Proactive control of feature gain is too weak to prevent the capture of attention by salient feature singletons
Speaker: Dr Giorgio Fuggetta
10th August 2023
Abstract: A debate exists whether it is possible to prevent attentional capture by salient singletons that automatically generate priority signals. Stimulus-driven capture, goal-dependent contingent capture and hybrid models, such as the signal suppression hypothesis of controlled attention capture, have been put forward to try solving the controversy on how salient-but-irrelevant features guide visual search through the phenomenon of attentional capture. With the current study, we tested the hypothesis whether increasing task demands with Visual Working Memory (VWM) load enhances top-down proactive feature-based control to avoid attention to be captured by salient singletons. Thirty-nine participants performed a modified delayed-match-to-sample task where a salient, yet task-irrelevant, singleton object was interleaved in-between a sample array of one to four abstract shapes to encode in VWM and the target abstract shape during the retention interval. The salient feature singleton was part of a distractor array with a high set size of 16 items and designed to induce the singleton detection mode. Using a block design, the singleton distractor object was either part of the current task set, being a same or different abstract shape as compared to incoming target, or not part of the task set, being a Chinese character. Also, these singletons were lateralised and could appear unpredictably either in the same or different visual field as compared to incoming lateralised target location. Participants were instructed to fixate a central fixation and ignore the distractor array while performing the delayed match-to-sample task. Attentional capture by the singleton objects was primarily evaluated in terms of event-related potential (ERP) components as markers of attentional selection and the extent to which distractor array is processed in memory. Electrophysiological results revealed that the magnitude of early lateralised Pd component, a neural index of proactive inhibition of feature gain, was constant across the VWM load conditions. Also, a gradual reduction of amplitude of P1, N1, anterior P2, N2pc and P3b components was observed with VWM capacity load, indicating top-down feature suppression of salient singletons. Crucially, the reduced magnitude of N2pc indicated a gradual depletion, but not elimination, of attentional capture towards the peripheral singletons with increased VWM load. Moreover, the anterior N2 showed an opposite enhanced amplitude at higher VWM load, suggesting an increased top-down cognitive control and detection of conflicting singleton features with the perceptual template of sample array stored in VWM. Lastly, the decreased P3b peak amplitude with VWM capacity load demonstrated that when VWM was strained, there was an enhanced top-down cognitive control with greater reactive suppression for to-be-ignored peripheral stimuli along with disengagement and hyperfocusing of spatial attention to the central fixation, the most parsimonious location waiting for the incoming unpredictable lateralised target to appear. The ERP results of this study shed new light into the attentional capture debate. In particular, the early Pd, P1, N1 and anterior P2 components showed that proactive inhibition of feature gain of peripheral singletons took place as predicted by the signal suppression hypothesis. However, the N2pc and anterior N2 results revealed that top-down proactive control was not strong enough to prevent attentional capture and the detection of conflicting features of these salient stimuli, including Chinese characters which were not part of the current task set. Consequently, the stimulus-driven hypothesis has demonstrated to have a stronger predictive capacity, as compared to the contingent capture and signal suppression hypotheses, in explaining the results of this study. Hence, salient, yet task-irrelevant, singleton objects always attract attention if part of a large distractor array, irrespective of current selection intentions.
Bio: Giorgio Fuggetta obtained his BSc and MSc in Psychology from University of Padua, Italy (2000), and received his Ph.D. in Neuroscience from the University of Verona, Italy (2006). He was a postdoctoral research fellow at the Institute of Cognitive Neuroscience (ICN), University College London (UCL), UK (2006-2007), and then a Lecturer in Psychology at the University of Leicester, UK (2007-2016), before being appointed to the University of Roehampton, London, UK in 2016 as a Senior Lecturer in Psychology. His current research explores the neural mechanisms underlying visual working memory and attentional control processes using electroencephalography (EEG) and Transcranial Magnetic stimulation (TMS) techniques. He is also conducting translational research in cognitive and educational psychology, studying the association between individual differences in executive function (EF) skills and academic achievement in adolescence. I teach techniques in cognitive neuroscience to MSc Applied Cognitive Neuroscience students and research methods, statistics and neuropsychology to BSc Psychology students.