeducation, are given for comparison.

2.2. Materials

An action naming task, consisting of 60 items wasadministered. The verbs in the test were controlled for twolinguistic factors, i.e., instrumentality and name relation to anoun. A comparison will be made with the naming data ofan object naming test that also contained 60 items. Forty ofthe sixty nouns that are used for object naming are theinstruments that belong to the instrumental verbs. Next tothe 40 instruments, 20 objects that are semantically relatedto the non-instrumental verbs were used (e.g., to climb—tree). A list of items is provided in Appendix B.

2.3. Instrumentality

Instrumental verbs were deWned as referring to actionsfor which an instrument (not being a body part) is required.For example, to brush, to sledge, and to dig were regarded asinstrumental verbs, whereas to kneel, to climb, and to fold,were considered to be non-instrumental verbs. From the 60verbs in the action naming test, 40 were instrumental and20 were non-instrumental.

2.4. Name relation to a noun

Within the class of instrumental verbs a division wasmade between verbs that are related in name to the instru-ment (to bath, to brush, and to sledge), and those that arenot (to bake, to row and to dig).2 Both groups of verbs con-sisted of 20 items.

2.5. Transitivity

The verbs in the action naming test were controlled fortransitivity. Half of the verbs were transitive, i.e., verbs thatrequire a direct object, like to grind or to scratch. The otherhalf were intransitive verbs, i.e., verbs that cannot have adirect object, like to fence and to swim. The eVects of transi-tivity on verb retrieval were reported in a former study(Jonkers, 2000). For this study the same materials wereused. Every single group of 20 verbs consisted of 10 transi-tive and 10 intransitive verbs (i.e., 10 non-instrumentalverbs are transitive and 10 are intransitive etc.). The verbsthat were used were not speciWcally controlled for the the-matic structure of the verb. All items, however, concernaction verbs, which means that for most of the verbs thethematic structure is either verb(+agent, theme/patient)

2 In Dutch, the name-related instrumental verb is not identical in formto the instrument. The inWnitive is noun + en (e.g., Wets (noun: a cycle)—Wetsen (verb: to cycle)).



ARTICLE IN PRESS

(transitive verbs) or verb(+agent) (intransitive verbs). Allbut two verbs in the test were irreversible verbs.

2.6. Frequency

The verbs in the action naming test were mutuallymatched for word-frequency using the CELEX-frequencylist for Dutch (Burnage, 1990). This means that the verbs inthe diVerent subgroups have comparable frequencies. Thesame holds for the nouns in the object naming test andthese are also matched for frequency with the verbs that areused in action naming.

2.7. Imageability

Bird, Howard, and Franklin (2000) have argued thatverb retrieval problems are mainly caused by the fact thatverbs have a lower imageability than nouns. Post hoc,imageability ratings were obtained for most of the items inthe action naming test using the imageability ratings forDutch by van Loon-Vervoorn (1985).3 Non-instrumentalverbs had a mean rating of 6.29 (nD14), non-name-relatedinstrumental verbs had a mean rating of 6.47 (nD 17) andname-related instrumental verbs had a mean rating of 6.31(nD14). A one-way ANOVA revealed no signiWcant diVer-ence between the group ratings: (F(2,42)D 1.13, p > .05).

2.8. Reliability

Reliability of the tests was calculated on the basis of thescores of 15 aphasic speakers, randomly selected from thegroup of 66 subjects who were tested with the action nam-ing test. Reliability was calculated using the odd–evencoeYcient. The odd–even reliability was rD0.96 (p < .001)for both action and object naming.

2.9. Procedure

The subjects were tested according to the following pro-cedure: a picture was shown and the subject was asked tosay in one word what the person on the picture was doing(action naming) or what was on the picture (object nam-ing). The tests started with two examples. Subjects whogave an inadequate response to one of the examples werecorrected. If a subject produced (a part of) a sentence forthe examples, he was asked to react in just one word. Afterthe examples, no more feedback was given. There was notime limit, self-corrections were allowed and scoring wasbased on the Wnal answer.4

3 Van Loon-Vervoorn obtained imageability ratings for 1000 verbs, basedon ratings of non-language disturbed subjects on a 7 point scale (1, low im-ageability; 7, high imageability). Ratings were available for 45 of the 60 items.

4 The Wnal answer was scored because it is assumed that, next to the Wrstanswer, this is the most objective way to score the reactions. As the study fo-cused on the ability to retrieve the verb more than on the speed of retrieval,the Wnal answer was used in order to be able to score repairs as well.


2.10. Scoring

The elicited speech was tape-recorded. In scoring it wasonly counted if the target verb was produced, not only ifthese were produced as single words. This means that whena subject produced the correct verb within a (part of a) sen-tence, then the reaction was counted as correct, for exam-ple: this man is mowing the grass or mowing the grassinstead of only saying mowing.

3. Results

3.1. Action naming

The results for the eVect of instrumentality and namerelation to a noun in action naming are depicted in Table 1.Individual scores are provided in Appendix C.

The control subjects showed superior performance on alltypes of verbs as compared to the aphasic groups (controlsubjects vs. anomic aphasic speakers: non-instrumentalverbs: t(32)D9.6, p < .001; non-name-related instrumentalverbs: t(32)D 8.1, p < .001; name-related instrumental verbs:t(32)D6.9, p < .001; control subjects vs. Broca’s aphasicspeakers: non-instrumental verbs: t(28)D 10.2, p < .001;non-name-related instrumental verbs: t(28)D10.7, p < .001;name-related instrumental verbs: t(28)D8.6, p < .001).

