Pointing with the index finger is a universal behavior. However, the functional significance of indexical pointing has not been examined empirically. We examined the efficacy of various pointing gestures in evoking viewer's attentional shifts. After viewing the gesture cue, observers quickly reported the location of a visual target. With a short cue-target delay, reaction times were generally shorter for the target at the location where gesture cues pointed, but not with a long cue-target delay. Moreover, the indexical pointing gesture produced a significantly larger cueing effect than the other gestures. Our control experiments indicated that the index-finger advantage is tightly linked to the proper morphological shape (i.e. length and position of the index finger) of the indexical pointing and is not explained by the directional discriminability of the gesture. The visual system seems to use mechanisms that are partially independent of the directional discrimination of gestures, in order to quickly modulate the viewer's attention.Key words: visual attention, attentional shift, indexical pointing.Referential pointing with the index finger 2 is found cross-culturally (Eibl-Eibesfeldt, 1989). It is so ubiquitous that many people include the extension of the index finger in the definition of pointing. However, there is no obvious reason for this exclusive use of the index finger in pointing. In this paper, we refer to a directional hand gesture as "pointing" and a pointing gesture with the index finger extended as "indexical pointing."Despite its omnipresence, the functional significance of indexical pointing has not been examined. Accumulating evidence suggests that socially relevant stimuli are automatically processed and used to modulate the viewer's attentional state. Eye-gaze direction, head orientation, and body orientation have been shown to control the direction of attention of the viewer (Driver, Davis, Ricciardelli, Kidd, Maxwell, & Baron-Cohen, 1999;Friesen & Kingstone, 1998;Hood, Willen, & Driver, 1998;Kingstone, Friesen, & Gazzaniga, 2000;Langton & Bruce, 1999, 2000Langton, Watt, & Bruce, 2000). showed that hand and head cues interfere with each other in a Stroop paradigm and suggested that visual directional cues with social significance are processed in parallel in deciding the direction of another individual's attention.