What “learned distractor rejection” actually means in these experiments
In a typical experiment, participants search for a target shape while a brightly colored distractor appears on every trial. The distractor’s color stays the same within a block of trials but changes at the start of each new block. Over time, people learn to ignore that specific color, so it captures their attention less. This is called learned distractor rejection [1][4].
The learning is not slow. When researchers looked at very small windows of trials, they found that efficient rejection emerged after only two or three encounters with the distracting color [4]. An earlier study using a finer-grained analysis observed the initial capture and following attenuation within the first eight trials of a block [1]. In other words, the system adapts almost immediately once it has learned that a particular color is consistently irrelevant.
A new cue can wipe out the learned rejection effect
The clearest evidence about generalization comes from an experiment that first trained participants to ignore one color, then added a trial-by-trial cue. On neutral-cue trials, the learned rejection effect persisted: people still avoided the trained distractor color [2]. But when the cue explicitly marked the learned distractor color as something to hold in mind, the learned rejection effect was eliminated—participants no longer avoided that color [2].
This suggests that learned rejection is not an isolated, automatic filter that operates regardless of current goals. Instead, guidance from long-term memory (the learned distractor color) and guidance from visual working memory (the current cue) combine to determine where attention goes [2]. When the two conflict, the new cue can override the old learning. That is a form of context sensitivity: the learned rejection does not simply transfer unchanged when the task demands change.
Fast re-learning is not the same as generalizing to new stimuli
Several studies show that once someone has learned to reject one distractor color, they re-learn rejection of a new color very rapidly in the next block [1][4]. This might look like generalization, but it is better described as fast re-learning within the same task and context. The participant is not applying the old filter to a new color; they are quickly building a new filter for the new color.
The distinction matters. If learned rejection truly generalized to new stimuli, a person trained to ignore red should immediately ignore blue without any new learning. The experiments do not show that. Instead, they show that the learning process itself is efficient and can restart almost instantly when the distractor color changes [1][4]. One paper explicitly notes that it did not address whether learned mechanisms generalize to entirely new stimuli or contexts [1].
What would count as real generalization—and why we don’t yet know
To test generalization properly, an experiment would need to train rejection in one context (say, a particular task or display type) and then measure whether rejection appears in a genuinely different context without new learning. None of the studies described here did that. One study examined learned rejection alongside strong target guidance and found that the two can operate together [3], and another identified a proactive oculomotor avoidance mechanism that avoids distractors whenever possible [5]. But neither tested transfer to new stimuli or contexts.
So the honest answer is that learned distractor rejection is robust, rapid, and sensitive to current cues, but the evidence for broad generalization is indirect at best. The one study that manipulated a new cue found that the learned effect could be eliminated [2], which argues against a simple, context-free transfer. Until an experiment directly trains rejection in one setting and tests it in another, the safest conclusion is that learned rejection is powerful within the learning context but should not be assumed to generalize automatically.
Sources used in this answer
Determinants of Learned Distractor Rejection
Re-learning of distractor rejection after a color change can occur extremely rapidly, with initial capture and attenuation observed within the first eight trials of a block, but the study did not test generalization to entirely new stimuli or contexts.
Learned Distractor Rejection Falls Prey to the Attentional White Bear
A learned distractor rejection effect persisted on neutral-cue trials but was eliminated when a trial-by-trial negative cue specified the learned distractor color, indicating that guidance from visual working memory and long-term memory combine to determine attentional priority.
Learned distractor rejection in the face of strong target guidance.
Learned distractor rejection can operate alongside strong target guidance in visual search, but the study did not address whether learned rejection generalizes to new stimuli or contexts.
Learned distractor rejection: Robust but surprisingly rapid
Using finer-grained analyses, learned distractor rejection emerged after only two or three encounters with the distracting item, much faster than previously assumed, but the study did not test generalization to new stimuli or contexts.
Testing the underlying processes leading to learned distractor rejection: Learned oculomotor avoidance
Learned distractor rejection is supported by a proactive attentional control mechanism termed learned oculomotor avoidance, with fewer distractor fixations and shorter dwell times, but the study did not address generalization to new stimuli or contexts.