An ANOVA revealed no eVect of verb type for theBroca’s aphasic speakers and the control subjects (Broca’saphasic speakers: F(1,43)D1.05, p > .05; control subjects:

Table 1Mean and standard deviation (SD) of the action and object naming scoreswith respect to the factors instrumentality and name relation to a noun instudy 1 (¡I, non-instrumental; +I ¡ N, instrumental, non-name-related;+I + N, instrumental, name-related; anomic, anomic aphasic speakers;Broca, Broca’s aphasic speakers; control, control subjects)

¡I I ¡ N I + N

Instrumentality and name relation

AnomicAction naming 7.4 (4.0) 10.0 (3.8) 10.3 (4.5)Object naming 14.4 (4.4) 12.5 (4.5) 11.8 (5.0)

BrocaAction naming 6.6 (4.0) 7.6 (3.6) 8.4 (4.4)Object naming 11.7 (3.1) 11.7 (3.8) 12.1 (3.3)

ControlAction naming 17.8 (1.5) 18.2 (1.2) 18.6 (1.2)Object naming 18.1 (1.5) 18.6 (1.2) 18.7 (1.1)

Total

AnomicAction naming 27.6 (11.6)Object naming 41.0 (11.2)

BrocaAction naming 22.6 (10.8)Object naming 35.7 ( 9.0)

ControlAction naming 54.6 (2.9)Object naming 55.3 (2.9)



ARTICLE IN PRESS

F(1,43)D 2.10, p > .05). The scores of the anomic aphasicsubjects, however, were inXuenced by the type of verb thatwas used in the test. An ANOVA revealed an eVect of verbtype (F(1, 55)D5.53, p < .05). A post hoc analysis showed abetter performance for the instrumental than the non-instrumental verbs (tD2.35, p < .05). No eVect of name rela-tion was found (tD 0.24, p > .05). The interaction betweeninstrumentality and name relation reached signiWcance(F(2, 54)D 2.74, pD .07).5

3.2. Object naming

Object naming scores can be found in Table 1, individualscores are provided in Appendix C. No diVerence wasfound between action naming and object naming in thecontrol group (t(14)D1.28, p > .05). Both aphasic groups,however, were better in object naming than in action nam-ing. (Broca’s aphasic subjects t(14)D 8.25, p < .001; anomicaphasic subjects: t(18)D6.80, p < .001). Individually, all butone aphasic subject had higher scores on object namingthan on action naming.

Within the group of objects a division was made withrespect to the factors instrumentality and name relation. NoeVects of noun-type were found for the control subjects andthe Broca’s aphasic subjects (control subjects: F(1,43)D1.74,p>.05; Broca’s aphasic subjects: F(1,43)D0.19, p>.05). How-ever, the type of noun played a role in object naming in theanomic aphasic subjects (F(1,55)D4.07, p<.05). A post hocanalysis revealed a signiWcantly worse performance on instru-ments than on objects that were not an instrument (tD2.02,p<.05) No eVect of name relation was found (tD0.48, p>.05).Also no interaction between instrumentality and name relationwas found (F(2,54)D2.13, p>.05).

4. Discussion

Instrumentality had an eVect on verb and noun retrievalin the anomic aphasic speakers, but not in the Broca’saphasic speakers or the control subjects. No eVect of namerelation on action or object naming was found in either

5 One reviewer suggested that the eVect of verb type might relate to theseverity of the verb impairment rather than to the type of aphasia. In orderto test this, all aphasic subjects were divided in two groups, those that hadat least a 25% lower score on action naming than on object naming(nD 22) and those with a smaller diVerence between the action and objectnaming scores (n D 12). An eVect of instrumentality was found in the Wrstgroup (F(1, 66) D 5.50, p < .05) and not in the second group (F(1, 34) D 2.43,p > .05). This might indeed lead to the conclusion that the eVect of verbtype relates to the severity of the verb impairment. However, in both theverb-impaired and in the non-verb-impaired group an eVect of instrumen-tality was only found for the anomic aphasic speakers and not for theBroca’s aphasic speakers that were included in these groups (non-verb-im-paired: Broca’s aphasic speakers: z D 0.45, p > .05; anomic aphasic speak-ers: z D 2.21, p < .05); verb-impaired: Broca’s aphasic speakers:t(10)D 1.14, p > .05; anomic aphasic speakers: t(10)D 3.75; p < .01. Thisshows that the eVect of instrumentality at least also relates to the type ofaphasia.


group. This eVect for action naming is in line with the Wnd-ings for Greek of Kambanaros and van Steenbrugge (2006).

As mentioned above, Kemmerer and Tranel (2000)reported a positive eVect of name relation between verbs andnouns on action naming in a group of aphasic speakers. Inthe current study and in the study of Kambanaros and vanSteenbrugge (2006), the eVect of name relation was investi-gated within a group of instrumental verbs. The instrumen-tal verbs were easier to retrieve for the anomic aphasicspeakers than the non-instrumental verbs. This might,therefore, have confounded the results with respect to aneVect of name relation as such: name relation could have apositive eVect on verb retrieval, as reported by Kemmererand Tranel (2000), but only if it is studied in isolation. Thefact that the interaction between instrumentality and namerelation in action naming almost reached signiWcance in theanomic aphasics group already shows that name relationmight also have a positive eVect on the action namingscores. Therefore, a second study was performed investigat-ing name relation, only using verbs with or without a namerelation to a noun. In line with the results of Kemmerer andTranel, a positive eVect of name relation was expected.

