The neural and behavioral basis of statistical learning of target facilitation in visual search

Yu, Hao (2024): The neural and behavioral basis of statistical learning of target facilitation in visual search. Dissertation, LMU München: Fakultät für Psychologie und Pädagogik

Abstract

Static statistical regularities in the placement of targets and salient distractors within the search display can be learned and used to optimize attentional guidance. Whether statistical learning also extends to dynamic regularities governing the placement of targets and distractors on successive trials remains controversial. Here, we applied the same dynamic cross-trial regularity – one-step shift of the critical item in clock-/counterclockwise direction – to either the target or a distractor. In two experiments, we found and replicated robust learning of the predicted target location: processing of the target at this location was facilitated, compared to random target placement. But we found little evidence of proactive suppression of the predictable distractor location – even in a close replication of Wang et al. (2021), who had reported a dynamic distractor suppression effect. Facilitation of the predictable target location was associated with explicit awareness of the dynamic regularity, whereas participants showed no awareness of the distractor regularity. We propose that this asymmetry arises because, owing to the target’s central role in the task set, its location is explicitly encoded in working memory, enabling the learning of dynamic regularities. In contrast, the distractor is not explicitly encoded; so, statistical learning of dynamic distractor locations is more precarious. People can learn and use both static and dynamic (cross-trial) regularities in the positioning of target items during parallel, ‘pop-out’ visual search. Static target-location learning also works in serial search, however, acquiring dynamic regularities is hindered by the demands of item-by-item scanning. Also, questions have been raised regarding whether explicit awareness is necessary for using dynamic regularities to optimize performance. The present study re-examined if dynamic regularities can be learned in serial search when regular shifts of the target location occur frequently, and if such learning correlates with awareness of the dynamic rule. We adopted the same regularity used by Yu et al. (2023) to demonstrate dynamic learning in parallel search: a cross-trial shift of the target location in a (counter-)clockwise direction within a circular array in 80% of the trials, compared to irregular shifts in the opposite direction (10%) or some other random direction (10%). The results showed that about 70% of participants learned the dynamic regularity, with performance gains correlating with awareness: the more accurately they estimated the likelihood of the target shifting in the frequent direction, the greater their gains. Importantly, part of the gains accrued already early during the search: a large proportion of the very first and short-latency eye movements were directed to the predicted location, regardless of the target appeared there. We discuss whether this rule-driven behavior is causally mediated by conscious control. Target facilitation can be achieved via spatial statistical learning of the target. Although many behavioral studies have found participants' performance was improved when targets occur at frequent locations relative to infrequent, the electrophysiological correlates of statistically learned target enhancement, however, have not been well studied. Moreover, a recent behavioral study (Turatto and Valsecchi 2022) has proved that the learned distractor suppression could cause long-lasting changes in the priority map, which affects the future computation of target salience at the same location to facilitate attention selection. It remains to be seen whether the learned target enhancement can also cause long-term changes in the priority map. Thus in the current study, the EEG data were collected while participants searched for a target in a serial search task with an unbalanced distribution of targets. N2pc (lateralized event-related potentials) and lateralized alpha (8–12 Hz) power were used to track the temporal dynamics of effects. The experiment consisted of two phases: the probability manipulation of target location was implemented in the training phase; the probability manipulation of target location was removed in the testing phase, and the distractor was also introduced which shared a similar feature as the target. Behaviorally, we found robust learning of predicted target locations in the training phase: processing of the target at frequent locations was facilitated, compared to infrequent target placement. In the testing phase, we found that the long-term learning even persisted when the target in the training phase became the distractor. Electrophysiologically, the learned target enhancement is not associated with lateralized alpha power during the pre-stimulus period. Yet the targets at frequent locations induced larger N2pc amplitude and earlier N2pc latency in the training phase. Thus, we confirm that facilitation resulting from target spatial statistical learning can cause plastic changes on the priority map.

• Dokumententyp: Dissertationen (Dissertation, LMU München)
• Keywords: visual search, statistical learning, eyetracking, eeg, explicit awareness
• Themengebiete: 100 Philosophie und Psychologie ; 100 Philosophie und Psychologie > 150 Psychologie
• Fakultäten: Fakultät für Psychologie und Pädagogik
• Sprache der Hochschulschrift: Englisch
• Datum der mündlichen Prüfung: 15. Juli 2024
• 1. Bericht­erstatter:in: Shi, Zhuanghua