Effects of distractor interference cannot be mitigated by predictive cues

The contents of visual working memory (VWM) guide daily behaviours. However, VWM is severely capacity limited, making it critical to protect its contents from distracting information. VWM contents are subject to interference in numerous ways, one of which is a bias referred to as “attractive pull”, wherein reports of remembered features are biased toward distractor features (Huang & Sekuler, 2010; Rademaker, Bloem, De Weerd, & Sack, 2015). Here we investigated if we can protect VWM contents against such interference by making distractor presence predictable. Participants remembered the orientation of a target Gabor across a short delay, and we sometimes presented a distractor Gabor during this delay. To assess the effects of attractive pull, we manipulated the orientation difference between the target and distractor Gabors, with larger differences expected to increase the “pull” of the distractor. We also manipulated target encoding time by varying target/distractor stimulus onset asynchrony (SOA), with increased “pull” expected for shorter encoding times. To assess for control over attractive pull, some blocks included a cue at the beginning of the trial to indicate whether or not a distractor would be presented on that trial (predictive blocks) while other blocks did not provide any information (non-predictive blocks). As expected, attractive pull increased with decreasing SOAs, and with larger differences between the target and distractor orientations. However, the predictive cue was not able to mitigate these effects. We suggest that even if presented with a predictive cue prior to target feature encoding, participants cannot effectively prepare to protect VWM contents.