5. Methods study 2

5.1. Participants

The study included 10 anomic aphasic speakers and 10Broca’s aphasic speakers, all native speakers of Dutch. Thesewere diVerent participants than the participants who weretested in the Wrst study. The aphasic speakers were takenfrom a larger group of 35 aphasic subjects that were testedfor the standardization of the Werkwoorden-en Zinnentest(WEZT; Bastiaanse, Maas, & Rispens, 2000), the Dutch ver-sion of the Verb and Sentence Test (VAST; Bastiaanse,Edwards, & Rispens, 2002). Aphasic subjects who were notsuVering from one of the two mentioned types of aphasiawere excluded from the study. The diagnosis of anomic apha-sia or Broca’s aphasia was based on the classiWcation of theDutch version of the Aachen Aphasia Test (Graetz et al.,1992) and was conWrmed by clinical judgment of the sponta-neous speech of the aphasic subjects by someone who did notknow them.6 Subject data are presented in Appendix A.Nineteen aphasic speakers suVered from a single stroke, onefrom closed-head injury. Mean age of the anomic aphasicspeakers was 55 (range 37–75). For the Broca’s aphasicspeakers the mean age was 47 (range 22–68). The aphasicsubjects did not have major auditory or visual problems.

5.2. Materials

The action naming task of the WEZT was administered.This test consists of 40 items. Twenty verbs have no name

6 Three anomic aphasic speakers were classiWed as rest aphasic on thebasis of the AAT scores. In these cases clinical judgement determined theclassiWcation anomic aphasia.



ARTICLE IN PRESS

relation to a noun that is depicted on the picture (e.g., topick) and 20 verbs are name-related to a noun (e.g., to peel-apeel). Verbs were matched for frequency and controlled fortransitivity. A group of 40 non-brain-damaged speakers,selected to match the aphasic groups in age, gender, andeducation, scored more than 95% correct on this test. A listof items is provided in Appendix B.

5.3. Procedure and scoring

The test was administered and scored following the sameprocedure as described for the Wrst study.

5.4. Imageability

Imageability ratings were obtained post hoc based on 33of the 40 items of the action naming test using the image-ability ratings for Dutch by van Loon-Vervoorn (1985).Name-related verbs had a mean rating of 6.43 (nD 17), non-name-related verbs had a mean rating of 6.51 (nD16). ThisdiVerence is not signiWcant: t(15)D 0.60, p > .05.

5.5. Reliability

Reliability of the test was calculated on the basis of thescores of 20 aphasic speakers, randomly selected from thetotal group of subjects who were tested with the WEZT.Reliability was calculated using the odd–even coeYcients.The odd–even reliability for action naming was rD0.94(p < .001).

6. Results

The results are depicted in Table 2. Individual scores areprovided in Appendix C. Name-related verbs were betterretrieved by the anomic aphasic speakers than non-name-related verbs (t(9)D 3.88, p < .01). For the Broca’s aphasicspeakers comparable scores were found for name-relatedand non-name-related verbs (t(9)D 0.66, p > .05).

7. General discussion

The present study revealed an eVect of the conceptualfactor instrumentality on verb and noun retrieval and of thelexical/phonological factor name relation to a noun on verbretrieval in anomic aphasic speakers, but not in Broca’s

Table 2Mean and standard deviation (SD) of the action naming scores withrespect to the factor name relation to a noun in the second study

Non-name-related Name-related

Name relation to a noun

Anomic aphasic speakersMean (SD) 10.7 (4.1) 13.3 (3.8)

Broca’s aphasic speakersMean (SD) 9.6 (4.1) 10.1 (4.3)


aphasic speakers. The eVect of name relation was onlyfound when it did not interfere with the eVect of instrumen-tality. Within the group of instrumental verbs name relationto the instrument did not play a role (experiment 1). How-ever, when name relation as such was studied, it had a posi-tive eVect on verb retrieval in anomic aphasic speakers(experiment 2). This could explain why a positive eVect ofname relation was not found by Kambanaros and van Stee-nbrugge (2006), as, like in the Wrst experiment, they alsostudied the eVect of name relation within the group ofinstrumental verbs.7

In Section 1 it was mentioned that the absence of aneVect of name relation would be diYcult to explain withinthe spreading activation theory of Dell (1986). Therefore,Kambanaros and van Steenbrugge (2006) explained theirresults in terms of the two-stage model of Levelt (1989). Asthe present study, however, shows that action naming canbe inXuenced by name relation if it is studied as such, thespreading activation theory (Dell, 1986) seems to be themost plausible model to explain the outcomes of this study.

Within the spreading activation framework it can beexplained that instrumentality and name relation between averb and a noun play a role in verb retrieval. In Section 1 ithas already been discussed that the conceptual representa-tion of an instrumental verb contains the instrument. Con-sidering the spreading activation theory, this means thatwith an instrumental verb not only the lemma of the verb isactivated, but also the lemma of the instrument. This isdepicted in Fig. 1.

