[PS-1.11] Satiating categorical meaning: Simulating the reversal of semantic priming into semantic satiation in a category matching task

GALMAR, B. 1 & CHEN, J. 1, 2

1 Institute of Education, National Cheng Kung University, Tainan, Taiwan
2 Institute of Cognitive Science, National Cheng Kung University, Tainan, Taiwan

Verbal satiation is commonly characterized by a temporary loss of the meaning of a word following its prolonged viewing (e.g., 3 min.) or its active repetition (oral or written, e.g., for 30 times). In a category-matching task experiment, we asked participants to judge if two exemplars (a prime and a target) belonged to a same category in blocks of 40 trials each. Within a block, 20 exemplars of one category made half of the primes (repeated categorical meaning trials) while only two exemplars of each of the remaining 10 categories served as the second half of the primes (baseline categorical meaning trials). We found that repetitive processing of exemplars of a same category without repetition of the form of the category label induced verbal satiation. The locus of such a verbal satiation is semantic as only meaning of the category was repeatedly processed. The observed time course of verbal satiation conforms to the classical habituation model (Rankin, et al., 2009): sensitization (facilitatory effects of semantic priming) before habituation (refractory effects of semantic satiation). We simulated the time course of semantic satiation in our category matching task. The categorical meaning to be satiated was modeled as a single neuron unit with an excitatory, recurrent connection. The categorical meaning neuron unit responded to an input signal modeling a sequence of pairs of words (exemplars from the repeated category and other categories). A rule implementing a simple novelty detector controlled the increment or the decrement of the input weight to the input signal over a time-steps window corresponding to the verbal working memory span. Simulated response times were derived from the response of the neuron unit. We finally proposed to bridge our simple cognitive model to previous neurocomputational simulations of the relationships among repetition suppression, neural synchronization, and behavioral priming (Gotts, 2003).