University of Glasgow
Department of Psychology

 

Language dependency


The model does not distinguish among languages and, therefore, predicts the same response latencies irrespective of the language used. Carpenter and Just (1975) reported in short that the model makes correct predictions for Norwegian, which shows a different surface structure with the negative morpheme in a different position compared with English. To be precise, the constituent comparison model specifies, as Carpenter and Just state, a matching process between semantic representations and, thus, results in the same prediction as for English. This is not very astonishing as it was earlier pointed out by e.g. Blumenthal and Boakes (1967) that the deep structure is more influential in sentence comprehension than the surface structure. This was even confirmed by Carpenter and Just (1975; p. 53) themselves. To confirm this universal claim further languages have to be tested. No reports were found so far that examined this model with the same experimental design across languages. ...
 

   ...German data pattern are different from English ones. From some experiments it appears that German native speakers tested in German show different response latencies in some conditions (Glanzmann, 1981; Kron, 1987). There is no stringent evidence for this statement yet because too many factors varied between experimental designs but if one contrasts the data of the Clark and Chase (1972) study and Glanzmann’s study (1981) using German native speakers it becomes obvious that there are differences. Both experiments presented pictures first and used the predicates "above"/"über" and "below"/"unter", thus being at least similar enough in their structure to allow a comparison of the data pattern. From the above mentioned experiments (Clark & Chase, 1972; Glanzmann, 1981; Kron, 1987) it appears that German native speakers respond differently when the predicates "below"/"unter" are used but similar when the predicates "above"/"über" are used. For "above"/"über" predicates there may be a difference under the condition TN with German speakers reacting faster than English speakers. However, there are three possibilities what this can mean. It may represent a difference in processing, an artefact or the influence of a confounded variable, e.g. the use of different strategies (Roberts, Wood & Gilmore, 1994).

Previous page

Next page
 
 

Back to the main page.
 


(c) Dirk Köster; first posted January 06/01                                                                                                                           last modified January 08/01