Co-activation of the instrument might play a role in acti-vating the phonological form of the verb. According to thespreading activation theory, a lemma deactivates, after ithas reached its threshold and got the phonological formactivated, while other co-activated lemmas remain active.

7 Kambanaros and van Steenbrugge (2006) reported a negative eVect ofname relation in their bilingual patient group for their second languageEnglish. The authors explained this Wnding by suggesting weaker connec-tions between lemmas and phonological forms in L2 and therefore an in-creased competition of phonological forms that are co-activated with aninstrumental verb.

Fig. 1. The activation of the verb to skate according to the theory ofspreading activation (cf. Dell, 1986).

Concept

Lemma skateN skateV skiV

Phonological form skateV rate



ARTICLE IN PRESS

These co-activated lemmas could reactivate the targetlemma in case of retrieval problems. The assumption is thatin the anomic aphasic speakers, the lemma of the verb deac-tivates too quickly to fully activate the phonological formthat corresponds to this lemma. This means that these sub-jects are able to activate the lemma of a verb that belongs toits conceptual representation, but they do not always suc-ceed in activating the corresponding phonological form.Instrumental verbs are easier to retrieve than non-instru-mental verbs as co-activation of the lemma of the instru-ment helps to reactivate the verb lemma. Co-activation ofthe instrument is furthermore strengthened by the depictionof the instrument in the picture.8

Co-activation of a noun can only be helpful if nouns areeasier to retrieve than verbs. From the current study it is clearthat this is the case. However, within the group of nouns,instrumental nouns were more diYcult to retrieve than non-instrumental nouns, again only for the anomic aphasicspeakers. Although the scores for the instrumental nouns arestill higher than scores for the instrumental verbs, this showsthat there exists a tighter relationship between instrumentalverbs and their corresponding instruments than to verbs andsemantically related nouns in general. This is also in line withthe Wndings of Ferretti, Mc Rae, and Hatherell (2001). Theyfound priming eVects in a verb reading and lexical decisiontask for verb-instrument pairs, demonstrating that subjectsimmediately activate knowledge of instruments after readingan instrumental verb. Comparable results were not found forverb-location pairs, showing that information about locationis less salient.

Not only at the conceptual level may the availability of arelated noun be helpful in retrieving the verb. Name relationbetween a verb and a noun was also shown to have a positiveeVect on verb retrieval in anomic aphasic speakers. Asalready mentioned, in the current study the anomic aphasicspeakers were better in noun retrieval than in verb retrievalin general. If some verbs share most of their phonologicalinformation with a noun, as is the case for name-relatedverbs and nouns, then it is expected that these verbs are alsoeasier to retrieve than other verbs.

The question remains however, why such a Wnding wasnot found in the Broca’s aphasic speakers. This group isassumed to have problems in verb lemma retrieval and notin verb form retrieval (cf. Jonkers, 1998). Therefore Broca’saphasic speakers do not proWt from the fact that someverbs and their name-related nouns share most of theirphonological information, notwithstanding a better perfor-mance on object naming than on action naming.

A deWcit in activating the verb lemma in Broca’s aphasicspeakers also explains why instrumentality did not inXuencethe action naming scores in these subjects. It was assumedthat the co-activated instrument facilitates verb retrieval

8 An eVect of instrumentality might be fully retraced to the visual repre-sentation of the action. The absence of this eVect in the Broca’s aphasicspeakers is an indication that this does not (solely) explain the eVect of thisfactor in the anomic aphasic speakers.


when the phonological form cannot be activated. At thattime the verb lemma has already been activated. Co-activa-tion of the lemma of the instrument does not seem to facili-tate lemma retrieval and therefore it does not have an eVecton verb retrieval in Broca’s aphasic speakers.

The results of this study have implications for theories onverb retrieval deWcits. Some years ago, a discussion hasstarted on the role of imageability in verb and noun retrieval.Bird et al. (2000) stated that deWcits in verb production arenot category speciWc, but simply an eVect of verbs having alower imageability than nouns. This view has been disputedby Shapiro and Caramazza (2001) and Berndt et al. (2002)(see also Bird, Howard, & Franklin, 2001).

With respect to the current study, it can be said that iflow imageability is the decisive factor for verb Wnding prob-lems, this would mean, that verbs that are instrumental orname-related to a noun should be easier to retrieve thannon-instrumental and non-name-related verbs, because ofthe higher imageability of the former verbs. However, in thecurrent study, comparable imageability ratings wereobtained for the diVerent types of verbs. This means that,without claiming that imageability does not play a role inverb retrieval at all (see for example the study of Luzzattiet al., 2002; revealing a clear eVect of imageability on verbretrieval), it may not exclusively account for the perfor-mance found for the retrieval of diVerent types of verbs.

A second implication of this study relates to the diVerentexplanations for verb retrieval deWcits that have beendescribed in the literature. Some researchers claim that verbretrieval deWcits connect with the fact that verbs and nounsare stored separately in the phonological lexicon (Miceliet al., 1984, 1988; Williams & Canter, 1987). Agrammaticspeakers (Miceli et al., 1984, 1988) or aphasic speakers ingeneral (Williams & Canter, 1987) are considered to have animpairment in (the access to) the verb component of the lexi-con. Assuming diVerent subcomponents for verbs and nounsin the lexicon cannot explain the eVect of name relation to anoun in subjects suVering from a verb form retrieval deWcit.However, assuming that verbs and name-related nouns sharemost of their phonological information can account for thedata of the present study. Therefore, in the light of the pres-ent study, a shared phonological lexicon is assumed for thestorage of nouns and verbs.

