ASA 130th Meeting - St. Louis, MO - 1995 Nov 27 .. Dec 01

5aSC20. Closed-set response formats: An evaluation within the framework of the neighborhood activation model.

Rebecca S. Prohofsky

Mitchell S. Sommers

Dept. of Psych., Washington Univ., St. Louis, MO 63130

It has been suggested that open- and closed-set response formats engage distinct perceptual mechanisms during spoken word recognition. One finding consistent with this hypothesis is that lexical difficulty (neighborhood density and neighborhood frequency) reduce identification performance in open- but not closed-set formats [Kirk et al., Ann. Otol. Rhinol. Laryngol. (in press)]. An alternative explanation for the differential effects of response format on lexical difficulty in this earlier study, however, is that the response alternatives were not systematically selected to maximize confusability with the target items. Therefore, the present investigation compared the effects of lexical difficulty in open- and closed-set response formats when the response alternatives in the closed-set condition were the five words most confusable with the target item. Preliminary results indicate that, relative to the open-set condition, when the difficulty of alternatives is controlled, identification within the closed-set format is improved for hard, but not for easy, words. It is hypothesized that the improved word recognition for hard words in closed-set formats is the result of reducing the viable candidate set. In contrast, the already small candidate set of easy words minimizes the benefits of reducing the number of viable candidate items.