Visual Thinking Lab

We study visual thinking:
how it works, and how
educaton + design can
make it work better.

Assigning visual structure within an object

Another new line of work, led by Ph.D. student Lucie Xu, examines a potential role for the location of visual attention as a marker for structure within objects. For example, objects in the world do not have a surface that can be objectively labeled the 'front'. We impose this designation on one surface of an object according to several cues, including which surface is associated with the most task-relevant information (e.g. who sits where determines the 'head' of the table) or the direction of motion of an object (e.g. a skateboard's front is determined by its motion). However, when these cues are competing, weak, or absent, we can also flexibly assign one surface as the 'front'. The figure to the right shows an example of this flexibility. We are testing the novel hypothesis that one important tool in our suite of object recognition abilities is the location of the 'spotlight' of attention, where the attended region becomes the front.

 

In the first study, Lucie uses an electrophysiological 'attention tracker' to measure its distribution over the left vs. right side of a display, and shows that within an object with ambiguous structure, the 'front' of the object is the attended surface (figure on the right) (Xu & Franconeri, 2012). This work complements recent work from another lab using a different neuroimaging techniques (fMRI), while offering a clearer picture of how attention and structure are closely temporally linked by using a high-temporal resolution technique. In a second project, Lucie uses behavioral techniques to show manipulations of attention lead to changes in structure, suggesting a causal role for the location of attention (Xu & Franconeri, in preparation).

 

Another line of this work explores a potential role of attention in maintaining object identity over time. Ph.D. student Lucie Xu has constructed displays where two objects rapidly change position within a display, leaving ambiguity about which went where, a phenomenon know as 'ambiguous apparent motion'. An observer can either see two objects staying in their places, or switching places. Using her electrophysiological 'attention tracker' (e.g. Figure 7), she has found exciting data that suggest that movements of the attentional 'spotlight' may underlie these two percepts. When seeing two objects that stay in place, attention does not move. But when seeing the objects swap positions, attention tracks the more salient object in perfect time with the percept, suggesting that its position helps maintain object continuity over time (Xu & Franconeri, in preparation). In a related project, some of my collaborative research has shown that in similar dynamic displays, any feature (including colors and shapes) that cause attention to follow an object, can lead to that object to be counted at the 'same' over time by the visual system (Hollingworth & Franconeri, 2009).


(a) For some objects, the location of the "front" is ambiguous. For the box, one can impose at least three organizations: surface 1 being the front (e.g., a remote control), surface 2 being the front (e.g. a clock radio), or surface 3 being the front (e.g. a security camera). For the cheese grater, the 'front' is determined by the given task. For the table, we can designate an arbitrary position as 'head' of the table. In these cases, the 'front' of an object may be guided by the position of the attentional 'spotlight'.

(b) We asked observers to view a cube with ambiguous structure (either the left or right side could be perceived as the front). Viewing the cube produces a percept that alternates between 1 & 2.

(c) During an extended viewing of the cube, electrophysiological measurements reveal that attention was directed toward the perceived front (plotted as the right side here) before and after (significant from -800ms to +600ms) a buttonpress report (time 0 on vertical axis) indicating that that side had 'popped' to the front.