The results of the current study are in line with theresults found by Breedin et al. (1998). They showed thatsemantically complex verbs were easier to retrieve thansemantically simpler verbs for aphasic subjects suVeringfrom verb retrieval problems. As mentioned before, suchresults do not Wt in a theory that assumes a lexical organiza-tion on the basis of grammatical category as well. Withrespect to Breedin et al.’s results we assume that instrumen-tal verbs might be seen as semantically more complex thannon-instrumental verbs, due to the presence of an instru-ment in the conceptual representation of the verb.

In general, it is stressed that eVects of verb type might befound in verb retrieval of both anomic and Broca’s aphasicspeakers. Which eVects play a role primarily depends on the



ARTICLE IN PRESS

locus of the functional deWcit that is assumed for the diVerenttypes of aphasia. Grammatical factors seem to play a role inBroca’s aphasic speakers, suVering from a grammatical deWcit(e.g., Kim & Thompson, 2000; Jonkers, 2000), whereas concep-tual and phonological factors inXuence verb retrieval in ano-mic aphasic speakers, suVering from a lexical retrieval deWcit.

Acknowledgments

The Wrst study reported in this article was part of thePh.D. project of the Wrst author (Jonkers, 1998). This pro-ject was sponsored by the Dutch Foundation for ScientiWcResearch (NWO). We thank the aphasic and non-aphasicparticipants for their co-operation. Furthermore, we thankthe Aphasia Research Group Groningen for their valuablecomments on an earlier version of this article.

Appendix A. Patient data

Study 1

Subject data of the aphasic speakers in the Wrst study (A, anomicaphasic speaker; B, Broca’s aphasic speaker; CVA, cerebro-vascularaccident; L, left hemisphere; TBI, traumatic brain injury; enceph.,encephalitis; MPO, months post onset; m, male; f, female; ¤ subject isleft-handed).

Age Aetiology MPO Gender

A1 68 CVA-L 36 fA2 50 CVA-L 13 fA3 52 CVA-L 18 mA4 73 CVA-L 49 mA5 66 CVA-L 7 mA6 52 CVA-L 9 mA7 55 CVA-L 7 mA8 74 CVA-L 12 mA9 79 CVA-L 15 mA10 74 CVA-L 11 fA11 68 CVA-L 8 mA12 35 TBI 7 mA13 47 enceph. 3 mA14 62 enceph. 4 mA15 48 CVA-L 6 mA16 24 CVA-L 4 fA17 35 TBI 28 mA18 62 CVA-L 11 mA19 61 CVA-L 5 m

Mean 57.1 13.3

B1 45 CVA-L 77 mB2 78 CVA-L 29 fB3 38 CVA-L 154 fB4 54 CVA-L 34 mB5 61 CVA-L 11 mB6 72 CVA-L 89 mB7 65 CVA-L 5 mB8 53 CVA- L 12 mB9 66 CVA-L 37 mB10 51 CVA-L 10 fB11 63 CVA-L 125 fB12 47 CVA-L 38 mB13 48 CVA-L 15 fB14 62 CVA-L 56 mB15 71 CVA-R¤ 9 m

Mean: 58.3 46.7


Study 2

Subject data of the aphasic speakers in the second study (A, anomic apha-sic speakers; B, Broca’s aphasic speakers; CVA, cerebro-vascular accident;L, left hemisphere; TBI, traumatic brain injury; MPO, months post onset;m, male; f, female; ¤ subject is left-handed).

Appendix B. Test items

Items test 1

Age Aetiology MPO Gender

A1 63 CVA-L 38 mA2 66 CVA-L 11 mA3 46 CVA-L 6 fA4 77 CVA-L 50 fA5 41 CVA-L 5 fA6 51 CVA-L 7 mA7 51 CVA-L 5 fA8 40 CVA-L 9 fA9 57 CVA-L 18 fA10 62 CVA-L 15 m

Mean 55.4 16.4

B1 50 CVA-L 3 fB2 51 CVA-L 10 mB3 40 CVA-L 8 mB4 22 TBI 53 mB5 51 CVA-L 26 mB6 55 CVA-R¤ 21 mB7 48 CVA-L 25 mB8 44 CVA-L 13 mB9 39 CVA-L/TBI 3 fB10 68 CVA-L 8 m

Mean 46.8 17.0

¡I +I¡ N +I + N

ActionsWorstelen

(wrestling)Badmintonnen (playing badminton)

Sleeën (sledging)

Duiken (diving) Hockeyen (playing hockey)

Baden (taking a bath)

Hurken (squatting) Stijgen (taking oV) Fluiten (playing the Xute)

Toveren (practicing witchcraft)

Dobbelen (dicing) Tollen (playing with a top)

Springen (leaping) Vissen (Wshing) Skiën (skiing)Bedelen (begging) Biljarten (playing

billiards)Puzzelen (doing a puzzle)

Dalen (descending) Schaken (playing chess) Hoepelen (rolling a hoop)

Turnen (doing gymnastics)

Varen (navigating) Wippen (seesawing)

Zwaaien (waving) Fotograferen (photographing)

Bellen (ringing a bell)

