The Semantic Processing of Motion Verbs

This experiment investigated whether aspectually flexible motion verbs like "stroll" and "run" are lexically specified as atelic, or whether they are underspecified.  Participants  Forty-eight students from the University of Tübingen participated in the study and received a monetary reimbursement. They were all native speakers of German and were naïve to the purpose of the study.  Method  We implemented a self-paced reading study combined with an acceptability judgment task. The 2x2-design was composed of the two within-factors directional phrase (telic vs. ambiguous) and temporal adverbial (completive vs. durative). The resulting four conditions are abbreviated as telcompl, teldur, alpha-compl, alpha-dur.  Material  Materials consisted of thirty-two experimental sentences that were realized in all four conditions. Experimental sentences were split into six regions. The regions of interest were the noun at the end of the directional phrase (region 4), the spill-over region 5 'and namely', and the temporal adverbial (region 6). Sixteen motion verbs were used: fahren 'drive', fliegen 'fly', gehen 'go', klettern 'climb', kriechen 'crawl', laufen 'walk', marschieren 'march', radeln 'cycle', rasen 'speed', reiten 'ride (a horse)', rennen 'run', schleichen 'sneak', schlendern 'amble', segeln 'sail', spazieren 'stroll', wandern 'hike'. Each verb was used in two experimental sentences. The experimental sentences were combined with sixty-four filler sentences, which were designed to resemble the experimental sentences.   Procedure  The experiment was programmed using the E-Prime 2.0 Professional software (Psychology Software Tools, Inc.). Participants were tested individually at a PC. The main experiment was preceded by four practice trials. Sentences were presented to participants with a moving window technique. Presentation of regions was self-paced. After the first press of the space bar, the trial began with a display showing the whole stimulus sentence completely masked, i.e., all characters including spaces were substituted with dashes. With the second press of the space bar, region 1 was shown by replacing the dashes with the corresponding characters. Any further press of the space bar led to the remasking of the current region and a concurrent demasking of the subsequent region. After the presentation of a sentence was completed, a final press of the space bar revealed a prompt for an acceptability judgement on a five-point scale from 1 (sehr schlecht 'very bad') to 5 (sehr gut 'very good'). In order to encourage participants to read the sentences carefully, one third of the trials ended with a yes-no comprehension question, which was answered by means of two designated keys. Participants were instructed to read through the sentences at a natural pace and to read closely enough to be able to answer the questions. Experimental sessions lasted less than thirty minutes.