In recent years, the cross-linguistic plasticity of perceptual boundaries has drawn a lot of attention. It is often assumed that natural phonetic boundaries in the pre-linguistic child are replaced by languagespecific ones. For consonantal perception, the most prominent empirical domains encompass research on voicing and stop place of articulation perception. Specifically, for place perception, a vast amount of studies on the transitional cues has evolved in the tradition of Delattre et al. (1955, for a review see Sussman et al., 1998). In our view, it is noteworthy that this research has almost exclusively focused on the three oral stop place categories velar, alveolar and bilabial, leaving the palatal place of articulation aside.

The focus on this particular issue emerged as the result of our previous research (Carré et al., 2002, Serniclaes et al., 2003): According to these studies, perception of consonant place of articulation is organized around a central reference given by the neutral vowel. In the neutral vocalic context, place boundaries correspond to flat F2-F3 transitions in stimuli where the latter covary. When separately modified, the perceptual F2-F3 space is divided into four distinct regions. In Serniclaes et al. (2003), we have investigated the partitioning of this space into the three phonologically relevant stop place categories present in the French language. Results indicated that even when natural boundaries are located inside phonological categories, they can still affect consonant discrimination. Furthermore, nonphonological categories remained perceptible for some adult subjects: A subset of the participants exhibited above threshold (50 %) identification responses for a category absent in their language. The purpose of the present work is to see whether it is possible to relate this additional category to the palatal stop place of articulation of a four categorylanguage. To achieve this, we will conduct a replication of the identification and discrimination experiments reported in Serniclaes et al.(2003). Positive results would substantiate the palatal place of articulation to be grounded in a natural, phonetically motivated category. Negative results, on the contrary, would indicate that the palatal stop does not rely on the information present in the transitions of the second and third formants: The perception of the palatal place of articulation would then not mainly rely on F2-F3 formant transitions. Rather, the burst characteristics, or still other acoustic cues, would play a major role in the identification of palatal consonants.