Naming abilities in spontaneous speech in Parkinson and Alzheimer’s disease R. Pignatti a , F. Ceriani a , L. Bertella a , I. Mori a , C. Semenza a,b,* a Istituto Auxologico Italiano, Verbania, Italy b Department of Psychology, University of Trieste, Italy Accepted 6 July 2006 Introduction Different patterns of impairment in naming abilities are reported in lit- erature for Parkinson’s disease (PD) and Alzheimer’s disease (DAT). While PD usually experience harder difficulties in verb rather than in object name retrieval, AD may rather show the opposite pattern (Daniele, Giustolisi, Silveri, Colosimo, & Gainotti, 1994; Bertella et al., 2002). Verb- naming difficulties in PD are usually related to functional damage of a cor- tical/subcortical system including frontal structures related to Broca’s area; in contrast, object-naming difficulties in DAT are thought to result mainly from deterioration to their temporal lobes. Although it has been recently recognised that data from spontaneous connected speech may importantly distinguish degenerative pathologies (Chapman et al., 2002) no information is available thus far on the spontaneous production of nouns and verbs for both PD and DAT. Participants and methods Thirty-six participants (18 PD and 18 DAT) with a mild level of cog- nitive impairment (MMSE mean score: 24) were recruited for this study. The two groups were age and education matched and fully comparable in their cognitive level, as they demonstrated similar results on several neu- ropsychological screening tests (MMSE, Attentional Matrices, Phonemic and Semantic Fluency, Raven’s Colored Progressive Matrices). They were administered a picture naming test (Miceli, Laudanna, & Burani, 1990), requiring naming of both objects and actions. They were then administered the ‘‘spontaneous speech’’ section of aachener aphasie test (AAT). Every occurrence of names, verbs and other categories of words utilized in the patients’ speech during the first 300 words, was counted. As suggested by Semenza Panzeri, and Re (1989) two types of count were performed: a ‘‘token’’ count, including all occurrences, and a ‘‘type’’ count, where repeated items were counted only once. A type count rather than a token count is held to be a better indication of the richness of the speaker’s lexicon. Results Table 1 reports the main results for the production of nouns and verbs in the two experimental conditions as well as basic neuropsycho- logical assessment and demographic data. ANOVAS were used for sta- tistical treatment. No significant difference was found between PD and DAT considering the outcome of the naming test separately for objects and actions. However, when the object/action rate was considered, a difference emerged in the expected direction: in fact PD displayed an object/verb ratio significantly higher than DAT. The analysis of sponta- neous speech production showed that the two pathological groups did not differ from normal controls in the occurrence of nouns and verbs over all parts of speech. However DAT produced significantly less verbs than PD in the type count, no other comparison yielding significant differences. Discussion This investigation compared the production of nouns and verbs in pic- ture naming and spontaneous speech in PD and DAT. Importantly the two groups were tested at an early stage of the disease and at a comparable degree of cognitive deterioration. Picture naming confirms previous find- ings insofar it shows that, relative to DAT, verbs are less easily produced in PD. However this seems to be true only for an explicit task, i.e. con- strained production. Data from spontaneous speech show instead that, doi:10.1016/j.bandl.2006.06.071 * Corresponding author. E-mail address: semenza@univ.trieste.it (C. Semenza). www.elsevier.com/locate/b&l Brain and Language 99 (2006) 8–219