Auxiliary Scene-Context Information Provided by Anchor Objects Guides Attention and Locomotion in Natural Search Behavior

Abstract

Successful adaptive behavior requires efficient attentional and locomotive systems. Previous research has thoroughly investigated how we achieve this efficiency during natural behavior by exploiting prior knowledge related to targets of our actions (e.g., attending to metallic targets when looking for a pot) and to the environmental context (e.g., looking for the pot in the kitchen). Less is known about whether and how individual nontarget components of the environment support natural behavior. In our immersive virtual reality task, 24 adult participants searched for objects in naturalistic scenes in which we manipulated the presence and arrangement of large, static objects that anchor predictions about targets (e.g., the sink provides a prediction for the location of the soap). Our results show that gaze and body movements in this naturalistic setting are strongly guided by these anchors. These findings demonstrate that objects auxiliary to the target are incorporated into the representations guiding attention and locomotion.

Learned associations serve as target proxies during difficult but not easy visual search

“…However, when target objects are hard to find, it may be advantageous to use other information that is predictive of the target's location to help guide attention. The spatial structure of scenes is one source of information; when looking for a coffee machine in a kitchen, one tends to look on countertops (Biederman, 1981;Boettcher et al, 2018;Castelhano & Heaven, 2011;Castelhano & Henderson, 2007;Helbing et al, 2022;Mack & Eckstein, 2011;Malcolm & Henderson, 2009;Malcolm, Rattinger, et al, 2016). Another source of predictive information comes from other objects that reliably appear near or with the target of interest.…”

“…"Anchor objects" are a special case of this in which one object in a co-occurring pair (e.g., a bathroom sink) is much larger than its associate (e.g., toothbrush). Thus, when searching for a toothbrush, the sink may serve as a more visible and highly predictive spatial cue that guides attention towards the target (Helbing et al, 2022;Mack & Eckstein, 2011;Vo et al, 2019;Vo & Wolfe, 2013). Boettcher et al (2018) argued that the anchor object might even be an auxiliary part of the target template because the two are automatically associated in memory.…”

Hierarchical organization of objects in scenes is reflected in mental representations of objects

“… 2 ; for a review see 16 ) or on the relationship between anchors and related local objects (e.g. 26 , 27 ). Here, we aimed at bridging the gap between these two levels considering the role of meaningful clusters of objects (“phrase” level) as an intermediate structure within the hierarchy.…”

“…The important role of anchor objects in visual search has been further corroborated by a series of eye-tracking experiments where the absence of anchor objects (e.g., the toilet being replaced by a washing machine) resulted in less efficient search performance as seen in faster RTs and reduced gaze coverage of the scene 26 . These results were then replicated in more ecologically valid and immersive setting provided by virtual reality (VR 27 ). Participants had to search for target local objects within virtual environments that either displayed anchor objects or anchors replaced by gray cuboids in the same position.…”

Viewpoint-Dependence and Scene Context Effects Generalize to Depth Rotated 3D Objects

“…Here, in order to increase the external validity of these findings (Draschkow, 2022), we aimed to generalize the insights from 2D photographic images to 3D models of objects (Biederman & Gerhardstein, 1993;Gauthier et al, 2002;Logothetis et al, 1994;Poggio & Edelman, 1990;Zisserman et al, 1995). Recent work using 3D immersive environments has highlighted the importance of studying vision under more naturalistic constraints in order to investigate cognitive processes in the context of natural behavior (Draschkow et al, 2021;Helbing et al, 2020Helbing et al, , 2022Kristjánsson & Draschkow, 2021). An additional benefit of using 3D models is that we could probe a broad range of viewpoints and empirically establish accidental and canonical viewpoints, allowing for a broader representation of the viewpoints we encounter in our natural environment.…”