Salience and set size functions in feature search

Michael Zehetleitner

Ludwig-Maximilians-Universität,

Munich Germany


In feature search, targets very dissimilar to distracters can be found efficiently independent of
the set size. The more similar the target is to the distracters, the worse search performance
becomes. This drop in performance has previously been attributed to the search becoming
more 'inefficient'. That is, the slopes of search functions are supposed to increase as target-
distracter difference decreases. Indeed, feature singletons very similar to distracters have been
demonstrated to produce inefficient search. However, search functions have not yet been
examined for parametric manipulations of target-distracter similarity.

As it turns out, the drop in search performance with increasing target-distracter similarity cannot
be attributed entirely to increased search slopes. Interestingly, when continuously increasing
target-distracter similarity, initially intercepts increase and search slopes take off only for targets
very similar to distracters. That is, feature contrast affects the two components of search
functions differently and at different time courses.
Additionally, observer-based factors and task demands modulate these differential effects of
feature contrast on slopes and intercepts of search functions.