Niezen (sneezing) Timmeren (hammering) Fietsen (cycling)Plukken (picking) Hakken (choping) Tappen (drawing

beer)Melken (milking) Poetsen (polishing) SchoVelen (hoeing)Vouwen (folding) Naaien (sewing) Kammen (combing)Feliciteren

(congratulating)Snijden (cutting) Ploegen (ploughing)

Vlechten (braiding) Roeren (stiring) Dweilen (mopping)Stuiteren (bouncing) Sproeien (watering) Lijmen (gluing)Krabben (scratching) Zeilen (sailing) Zeven (sieving)Wringen (wringing) Verven (painting) Schaven (planing)

(continued on next page)



ARTICLE IN PRESS

¡I, non-instrumental verb; +I ¡ N, instrumental verb without a namerelation with the instrument; +I + N, instrumental verb with a name rela-tion with the instrument (the Wrst 10 verbs are all intransitive, the second10 verbs are transitive).

¡I, non-instrumental noun; +I ¡ N, instrument without a name relationwith the instrumental verb; +I + N, instrument with a name relation withthe instrumental verb.

Items test 2

¡N, verb without a name relation with a noun; +N, verb with a namerelation with a noun.

Appendix B (continued)

¡I +I ¡ N +I + N

Gooien (throwing) Bakken (frying) Vijlen (Wling)Masseren

(massaging)Spitten (digging) Zagen (sawing)

¡I +I ¡ N +I + N

ObjectsBeen (leg) Roller (roller brush) Vijl (Wle)Nagel (nail) Slang (hose) Lijm (glue)Balk (beam) Hengel (Wshing-rod) Puzzel (jig saw puzzle)Knie (knee) Zeilboot (sailboat) Wipwap (seesaw)Sok (sock) Koekepan

(frying pan)Bel (bell)

Hond (dog) Bijl (axe) Hoepel (hoop)Bedelaar (beggar) Keu (cue) Fluit (Xute)Vuist (Wst) Schaakbord

(chessboard)Zeef (sieve)

Bloem (Xower) Vliegtuig (airplane) Schaaf (plane)Stoep (pavement) Boot (boat) Bad (bath)Kado (present) Lepel (spoon) Zaag (saw)Haarknip

(hairpin/-clip)Hamer (hammer) Ski (ski)

Tovenaar (wizard) Mes (knife) Ploeg (plough)Berg (mountain) Dobbelstenen (dice) Dweil (mop)Bal (ball) Stick ((hockey)stick) Tap (tap)Duim (thumb) Racket (racket) Tol (top)Trein (train) Fototoestel (camera) Slee (sleigh)Papier (sheets) Schop (shovel) Fiets (bicycle)Zakdoek

(handkerchief)Doekje (cloth) SchoVel (hoe)

Duikplank (diving board)

Naald (needle) Kam (comb)

¡N +N

Stofzuigen (vacuuming) Pompen (pumping)Plukken (picking) Rolschaatsen (roller-skating)Schaken (playing chess) Biljarten (playing biliards)Roeien (rowing) Vegen (sweeping)Strijken (ironing) Tollen (playing with a top)Drinken (drinking) Persen (pressing)Metselen (bricklaying) Stempelen (stamping)Knijpen (pinching) Schommelen (swinging)Gooien (throwing) Kammen (combing)Dammen (playing at draughts) Vijlen (Wling)Schermen (fencing) Schaatsen (skating)Dobbelen (dicing) Boren (drilling)Masseren (massaging) Fietsen (cycling)Kruipen (crawling) Harken (raking)Hinkelen (hopping) Schillen (peeling)Zwemmen (swimming) Verven (painting)Maaien (mowing) Skieën (skiing)Varen (navigating) Baden (taking a bath)Blussen (extinguishing) Duiken (diving)


Appendix C. Individual scores

Study 1

Individual scores study 1 for action and object naming: B, subjects suVer-ing from Broca’s aphasia; A, subjects suVering from anomic aphasia; C,control subjects; ¡I, non-instrumental; +I ¡ N, instrumental, non-name-related; +IN, instrumental, name-related.

¡I (n D 20) +I ¡ N (n D 20) +IN (n D 20) Total

Actions Objects Actions Objects Actions Objects Actions Objects

Actions and objectsB1 7 12 11 9 10 8 28 29B2 5 10 5 9 4 7 14 26B3 16 19 15 19 15 17 46 55B4 3 12 8 9 10 9 21 30B5 3 12 5 8 3 13 11 33B6 4 11 10 16 15 15 29 42B7 8 15 12 17 10 18 30 50B8 9 12 3 13 6 15 18 40B9 13 14 11 17 16 15 40 46B10 7 11 7 9 11 12 25 32B11 4 11 7 9 6 10 17 30B12 3 8 4 9 3 9 10 26B13 6 9 4 13 6 11 16 33B14 9 14 9 10 7 12 25 36B15 2 6 3 9 4 12 9 27

