Distinct, but top-down modulable color and positional priming mechanisms in visual pop-out search

This alternative explanation may, in principle, also hold for the pattern of positional priming effects found in Experiments 1–3: positional priming effects may have been reduced in Experiments 1 (baseline) and 3 (predictable target color) relative to Experiment 2 (predictable target position) because, in the latter, the probability of an intervening trial (between the critical trials N − j and N) containing a same-position target was higher than the probability of a different-position target [Experiments 1 and 3: p(same-position) = 0.04, p(different position) = 0.96; Experiment 2: p(same-position) = 0.40, p(different position) = 0.60; values for Experiment 2 collapsed across trials N − 1 through N − 5]. Unfortunately, an analogous analysis to that of color priming effects (with pure sequences of same-color trials) could not be conducted for the position priming effects, because the number of pure position repetition sequences was too low in Experiments 1 and 3 to permit statistical examination. However, previous findings suggest that the relative proportions of same- and different-position trials do not affect the temporal extension of positional priming (Maljkovic and Nakayama 1996). Specifically, Maljkovic and Nakayama (1996; Experiment 2) showed that positional priming extended back 5–8 trials – importantly, independently of whether there were 6 or 12 possible target locations. That is, doubling the probability with which a given target position is repeated did not change the temporal extension of positional priming.

Of course, this does not rule out other accounts of inter-trial priming effects in terms of the retrieval of task-relevant episodic memories attributing repetition benefits at stages following focal attentional selection (i.e., stimulus-response translation; e.g., Logan, 1990, 2002; Neill, 1997; Waszak, Hommel, & Allport, 2003).