Getting the closer object? An information-based dissociation between vision for perception and vision for movement in early infancy

Exploring perceptual grouping by proximity principle in multistable dot lattices: Dissociation between vision-for-perception and vision-for-action

Perceptual grouping, a fundamental mechanism in our visual system, significantly influences our interpretation of and interaction with the surrounding world. This study explores the impact of the proximity principle from the perspective of the Two Visual Systems (TVS) model. The TVS model argues that the visual system comprises two distinct streams: the ventral stream, which forms the neural basis for "vision-for-perception," and the dorsal stream, which underlies "vision-for-action." We designed a perceptual grouping task using dot lattices as well as a line-orientation discrimination task. Data were collected using vocal and mouse methods for the vision-for-perception mode, and joystick and pen-paper methods for the vision-for-action mode. Each method, except for vocal, included separate blocks for right and left hands. The proximity data were fitted using exponential and power models. Linear mixed-effects models were used for the statistical analyses. The results revealed similar line-orientation discrimination accuracy across all conditions. The exponential model emerged as the best fit, demonstrating adherence to the Pure Distance Law in both perceptual modes. Sensitivity to the proximity principle was higher in the vision-for-action mode compared to the vision-for-perception. In terms of orientation biases, a strong preference for vertical orientation was observed in the vision-for-perception mode, whereas a noticeable preference toward either of the oblique orientations was detected in the vision-for-action mode. Analysis of free-drawn lines demonstrated an affordance bias in the vision-for-action mode. This suggests a remarkable tendency to perceive organizations within specific orientations that offer more affordances due to the interaction between the body postures and tools.

Catching moving objects: Differential effects of background motion on action mode selection and movement control in 6- to 10-month-old infants

In human adults the use of visual information for selecting appropriate modes for action appears to be separate from the use of visual information for the control of movements of which the action is composed (Milner & Goodale, [1995] The visual brain in action; [2008] Neuropsychologia 46:774-785). More specifically, action mode selection primarily relies upon allocentric information, whereas movement control mainly exploits egocentric information. In the present study, we investigated to what degree this division is already present in 6- to 10-month-old infants when reaching for moving objects; that is, whether allocentric information is uniquely exploited for action mode selection (i.e., reaching with one or the other hand) or whether it is also used for movement control (i.e., reaching kinematics). Infants were presented with laterally approaching objects at two speeds (i.e., 20 and 40 cm/s) against a stationary or moving background. Background motion affects allocentric information about the object's velocity relative to its background. Results indicated that object speed constrained both infants' action mode selection and movement control. Importantly, however, the influence of background motion was limited to action mode selection and did not extend to movement control. The findings provide further support for the contention that during early development information usage is--at least to some degree--separated for action mode selection and movement control.