No abstract

BackgroundMany bats vocalizing through their nose carry a prominent noseleaf that is involved in shaping the emission beam of these animals. To our knowledge, the exact role of these appendages has not been thoroughly investigated as for no single species both the hearing and the emission spatial sensitivities have been obtained. In this paper, we set out to evaluate the complete spatial sensitivity of two species of New World leaf-nosed bats: Micronycteris microtis and Phyllostomus discolor. From an ecological point of view, these species are interesting as they belong to the same family (Phyllostomidae) and their noseleaves are morphologically similar. They differ vastly in the niche they occupy. Comparing these species allows us to relate differences in function of the noseleaf to the ecological background of bat species.Methodology/Principal FindingsWe simulate the spatial sensitivity of both the hearing and the emission subsystems of two species, M. microtis and P. discolor. This technique allows us to evaluate the respective roles played by the noseleaf in the echolocation system of these species. We find that the noseleaf of M. microtis focuses the radiated energy better and yields better control over the emission beam.ConclusionsFrom the evidence presented we conclude that the noseleaves serve quantitatively different functions for different bats. The main function of the noseleaf is to serve as an energy focusing mechanism that increases the difference between the reflected energy from objects in the focal area and objects in the periphery. However, despite the gross morphological similarities between the noseleaves of the two Phyllostomid species they focus the energy to a different extent, a capability that can be linked to the different ecological niches occupied by the two species.

850Probably the most controlled variable in the literature on word recognition and production is word frequency. Participants respond faster to high-frequency (HF) words than to low-frequency (LF) words in almost any lexical processing task, including lexical decision, reading aloud, semantic categorization, and picture naming (e.g., Carroll & White, 1973;Forster & Chambers, 1973;Whaley, 1978).Although there is little doubt that the word frequency effect (FE) reflects an important property of the organization of the mental lexicon, there is debate about the specific locus of this effect. Some models consider the FE a result of implicit learning. In this view, lexical representations are strengthened by repeated exposure-for example, by lowering a recognition threshold (e.g., the logogen model of Morton, 1970). HF words are recognized faster than LF words are, because their lexical representations reach and surpass the threshold faster. A similar mechanism is also responsible for FEs in localized and distributed connectionist models of visual word recognition (e.g., McClelland & Rumelhart, 1981). In contrast, the influential "rank hypothesis" proposed by Murray and Forster (2004) assumes that the lexicon is organized into frequency-ordered bins that are searched serially during visual word recognition. According to that account, HF words are recognized faster than LF words are, because the search process considers them earlier. Contrasting with the learning account, FEs should thus stay the same as long as the relative frequency does not change, even if the absolute number of exposures increases.Surprisingly, no study has ever explicitly focused on FEs in visual word recognition in a second language (L2). This offers a complex but interesting case for these accounts of the FE, because words in L2 are encountered less often than L1 words, even if these words have the same "objective" frequency-namely, the frequency determined from "monolingual" language corpora-and even though the relative frequency ranking within L2 is probably very similar to that in L1. In this view, it is a challenging open question whether the L2 FE is different from the L1 FE. Therefore, it is the aim of the present study to directly compare L1 and L2 FEs in word recognition by bilinguals, and to explore the resulting constraints for future modeling of (bilingual) word recognition.Interestingly, in the production domain, two studies have recently adopted the same approach. Gollan, Montoya, Cera, and Sandoval (2008) found that English-dominant Spanish-English bilinguals showed a larger FE in picture naming in the nondominant language than they did in the dominant language. Gollan et al. (2008) interpreted these results within a version of the rank model adopted for wordThe frequency effect in second-language visual word recognition WOUTER DUYCKGhent University, Ghent, Belgium DIETER VANDERELST University of Antwerp, Antwerp, Belgium AND TIMOTHY DESMET AND ROBERT J. HARTSUIKER Ghent University, Ghent, BelgiumA lexical decision experiment with ...

No abstract