A1 13 16 14 14 18 13 45 43A2 14 20 18 16 18 18 50 54A3 5 14 12 12 9 14 26 40A4 2 14 7 11 6 11 15 36A5 8 14 8 12 11 13 27 39A6 1 4 5 5 4 3 10 12A7 11 18 11 15 17 18 39 51A8 4 14 3 8 5 4 12 26A9 5 17 7 18 8 15 20 50A10 7 18 9 14 9 16 25 48A11 11 19 13 16 14 17 38 52A12 6 14 10 16 13 16 29 46A13 11 15 14 15 15 17 40 47A14 5 15 11 10 10 9 26 34A15 8 16 11 15 7 9 26 40A16 11 19 13 13 11 13 35 45A17 1 9 4 7 4 6 9 22A18 11 15 11 12 9 13 31 40A19 7 17 8 19 7 17 22 53

C1 17 20 17 18 20 18 54 56C2 19 19 20 19 20 20 59 58C3 19 18 19 20 17 19 55 57C4 19 19 19 19 20 19 58 57C5 16 17 16 18 18 18 50 53C6 16 15 18 19 17 18 51 52C7 18 20 19 18 19 17 56 55C8 19 17 19 19 19 20 57 56C9 20 20 18 20 20 20 58 60C10 19 18 18 19 17 18 54 55C11 17 18 17 20 18 18 52 56C12 19 19 18 19 20 20 57 58C13 16 17 20 19 18 20 54 56C14 18 17 18 16 17 17 53 50C15 15 16 17 16 19 19 51 51



ARTICLE IN PRESS

Study 2

Individual scores study 2: B, subjects suVering from Broca’s aphasia; A,subjects suVering from anomic aphasia.

References

Bastiaanse, R. (1991). Naming of instrumental verbs in aphasia: an explor-ative study. Clinical Linguistics and Phonetics, 5, 355–368.

Bastiaanse, R., Edwards, S., & Rispens, J. (2002). Verb and sentence test(VAST). SuVolk: Thames Valley Test Company.

Bastiaanse, R., Maas, E., & Rispens, J. (2000). Werkwoorden-en zinnentest(WEZT). Lisse: Swets Test Services.

Bastiaanse, R., & van Zonneveld, R. (2004). Broca’s aphasia, verbs and themental lexicon. Brain and Language, 90, 198–202.

Bastiaanse, R., & van Zonneveld, R. (2005). Sentence production withverbs with alternating transitivity in agrammatic Broca’s aphasia. Jour-nal of Neurolinguistics, 18, 57–66.

Barde, L. H. F., Schwartz, M. F., & Boronat, C. B. (2006). Semantic weightand verb retrieval in aphasia. Brain and Language, 97, 266–278.

Berndt, R. S., Haendiges, A. N., Burton, M. W., & Mitchum, C. (2002).Grammatical class and imageability in aphasic production: their eVectsare independent. Journal of Neurolinguistics, 15, 353–371.

Bird, H., Howard, D., & Franklin, S. (2000). Why is a verb like an inani-mate object? Grammatical category and semantic category deWcits.Brain and Language, 72, 246–309.

Bird, H., Howard, D., & Franklin, S. (2001). Noun–verb diVerences? Aquestion of semantics: a response to Shapiro and Caramazza. Brainand Language, 76, 213–222.

Black, M., Nickels, L., & Byng, S. (1991). Patterns of sentence processingdeWcit: processing simple sentences can be a complex matter. Journal ofNeurolinguistics, 6, 79–101.

Breedin, S. D., & Martin, R. C. (1996). Patterns of verb impairment inaphasia: an analysis of four cases. Cognitive Neuropsychology, 13, 51–91.

Breedin, S. D., SaVran, E. M., & Schwartz, M. F. (1998). Semantic factorsin verb retrieval: an eVect of complexity. Brain and Language, 63, 1–31.

Burnage, G. (1990). A guide for users. Nijmegen: CELEX, Centre for Lexi-cal Information.

Byng, S. (1988). Sentence processing deWcits: theory and therapy. CognitiveNeuropsychology, 5, 629–676.

Caramazza, A., & Hillis, A. (1991). Lexical organization of nouns andverbs in the brain. Nature, 349, 788–790.

De Bleser, R., & Kauschke, C. (2003). Acquisition and loss of nouns andverbs: parallel or divergent patterns? Journal of Neurolinguistics, 16,213–229.

Name-related (nD 20) Non-name-related (n D 20)

A1 13 9A2 14 12A3 10 9A4 9 3A5 14 10A6 14 15A7 15 13A8 18 16A9 19 14A10 7 6

B1 10 7B2 11 14B3 15 15B4 7 7B5 4 2B6 15 11B7 7 9B8 17 14B9 8 10B10 7 7


Dell, G. S. (1986). A spreading activation theory of retrieval in sentenceproduction. Psychological Review, 93, 283–321.

Dell, G. S., Schwarz, M. F., Martin, N., SaVran, E. M., & Gagnon, D. A.(1997). Lexical access in aphasic and non-aphasic speakers. Psychologi-cal Review, 104, 801–838.

Ellis, A. W., & Young, A. W. (1988). Human cognitive neuropsychology.London: Lawrence Erlbaum Associates.

Ferretti, T. R., Mc Rae, K., & Hatherell, A. (2001). Integrating verbs, situ-ating schemas, and thematic role concepts. Journal of Memory andLanguage, 44, 516–547.

Foygel, D., & Dell, G. S. (2000). Models of impaired lexical access inspeech production. Journal of Memory and Language, 43, 182–216.

Graetz, P., De Bleser, R., & Willmes, K. (1992). Akense afasie test. Lisse:Swets & Zeitlinger.

