Sentence and text processing
Friday, September 30th, 2011 [17:20 - 19:20]
PS_1.105 - Effect of narrative points of view on processing and accessing spatial text information
Fahrat, S. & Tapiero, I.
Université Lumière Lyon 2
Previous experiments showed that some conditions like: memorization of a map before reading a text (Zwaan, Radvansky, Hilliard, & Curiel, 1998) or explicit instructions that elicit the reader to focus on spatial information, (Zwaan & Van Oostendorp, 1994) seemed necessary for the reader to take into account spatial dimension. In line with these experiments, we investigated how narrative perspectives (Cohn, 1978) might influence the accessibility and the processing of spatial information while reading a text. We contrasted three types of narrative points of view (internal, external and omniscient) and measured the effect of inconsistent spatial information on reading times according to the three narrative perspectives. We observed longer reading times for inconsistent information in the internal perspective compared to the two other. Our results confirmed that different perspectives have an effect on accessibility of information. More specifically, that internal perspective allowed a more specific representation of spatial dimension.
PS_1.106 - Emotional valence effects during the comprehension of causal and adversative sentences
Morera, Y. 1 , León, J. A. 2 & de Vega, M. 1
1 Universidad de La Laguna
2 Universidad Autónoma de Madrid
Connectives are text devices that signal the relation between adjacent sentences. In a double task paradigm, participants listened to first clause sentences in which a causal or an adversative connective was provided and an emotional positive or negative connotation was varied: a) Because/Although he was a very talent artist… b) Because/Although he was not a very talent artist… Then, an emotional icon was presented in the centre of the screen, which could be either a “happy” or a “sad” icon. After that, two words appeared on the screen and participants had to choice the most congruent with the sentence meaning (e.g., 1) He triumphed or 2) He failed). When there was no delay between the sentence and the emotional icon (Exp.1), a match effect with respect to the first clause meaning occurred (faster responses in a) sentences + happy-icon and in b) + sad-icon, compare to mismatched conditions: a) + sad-icon and b) + happy-icon). However, when the delay was 1000 ms (Exp.2), responses were faster when the emotional icon matched the second clause meaning (1) + happy-icon and 2) + sad-icon). The results are related with the role of connectives in activating emotional inferences.
PS_1.107 - Recursion in grammar and in the parser
Lobina, D. & Garcia-Albea, J.
Departament de Psicologia/CRAMC, Universitat Rovira i Virgili, Tarragona, Spain
Recursion is a property of the language faculty at various levels. First, it is at the core of its computational system. Secondly, it is a property of the generated structures, in the sense that all phrases appear to follow the same geometry: a [Specifier [Head-Complement(s)] configuration. Thus, a sentence is a combination of embedded phrases of this type and we here investigate if this results in a recursive application of the parser. As a metric, the memory load of assembling phrases in mono-clausal Subject-Verb-Object sentences was assessed by employing the click-detection paradigm -a technique sensitive to cognitive loads within and between clauses. 60 pairs of Spanish sentences, containing a longer phrase in either the Subject or Object position were employed, and three click positions (controlled for length) were determined. Results show that reaction times (RTs) are slower at the beginning of sentences, but there is a robust linear decrease in RTs between positions. An ANOVA analysis determines that both the sentence type and the click position factors are significant, but there is no interaction effect. Nevertheless, all within-sentence-type comparisons were significant, while only the second position -the S-HC frontier- proved to be significant across sentence-type, suggesting a recursive process.
PS_1.108 - Processing grammatical gender of role nouns: Further evidence from eye-movements
Irmen, L. & Schumann, E.
Department of Psychology, University of Heidelberg
Two eye-tracking experiments investigated the effects of masculine vs. feminine grammatical gender of role nouns on establishing co-reference relations in German. Participants were presented with sentences of the basic structure My <kinship term> is a <role noun> <prepositional phrase> (e.g. My brother is a singer in a band). Role nouns were either masculine or feminine. Kinship terms were lexically male or female and thereby specified referent gender. Experiment 1 tested a fully crossed design including items with an incorrect combination of lexically male kinship term and feminine role noun (brother - [female] singer). Experiment 2 tested only correct items to control for possible effects of incorrect materials in Experiment 1. In early stages of processing, feminine role nouns, but not masculine ones, were fixated longer in case of a mismatch between grammatical and referential gender. In later stages of sentence wrap-up, sentences with masculine role nouns were fixated longer than those with feminine ones, irrespective of referential gender. Both experiments indicate that, for feminine role nouns, cues to referent gender are integrated immediately, whereas a late integration obtains for masculine forms. Our findings are discussed with regard to the different morphological and referential features of masculine and feminine gender in German.