Subliminal spatial cues capture attention and strengthen between-object link

Abstract

According to the spreading hypothesis of object-based attention, a subliminal cue that can successfully capture attention to a location within an object should also cause attention to spread throughout the whole cued object and lead to the same-object advantage. Instead, we propose that a subliminal cue favors shifts of attention between objects and strengthens the between-object link, which is coded primarily within the dorsal pathway that governs the visual guidance of action. By adopting the two-rectangle method and using an effective subliminal cue to compare with the classic suprathreshold cue, we found a different result pattern with suprathreshold cues than with subliminal cues. The suprathreshold cue replicated the conventional location and object effects, whereas a subliminal cue led to a different-object advantage with a facilitatory location effect and a same-object advantage with an inhibitory location effect. These results support our consciousness-dependent shifting hypothesis but not the spreading hypothesis.

Introduction

To recognize an object in a multi-object scene, our brain needs to calculate the relation of properties—shape, color, configuration, and so on—within objects. For example, a pail with a curvature on the side can be a mug, but if the curvature is on the top of the pail, it is more likely to be a bucket (Biederman, 1987). Visual attention can facilitate processing of properties belonging to the same object, that is, object-based attention (Duncan, 1984, Egly et al., 1994), and this kind of object-based attention may be achieved by strengthening the within-object link that is critical for object recognition. However, to act in a multi-object environment, our brain needs to calculate the relation of properties—orientation, size, and distance—between objects. For example, to hit a baseball, it is critical to know the moment-by-moment distance between the ball and the bat. In this case, it is likely that attention helps action execution by strengthening the between-object link (Davis, 2001, Humphreys, 1998) that is important for visually guided action.

Indeed, two visual pathways have been identified for the two main functions of vision: object recognition and action (Goodale & Milner, 1992). The ventral pathway—from visual primary cortex (V1) to temporal cortex—is mainly involved in object recognition, whereas the dorsal pathway—from V1 to frontal-parietal cortex—is mainly involved in the visual guidance of action (Goodale et al., 1991, Kluver and Bucy, 1938). The double-dissociation demonstrated by neuropsychological patients provides evidence for the two-pathway theory. On one hand, patients with lesion areas in the ventral pathway lose conscious vision for object recognition but not the unconscious vision to act (Goodale & Milner, 2004). For example, Patient DF cannot report the orientation of a pencil, but she can posture her hand correctly as she reaches out to grasp it (Goodale et al., 1991). On the other hand, patients with lesions in the dorsal pathway have intact object recognition but impaired visually guided action. These optic ataxia patients are able to report the orientation of a slot cut in a disk, but they cannot reach out and pass their hand through it (Perenin & Vighetto, 1988). Contrary to Patient DF, the optic ataxia patients have conscious vision for object recognition but they cannot use this vision to guide their action. This double-dissociation of conscious and unconscious vision revealed by neuropsychological patients with damage in ventral and dorsal pathways, respectively, hints at the possibility that manipulating normal participants’ consciousness of the stimuli can dissociate the two pathways and show their difference in affecting performance. This is the goal of the current study.

In a seminal paper, Egly and colleagues (1994) used a cueing paradigm with a double-rectangle display to demonstrate the existence of within-object link. They presented two outlined rectangles, with one end of one rectangle brightened as a cue to indicate the possible location of a target. The target was a small solid square, shown subsequently within one end of a rectangle. Within-object link was indicated by the same-object advantage: RTs were shorter when the target appeared at the uncued end of the cued than at the uncued rectangle, with an equal cue-to-target distance between the two. Egly, Driver, and Rafal (1994) suggest that there is a cost of shifting attention between objects (see also Lamy & Egeth, 2002). As with Egly et al. (1994), a series of studies showing this same-object advantage have used suprathreshold stimuli that supposedly trigger conscious vision in the ventral pathway (e.g., Abrams and Law, 2000, Lamy and Tsal, 2000, Moore et al., 1998). Here we suggest that vision for action, which is triggered by unconscious vision, may favor shifts of attention between objects relative to shifts of attention within objects. Therefore, we hypothesize that if subliminal stimuli are used, different-object advantage—that is, faster response to a target within an uncued object than within a cued object—should be obtained instead because unconscious vision involved in the dorsal pathway is primarily for action, and action requires a between-object link (cf. Davis, 2001).

In contrast to this consciousness-dependent shifting hypothesis, the influential spreading hypothesis of object-based attention would make an opposite prediction (Richard, Lee, & Vecera, 2008). The spreading hypothesis states that when attention is cued to a location within an object, attention will spread automatically from the cued location to the whole object. Consequently, a subliminal cue that can successfully capture attention to a specific location within an object should also cause attention to spread throughout the whole cued object. In sum, regardless of the participant’s awareness of the cue, a conventional same-object advantage is expected.

To test these two hypotheses, we designed four experiments orthogonally by crossing the cue type (subliminal, suprathreshold) with the cue-to-target stimuli-onset-asynchrony (SOA; 100 ms, 1000 ms). A subliminal cue was followed by suprathreshold cue in each pair of experiments. The experiments were structured as follows:

• Experiment 1: subliminal cue, 100-ms SOA.

• Experiment 2: suprathreshold cue, 100-ms SOA.

• Experiment 3: subliminal cue, 1000-ms SOA.

• Experiment 4: suprathreshold cue, 1000-ms SOA.

Manipulation of SOA allows us to examine the object effects induced by the spatial cue across different time courses. Past studies using suprathreshold cues have shown that one’s attention is attracted first to the cued location but then is inhibited from going there again, as indicated by an early facilitation (faster RT) followed by late inhibition (slower RT) at the cued location (for a review, see Klein, 2000). Bennett and Pratt (2001) examined the spatial distribution of the late-inhibition component and found facilitation in the quadrant opposite to the cued (inhibited) location. Assuming that attention relocates to the opposite quadrant in the long-SOA condition, the uncued object in the current study becomes “attended-object,” and a reversed object effect should be obtained. Indeed, with suprathrehold cues, results opposite to that obtained in the short SOAs were found instead (Jordan & Tipper, 1999).

A recent study by Mulckhuyse, Talsma, and Theeuwes (2007) has shown the same bi-phasic mode of early facilitation and late inhibition with a subliminal spatial cue. Thus, we hypothesize that opposite object effects should be obtained for a long-SOA condition compared with a short-SOA condition for the subliminal cue as well. That is, with subliminal cues, we should expect to find a different-object advantage for a short SOA and a same-object advantage for a long SOA. In contrast, the spreading hypothesis (Richard et al., 2008) predicts conventional object effects—same-object advantage for a short SOA (Egly et al., 1994) and different-object advantage for a long SOA (Jordan & Tipper, 1999)—as long as a subliminal cue attracts attention to its location.