[PS-1.15] Distinguishing among types of transitions between words

Marjanovic, K. , Trippa, M. , Baldan, E. & Treves, A.

International School for Advanced Studies (SISSA)

Understanding the associative neuronal mechanisms that underlie language processing remains a challenge. To address this issue, we have developed an ERP paradigm to distinguish among the brain signatures of different types of transitions between words, which have been hypothesised to reflect transitions between discrete brain states, and described with models of latching dynamics.

Design: Participants are presented with 12 rounds of a computer game. Each game includes 28 words (displayed in a 7x4 grid), which latch on one another through one of 7 different types of transition: letter-addition, -omission, -change, anagram, antonym, synonym, semantic relation. The goal of the game is to find the (only) correct sequence, and move quickly to the next round. A pilot behavioural experiment has been conducted in order to test the validity of the task.

Results: Preliminary results reveal that there is considerable variability in the mean reaction times among types of transitions, also among semantic and word-form subtypes taken separately. There is a factor of 2 between the mean time required for the fastest (letter-omission, ca. 4s) and for the slowest (anagram, ca. 8s) type of latching.

Discussion: We present a novel paradigm to disentangle distinct associative mechanisms that contribute to language processing.