JackendoV, R. (1990). Semantic structures. Cambridge: MIT Press.Jones, E. V. (1984). Word order processing in aphasia: EVect of verb

semantics. In Advances in Neuropsychology. New York: Raven Press.Jonkers, R. (1998). Comprehension and Production of Verbs in aphasic

Speakers. Groningen, GRODIL (Groningen Dissertations in Linguis-tics) 25.

Jonkers, R., & Bastiaanse, R. (1998). How selective are selective word classdeWcits? Two case studies of action and object naming. Aphasiology, 12,245–256.

Jonkers, R., & Bastiaanse, R. (2006). The inXuence of instrumentality andname relation to a noun on verb comprehension in Dutch aphasicspeakers. Aphasiology, 20, 3–16.

Jonkers, R. (2000). Verb Wnding problems in Broca’s aphasic speakers: TheeVect of transitivity. In R. Bastiaanse & Y. Grodzinsky (Eds.), Gram-matical disorders in aphasia: A neurolinguistic perspective. London:Whurr Publishers.

Kambanaros, M., & van Steenbrugge, W. (2006). Noun and verb process-ing in Greek-English bilingual individuals with anomic aphasia and theeVect of instrumentality and verb–noun name relation. Brain and Lan-guage, 97, 162–177.

Kay, J., & Ellis, A. W. (1987). A cognitive neuropsychological case study ofanomia. Implications for psychological models of word retrieval.Brain, 110, 613–629.

Kemmerer, D., & Tranel, D. (2000). Verb retrieval in brain-damaged sub-jects: 1. Analysis of stimulus, lexical, and conceptual factors. Brain andLanguage, 73, 347–392.

Kim, M., & Thompson, C. K. (2000). Patterns of comprehension and pro-duction of nouns and verbs in agrammatism: implications for lexicalorganization. Brain and Language, 74, 1–25.

Kiss, K. (2000). EVect of verb complexity on agrammatic aphasic speakers’sentence production. In R. Bastiaanse & Y. Grodzinsky (Eds.), Gram-matical disorders in aphasia: A neurolinguistic perspective. London:Whurr Publishers.

Kohn, S. E., Lorch, M. P., & Pearson, D. M. (1989). Verb Wnding in apha-sia. Cortex, 25, 57–69.

Levelt, W. (1989). Speaking. Cambridge: MIT Press.Levelt, W. J. M., Roelofs, A., & Meyer, A. S. (1999). A theory of lexical

access in speech production. Behavioural and Brain Sciences, 22, 1–75.van Loon-Vervoorn, A. (1985). Voorstelbaarheidswaarden van Nederlandse

woorden : 4600 substantieven, 1000 verba en 500 adjectieven. Lisse:Swets en Zeitlinger.

Luzzatti, C., Raggi, R., Zonca, G., Pistarini, C., Contardi, A., & Pinna, G.-D. (2001). On the nature of selective impairment of verb and nounretrieval. Cortex, 37, 724–726.

Luzzatti, C., Raggi, R., Zonca, G., Pistarini, C., Contardi, A., & Pinna, G.-D. (2002). Verb–noun double dissociation in aphasic lexicalimpairments: the role of word frequency and imageability. Brain andLanguage, 81, 432–444.

Martin, N., & SaVran, E. M. (2002). The relationship of input and outputphonological processing: an evaluation of models and evidence to sup-port them. Aphasiology, 16, 107–150.

Miceli, G., Silveri, M. C., Nocenti, U., & Caramazza, A. (1988). Patterns ofdissociation in comprehension and production of nouns and verbs.Aphasiology, 2, 351–358.



ARTICLE IN PRESS

Miceli, G., Silveri, M. C., Villa, G., & Caramazza, A. (1984). On the basis of theagrammatics’ diYculty in producing main verbs. Cortex, 20, 207–220.

Shapiro, K., & Caramazza, A. (2001). Sometimes a noun is just a noun:comments on Bird, Howard, and Franklin (2000). Brain and Language,76, 202–212.

Shapiro, K., & Caramazza, A. (2003). Looming a loom: evidence for inde-pendent access to grammatical and phonological properties in verbretrieval. Journal of Neurolinguistics, 16, 85–111.

Thompson, C. K. (2003). Unaccusative verb production in agrammaticaphasia: the argument structure complexity hypothesis. Journal of Neu-rolinguistics, 16, 151–167.


Thompson, C. K., Lange, K. L., Schneider, S. L., & Shapiro, L. P. (1997).Agrammatic and non-brain-damaged subjects’ verb and verb argu-ment structure production. Aphasiology, 11, 473–490.

Thompson, C.K., Shapiro, L.P., Li, L., & Schendel, L. (1994). Analysis ofverbs and verb argument structure: a method for quantiWcation ofagrammatic language production. In P. Lemme (Ed.) Clinical Aphasi-ology, vol. 23. Austin, Pro-Ed.

Williams, S. E., & Canter, G. J. (1987). Action-naming performance in foursyndromes of aphasia. Brain and Language, 32, 124–136.

Zingeser, L. B., & Berndt, R. S. (1990). Retrieval of nouns and verbs inagrammatism and anomia. Brain and Language, 39, 14–32.


ARTICLE IN PRESS Action naming in anomic aphasic speakers: EVects of instrumentality and name...

Documents

Transcript of ARTICLE IN PRESS Action naming in anomic aphasic speakers: EVects of instrumentality and name...