Christiane Pons

The processing of interfixed German compounds

This postulate by Goethe (*1749), the first protagonist of a new discipline of morphology (albeit first only within biology), confronts us with the main problem of processing studies of morphology: Are morphological constructions processed as wholes or with regard to their parts or, if both, under which conditions? This question has been of central concern in the psycholinguistic literature on lexical processing over the past quarter century. The debate in this area was initiated by the provocative claim put forward by Taft and Forster (1975; 1976) that multimorphemic words are represented in the mental lexicon in terms of their constituents and that multimorphemic word recognition routinely involves a morphological decomposition procedure. Subsequent experimentation, however, has pointed to the view that neither this strong position nor the strong contrary position advocated by Butterworth (1983) accounts for the performance of language users across languages, task types, and stimulus categories (see McQueen and Cutler (1998) for a recent review). Even within individual categories of morphological construction, experimental results have led to a rather complex view of the role of morphology in lexical processing. Compound word processing, for example, has been shown to provide evidence for both whole word representation and constituent activation. In general, semantically transparent compounds show constituent activation, whereas semantically opaque compounds show greater evidence of whole word activation (Libben 1998; Sandra 1990; Zwitserlood 1994). Recent work by Libben, Derwing and de Almeida (1999) has also suggested that the processing of compounds may involve the creation of multiple representations that are simultaneously computed and evaluated. Libben et al. (1999) claim that the processing of compounds is not guided by a principle of parsing efficiency but rather by a mechanism that uncovers the maximum number of morphemes within a multimorphemic string.

Triangulating the Effects of Interfixation in the Processing of German Compounds

The German language shows a high degree of compounding. The process is very productive in the language and compounds may be formed from the combination of four or more roots. In addition, German compounding also shows the presence of interfixes between constituents in about 35% of all compounds. We report on a compound composition experiment that investigates the role of these interfixes in on-line compound processing. The study reveals that interfixation carries with it a processing cost in German and that this processing cost is elevated in cases in which the interfix attaches to a truncated form of the initial compound constituent. Moreover, we find that response times are increased when initial compound constituents show inconsistent interfixation patterns across the language. These results support the view that German compounds are represented in terms of their constituents and that interfixation choices are made on-line.

Integrating face-to-face language therapy with virtual reality applications for persons with aphasia

Under optimal conditions, providing language therapy to persons with aphasia (PWA) is an intensive and dynamic process which results in improved verbal communication and greater participation in everyday life. However, due to cuts in health care spending, PWAs are not receiving the necessary amount of language therapy to achieve significant gains in language abilities. Recent developments of computerized language programs and virtual reality applications are promising as they enable a PWA to attain an adequate dosage of language therapy. In this paper, challenges of integrating face-to-face therapy with computerized and virtual applications will be exemplified by the ELA® Virtual House.

Textpragmatic impairments of figure-ground distinction in right-brain damaged stroke patients compared with aphasics and healthy controls

Abstract Following previously published work on aphasic texts and unpublished pilot studies on textpragmatic impairments in right-brain damaged stroke patients (RBDs), we present an exploratory study on textpragmatic dimensions of several text classes produced by RBDs as compared with aphasics and healthy controls. In contrast to the main thrust of published research on (text)pragmatic impairments in RBDs, we assume for them the existence of a specific major impairment of figure-ground segregation, with consequences in defective textual organization, in inferencing, and in metalinguistic behavior. In this way we explain the RBDs’ deficits in producing (but not in comprehending) the essential textual core and emotional elements, and essential inferences. Difficulties in figure-ground distinction also result in RBDs’ insecurity about how well they solve discursive problems. These deficits appeared in the tests that we administered to 4 RBDs, 5 controls, 3 Broca and 2 Wernicke aphasics. Subjects had to carry out the following tasks: (a) oral production of a longer and a shorter picture story, (b) oral and written reproduction of a longer and a shorter oral narrative, (c) productions of telegraphic versions, of titles and of punch lines for (a) and (b). Differences in the RBDs’ performances in these distinct tasks are traced back to different cognitive demands on oral vs. written vs. reductive reproduction vs. picture story telling.

Applying the principles of experience-dependent neural plasticity: building up language abilities with ELA ® -Computerized language modules

In this paper, a computerized language therapy program that aims at supplying the required dose of practice for PWAs will be presented, namely the ELA®-Language Modules. The rationale and underlying principles for each linguistic level and the linguistic structure of the language tasks for the word, sentence and text level and for dialogues will be explained and how the compo-nents of the ELA®-Language Modules adhere to the principles of experience-dependent neural plasticity. First pilot applications of the ELA®-Language Modules with PWAs are discussed in terms of the principles of experience-dependent neural plasticity and usability.