The system of the present invention provides actual interactive choices to members of an audience in the form of three dimensional (3D) objects in space, and after each member of the audience reaches out at, points at, or looks at the 3D object of choice, the system responds and alters the audio and/or video feedback to each member. Specifically, a 3D video production is presented on a screen and at several points during the presentation a character or person, appearing on the screen, requests each user to respond to a question by physically selecting one of several possible 3D objects, the 3D objects appearing through the 3D effect to be floating directly in front of each user. The 3D answer options could be three 3D characters, for example. Selection of one of these objects is facilitated through a unique user interface unit, embodied in either a helmet, gun, or wand configuration. Each of these alternative user interface units allow the interactive selection to be made by physical gestures. The user interface unit is comprised of a detection device, a processor and associated circuitry, and attached headphones. Immediately upon making a selection, the user interface unit determines the choice made by the user, matches the choice with a distinct prerecorded audio feedback response, and presents to the user a distinct and individualized audio response. These personalized feedback responses are seamlessly communicated to each user through a set of headphones, and thus, amount to minimally perceived alterations from the common audio presentation. Interactive realism is enhanced through the preferred lip synching of the audio feedback response to the video character's mouth movements.