Abstract: Introduction The mental models of a text are representations of its situations -- what it is about rather than the text itself. Franklin & Tversky (1990) investigated when people constructed the spatial situation and how they discriminated objects on different directions. the direction effect was found. The researcher used the “Egocentric Framework” to explain it. It is now commonly agreed that the working memory plays an important role in the cognitive process. Therefore, we investigated, in the present study, how the working memory span and direction affected the process of spatial situation model constructing and updating.
Method Participants: The participants were 16 low-span and 16 high-span adults. Materials: 6 texts. There were 5 sentences in each text that described a spatial model. Design: The first independent variable, span -- high or low, was manipulated between the subjects. The second independent variable, time pattern -- unlimited-time or limited-time, was manipulated within the subjects. The third independent variable, direction -- left-right or front-back, was manipulated within the subjects. The number of the correct-or-wrong-judgment sentences was the dependent variable. Procedure: Experimental texts were presented in two time patterns – the limited-time pattern and the unlimited pattern. The former was defined as 4000ms and the latter meant the subjects could study the text as long as they wanted. Once they finished the text study, they should judge whether each sentence about the text was correct or not by keying in “J” for “correct” or “F” for “wrong”.
Results Repeated-measure ANOVAs were conducted for Experiments 1 and 2. In the analysis of Experiment 1 with the performance of the spatial situation model constructing task used as the dependent variable, there were main effects of span, F(1,30)= 4.800, p< 0.05, direction, F(1,30)=61.294, p < 0.001, and time pattern, F(1,30)=29.925, p < 0.001, and an interaction between direction and time pattern, F(1,30)=23.508, p< 0.001. There was no obvious difference between the two time patterns on the front-back judgments, but the performance of the unlimited-time task was better than that of the limited-time task in the process of left-right judgments. In the analysis of Experiment 2 with the performance of the spatial situation model updating task used as the dependent variable, there were a main effect of time pattern, F(1,30)=18.274, p < 0.001 and an interaction between direction and time pattern, F(1,30)=7.226, p< 0.05),. There was no significant difference between the two time patterns on the front-back judgments, but the performance of the unlimited-time task was better than that of the limited-time task in the process of left-right judgments. No other main effects or interactions were significant.
Conclusions On the basis of the results, we draw conclusions as follows. First, the visuospatial working memory span affects the process of the spatial model constructing, but does not affect the process of the spatial situation model updating. Second, the time pattern affects the process of both the model constructing and updating. Finally, the direction only affects the process of the spatial situation model constructing. During the processes of the spatial situation model constructing and updating, the judgment of the front-back position is not affected by the time-limited pattern, but nevertheless the judgment of the right-left position is affected by the time-limited pattern. The effect of the front-back position is stable and superior

Key words: spatial situation model, direction effect, visuospatial working memory span, time pattern