Looking into cognate recognition in intercomprehension: an eye tracking study

The present study aims to investigate cognate recognition in the context of intercomprehension, a communicative mode in which the proximity of the respective languages allows interlocutors to understand each other despite their speaking their own L1s (European Commission 2012). The experiment relies on an eye-tracking protocol, a technique that so far has been hardly applied to the study of intercomprehension. Indeed, while most existing research (e.g. Berthele 2011; Möller & Zeevaert 2015; Saturno 2023) is mainly concerned with the final output of the phenomenon, the present study rather aims to describe the process, with the goal to investigate whether cognate recognition in the context of intercomprehension activates a sequential or holistic reading mechanism (Grainger 2022). In the former scenario, target items are compared to similar items from known languages in search of correspondences and divergences. One would thus expect longer fixation times on those letters of the target item that diverge from similar items belonging to other languages; at the same time, different types of interlingual correspondences are expected to lead to different recognition rates. In the holistic reading scenario, on the other hand, fixation times should equally distribute across all letters of the stimulus item, with no observable difference attributable to the specific type of interlingual correspondences. An artificial language was developed based on the assumption that the recognition of a word is significantly hindered by modifications concerning consonants, but not vowels (Schmandt et al. 2022; Berrebi et al. 2023). The stimuli were designed on the basis of 12 Italian words, selected from the PhonItalia corpus (Goslin et al. 2014) in such a way as to balance overall frequency and phonological neighborhood size and frequency. These words were then modified with respect to one, two or three articulatory traits (place of articulation, manner of articulation, voicing) of the initial consonant, e.g. It. domanda > *tomanda, *momanda etc. 18 distractors were also devised in which no systematic relation to the Italian counterpart could be identified. The planned experiment will be carried out as follows. The stimuli (target items and distractors) will be presented to 20 L1 Italian participants in a sufficiently large font size for the eye-tracker (Tobii Pro Nano - 60 Hz) to detect fixations on individual letters. Participants will be asked to translate the unknown item into Italian. Fixation times (first fixation duration, first pass reading time, total reading time, total fixation count) will be computed for three regions of interest, i.e. a) the modified letter, b) the whole word, and c) the whole word with the exclusion of the first letter. Data collection and analysis will be conducted with iMotions 9.3.0.5 software. The poster discusses the rationale of the experiment and presents its preliminary results. It is expected that longer fixation times will correspond to initial consonants modified with respect to a greater number of articulatory parameters. Should it be confirmed, this observation would bring support to the hypothesis that cognate recognition in intercomprehension relies on sequential reading.